**The Blend Ethanol/Gasoline and Emission of Gases**

Antonio Carlos Santos *Empresa de Pesquisa Energética Brazil* 

## **1. Introduction**

32 Greenhouse Gases – Emission, Measurement and Management

Zaman, M. Nguyen, M.L. & Saggar, S. (2008c). N2O and N2 emissions from pasture and

Zart, D. & Bock, E. (1998). High rate of aerobic nitrification and denitrification by

laboratory condition. *Australian Journal of Soil Research,* 46, pp. 526-534. Zaman, M. Saggar, S. Blennerhassett, J.D. & Singh, J. (2009). Effect of urease and nitrification

41, pp. 1270-1280.

*Fertility of Soils*, 34, pp. 79-84.

wetland soils with and without amendments of nitrate, lime and zeolite under

inhibitors on N transformation, gaseous emissions of ammonia and nitrous oxide, pasture yield and N uptake in grazed pasture system. *Soil Biology and Biochemistry,*

Nitrosomonas eutropha grown in a fermentor with complete biomass retention in the presence of gaseous N2O or NO. *Archives of Microbiology,* 169, pp. 282-286. Zerulla, W. Barth, T. Dressel, J. Erhardt, K. Locquenghien, K.H.V. Pasda, G. Radle, M. &

Wissemeier, A.H. (2001). 3,4-Dimethylpyrazole phosphate DMPP-a new nitrification inhibitor for agriculture and horticulture, an introduction. *Biology and* 

> By simulation, the sparkling ignition internal combustion engine, mechanic Otto cycle, four strokes, can be approximated to a thermodynamic cycle made of four different phases such as intake, compression (combustion), expansion and exhaustion.

> According to Heywood (1988), the combustion reactions from those engines are characterized by a very rapid detonation, with the following characteristics: high pressure, due to the compression phase; a reaction zone extremely thin; the lack of chemical balance because of a very short residence time; a drop in temperature next to the gas valves outlets, due to expansion in the exhaust and then to the environment.

> In a stoichiometry, the relative amount of the resulting products to the atmosphere depends on the chemical composition of the fuel, engine design, conditions of operation, the intermediate species that are formed during the process and the presence of a catalyst in the exhaust.

> The blended ethanol gasoline has major effects on reducing emissions of greenhouse gases, while the engine design, as well as the operation conditions, and the presence of a catalyst in the exhaust, decrease pollutants and greenhouse gases.

> The preliminary combustion processes are explained by reactions mechanisms and chemical kinetics, and are important to show how these gases are formed and to determine the emission levels.

> The engine has a decisive role in the efficient burning; in this case, the physical and chemical properties of the fuel must be specified in accordance with the requirements of your project.

> On the other hand, in Otto cycle engines, adding ethanol to gasoline, it reduces emissions of greenhouse gases like carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O), and pollutants like carbon monoxide (CO), volatile organic compounds (VOC). Though, there is an increase in nitrogen oxides (NOx), and aldehydes (RCHO).

> The purpose of this study is to compare the emissions of pollutant and greenhouse gases, when using the blend gasoline/ethanol or hydrated ethanol; understanding the differences between the combustion of two types of vehicles, and reviewing mechanisms that allow the increments and reduction of pollutant and the greenhouse gases.

The Blend Ethanol/Gasoline and Emission of Gases 35

The flue gas analysis carried out shows that many organic compounds found in the exhaust are not originally in the fuel, indicating that significant changes occur during the process.

The observations made in these studies concluded that the reactions of decomposition (pyrolysis) are predominant at high temperatures, above 900K, and the reactions at low

In a stoichiometric reaction, the relative amount of the resulting products into the atmosphere depends on the chemical composition of fuel, engine design, conditions of operation, the intermediate species that formed during the process and the presence of a

The surveys reveal that at high temperatures occur decomposition on fuel molecules,

At this stage, there is a predominance of decomposition reactions and extraction of hydrogen which, according to Curran et al, responds for 50% of fuel consumption, as the

�� � �� � ��� � ��<sup>∙</sup>

�� � � � ��� � ��<sup>∙</sup>

At temperatures below 900 C occur the following reactions keys: extraction of the H atom of alkanes, cyclic ether, aldehyde or ketone, addition of radicals to oxygen, split hemolytic CC and OO and decomposition of several radicals, radical reaction of the peroxide alkyl with HO2 and H2O2 isomerization of the radical peroxide and hydroperoxide alkyl, scission of hydroperoxy alkyl radical, alkyl hydroperoxide radical oxidation and formation of cyclic ether hydroperoxido alkyl radical and hydroperoxide radical. Many reactions have the

<sup>∙</sup> � ���� � ���� � ���

�� ��<sup>∙</sup> � ����

C�H��H � � � CH�CH��H � �H (6)

C�H��H � � � CH�CH�H � �H (7)

C�H��H � � � CH�CH�� � �H (8)

����<sup>∙</sup> � ������ � ���

Marinov (1998), citing Borison and Norton and Dryer et al, modeling studies shows that the

(1)

(2)

∙ (3)

∙ (4)

∙ (5)

extraction of hydrogen atoms, oxidation, isomerization and addition to double bond.

Fuel H2 CH4 C3H8 C6H14 C16H34 Species number 7 30 100 450 1200 Reactions Number 25 200 400 1500 7000

Table 1. Chemical species formed in combustion.

catalyst in the exhaust.

example below:

product OH and HO2.

temperatures, below 900 K, dominate the oxygen addition reactions.

���

pyrolysis of ethanol forms the three major isomers radicals.

����<sup>∙</sup>

## **2. Combustion in Otto cycle engines**

The amount of the fuel components oxidation depends on temperature, oxygen concentration and how they mix with the gases. For these conditions occur, it is necessary that: the air/fuel ratio to obey stoichiometry; fuel spray be made and, the devices that perform this operation, the carburetor (older system) and electronic fuel injection, are in perfect conditions of operation.

The air / fuel ratio should be closer to stoichiometric. Inadequate concentration of oxygen promotes the partial oxidation of hydrocarbons during the preliminary process of combustion, the expansion and the exhaustion, with the CO, VOC and particulates.

The unfinished spray of fuel provides the accumulation of fuel in the cylinder walls. This occurs either from non calibrated equipments or at the start moment, when the engine is cold. In both cases, the fuel not burned, or partially burned, can form a thin layer on the wall. There is flame extinction and, as a result, we get an incomplete combustion and CO, VOC and particulates emissions. This problem is avoided by the adoption of multipoint injection system, because it is more efficient than the previous one.

Incomplete combustion can occur during startup when the engine is cold, or the spark timing in relation to air / fuel mixture cannot be adequately controlled. In both cases, there may have misfire or partial burning.

The formation of gaps between the cylinder and piston engines, as they get older, allows its filling, due to the flame at the entrance slit, and the mixture escapes from the preliminary process of combustion. The gas coming out of these cracks during the expansion process is a source of CO, VOC and particulates.

The engine oil can form a thin film on the cylinder wall, piston and cylinder head. The layers of oil can absorb hydrocarbons during the compression process and, during the expansion, it could release these substances to the gases formed. This mechanism allows a fraction of the fuel to exhaust of the preliminary process of combustion.

#### **2.1 Mechanisms of chemical reactions in the engines**

The detailed chemical mechanisms of pollutant formation and kinetics of these processes is important to determine the emission levels. The breakdown of molecules (pyrolysis) predominates during the preliminary process of combustion while the oxidation of hydrocarbons (which escape the process) occurs during the expansion and exhaust.

For nitrogen oxides, the formation and destruction of intermediate chemical species are not part of the combustion process; but the reactions that produce them occur in this environment.

The combustion reaction mechanisms of several hydrocarbons, including heptane and isooctane were studied by Curran et al (1997-98) and the pure ethanol by Marinov (1998), both at Lawrence Livermore National laboratory in California.

Westbrook et al (1999) studied in the chemical kinetics model, the number of chemical species involved in combustion reactions. (Table 1)

The amount of the fuel components oxidation depends on temperature, oxygen concentration and how they mix with the gases. For these conditions occur, it is necessary that: the air/fuel ratio to obey stoichiometry; fuel spray be made and, the devices that perform this operation, the carburetor (older system) and electronic fuel injection, are in

The air / fuel ratio should be closer to stoichiometric. Inadequate concentration of oxygen promotes the partial oxidation of hydrocarbons during the preliminary process of

The unfinished spray of fuel provides the accumulation of fuel in the cylinder walls. This occurs either from non calibrated equipments or at the start moment, when the engine is cold. In both cases, the fuel not burned, or partially burned, can form a thin layer on the wall. There is flame extinction and, as a result, we get an incomplete combustion and CO, VOC and particulates emissions. This problem is avoided by the adoption of multipoint

Incomplete combustion can occur during startup when the engine is cold, or the spark timing in relation to air / fuel mixture cannot be adequately controlled. In both cases, there

The formation of gaps between the cylinder and piston engines, as they get older, allows its filling, due to the flame at the entrance slit, and the mixture escapes from the preliminary process of combustion. The gas coming out of these cracks during the expansion process is a

The engine oil can form a thin film on the cylinder wall, piston and cylinder head. The layers of oil can absorb hydrocarbons during the compression process and, during the expansion, it could release these substances to the gases formed. This mechanism allows a

The detailed chemical mechanisms of pollutant formation and kinetics of these processes is important to determine the emission levels. The breakdown of molecules (pyrolysis) predominates during the preliminary process of combustion while the oxidation of

For nitrogen oxides, the formation and destruction of intermediate chemical species are not part of the combustion process; but the reactions that produce them occur in this

The combustion reaction mechanisms of several hydrocarbons, including heptane and isooctane were studied by Curran et al (1997-98) and the pure ethanol by Marinov (1998), both

Westbrook et al (1999) studied in the chemical kinetics model, the number of chemical

hydrocarbons (which escape the process) occurs during the expansion and exhaust.

combustion, the expansion and the exhaustion, with the CO, VOC and particulates.

injection system, because it is more efficient than the previous one.

fraction of the fuel to exhaust of the preliminary process of combustion.

**2.1 Mechanisms of chemical reactions in the engines** 

at Lawrence Livermore National laboratory in California.

species involved in combustion reactions. (Table 1)

**2. Combustion in Otto cycle engines** 

perfect conditions of operation.

may have misfire or partial burning.

source of CO, VOC and particulates.

environment.


Table 1. Chemical species formed in combustion.

The flue gas analysis carried out shows that many organic compounds found in the exhaust are not originally in the fuel, indicating that significant changes occur during the process.

The observations made in these studies concluded that the reactions of decomposition (pyrolysis) are predominant at high temperatures, above 900K, and the reactions at low temperatures, below 900 K, dominate the oxygen addition reactions.

In a stoichiometric reaction, the relative amount of the resulting products into the atmosphere depends on the chemical composition of fuel, engine design, conditions of operation, the intermediate species that formed during the process and the presence of a catalyst in the exhaust.

The surveys reveal that at high temperatures occur decomposition on fuel molecules, extraction of hydrogen atoms, oxidation, isomerization and addition to double bond.

At this stage, there is a predominance of decomposition reactions and extraction of hydrogen which, according to Curran et al, responds for 50% of fuel consumption, as the example below:

$$RH + O\_2 \rightarrow R\_1H + OH^\cdot \tag{1}$$

$$RH + O \to R\_1H + OH^\cdot \tag{2}$$

At temperatures below 900 C occur the following reactions keys: extraction of the H atom of alkanes, cyclic ether, aldehyde or ketone, addition of radicals to oxygen, split hemolytic CC and OO and decomposition of several radicals, radical reaction of the peroxide alkyl with HO2 and H2O2 isomerization of the radical peroxide and hydroperoxide alkyl, scission of hydroperoxy alkyl radical, alkyl hydroperoxide radical oxidation and formation of cyclic ether hydroperoxido alkyl radical and hydroperoxide radical. Many reactions have the product OH and HO2.

$$\rm{R}O\_2^{\cdot} + \rm{H}\_2O\_2 \rightarrow \rm{ROOH} + \rm{HO\_2^{\cdot}} \tag{3}$$

$$ROOH^{\cdot} \rightarrow alceno + HO\_{2}^{\cdot}\tag{4}$$

$$RCOOH^{\cdot} \rightarrow RO^{\cdot} + HO\_{2}^{\cdot} \tag{5}$$

Marinov (1998), citing Borison and Norton and Dryer et al, modeling studies shows that the pyrolysis of ethanol forms the three major isomers radicals.

$$\rm{C}\_{2}\rm{H}\_{5}\rm{OH} + \rm{X} \rightarrow \rm{CH}\_{2}\rm{CH}\_{2}\rm{OH} + \rm{XH} \tag{6}$$

$$\text{CH}\_2\text{H}\_5\text{OH} + \text{X} \rightarrow \text{CH}\_3\text{CHOH} + \text{XH} \tag{7}$$

$$\rm{C}\_{2}\rm{H}\_{5}\rm{OH} + \rm{X} \rightarrow \rm{CH}\_{3}\rm{CH}\_{2}\rm{O} + \rm{XH} \tag{8}$$

$$CO + OH^{\cdot} \rightarrow CO\_{2} + H^{\cdot} \tag{9}$$

$$\text{CHO} \rightarrow \text{CO} + \text{H} \tag{10}$$

$$\text{CH}\_3\text{CO} \rightarrow \text{CO} + \text{H} \tag{11}$$

$$\text{CH}\_3\text{CO} \rightarrow \text{CH}\_3 + \text{CO} \tag{12}$$

$$\text{C}\_2\text{H}\_5\text{CO} \rightarrow \text{C}\_2\text{H}\_5 + \text{CO} \tag{13}$$

$$\rm{CHO} + \rm{O}\_2 \rightarrow \rm{CO} + \rm{HO}\_2^\cdot \tag{14}$$

$$\rm C\_3H\_7CO \to CO + nC\_3H\_7 \tag{15}$$

$$CO + HO\_2^\cdot \rightarrow CO\_2 + OH^\cdot \tag{16}$$

$$\text{ClO} + \text{OH}^{\cdot} \leftrightarrow \text{ClO}\_{2} + \text{H}^{\cdot} \tag{17}$$

$$\rm OH^{\cdot} + \rm CH\_{2}CO \rightarrow \rm CO\_{2} + \rm CH\_{3} \tag{18}$$

$$CH\_2O + CH\_3 \rightarrow CH\_4 + CHO\tag{19}$$

$$\rm C\_2H\_4 + \rm CH\_3 \to \rm CH\_4 + \rm COO \tag{20}$$

$$CH\_3^- + H^\cdot \to \cdot CH\_4 \tag{21}$$

$$CH\_4 + O \rightarrow CH\_3 + OH^\cdot \tag{22}$$

$$CH\_4 + H^\cdot \to \cdot CH\_3^\cdot + H\_2^\cdot \tag{23}$$

$$CH\_4 + OH^- \rightarrow CH\_3^\cdot + H\_2O \tag{24}$$

The Blend Ethanol/Gasoline and Emission of Gases 39

According to Heywood, in the conditions prevailing in the exhaust, the following secondary

ሱሮ ܱܰ (28)

ርۛۛۛۛۛۛۛۛۛۛሮܰଶ (29)

ܥܱ ܱܰ ՜ ܰଶܱ ܥܱଶ (30)

ܰଶܱ ାை

ethanol, reduces the concentration of CO (equation 13) and prevents that:

In this case, there is an increase in emissions of NOx and N2O reduction.

**3. Main variants in the formations of greenhouse gases** 

inverse relation with the NOx formed (equation 30).

of old cars and supervise drivers.

pollutants than newer ones).

**3.1 Characteristics of manufactured vehicles** 

vehicle and automotive technology applied.

ܰଶܱ ାுǡାைுǡାைǡାெ

It is observed that an increase in the concentration of OH, due to addition of anhydrous

Due to a certain amount of NO formed at high temperature, it can be concluded that at low temperatures the concentrations of N2O in the output gas into the atmosphere are in an

Reducing emissions of greenhouse gases and pollutants gases is closely correlated to fuel consumption, so that the vehicle efficiency is reflected in the amount of emitted gases.

The efficiencies and emissions measured in laboratory depend on two factors only: characteristics of manufactured vehicles and used fuel. On the other hand, when the vehicle is in use, the actual emissions vary depending on a number of factors, including the driving

Good conditions of public roads, no traffic jams in the city, and conscious motorists improve the driving cycle and then reduce vehicle gases emissions. In addition to these, regular maintenance and old vehicles removing is necessary. These actions depend on investments to keep roads in good conditions, also depend on public policies to encourage the exchange

The lower emissions can be achieved by using low pollution potential fuels. The engine design must be compatible with the physicochemical properties of the fuel, so that the performance can be improved by reducing consumption and mechanical maintenance.

The characteristics of manufactures vehicles depend on consumer preference for one type of

In developed countries, like United States, during the economic boom time, consumers prefer bigger and more powerful engines, increasing fuel consumption and emission of greenhouse gases. On countries with less power of purchase, such as Brazil, they have a fleet of more compact vehicles and powerless engines. Thus, they don't pollute that much.

Considering, however, the average age of fleet vehicles in countries with higher power of purchase, they have lower emissions, due to their lower age (older vehicle emit more

cycle, maintenance, age and environmental conditions (temperature, humidity).

reactions occur:

allows the formation of NO which is a NO2 and N2O precursor; the interactions with compounds from combustion processes explain their processes of destruction. Thus, the two processes are linked.

There are four proposed mechanisms to explain the formation of NO: the Zeldovich, or thermal NO, Fenimore, or "prompt", NO, from the formation of N2O, and the decomposition, of organic compounds of nitrogen.

In internal combustion engines, using gasoline, the typical temperatures lie around 2000 K. In this case, the prompt Zeldovich mechanism explains the formation of approximately 95% NO. The contribution due to the prompt NO is estimated in less than 5% and the rest are very small.

Breaking the triple bond of nitrogen from the air requires an extremely hugh activation energy, around 220 Kcal/mol. An oxygen molecule (O2) is not able to direct reaction between molecular nitrogen and molecular oxygen is too slow. To explain the formation of NO, Zeldovich proposed the following mechanism:

$$NO + N\_2 \to NO + N\tag{25}$$

$$O\_2 + N \to NO + O\tag{26}$$

Later, another reaction was added:

$$OH^{\cdot} + N \to NO + H^{\cdot} \tag{27}$$

This proposed mechanism consisted of three reactions mentioned above is known as the extended Zeldovich mechanism. NO concentrations correspond to equilibrium at temperature conditions at the exit of the cylinders.

Due to the characteristics of combustion reactions, the formation of NO does not reach chemical equilibrium. As the temperature of exhaust gases falls during the course of expansion, the reactions that involve NO stuck and its concentration remains at levels that correspond to equilibrium under conditions of exhaustion.

The combustion reactions under the conditions prevailing in the cylinder at adiabatic flame temperature, in an internal combustion engine, are decisive in the formation of NO and by extension of NOx and N2O.

The NO formation rate, according to the Zeldovich mechanism, is strongly accelerated above 2000 K and negligible if the temperature is below 1700 K. It is concluded that the exhaust gases to the atmosphere there is no more formation of NO, only reactions with the other gases.

Another factor that may increase the concentration of NO is the contribution from the pyrolysis of ethanol to the concentration of OH and HO2 radicals. Preferably they react with CO, reducing its emissions, but an increase in concentration should also contribute to the increase of NO (Zeldovich equation of 3).

The residence time of combustion gases in the engine is another variable to be observed. Below 10 seconds, the formation of NO is small; above this value, it is accelerated.

allows the formation of NO which is a NO2 and N2O precursor; the interactions with compounds from combustion processes explain their processes of destruction. Thus, the two

There are four proposed mechanisms to explain the formation of NO: the Zeldovich, or thermal NO, Fenimore, or "prompt", NO, from the formation of N2O, and the

In internal combustion engines, using gasoline, the typical temperatures lie around 2000 K. In this case, the prompt Zeldovich mechanism explains the formation of approximately 95% NO. The contribution due to the prompt NO is estimated in less than 5% and the rest are

Breaking the triple bond of nitrogen from the air requires an extremely hugh activation energy, around 220 Kcal/mol. An oxygen molecule (O2) is not able to direct reaction between molecular nitrogen and molecular oxygen is too slow. To explain the formation of

��<sup>∙</sup> � �� � �� � �<sup>∙</sup>

This proposed mechanism consisted of three reactions mentioned above is known as the extended Zeldovich mechanism. NO concentrations correspond to equilibrium at

Due to the characteristics of combustion reactions, the formation of NO does not reach chemical equilibrium. As the temperature of exhaust gases falls during the course of expansion, the reactions that involve NO stuck and its concentration remains at levels that

The combustion reactions under the conditions prevailing in the cylinder at adiabatic flame temperature, in an internal combustion engine, are decisive in the formation of NO and by

The NO formation rate, according to the Zeldovich mechanism, is strongly accelerated above 2000 K and negligible if the temperature is below 1700 K. It is concluded that the exhaust gases to the atmosphere there is no more formation of NO, only reactions with the

Another factor that may increase the concentration of NO is the contribution from the pyrolysis of ethanol to the concentration of OH and HO2 radicals. Preferably they react with CO, reducing its emissions, but an increase in concentration should also contribute to the

The residence time of combustion gases in the engine is another variable to be observed.

Below 10 seconds, the formation of NO is small; above this value, it is accelerated.

���� � �� � � (25)

�� � � � �� � � (26)

(27)

processes are linked.

very small.

decomposition, of organic compounds of nitrogen.

NO, Zeldovich proposed the following mechanism:

temperature conditions at the exit of the cylinders.

correspond to equilibrium under conditions of exhaustion.

Later, another reaction was added:

extension of NOx and N2O.

increase of NO (Zeldovich equation of 3).

other gases.

According to Heywood, in the conditions prevailing in the exhaust, the following secondary reactions occur:

$$N\_2O \xrightarrow{+O} NO \tag{28}$$

$$N\_2O \xrightarrow{+H, + OH, +O, +M} N\_2 \tag{29}$$

It is observed that an increase in the concentration of OH, due to addition of anhydrous ethanol, reduces the concentration of CO (equation 13) and prevents that:

$$\cdot CO + NO \rightarrow N\_2O + \cdot CO\_2 \tag{30}$$

In this case, there is an increase in emissions of NOx and N2O reduction.

Due to a certain amount of NO formed at high temperature, it can be concluded that at low temperatures the concentrations of N2O in the output gas into the atmosphere are in an inverse relation with the NOx formed (equation 30).

#### **3. Main variants in the formations of greenhouse gases**

Reducing emissions of greenhouse gases and pollutants gases is closely correlated to fuel consumption, so that the vehicle efficiency is reflected in the amount of emitted gases.

The efficiencies and emissions measured in laboratory depend on two factors only: characteristics of manufactured vehicles and used fuel. On the other hand, when the vehicle is in use, the actual emissions vary depending on a number of factors, including the driving cycle, maintenance, age and environmental conditions (temperature, humidity).

Good conditions of public roads, no traffic jams in the city, and conscious motorists improve the driving cycle and then reduce vehicle gases emissions. In addition to these, regular maintenance and old vehicles removing is necessary. These actions depend on investments to keep roads in good conditions, also depend on public policies to encourage the exchange of old cars and supervise drivers.

The lower emissions can be achieved by using low pollution potential fuels. The engine design must be compatible with the physicochemical properties of the fuel, so that the performance can be improved by reducing consumption and mechanical maintenance.

#### **3.1 Characteristics of manufactured vehicles**

The characteristics of manufactures vehicles depend on consumer preference for one type of vehicle and automotive technology applied.

In developed countries, like United States, during the economic boom time, consumers prefer bigger and more powerful engines, increasing fuel consumption and emission of greenhouse gases. On countries with less power of purchase, such as Brazil, they have a fleet of more compact vehicles and powerless engines. Thus, they don't pollute that much.

Considering, however, the average age of fleet vehicles in countries with higher power of purchase, they have lower emissions, due to their lower age (older vehicle emit more pollutants than newer ones).

The Blend Ethanol/Gasoline and Emission of Gases 41

There are limitations in efforts to completely eliminate emissions gases from oil products responsible for global warm. The existence of fossil carbon in its chemical composition will necessarily have compounds like HC, CO2, CH4 and CO, reducing of emissions can be done

Gasoline, without the addition of ethanol, is primarily made of branched and unbranched aliphatics, cyclic (naphthenic), and aromatics, besides substances that contain atoms of sulfur, nitrogen, metals and oxygen. Table 7 shows the origin and properties of the main substances and blends used in the formulation of gasoline and the processes to obtain it, site

Currently, the refinery gasoline is made of carefully balanced mixtures of hydrocarbons, to meet the performance requirements on engines, and to avoid the emissions of pollutants.

The gasoline chemical composition control is made from their physicochemical properties, such as: ASTM distillation; octane index; volatility; calorific value; and Reid vapor pressure. When analyzing the properties of gasoline manufactured in Brazil, one has to consider that it currently contains 25% of anhydrous ethanol (according to applicable law it may vary

The ASTM distillation of petroleum products at atmospheric pressure is one of the tests that evaluate the range and boiling oil and its derivatives to ensure the correct specification of

Alkylate Alkylation 40 - 150 90 – 100 Light naphtha Crude oil distillation. 30 - 120 50 – 65 Heavy naptha Crude oil distillation. 90 - 220 40 – 50

Cracker naphtha Fluid catalytic cracking process. 40 - 220 78 – 80

polymer Olefin polymerization. 60 - 220 80 – 100

Naphtha reform Catalytic reforming 40 - 220 80 – 85

In Brazil, the application of ASTM distillation (in gasoline) gets ranges from 30 to 215 ° C. The content of 10% gasoline evaporated in the ASTM distillation is related to the minimum amount required for initial ignition to occur; the 50% of the distillate is related to the heating

Process Boiler point

Technology 40 - 220 80 – 85

Technology. 30 - 150 70 – 76

range (°C)

Zero 27

Motor octane number.

> 101 75

**3.2 Characteristics of fuels and emissions** 

by the use of biofuels and catalysts in the exhaust.

the final product and the refining processes control.

Coker naphtha Delayed Coking Process

Crude oil distillation and transformation processes

Hydrocracking Process

**3.2.1 Gasoline** 

Petrobras (2011).

between 18 and 25).

Butane. Isopentane

Hydrocracking naphtha

Table 2. Gasoline.

Naphtha

De Cicco and Ross (1993) apud Azuaga (2000) bring together the technical approaches in the manufacture of the vehicle in order to increase the efficiency of vehicles, in three parts: the engine, transmission and load (vehicle weight, size and aerodynamics). In addition, one can cite the final control systems.

The use of advanced technologies in motor improves combustion efficiency, because the incomplete reactions are related to the emissions of some greenhouse gases. Three aspects are discussed: improving the fuel air mixture, for a better dispersion of fuel in the air; the closer this ratio to the stoichiometric equations provided in the reaction, and a shorter residence time of gases in the cylinder.

The most notable improvements regarding to this were the exchange of electronic carburetors by fuel injection, electronic ignition mapped and the adopting of more than one valve per cylinder. The first two technologies, as observed by DeCicco and Ross (1993) apud Azuaga (2000), improved the dosage to be introduced into the combustion chamber through load sensors, speed, temperature and pressure, and the third dispersion of fuel in the air that feeds the engine.

Important to note that improvements have been made to the engine to reduce friction between parts, with the introduction of variable valve timing, a mechanism that allows control of the position according to the vehicle operating conditions and thus, efficiently manage the processes of induction and exhaust (De Cicco and Ross 1993 cited in IPCC 1999).

Combustion gases leaving the exhaust of a vehicle, despite the technological efforts, contain certain amount of greenhouse gases. To eliminate them the auto industry has implemented a converter in the exhaust in order to promote a catalytic chemical reaction between them.

To reduce the loss of the mechanical efficiency of the transmission, which occurs when there is no efficient synchrony between it and the engine, new technologies have been produced, including front-wheel drive, automatic transmission and variable torque control.

This control allows the engine to operate at a lower speed under a given load condition and increases the speed when more power is required, as noticed by De Cicco and Ross (1993), cited Azuaga (2000).

The adoption of front wheel drive cars eliminates the heavy steering shaft providing energy savings by reducing the vehicle weight although it is not suitable for most models of light commercial vehicles.

The load reducing, by reducing the size of the compartments (truck, rear seats etc.) and components weight (through a careful redesign of its parts), and the use of new materials such as plastic, aluminum and mild steel, permit lighter vehicles been manufactured and then, there is a decrease in fuel consumption.

 A technologically advanced aerodynamics reduces the consumption because it allows less air resistance. This is possible with the elimination of acute angles in the side panels, between the hood and windshield.

The use of this technology, however, is restricted to passenger cars. It does not apply to light commercial vehicles, which requires open and load spaces, and specific height from the ground to run off the road.

## **3.2 Characteristics of fuels and emissions**

There are limitations in efforts to completely eliminate emissions gases from oil products responsible for global warm. The existence of fossil carbon in its chemical composition will necessarily have compounds like HC, CO2, CH4 and CO, reducing of emissions can be done by the use of biofuels and catalysts in the exhaust.

## **3.2.1 Gasoline**

40 Greenhouse Gases – Emission, Measurement and Management

De Cicco and Ross (1993) apud Azuaga (2000) bring together the technical approaches in the manufacture of the vehicle in order to increase the efficiency of vehicles, in three parts: the engine, transmission and load (vehicle weight, size and aerodynamics). In addition, one can

The use of advanced technologies in motor improves combustion efficiency, because the incomplete reactions are related to the emissions of some greenhouse gases. Three aspects are discussed: improving the fuel air mixture, for a better dispersion of fuel in the air; the closer this ratio to the stoichiometric equations provided in the reaction, and a shorter

The most notable improvements regarding to this were the exchange of electronic carburetors by fuel injection, electronic ignition mapped and the adopting of more than one valve per cylinder. The first two technologies, as observed by DeCicco and Ross (1993) apud Azuaga (2000), improved the dosage to be introduced into the combustion chamber through load sensors, speed, temperature and pressure, and the third dispersion of fuel in the air

Important to note that improvements have been made to the engine to reduce friction between parts, with the introduction of variable valve timing, a mechanism that allows control of the position according to the vehicle operating conditions and thus, efficiently manage the processes of induction and exhaust (De Cicco and Ross 1993 cited in IPCC 1999). Combustion gases leaving the exhaust of a vehicle, despite the technological efforts, contain certain amount of greenhouse gases. To eliminate them the auto industry has implemented a converter in the exhaust in order to promote a catalytic chemical reaction between them. To reduce the loss of the mechanical efficiency of the transmission, which occurs when there is no efficient synchrony between it and the engine, new technologies have been produced,

This control allows the engine to operate at a lower speed under a given load condition and increases the speed when more power is required, as noticed by De Cicco and Ross (1993),

The adoption of front wheel drive cars eliminates the heavy steering shaft providing energy savings by reducing the vehicle weight although it is not suitable for most models of light

The load reducing, by reducing the size of the compartments (truck, rear seats etc.) and components weight (through a careful redesign of its parts), and the use of new materials such as plastic, aluminum and mild steel, permit lighter vehicles been manufactured and

 A technologically advanced aerodynamics reduces the consumption because it allows less air resistance. This is possible with the elimination of acute angles in the side panels,

The use of this technology, however, is restricted to passenger cars. It does not apply to light commercial vehicles, which requires open and load spaces, and specific height from the

including front-wheel drive, automatic transmission and variable torque control.

cite the final control systems.

that feeds the engine.

cited Azuaga (2000).

commercial vehicles.

then, there is a decrease in fuel consumption.

between the hood and windshield.

ground to run off the road.

residence time of gases in the cylinder.

Gasoline, without the addition of ethanol, is primarily made of branched and unbranched aliphatics, cyclic (naphthenic), and aromatics, besides substances that contain atoms of sulfur, nitrogen, metals and oxygen. Table 7 shows the origin and properties of the main substances and blends used in the formulation of gasoline and the processes to obtain it, site Petrobras (2011).

Currently, the refinery gasoline is made of carefully balanced mixtures of hydrocarbons, to meet the performance requirements on engines, and to avoid the emissions of pollutants.

The gasoline chemical composition control is made from their physicochemical properties, such as: ASTM distillation; octane index; volatility; calorific value; and Reid vapor pressure. When analyzing the properties of gasoline manufactured in Brazil, one has to consider that it currently contains 25% of anhydrous ethanol (according to applicable law it may vary between 18 and 25).

The ASTM distillation of petroleum products at atmospheric pressure is one of the tests that evaluate the range and boiling oil and its derivatives to ensure the correct specification of the final product and the refining processes control.


Table 2. Gasoline.

In Brazil, the application of ASTM distillation (in gasoline) gets ranges from 30 to 215 ° C. The content of 10% gasoline evaporated in the ASTM distillation is related to the minimum amount required for initial ignition to occur; the 50% of the distillate is related to the heating

The Blend Ethanol/Gasoline and Emission of Gases 43

The aromatics (benzene and toluene are the most used) have the advantage of being chemically more stable than other unsaturated compounds; thus they are more resistant to auto ignition than iso-octane, during compression stroke. Benzene, for example, that has an

Ethanol has high octane without problems enthalpy and the boiling point presented by the aromatics compounds. You can reduce the volume of aromatics, replacing the oxygen,

The hydrocarbons have the advantage of having a high calorific value by reducing fuel

Ethanol has a low calorific value compared to hydrocarbons. Consequently, the mixture

A fuel of higher calorific value, however, has the disadvantage of forming NO, NO2 and

The carbon removed from the atmosphere by photosynthesis, and stored in biomass will serve as row material for bioethanol production, which will be burned in engines that emit

The process is closed, which net income is zero in terms of CO2 emissions, while fossil fuel burning releases into the atmosphere carbon stored over millions of years. The addition of

The fact of having 11% oxygen by mass results in better combustion, reducing

The ethanol carbon is produced by photosynthesis; so it does not contributes to the

 The presence of oxygen increases compounds like NOx and RCHO. The combination of these agents leads to an increased formation of 'photochemical smog' and tropospheric

The ethanol has high octane index, which allows a higher compression ratio and more

The reasons why emissions of internal combustion engines varied in the period 1980-2002

The emissions, in kg/vehicle, depend on the variables: the fleet, driving cycle, maintenance, fleet age, environmental conditions and traffic. The emission factors depend more on the technology used to improve the characteristics of manufactured vehicles and fuels. The engines did not have significant technological developments in 1980-1985. During this period, the increase in the percentage of ethanol in gasoline and the production of hydrated

index equal to 120 octanes, can be used as an agent to raise this ratio in fuels.

consumption due to energy availability increase.

**3.2.2 Ethanol hydrated** 

global warming.

powerful engine.

**4. Emissions of gases** 

ozone.

CO2 engines.

ethanol/gasoline has a lower calorific value than pure gasoline.

N2O by extension, due to a higher adiabatic flame temperature.

ethanol brings the following advantages and disadvantages:

The calorific value of ethanol is about 30% lower than gasoline.

were described in "Main variants in the formations of greenhouse gases".

emissions of gases such as CO, CH4 and HC.

ethanol engines were responsible for the variations.

without prejudice to the octane number of gasoline without increasing emissions.

and engine performance, while the remaining 90% is intended to minimize deposits formation, which is a source of emissions.

Volatility is the fuel agility to turn from liquid to vapor and this property is related to the 10% evaporated ASTM distillation. A fuel with high volatility promotes HC emissions, both in cold start and in the tank.

Currently there is a strong trend in reducing the volatility parameters, without compromising the ignition of the fuel. The way we can reduce this property is decreasing the amount of butane in gasoline in Brazil (see Table 1).

The Reid vapor pressure is the absolute pressure practiced by a mixture at 37.8 ° C with a rate of vapor/liquid of 4 to 1. The parameter is often used to characterize the volatility of gasoline and crude oils.

This test is mainly used to indicate the requirements that must be met for the products transportation and storage, including vehicles, preventing accidents and minimizing the losses.

It is observed that the higher the temperature, the greater is the evaporative emissions, and then the need to specify the volatility according to the region where it will be used and the season of the year.

A fuel with a low heat of vaporization and vapor pressure allows more complete combustion; the opposite can cause incomplete vaporization and a lack of control in the air/fuel relation making the burning process very difficult.

The aromatics have a high heat of vaporization and boiling point, then the difficulty of evaporation. These properties limit their use because they difficult the formation of homogeneous mixture air/fuel. As a result, there is a partial burning with the formation of CO and VOCs, especially during cold start.

Anhydrous ethanol also has a high heat of vaporization, but this property is compensated by its low boiling temperature, 10 ° C. In this case there is no difficulty of forming an air / fuel mixture as occurs with the aromatic.

The octane index measures the ability of a chemical compound to withstand high pressures without detonating. In the event of this fact, there will be a partial burning of fuel.

Engine designers take the octane into account to determine compression ratio, ignition advance curves and injection time.

The linear hydrocarbons have little resistance to compression and, for this reason, its concentration in gasoline should be controlled; the n-heptane, for example, has an octane rating of zero.

Branched hydrocarbons resist high pressures inside the cylinders, without detonation and consequent win of power; iso-octane, for example, has an octane rating of 100.

Hydrocarbons containing double bonds are desirable because they generally have high octane, but there are some restrictions on its use in engines, because they can easily bond with hydrogen to form paraffin. Besides, it is revealed that the decrease of its content decreases the formation of volatile organic compounds.

The aromatics (benzene and toluene are the most used) have the advantage of being chemically more stable than other unsaturated compounds; thus they are more resistant to auto ignition than iso-octane, during compression stroke. Benzene, for example, that has an index equal to 120 octanes, can be used as an agent to raise this ratio in fuels.

Ethanol has high octane without problems enthalpy and the boiling point presented by the aromatics compounds. You can reduce the volume of aromatics, replacing the oxygen, without prejudice to the octane number of gasoline without increasing emissions.

The hydrocarbons have the advantage of having a high calorific value by reducing fuel consumption due to energy availability increase.

Ethanol has a low calorific value compared to hydrocarbons. Consequently, the mixture ethanol/gasoline has a lower calorific value than pure gasoline.

A fuel of higher calorific value, however, has the disadvantage of forming NO, NO2 and N2O by extension, due to a higher adiabatic flame temperature.

## **3.2.2 Ethanol hydrated**

42 Greenhouse Gases – Emission, Measurement and Management

and engine performance, while the remaining 90% is intended to minimize deposits

Volatility is the fuel agility to turn from liquid to vapor and this property is related to the 10% evaporated ASTM distillation. A fuel with high volatility promotes HC emissions, both

Currently there is a strong trend in reducing the volatility parameters, without compromising the ignition of the fuel. The way we can reduce this property is decreasing

The Reid vapor pressure is the absolute pressure practiced by a mixture at 37.8 ° C with a rate of vapor/liquid of 4 to 1. The parameter is often used to characterize the volatility of

This test is mainly used to indicate the requirements that must be met for the products transportation and storage, including vehicles, preventing accidents and minimizing the

It is observed that the higher the temperature, the greater is the evaporative emissions, and then the need to specify the volatility according to the region where it will be used and the

A fuel with a low heat of vaporization and vapor pressure allows more complete combustion; the opposite can cause incomplete vaporization and a lack of control in the

The aromatics have a high heat of vaporization and boiling point, then the difficulty of evaporation. These properties limit their use because they difficult the formation of homogeneous mixture air/fuel. As a result, there is a partial burning with the formation of

Anhydrous ethanol also has a high heat of vaporization, but this property is compensated by its low boiling temperature, 10 ° C. In this case there is no difficulty of forming an air /

The octane index measures the ability of a chemical compound to withstand high pressures

Engine designers take the octane into account to determine compression ratio, ignition

The linear hydrocarbons have little resistance to compression and, for this reason, its concentration in gasoline should be controlled; the n-heptane, for example, has an octane

Branched hydrocarbons resist high pressures inside the cylinders, without detonation and

Hydrocarbons containing double bonds are desirable because they generally have high octane, but there are some restrictions on its use in engines, because they can easily bond with hydrogen to form paraffin. Besides, it is revealed that the decrease of its content

without detonating. In the event of this fact, there will be a partial burning of fuel.

consequent win of power; iso-octane, for example, has an octane rating of 100.

decreases the formation of volatile organic compounds.

formation, which is a source of emissions.

the amount of butane in gasoline in Brazil (see Table 1).

air/fuel relation making the burning process very difficult.

CO and VOCs, especially during cold start.

fuel mixture as occurs with the aromatic.

advance curves and injection time.

rating of zero.

in cold start and in the tank.

gasoline and crude oils.

season of the year.

losses.

The carbon removed from the atmosphere by photosynthesis, and stored in biomass will serve as row material for bioethanol production, which will be burned in engines that emit CO2 engines.

The process is closed, which net income is zero in terms of CO2 emissions, while fossil fuel burning releases into the atmosphere carbon stored over millions of years. The addition of ethanol brings the following advantages and disadvantages:


## **4. Emissions of gases**

The reasons why emissions of internal combustion engines varied in the period 1980-2002 were described in "Main variants in the formations of greenhouse gases".

The emissions, in kg/vehicle, depend on the variables: the fleet, driving cycle, maintenance, fleet age, environmental conditions and traffic. The emission factors depend more on the technology used to improve the characteristics of manufactured vehicles and fuels. The engines did not have significant technological developments in 1980-1985. During this period, the increase in the percentage of ethanol in gasoline and the production of hydrated ethanol engines were responsible for the variations.

The Blend Ethanol/Gasoline and Emission of Gases 45

gasoline, but the final figures may be overestimated. For this reason it is not possible to

To understand the role of ethanol in reducing the emission factor of CO, see equations 16 and 17, HC and CH4 according to equations 1 and 2. For NOx, see equations 25, 26, 27 and

CO, NOx emissions factors (left axis of the graphs) and HC (right axis) of the blend ethanol/gasoline are higher than hydrous ethanol in the period between 1980-1985 (Figures

The ethanol emissions factors of CO, HC, considering the reaction mechanisms, are lower than gasoline and RCHO (right axis of the graphs) higher. However, the lowest values for NOx, can only be understood assuming that the engine of ethanol has better technology

evaluate the effect of the blend.

2 and 3).

30. RCHO has the precursor products in equations 6 e 7.

than that of gasoline in the period studied (Figures 2 and 3).

Fig. 2. Emission factors of CO and HC 1980-1995.

Fig. 3. Emission factors of NOx and RCHO 1980-1995.

Note: the percentage increase data of ethanol in gasoline were obtained from the ordinances of Ministério de Agriculture, Pecuária e Abastecimento.

Fig. 1. Percentage of Ethanol in Gasoline 1978-2009.

The fall in oil prices and the ethanol supply crisis in 1986 completely changed the automotive scenery in Brazil. Gasoline vehicles, which in the past have had their production reduced, increased its share of the fleet and dominated the market at the expense of hydrated ethanol.

With a fleet reduction of hydrous ethanol in 1987, initially there was an increase in greenhouse gas emissions. They, however, began to decline because of the incentives given by the government for the manufacture of vehicles of 1.000 cc; the conversion of engines to use natural gas; and the use of ethanol vehicles for taxis. After 1992, to accomplish CONAMA Resolution Nº 010/1989, the companies have been introducing new technologies in vehicle manufacturing and fuels formulas (Main variants in the formations of greenhouse gases). These applications allow the reduction of emissions.

In this decade, there has been a considerable increase in metropolis fleet, generating large traffic congestion. In these conditions, there is an increase of emissions.

#### **4.1 Emission factors**

The determination of emission factors of Brazilian vehicles must be adapted to the characteristics of the fuels used in Brazil, hydrated ethanol and the blend gasoline/anhydrous ethanol.

The Conselho Nacional de Meio Ambiente (CONAMA) does not require the measurement of CH4 and N2O, but includes (HC) and RCHO a corresponding part of the volatile organic compound without methane. Both are not included in international data.

The emission factors of CO, HC, NOx and RCHO for the period 1980 to 2009, used in this work, are from Motor Vehicle Air Pollution Control Program (PROCONVE), while CO2 and CH4 are from the First National Atmospheric Emissions Inventory for Road Motor Vehicles of the Ministério do Meio Ambiente.

N2O emissions will not be calculated due to the lack of emission factors for blend gasoline/ethanol and hydrated ethanol. Some authors use the emission factors of pure

Note: the percentage increase data of ethanol in gasoline were obtained from the ordinances

The fall in oil prices and the ethanol supply crisis in 1986 completely changed the automotive scenery in Brazil. Gasoline vehicles, which in the past have had their production reduced, increased its share of the fleet and dominated the market at the expense of hydrated ethanol. With a fleet reduction of hydrous ethanol in 1987, initially there was an increase in greenhouse gas emissions. They, however, began to decline because of the incentives given by the government for the manufacture of vehicles of 1.000 cc; the conversion of engines to use natural gas; and the use of ethanol vehicles for taxis. After 1992, to accomplish CONAMA Resolution Nº 010/1989, the companies have been introducing new technologies in vehicle manufacturing and fuels formulas (Main variants in the formations of greenhouse

In this decade, there has been a considerable increase in metropolis fleet, generating large

The determination of emission factors of Brazilian vehicles must be adapted to the characteristics of the fuels used in Brazil, hydrated ethanol and the blend

The Conselho Nacional de Meio Ambiente (CONAMA) does not require the measurement of CH4 and N2O, but includes (HC) and RCHO a corresponding part of the volatile organic

The emission factors of CO, HC, NOx and RCHO for the period 1980 to 2009, used in this work, are from Motor Vehicle Air Pollution Control Program (PROCONVE), while CO2 and CH4 are from the First National Atmospheric Emissions Inventory for Road Motor Vehicles

N2O emissions will not be calculated due to the lack of emission factors for blend gasoline/ethanol and hydrated ethanol. Some authors use the emission factors of pure

of Ministério de Agriculture, Pecuária e Abastecimento. Fig. 1. Percentage of Ethanol in Gasoline 1978-2009.

gases). These applications allow the reduction of emissions.

**4.1 Emission factors** 

gasoline/anhydrous ethanol.

of the Ministério do Meio Ambiente.

traffic congestion. In these conditions, there is an increase of emissions.

compound without methane. Both are not included in international data.

gasoline, but the final figures may be overestimated. For this reason it is not possible to evaluate the effect of the blend.

To understand the role of ethanol in reducing the emission factor of CO, see equations 16 and 17, HC and CH4 according to equations 1 and 2. For NOx, see equations 25, 26, 27 and 30. RCHO has the precursor products in equations 6 e 7.

CO, NOx emissions factors (left axis of the graphs) and HC (right axis) of the blend ethanol/gasoline are higher than hydrous ethanol in the period between 1980-1985 (Figures 2 and 3).

The ethanol emissions factors of CO, HC, considering the reaction mechanisms, are lower than gasoline and RCHO (right axis of the graphs) higher. However, the lowest values for NOx, can only be understood assuming that the engine of ethanol has better technology than that of gasoline in the period studied (Figures 2 and 3).

Fig. 2. Emission factors of CO and HC 1980-1995.

Fig. 3. Emission factors of NOx and RCHO 1980-1995.

The Blend Ethanol/Gasoline and Emission of Gases 47

The manufacture of flex-fuel engines since 2003 has changed the pattern of emissions of the Brazilian fleet. The engine flex-fuel interrupted the downward trend of emission factors.

For CO and HC there is a wide variation (figure 6). Then, you cannot establish a pattern of

You can see, however, that NOx and RCHO ethanol is higher than gasoline, as expected in

The problem with the new technology is the compression ratio, which is not adjusted for the

behavior.

Fig. 6. Emission factors of CO and HC for fuel-flex.

the mechanisms of combustion reactions (Figure 7).

Fig. 7. Emission factors of NOx and RCHO for fuel-flex.

octane best value of both fuels.

The gasoline forms a smaller amount of oxygen. For this reason, their emissions of aldehydes (right axis) are smaller than ethanol (Figure 3), as expected in the mechanisms of combustion reactions.

After 1992, to meet the Resolution CONAMA 010/1989, Automakers and Refineries have introduced new technologies in engines that use a blend of ethanol/gasoline, and formulate new kinds of fuels (see Main variants in the formations of greenhouse gases).

Encouraged by the government, the of 1.000 cc engines manufacturers have participated with 71% of sales in 2001 (dropping to 52,7% in 2009) also contributed to the emissions reducing.

For ethanol engines, the reductions are much smaller, due to their manufacture gradual reduction, as in this case, there were no new technologies incorporated.

Fig. 4. Emission factors of CO and HC 1996-2009.

Fig. 5. Emission factors of NOx and RCHO 1996-2009.

The gasoline forms a smaller amount of oxygen. For this reason, their emissions of aldehydes (right axis) are smaller than ethanol (Figure 3), as expected in the mechanisms of

After 1992, to meet the Resolution CONAMA 010/1989, Automakers and Refineries have introduced new technologies in engines that use a blend of ethanol/gasoline, and formulate

Encouraged by the government, the of 1.000 cc engines manufacturers have participated with 71% of sales in 2001 (dropping to 52,7% in 2009) also contributed to the emissions

For ethanol engines, the reductions are much smaller, due to their manufacture gradual

new kinds of fuels (see Main variants in the formations of greenhouse gases).

reduction, as in this case, there were no new technologies incorporated.

Fig. 4. Emission factors of CO and HC 1996-2009.

Fig. 5. Emission factors of NOx and RCHO 1996-2009.

combustion reactions.

reducing.

The manufacture of flex-fuel engines since 2003 has changed the pattern of emissions of the Brazilian fleet. The engine flex-fuel interrupted the downward trend of emission factors.

For CO and HC there is a wide variation (figure 6). Then, you cannot establish a pattern of behavior.

Fig. 6. Emission factors of CO and HC for fuel-flex.

You can see, however, that NOx and RCHO ethanol is higher than gasoline, as expected in the mechanisms of combustion reactions (Figure 7).

The problem with the new technology is the compression ratio, which is not adjusted for the octane best value of both fuels.

Fig. 7. Emission factors of NOx and RCHO for fuel-flex.

The Blend Ethanol/Gasoline and Emission of Gases 49

In 2003, a new technology was introduced in Brazil, the engine "flex-fuel", becoming a bestseller. For this reason, hydrated ethanol and gasoline cars gradually ceased to be

The manufacture of flex-fuel engine has changed the pattern of emissions in the country. The gasoline engines have smaller greenhouse gases emissions, CO, HC and NOx RCHO, compared with ethanol. The reason is the compression ratio, which is not adjusted for the

manufactured in the country and they had to start being imported.

best value of the fuel octane.

Fig. 10. Emission of CO and HC for fuel-flex.

Fig. 11. Emission of NOx and RCHO for fuel-flex.

#### **4.2 Gases emissions by Otto cycle vehicles**

The variables that influence the emissions of pollutant and greenhouse gases were analyzed in Chapter 3.To analyze the impact of these action variables, first were used data from the National Atmospheric Emissions Inventory for Road Motor Vehicles of the Ministério do Meio Ambiente.

#### **4.2.1 Pollutant gases emissions by Otto cycle vehicles**

During the period of 1980-1995, gasoline engines have their emissions of CO, NOx (left axis), and HC, RCHO (right axis), influenced by the reduction of the fleet and the emission factors. Unlike the factors, the emissions follow the expected by the mechanisms of combustion reactions, ie, CO and HC are higher for gasoline, while NOx and RCHO are higher for ethanol.

Fig. 8. Emission of CO and HC.

After this period, the fleet reduction of ethanol, and the technological developments of the engines, driven by the demands of Resolution CONAMA 010/1989, started making a difference. Emissions of CO and HC are also higher for ethanol.

Fig. 9. Emission of NOx and RCHO.

The variables that influence the emissions of pollutant and greenhouse gases were analyzed in Chapter 3.To analyze the impact of these action variables, first were used data from the National Atmospheric Emissions Inventory for Road Motor Vehicles of the Ministério do

During the period of 1980-1995, gasoline engines have their emissions of CO, NOx (left axis), and HC, RCHO (right axis), influenced by the reduction of the fleet and the emission factors. Unlike the factors, the emissions follow the expected by the mechanisms of combustion reactions, ie, CO and HC are higher for gasoline, while NOx and RCHO are higher for ethanol.

After this period, the fleet reduction of ethanol, and the technological developments of the engines, driven by the demands of Resolution CONAMA 010/1989, started making a

difference. Emissions of CO and HC are also higher for ethanol.

**4.2 Gases emissions by Otto cycle vehicles** 

**4.2.1 Pollutant gases emissions by Otto cycle vehicles** 

Meio Ambiente.

Fig. 8. Emission of CO and HC.

Fig. 9. Emission of NOx and RCHO.

In 2003, a new technology was introduced in Brazil, the engine "flex-fuel", becoming a bestseller. For this reason, hydrated ethanol and gasoline cars gradually ceased to be manufactured in the country and they had to start being imported.

The manufacture of flex-fuel engine has changed the pattern of emissions in the country. The gasoline engines have smaller greenhouse gases emissions, CO, HC and NOx RCHO, compared with ethanol. The reason is the compression ratio, which is not adjusted for the best value of the fuel octane.

Fig. 10. Emission of CO and HC for fuel-flex.

Fig. 11. Emission of NOx and RCHO for fuel-flex.

The Blend Ethanol/Gasoline and Emission of Gases 51

It should be noted that the CO2 from burning of hydrous ethanol, is not counted in the emissions of greenhouse gases. The carbon removed from the atmosphere through photosynthesis and stored in biomass will serve as a feedstock for biofuel production, which in turn will be burned in engines. The process is closed, whose net income is zero in terms of CO2 emissions, without considering the contributions of land use disputes whose values are

Dedicated gasoline engines, in the period 1980-1995, have its emission of CH4 affected by the reduction of the fleet and the emission factor. After this period, the fleet reduction of ethanol, the technological developments of the engines, driven by the demands of

Fig. 14. Emissions of CH4 for fuel-flex

Resolution CONAMA 010/1989, start making a difference.

not resolved.

Fig. 15. Avoided Emissions.

#### **4.2.2 Greenhouse gas emissions vehicle Otto cycle**

Ethanol consumption by the national fleet allows emissions of greenhouse gases in the transportation sector are substantially below those that could be achieved if the only fuel used was pure gasoline.

In the period 1980/87 the decline in fossil CO2 emissions was due to the use of ethanol, while the increase occurred between 1988/97 was due to the reduction of its use. The decline from 1998/07 is due in part to a fleet of predominantly motor 1000 cc, the technological evolution of engines and from 2003, with the price drop and a greater supply, the market returned to using ethanol, figure 12.

Fig. 12. Emissions of CO2.

Fig. 13. Emissions of CH4.

Ethanol consumption by the national fleet allows emissions of greenhouse gases in the transportation sector are substantially below those that could be achieved if the only fuel

In the period 1980/87 the decline in fossil CO2 emissions was due to the use of ethanol, while the increase occurred between 1988/97 was due to the reduction of its use. The decline from 1998/07 is due in part to a fleet of predominantly motor 1000 cc, the technological evolution of engines and from 2003, with the price drop and a greater supply,

**4.2.2 Greenhouse gas emissions vehicle Otto cycle** 

the market returned to using ethanol, figure 12.

used was pure gasoline.

Fig. 12. Emissions of CO2.

Fig. 13. Emissions of CH4.

Fig. 14. Emissions of CH4 for fuel-flex

It should be noted that the CO2 from burning of hydrous ethanol, is not counted in the emissions of greenhouse gases. The carbon removed from the atmosphere through photosynthesis and stored in biomass will serve as a feedstock for biofuel production, which in turn will be burned in engines. The process is closed, whose net income is zero in terms of CO2 emissions, without considering the contributions of land use disputes whose values are not resolved.

Dedicated gasoline engines, in the period 1980-1995, have its emission of CH4 affected by the reduction of the fleet and the emission factor. After this period, the fleet reduction of ethanol, the technological developments of the engines, driven by the demands of Resolution CONAMA 010/1989, start making a difference.

Fig. 15. Avoided Emissions.

The Blend Ethanol/Gasoline and Emission of Gases 53

The advantage of adding ethanol to gasoline and the manufacturing of engines that use

Looking at CO2 emissions of a car that runs on a blend ethanol and gasoline compared to pure gasoline, there is a 22% reduction on average. This decrease reaches 100% when it

Further studies are needed to understand the influence of adding ethanol to gasoline and the use of hydrous ethanol, such as "Determination of emission factor of N2O", "Emissions of N2O by mixing gasoline / ethanol and hydrous ethanol", and "Comparison emissions

Considering that in the next year cellulosic ethanol could be a reality, these studies provide

Azuaga, Denise. Danos Ambientais Causados por Veículos Leves no Brasil. Dissertação de

California Environmental Protection Agency. Air Quality Impacts of the Use of Ethanol in

Heywood, John B. Internal Combustion Engine Fundamentals. Singapore: McGraw Hill

Curran, H. J.; Chen J.-S.; Litzinger, T. A. The Lean Oxidation of Iso-Octane at Elevated

Curran,, H.J.ET AL. A Comprehensive Modeling Study of n-Heptane Oxidation. Lawrence

Curran, H. J.et al. Comprehensive Oxidation of Automotive Primary Reference Fuels at

Intergovernmental Panel on Climate Change. 2006 IPCC Guidelines for National

docs/science\_and\_technology/chemistry/combustion/nc7.pdf .

Pressures. Lawrence Livermore National Laboratory, Livermore, CA 94551.

Elevated Pressures. Lawrence Livermore National Laboratory, CA 94551. United

Greenhouse Gas Inventories, Prepared by the National Greenhouse Gas Inventories

Mestrado. Universidade Federal do Rio de Janeiro, COPPE. Rio de Janeiro 2000.

California Reformulated Gasoline. AIR RESOURCES BOARD Planning and Technical Support Division. State of Califórnia Allen, Paul (Org.). Novembro 1999.

a basis for developing policies and technologies that reduce environmental impact.

http://www.ppe.ufrj.br/ppe/ production/tesis/dazuaga.pdf . Empresa de Pesquisa Energética. Balanço Energético Nacional, 2010. /MME.

http://www.arb.ca.gov/cbg/ regact/ethanol/ethfate.htm.

http://www.areaseg.com/conama/ 1989/010-1989.pdf.

https://e-reports-ext.llnl.gov/pdf/235217.pdf.

Livermore National Laboratory. Availabel at:

https://wwwpls.llnl.gov/data/docs/science\_and\_ technology/chemistry/combustion/prf\_paper.pdf

https://www-pls.llnl.gov/data/

States of America. Availabel at:

Programme. Published: IGES, Japan.

comes to cars that use ethanol only, without considering the life cycle analysis.

hydrated ethanol are in the emission of greenhouse gases.

between pure gasoline and blend gasoline/ethanol".

**6. References** 

Availabel at:

Availabel at:

Book Co. 1988.

Availabel at:

CONAMA. Resolução nº 010/1989. Availabel at:

## **4.2.3 Avoided emissions**

The replacement of fossil carbon by the one that comes from biofuels reduces CO2 emissions. We can estimate the difference between those from a hypothetical situation where the fleet of vehicles consumes only pure gasoline and the related to a real situation, where the fleet consists of vehicles that consume a mixture of gasoline/anhydrous ethanol and vehicles that require ethanol hydrated.

The Energy Research Company has estimated the avoided emissions (tCO2/toe) in the period of 2000 to 2010. For comparison the same figure plotted in fuel consumption (gasoline more ethanol), in the period.

## **5. Conclusions**

The changing profile of the Brazilian fleet allows you to compare the factors that contributed to the amount of emissions of pollutant and greenhouse gases, establishing a connection between the mechanisms of chemical reactions, the fuel composition, the evolution in technology of the vehicle and composition of gases coming out from the exhaust.

The values of the factors and emissions of CO, HC and NOx blend gasoline/ethanol and ethanol are influenced: by the presence of OH, HO2 radicals and the introduction of new technologies in vehicles manufacturing and fuel formulation.

The addition of ethanol, increasing the OH and HO2 concentrations, can be noted between 1980 and 1990, when vehicles did not have more advanced technologies. In this case, there was a reduction of the factors and emissions of CO and HC, while the factors for NOx decreased, and the emissions increased.

To meet the CONAMA Resolution No. 010/1989, which was enforced in 1992, vehicles manufacturers have introduced new technologies, including: more efficient engines with less power; catalyzers in the exhaust, reduction in fuel consumption due to an improved aerodynamics, reduction in vehicles weight and the petroleum refiners introduced new formulations.

After 1996, this set of measures began to influence, reducing the factors and the emissions of all gases. The hydrated ethanol vehicle had lower earnings due to lower technology incorporation and to a reduction in manufacturing.

The RCHO factors and emissions are much higher for hydrated ethanol than gasoline / ethanol mix throughout the entire period, including "flex-fuel" engines, showing that the formation of oxygenated hydrocarbons is strongly influenced by the OH and HO2 radicals.

With the success of "flex-fuel" sales, hydrated ethanol vehicles are no longer manufactured and gasoline is focusing on more power and deluxe segment.

Due to a unique compression ratio for fuels with different octane ratings, "flex-fuel" engines, using blend ethanol and ethanol/gasoline, had a reduction in efficiency and thus higher emissions.

The values of CO and NOx emissions for blend gasoline/ethanol and hydrous ethanol reveal the need of an evolution in technology of the engines "flex-fuel".

The advantage of adding ethanol to gasoline and the manufacturing of engines that use hydrated ethanol are in the emission of greenhouse gases.

Looking at CO2 emissions of a car that runs on a blend ethanol and gasoline compared to pure gasoline, there is a 22% reduction on average. This decrease reaches 100% when it comes to cars that use ethanol only, without considering the life cycle analysis.

Further studies are needed to understand the influence of adding ethanol to gasoline and the use of hydrous ethanol, such as "Determination of emission factor of N2O", "Emissions of N2O by mixing gasoline / ethanol and hydrous ethanol", and "Comparison emissions between pure gasoline and blend gasoline/ethanol".

Considering that in the next year cellulosic ethanol could be a reality, these studies provide a basis for developing policies and technologies that reduce environmental impact.

## **6. References**

52 Greenhouse Gases – Emission, Measurement and Management

The replacement of fossil carbon by the one that comes from biofuels reduces CO2 emissions. We can estimate the difference between those from a hypothetical situation where the fleet of vehicles consumes only pure gasoline and the related to a real situation, where the fleet consists of vehicles that consume a mixture of gasoline/anhydrous ethanol and vehicles that

The Energy Research Company has estimated the avoided emissions (tCO2/toe) in the period of 2000 to 2010. For comparison the same figure plotted in fuel consumption

The changing profile of the Brazilian fleet allows you to compare the factors that contributed to the amount of emissions of pollutant and greenhouse gases, establishing a connection between the mechanisms of chemical reactions, the fuel composition, the evolution in

The values of the factors and emissions of CO, HC and NOx blend gasoline/ethanol and ethanol are influenced: by the presence of OH, HO2 radicals and the introduction of new

The addition of ethanol, increasing the OH and HO2 concentrations, can be noted between 1980 and 1990, when vehicles did not have more advanced technologies. In this case, there was a reduction of the factors and emissions of CO and HC, while the factors for NOx

To meet the CONAMA Resolution No. 010/1989, which was enforced in 1992, vehicles manufacturers have introduced new technologies, including: more efficient engines with less power; catalyzers in the exhaust, reduction in fuel consumption due to an improved aerodynamics, reduction in vehicles weight and the petroleum refiners introduced new

After 1996, this set of measures began to influence, reducing the factors and the emissions of all gases. The hydrated ethanol vehicle had lower earnings due to lower technology

The RCHO factors and emissions are much higher for hydrated ethanol than gasoline / ethanol mix throughout the entire period, including "flex-fuel" engines, showing that the formation of oxygenated hydrocarbons is strongly influenced by the OH and HO2 radicals. With the success of "flex-fuel" sales, hydrated ethanol vehicles are no longer manufactured

Due to a unique compression ratio for fuels with different octane ratings, "flex-fuel" engines, using blend ethanol and ethanol/gasoline, had a reduction in efficiency and thus higher

The values of CO and NOx emissions for blend gasoline/ethanol and hydrous ethanol

technology of the vehicle and composition of gases coming out from the exhaust.

technologies in vehicles manufacturing and fuel formulation.

**4.2.3 Avoided emissions** 

require ethanol hydrated.

**5. Conclusions** 

formulations.

emissions.

(gasoline more ethanol), in the period.

decreased, and the emissions increased.

incorporation and to a reduction in manufacturing.

and gasoline is focusing on more power and deluxe segment.

reveal the need of an evolution in technology of the engines "flex-fuel".

Azuaga, Denise. Danos Ambientais Causados por Veículos Leves no Brasil. Dissertação de Mestrado. Universidade Federal do Rio de Janeiro, COPPE. Rio de Janeiro 2000. Availabel at:

http://www.ppe.ufrj.br/ppe/ production/tesis/dazuaga.pdf .

Empresa de Pesquisa Energética. Balanço Energético Nacional, 2010. /MME.

California Environmental Protection Agency. Air Quality Impacts of the Use of Ethanol in California Reformulated Gasoline. AIR RESOURCES BOARD Planning and Technical Support Division. State of Califórnia Allen, Paul (Org.). Novembro 1999. Availabel at:

http://www.arb.ca.gov/cbg/ regact/ethanol/ethfate.htm.

CONAMA. Resolução nº 010/1989. Availabel at:

http://www.areaseg.com/conama/ 1989/010-1989.pdf.

	- https://e-reports-ext.llnl.gov/pdf/235217.pdf.

technology/chemistry/combustion/prf\_paper.pdf

Intergovernmental Panel on Climate Change. 2006 IPCC Guidelines for National Greenhouse Gas Inventories, Prepared by the National Greenhouse Gas Inventories Programme. Published: IGES, Japan.

**3** 

**Greenhouse Gas Emissions** 

*2Wageningen University* 

*2The Netherlands*

*1Brazil* 

**from Hydroelectric Reservoirs: What** 

Raquel Mendonça1,2, Nathan Barros1, Luciana O. Vidal1, Felipe Pacheco1, Sarian Kosten2 and Fábio Roland1

*1Federal University of Juiz de Fora, Federal University of Rio de Janeiro* 

**Knowledge Do We Have and What is Lacking?** 

*The question that motivated us to write this chapter was "How does hydroelectricity production contribute to the global emissions of greenhouse gases?". Here, we present an overview of (i) scientific advances on the topic and (ii) aspects of hydroelectric reservoir ecology in the context of greenhouse* 

Electricity production is a challenging issue when it comes to mitigating greenhouse gases (GHG) emissions without risking development goals. Non-renewable energy sources, as fossil fuel burning, account for most of the global energy production (68% in 2007) and are responsible for most of the anthropogenic GHG emissions to the atmosphere (40%, International Panel on Climate Change - IPCC, 2007). Compared to fossil fuels, hydropower has been considered an attractive renewable energy source with the advantage of being less harmful in terms of GHG emissions (International Energy Agency - IEA, 2008). Hydropower currently provides about 16% of the world's electricity supply (IEA, 2008) with many countries depending on it for more than 90% of their supply. Indeed, hydropower is a proven, mature, predictable and typically price-competitive technology. Moreover, it has among the best conversion efficiencies of all known energy sources: 90% efficiency as opposed to up to 50% efficiency of e.g. fossil fuel burning. The historical perception of hydroelectricity as being GHG neutral (Hoffert et al., 1998), however, is now known to be flawed. The concern regarding the GHG emissions by hydroelectric reservoirs has steeply increased since the early 90's, even though it remains unclear what their actual emission is. Although inland water systems naturally produce and emit carbon to the atmosphere (Cole et al., 2007) different characteristics of hydroelectric reservoirs cause that they often produce and emit more than natural systems, especially in the first twenty years after flooding (Barros et al., 2011). This is mainly due to the usually excessive availability of decomposable organic matter in hydroelectric reservoirs. Not only large amounts of soil and terrestrial vegetation are flooded by damming rivers, but terrestrial organic matter derived from land erosion is continuously flushed into reservoirs as well. The usually high water residence

**1. Introduction** 

*gas production, consumption and emission.* 

Kilpinen, Pia. NOx Emission Formation in Marine Diesel Engines – towards a quantitative understanding. Availabel at:

http://www.umweltdaten.de/verkehr/downloads/ seeschifffahrt/mn2-03.pdf 1.

Marinov, Nick M. A Detailed Chemical Kinetic Model for High Temperature Ethanol Oxidation. Availabel at:

http://www..cmls.llnl.gov/data/docs/science\_and \_technology/chemistry/ combustion/ethanol\_paper.pdf.


kit.edu/deutschmann/img/content/05\_YMuharam\_PhD\_UniHD.pdf

PROCONVE/CETESB. Fatores de Emissão. Departamento de Tecnologia de Emissões de Veículos. Availabel at:

http://www.cetesb.sp.gov.br/Ar/emissoes/proconve3.asp.

Westbrook, Charles K. al al. The Internal Combustion at Work. Lawrence Livermore National Laboratory, Livermore United States of America. Availabel at: https://www.llnl.gov/str/Westbrook.html.

## **Greenhouse Gas Emissions from Hydroelectric Reservoirs: What Knowledge Do We Have and What is Lacking?**

Raquel Mendonça1,2, Nathan Barros1, Luciana O. Vidal1, Felipe Pacheco1, Sarian Kosten2 and Fábio Roland1 *1Federal University of Juiz de Fora, Federal University of Rio de Janeiro 2Wageningen University 1Brazil 2The Netherlands*

## **1. Introduction**

54 Greenhouse Gases – Emission, Measurement and Management

Kilpinen, Pia. NOx Emission Formation in Marine Diesel Engines – towards a quantitative

Ministério do Meio Ambiente. – Relatório de Referência – Emissões de Gases de Efeito Estufa no Setor Energético por Fontes Móveis. Brasília. Availabel at:

Muharam, Yuswan. Modelling of the Oxidation and Combustion of Large Hydrocarbons

kit.edu/deutschmann/img/content/05\_YMuharam\_PhD\_UniHD.pdf PROCONVE/CETESB. Fatores de Emissão. Departamento de Tecnologia de Emissões de

Westbrook, Charles K. al al. The Internal Combustion at Work. Lawrence Livermore National Laboratory, Livermore United States of America. Availabel at:

http://www.mma.gov.br/estruturas/182/\_arquivos/emissoes\_

Universidade de Heidelberg, 18 november 2005. Availabel at:

http://www.cetesb.sp.gov.br/Ar/emissoes/proconve3.asp.

http://www.umweltdaten.de/verkehr/downloads/ seeschifffahrt/mn2-03.pdf 1. Marinov, Nick M. A Detailed Chemical Kinetic Model for High Temperature Ethanol

http://www..cmls.llnl.gov/data/docs/science\_and \_technology/chemistry/

Using an Automatic Generation of Mechanisms. Dissertação de Mestrado,

understanding. Availabel at:

combustion/ethanol\_paper.pdf.

Oxidation. Availabel at:

veiculares\_182.pdf.

http://www.itcp.

Veículos. Availabel at:

https://www.llnl.gov/str/Westbrook.html.

*The question that motivated us to write this chapter was "How does hydroelectricity production contribute to the global emissions of greenhouse gases?". Here, we present an overview of (i) scientific advances on the topic and (ii) aspects of hydroelectric reservoir ecology in the context of greenhouse gas production, consumption and emission.* 

Electricity production is a challenging issue when it comes to mitigating greenhouse gases (GHG) emissions without risking development goals. Non-renewable energy sources, as fossil fuel burning, account for most of the global energy production (68% in 2007) and are responsible for most of the anthropogenic GHG emissions to the atmosphere (40%, International Panel on Climate Change - IPCC, 2007). Compared to fossil fuels, hydropower has been considered an attractive renewable energy source with the advantage of being less harmful in terms of GHG emissions (International Energy Agency - IEA, 2008). Hydropower currently provides about 16% of the world's electricity supply (IEA, 2008) with many countries depending on it for more than 90% of their supply. Indeed, hydropower is a proven, mature, predictable and typically price-competitive technology. Moreover, it has among the best conversion efficiencies of all known energy sources: 90% efficiency as opposed to up to 50% efficiency of e.g. fossil fuel burning. The historical perception of hydroelectricity as being GHG neutral (Hoffert et al., 1998), however, is now known to be flawed. The concern regarding the GHG emissions by hydroelectric reservoirs has steeply increased since the early 90's, even though it remains unclear what their actual emission is.

Although inland water systems naturally produce and emit carbon to the atmosphere (Cole et al., 2007) different characteristics of hydroelectric reservoirs cause that they often produce and emit more than natural systems, especially in the first twenty years after flooding (Barros et al., 2011). This is mainly due to the usually excessive availability of decomposable organic matter in hydroelectric reservoirs. Not only large amounts of soil and terrestrial vegetation are flooded by damming rivers, but terrestrial organic matter derived from land erosion is continuously flushed into reservoirs as well. The usually high water residence

Greenhouse Gas Emissions from Hydroelectric

Reservoirs: What Knowledge Do We Have and What is Lacking? 57

Fig. 1. Timeline of scientific advances regarding the role of hydroelectric reservoirs as

sources of GHG to the atmosphere.

time in reservoirs as compared to rivers, combined with high inorganic nutrient inputs, favors organic matter decomposition and, thus, the production of two major GHGs – carbon dioxide (CO2) and methane (CH4). The amount of CO2 and CH4 emitted varies (a) among reservoirs (as function of drainage basin characteristics, reservoir morphology, climate, etc.); (b) within reservoirs (along longitudinal gradients from the tributaries to the dam, before and after the dam, etc.); and (c) over time (with reservoir aging, seasonally, daily, with changes in anthropogenic activities in the drainage basin, and with dam operation depending on energy needs and precipitation regime). Attempts to estimate the amounts of CO2 and CH4 emitted to the atmosphere should consider such variability which makes it a complex task.

Despite the complexity of the GHG issue, the effect of damming rivers on the atmospheric GHG concentrations cannot be disregarded. The concern about the impacts of hydroelectric reservoirs on the global GHG budgets has been increasing in the same pace as the construction of new dams. Nowadays, there are at least 45,000 large hydroelectric reservoirs in operation worldwide (World Commission on Dams –WCD, 2000). Moreover, recent inventories estimate the total surface area of world's hydroelectric reservoirs at about 350,000 km2 (Barros et al., 2011). The substantial size of some hydroelectric projects and the extensive total surface area globally covered by reservoirs require that research determining the impacts of these systems be done at ever-increasing spatial and temporal scales.

This chapter focuses on the GHG emissions that are due to the landscape transformation (damming a river to form a reservoir) and to reservoir operation to produce electricity. First, the scientific advances towards understanding the role of hydroelectric reservoirs and their environmental effects as sources of GHG are delineated. Then, the metabolic processes involved in GHG production and consumption are described with focus on the two major interacting compartments: the water column and the sediment. Next, the external factors influencing the emission rates from reservoirs are discussed. Finally, an overlook of future perspectives in terms of GHG emissions from hydroelectric reservoirs is presented.

## **2. Important scientific advances**

The ever increasing global energy demand and the concern about the changes in environment have lead to an urge to assess the hydropower 'footprint' in terms of GHG emissions to the atmosphere. Since the early 90's the role of hydroelectric reservoirs as sources or, as the opposite, sinks of GHG has rapidly become a global topic of investigation (**Figure 1**). At least 85 globally distributed hydroelectric reservoirs have so far been studied with focus on GHG fluxes (Barros et al., 2011). The first scientific papers focused on reservoirs located in Canada (e.g. Rudd et al., 1993; Duchemin et al., 1995), Brazil (e.g. Rosa & Schaeffer, 1994; Fearnside, 1995, 1997), Panama (Keller & Stallard, 1994) and French Guiana (e.g. Galy-Lacaux et al., 1997; Galy-Lacaux et al., 1999). Later, reservoirs in Finland (e.g. Huttunen et al., 2002), USA (e.g. Soumis et al., 2004), Sweden (e.g. Aberg et al., 2004; Bergstrom et al., 2004) and Switzerland (e.g. Diem et al., 2007) were studied. Only very recently, GHG emissions from reservoirs located in China, the country with the largest installed capacity in the world, became focus of investigation (e.g. Chen et al., 2009; Wang et al., 2011; Zheng et al., 2011) (**Figure 2**).

#### Greenhouse Gas Emissions from Hydroelectric Reservoirs: What Knowledge Do We Have and What is Lacking? 57

56 Greenhouse Gases – Emission, Measurement and Management

time in reservoirs as compared to rivers, combined with high inorganic nutrient inputs, favors organic matter decomposition and, thus, the production of two major GHGs – carbon dioxide (CO2) and methane (CH4). The amount of CO2 and CH4 emitted varies (a) among reservoirs (as function of drainage basin characteristics, reservoir morphology, climate, etc.); (b) within reservoirs (along longitudinal gradients from the tributaries to the dam, before and after the dam, etc.); and (c) over time (with reservoir aging, seasonally, daily, with changes in anthropogenic activities in the drainage basin, and with dam operation depending on energy needs and precipitation regime). Attempts to estimate the amounts of CO2 and CH4 emitted to the atmosphere should consider such variability which makes it a

Despite the complexity of the GHG issue, the effect of damming rivers on the atmospheric GHG concentrations cannot be disregarded. The concern about the impacts of hydroelectric reservoirs on the global GHG budgets has been increasing in the same pace as the construction of new dams. Nowadays, there are at least 45,000 large hydroelectric reservoirs in operation worldwide (World Commission on Dams –WCD, 2000). Moreover, recent inventories estimate the total surface area of world's hydroelectric reservoirs at about 350,000 km2 (Barros et al., 2011). The substantial size of some hydroelectric projects and the extensive total surface area globally covered by reservoirs require that research determining the impacts of these systems be done at ever-increasing spatial and temporal

This chapter focuses on the GHG emissions that are due to the landscape transformation (damming a river to form a reservoir) and to reservoir operation to produce electricity. First, the scientific advances towards understanding the role of hydroelectric reservoirs and their environmental effects as sources of GHG are delineated. Then, the metabolic processes involved in GHG production and consumption are described with focus on the two major interacting compartments: the water column and the sediment. Next, the external factors influencing the emission rates from reservoirs are discussed. Finally, an overlook of future perspectives in terms of GHG emissions from hydroelectric reservoirs

The ever increasing global energy demand and the concern about the changes in environment have lead to an urge to assess the hydropower 'footprint' in terms of GHG emissions to the atmosphere. Since the early 90's the role of hydroelectric reservoirs as sources or, as the opposite, sinks of GHG has rapidly become a global topic of investigation (**Figure 1**). At least 85 globally distributed hydroelectric reservoirs have so far been studied with focus on GHG fluxes (Barros et al., 2011). The first scientific papers focused on reservoirs located in Canada (e.g. Rudd et al., 1993; Duchemin et al., 1995), Brazil (e.g. Rosa & Schaeffer, 1994; Fearnside, 1995, 1997), Panama (Keller & Stallard, 1994) and French Guiana (e.g. Galy-Lacaux et al., 1997; Galy-Lacaux et al., 1999). Later, reservoirs in Finland (e.g. Huttunen et al., 2002), USA (e.g. Soumis et al., 2004), Sweden (e.g. Aberg et al., 2004; Bergstrom et al., 2004) and Switzerland (e.g. Diem et al., 2007) were studied. Only very recently, GHG emissions from reservoirs located in China, the country with the largest installed capacity in the world, became focus of investigation (e.g. Chen et al., 2009; Wang et

complex task.

scales.

is presented.

**2. Important scientific advances** 

al., 2011; Zheng et al., 2011) (**Figure 2**).

Fig. 1. Timeline of scientific advances regarding the role of hydroelectric reservoirs as sources of GHG to the atmosphere.

Fig. 2. Global distribution of hydroelectric reservoirs (HR), expressed as the proportion of the total number of reservoirs globally constructed on each continent, the proportion of the global HR area on each continent and the number of HR studied on each continent. The blue gradient indicates the hydroelectric installed capacity per continent. The black circle sizes are proportional to the number of papers published dealing with GHG emissions from reservoirs located in each country. The percentages of total number and total area of hydroelectric reservoirs were calculated based on ICOLD (2007). The numbers of papers published approaching GHG emissions from hydroelectric reservoirs were extracted from Barros et al., 2011, Chanudet et al., 2011, Wang et al., 2011, and Zheng et al., 2011.  Greenhouse Gas Emissions from Hydroelectric

Reservoirs: What Knowledge Do We Have and What is Lacking? 59

Historically, the question to which extent hydropower is a GHG-friendly source of energy has been an important one. A first study focusing on reservoirs in northern Canada suggested that the amount of GHG emitted from reservoirs was substantial when compared to emissions from fossil fuel burning (Rudd et al., 1993). A subsequent assessment estimated that GHG emissions from hydropower plants (considering their full lifecycle) was 30 to 60 times less than the emissions from energy generated from fossil fuel (Gagnon & Van de Vate, 1997). This assessment, however, did not include data from hydroelectric reservoirs in tropical climates which were later found to have relatively high emissions. Yearly GHG emissions from a large reservoir located in the Brazilian Amazon (Tucuruí reservoir), for example, was argued to overcome the fossil fuel emissions from Brazil's largest city, São Paulo (Fearnside, 2002). Others criticized this finding and considered the reservoir GHG emissions to be largely overestimated (Rosa et al., 2004). This critic triggered a scientific debate between two groups with contrasting opinions (Fearnside, 2004; Rosa et al., 2004; Cullenward & Victor, 2006; Fearnside, 2006; Giles, 2006). Although the groups disagree on the amount of GHG emitted from hydropower in relation to other energy sources, they do agree that GHG emissions from tropical reservoirs are large. Emissions from the Brazilian Curuá-Una reservoir, for instance, were argued to overcome the emissions from oil generated electricity (Fearnside, 2005). Later, Brazilian hydroelectric reservoirs were shown to emit less carbon per energy production than thermonuclear power plants, with the exception of some cases of low power density, i.e. low energy production/flooded area ratio (Dos Santos et al., 2006). Finally, a recent inventory including several Brazilian reservoirs located both in the Amazon and in other biomes showed that the GHG emissions per energy produced (GHG/MWh) are lower in most reservoirs, regardless of their age (Ometto et al., submitted). This inventory showed that the highest GHG/MWh rates occur in large

Overall, hydropower may thus produce electricity with one of the lowest life-cycle GHG emissions (Weisser, 2007), especially in non-tropical regions (Barros et al., 2011). Nevertheless, the important role of hydroelectric reservoirs in the global GHG dynamics is unquestionable.

The high GHG emissions from hydroelectric reservoirs were originally argued to be due to the decomposition of flooded organic soil and vegetation (Rudd et al., 1993; Abril et al., 2005). This understanding of the role of reservoirs needs to take into account the balance between GHG emissions and consumption prior to and after the impoundment of a certain area (Teodoru et al., 2010). Net GHG emission was considered close to zero prior to impoundment, as emissions from rivers would be compensated by the sink of terrestrial photosynthesis. After flooding, the GHG neutral terrestrial system is replaced by a system

This focus on the flooded organic matter further lead to the understanding that with reservoir aging, the amount of decomposable flooded organic matter would be gradually reduced and, thus, GHG emissions from reservoir surfaces would decline. Indeed, a longterm assessment of GHG emissions from the tropical Petit Saut reservoir showed that CO2

4 emissions are high in the first two years after impoundment after which emissions

**2.1 GHG emissions – hydropower versus other electricity sources** 

Amazonian reservoirs with low energy production rates.

with net GHG emissions to the atmosphere (Kelly et al., 1994).

**2.2 Organic matter and GHG** 

and CH

Fig. 2. Global distribution of hydroelectric reservoirs (HR), expressed as the proportion of the total number of reservoirs globally constructed on each continent, the proportion of the global HR area on each continent and the number of HR studied on each continent. The blue gradient indicates the hydroelectric installed capacity per continent. The black circle sizes are proportional to the number of papers published dealing with GHG emissions from reservoirs located in each country. The percentages of total number and total area of hydroelectric reservoirs were calculated based on ICOLD (2007). The numbers of papers published approaching GHG emissions from hydroelectric reservoirs were extracted from Barros et al., 2011, Chanudet et al., 2011, Wang et

al., 2011, and Zheng et al., 2011.

## **2.1 GHG emissions – hydropower versus other electricity sources**

Historically, the question to which extent hydropower is a GHG-friendly source of energy has been an important one. A first study focusing on reservoirs in northern Canada suggested that the amount of GHG emitted from reservoirs was substantial when compared to emissions from fossil fuel burning (Rudd et al., 1993). A subsequent assessment estimated that GHG emissions from hydropower plants (considering their full lifecycle) was 30 to 60 times less than the emissions from energy generated from fossil fuel (Gagnon & Van de Vate, 1997). This assessment, however, did not include data from hydroelectric reservoirs in tropical climates which were later found to have relatively high emissions. Yearly GHG emissions from a large reservoir located in the Brazilian Amazon (Tucuruí reservoir), for example, was argued to overcome the fossil fuel emissions from Brazil's largest city, São Paulo (Fearnside, 2002). Others criticized this finding and considered the reservoir GHG emissions to be largely overestimated (Rosa et al., 2004). This critic triggered a scientific debate between two groups with contrasting opinions (Fearnside, 2004; Rosa et al., 2004; Cullenward & Victor, 2006; Fearnside, 2006; Giles, 2006). Although the groups disagree on the amount of GHG emitted from hydropower in relation to other energy sources, they do agree that GHG emissions from tropical reservoirs are large. Emissions from the Brazilian Curuá-Una reservoir, for instance, were argued to overcome the emissions from oil generated electricity (Fearnside, 2005). Later, Brazilian hydroelectric reservoirs were shown to emit less carbon per energy production than thermonuclear power plants, with the exception of some cases of low power density, i.e. low energy production/flooded area ratio (Dos Santos et al., 2006). Finally, a recent inventory including several Brazilian reservoirs located both in the Amazon and in other biomes showed that the GHG emissions per energy produced (GHG/MWh) are lower in most reservoirs, regardless of their age (Ometto et al., submitted). This inventory showed that the highest GHG/MWh rates occur in large Amazonian reservoirs with low energy production rates.

Overall, hydropower may thus produce electricity with one of the lowest life-cycle GHG emissions (Weisser, 2007), especially in non-tropical regions (Barros et al., 2011). Nevertheless, the important role of hydroelectric reservoirs in the global GHG dynamics is unquestionable.

### **2.2 Organic matter and GHG**

The high GHG emissions from hydroelectric reservoirs were originally argued to be due to the decomposition of flooded organic soil and vegetation (Rudd et al., 1993; Abril et al., 2005). This understanding of the role of reservoirs needs to take into account the balance between GHG emissions and consumption prior to and after the impoundment of a certain area (Teodoru et al., 2010). Net GHG emission was considered close to zero prior to impoundment, as emissions from rivers would be compensated by the sink of terrestrial photosynthesis. After flooding, the GHG neutral terrestrial system is replaced by a system with net GHG emissions to the atmosphere (Kelly et al., 1994).

This focus on the flooded organic matter further lead to the understanding that with reservoir aging, the amount of decomposable flooded organic matter would be gradually reduced and, thus, GHG emissions from reservoir surfaces would decline. Indeed, a longterm assessment of GHG emissions from the tropical Petit Saut reservoir showed that CO2 and CH4 emissions are high in the first two years after impoundment after which emissions

Greenhouse Gas Emissions from Hydroelectric

**for GHG emissions** 

Reservoirs: What Knowledge Do We Have and What is Lacking? 61

average emissions of 1400 mg m-2 d-1 of CO2 and 20 mg m-2 d-1 of CH4. Their estimated emissions from tropical reservoirs (3500 mg m-2 d-1 of CO2 and 300 mg m-2 d-1 of CH4), though, were based on data from a very small number of systems (four) and might have been overestimated due to the inclusion of young reservoirs (1-2 year old) which have high emissions. After estimating the global area occupied by all reservoirs types, the authors calculated the global emissions of GHG to be 273 Tg of CO2 and 48 Tg of CH4 per year. Considering that CH4 global warming potential is 25 times higher than that of CO2 (IPCC,

After 2000, there was an important increase in the amount of data on GHG emissions from reservoirs located in both temperate (e.g. Huttunen et al., 2002; Aberg et al., 2004; Bergstrom et al., 2004; Soumis et al., 2004; Tremblay et al., 2004; Duchemin et al., 2006) and tropical regions (e.g. Delmas et al., 2001; Fearnside, 2002; Rosa et al., 2004; Abril et al., 2005; Guerin et al., 2006 ; Guerin et al., 2007). Comparisons between emissions in different regions were applied as tools to understand the factors controlling emissions from reservoirs. For example, CO2 emissions from Swedish reservoirs were lower than those reported for other boreal regions, which was attributed to the fact that in Sweden often comparatively small areas with thin layers of organic soil are flooded for reservoir construction (Bergstrom et al., 2004). In 2011 a review of the achievements in 20 years of measurements of CH4 emission from tropical and equatorial reservoirs came out (Demarty & Bastien, 2011). The document claims that GHG emissions

2007), these emissions corresponded to 2,600 Tg of CO2-equivalents per year.

might have been underestimated in the tropics due to the neglect of CH4 emissions.

Finally, the latest global assessment of GHG emissions from hydroelectric reservoirs compiled data from 85 globally distributed systems which account for about 20% of the global area occupied by hydroelectric reservoirs (Barros et al., 2011). The authors estimated that hydroelectric reservoirs globally emit about 51 Tg of carbon per year (48 Tg per year as CO2 and 3 Tg per year as CH4 or 288 Tg of CO2-equivalents per year) which is low compared to the first global estimation (321 Tg of carbon per year, St Louis et al., 2000). This difference is argued to be caused (i) by the greater amount of data currently available and (ii) by the smaller area occupied by hydroelectric reservoirs (350.000 km2, Barros et al., 2011) when compared to the area occupied by all types of reservoirs (1.500.000 km2, St Louis et al., 2000). Furthermore, this latest assessment showed that GHG emissions are correlated to reservoirs age and latitude, with the highest emission rates occurring in the Amazon region.

**3. Environmental effects of hydroelectric reservoirs and the consequences** 

of changes which influence, directly or indirectly, the local GHG fluxes (**Figure 3**).

By definition, hydropower is a renewable source of electricity in which power is derived from the energy of water moving from higher to lower elevations. The amount of energy generated depends both on the accumulated water volume and on the difference in height between the water inlet and the outflow. While dams perform an important function, their effect on landscapes is remarkable: a fragment of river and its adjacent terrestrial environment are transformed in a new freshwater system, the reservoir. According to recent global inventories, 10,800 km3 of water were impounded in reservoirs (all kinds of reservoirs, e.g. irrigation, water supply, flood control, and aquaculture) in the last half century, causing the sea level to reduce by approximately 30 millimeters (Chao et al., 2008). The construction of reservoirs clearly represents, thus, one of the major human impacts on the hydrological cycle. The effects of such transformation, however, surpass the hydrological level. Impounded areas undergo a cascade

rapidly decline when the more labile flooded organic matter is decomposed (Galy-Lacaux et al., 1997; Galy-Lacaux et al., 1999). Emissions of CO2 from Brazilian reservoirs were predicted to be concentrated in the first 10 years after flooding (Fearnside, 1995, 1997). Later, evidence from Canadian systems showed a similar trend and suggested that CO2 emissions from reservoirs older than 10 years tend to equal the emissions from natural lakes and rivers (Tremblay et al., 2004). Similar results were also reported for a Swedish reservoir which was compared with a natural lake (Aberg et al., 2004). Moreover, a significant negative relationship between age and GHG emissions was registered for temperate reservoirs (St Louis et al., 2000) and for reservoirs located all over the globe (Barros et al., 2011).

Nevertheless, other sources of carbon to reservoirs besides flooded organic matter should be taken into consideration. During their complete lifetime, organic matter and nutrients from the drainage basin are continuously flushed into the systems through tributary rivers (Fearnside, 1995; Roland et al., 2010) and aquatic primary production rates tend to increase (Bayne et al., 1983). Once in the reservoirs, organic matter derived from the drainage basin and from aquatic primary production mineralizes at different rates, the latest being usually more labile (Kritzberg et al., 2005; Vidal et al., 2011). Most of the organic matter mineralization and, thus, most of the GHG production in reservoirs occurs in the sediment (Aberg et al., 2004; Abe et al., 2005). Furthermore, it has become clear that GHG emitted from reservoir surfaces is not necessarily produced within the system, as tributary rivers may export large amounts of GHG to reservoirs (e.g. Lima et al., 1998).

#### **2.3 Global emission estimates**

From 2004 on, the studies were marked by a more holistic approach incorporating emissions from water passing through the turbines and downstream of dams, as has been done earlier in the tropical Petit Saut reservoir (Galy-Lacaux et al., 1997). Continuous measurements from Petit Saut reservoir lead to a 10-year assessment of GHG emissions which showed that degassing downstream the turbines may represent the major pathway for CH4 emissions (Abril et al., 2005). High emissions downstream of dams were registered in other tropical (e.g. Guerin et al., 2006; Kemenes et al., 2007, 2011) and in temperate reservoirs as well (e.g. Soumis et al., 2004; Abril et al., 2005; Roehm & Tremblay, 2006). Motivated by the idea of mitigating CH4 emissions from reservoirs, a method was proposed to capture the CH4 emitted downstream dams and utilize it, for instance, to generate electricity during high demand periods (Bambace et al., 2007).

More recently, attention has turned to the spatial variability on GHG emissions within reservoirs. Measurements in tropical reservoirs suggested that neglecting the spatial variability in CO2 emission from reservoirs may lead to more than 25% error in estimations (Roland et al., 2010). In the same year, spatial variability in CO2 fluxes from temperate reservoirs was shown to decline with time after impounding (Teodoru et al., 2010). The importance of considering the spatial variability in CH4 emissions was also addressed based on data from Chinese reservoirs (Zheng et al., 2011).

The first global estimation of GHG emissions from reservoirs was published in 2000 (St Louis et al., 2000). This assessment considered emissions from reservoirs of all uses, including irrigation, water supply, energy generation and others. Based on 21 systems located in temperate climate (i.e. Canada, United States, and Finland), the authors calculated

rapidly decline when the more labile flooded organic matter is decomposed (Galy-Lacaux et al., 1997; Galy-Lacaux et al., 1999). Emissions of CO2 from Brazilian reservoirs were predicted to be concentrated in the first 10 years after flooding (Fearnside, 1995, 1997). Later, evidence from Canadian systems showed a similar trend and suggested that CO2 emissions from reservoirs older than 10 years tend to equal the emissions from natural lakes and rivers (Tremblay et al., 2004). Similar results were also reported for a Swedish reservoir which was compared with a natural lake (Aberg et al., 2004). Moreover, a significant negative relationship between age and GHG emissions was registered for temperate reservoirs (St

Nevertheless, other sources of carbon to reservoirs besides flooded organic matter should be taken into consideration. During their complete lifetime, organic matter and nutrients from the drainage basin are continuously flushed into the systems through tributary rivers (Fearnside, 1995; Roland et al., 2010) and aquatic primary production rates tend to increase (Bayne et al., 1983). Once in the reservoirs, organic matter derived from the drainage basin and from aquatic primary production mineralizes at different rates, the latest being usually more labile (Kritzberg et al., 2005; Vidal et al., 2011). Most of the organic matter mineralization and, thus, most of the GHG production in reservoirs occurs in the sediment (Aberg et al., 2004; Abe et al., 2005). Furthermore, it has become clear that GHG emitted from reservoir surfaces is not necessarily produced within the system, as tributary rivers

From 2004 on, the studies were marked by a more holistic approach incorporating emissions from water passing through the turbines and downstream of dams, as has been done earlier in the tropical Petit Saut reservoir (Galy-Lacaux et al., 1997). Continuous measurements from Petit Saut reservoir lead to a 10-year assessment of GHG emissions which showed that degassing downstream the turbines may represent the major pathway for CH4 emissions (Abril et al., 2005). High emissions downstream of dams were registered in other tropical (e.g. Guerin et al., 2006; Kemenes et al., 2007, 2011) and in temperate reservoirs as well (e.g. Soumis et al., 2004; Abril et al., 2005; Roehm & Tremblay, 2006). Motivated by the idea of mitigating CH4 emissions from reservoirs, a method was proposed to capture the CH4 emitted downstream dams and utilize it, for instance, to generate electricity during high

More recently, attention has turned to the spatial variability on GHG emissions within reservoirs. Measurements in tropical reservoirs suggested that neglecting the spatial variability in CO2 emission from reservoirs may lead to more than 25% error in estimations (Roland et al., 2010). In the same year, spatial variability in CO2 fluxes from temperate reservoirs was shown to decline with time after impounding (Teodoru et al., 2010). The importance of considering the spatial variability in CH4 emissions was also addressed based

The first global estimation of GHG emissions from reservoirs was published in 2000 (St Louis et al., 2000). This assessment considered emissions from reservoirs of all uses, including irrigation, water supply, energy generation and others. Based on 21 systems located in temperate climate (i.e. Canada, United States, and Finland), the authors calculated

Louis et al., 2000) and for reservoirs located all over the globe (Barros et al., 2011).

may export large amounts of GHG to reservoirs (e.g. Lima et al., 1998).

**2.3 Global emission estimates** 

demand periods (Bambace et al., 2007).

on data from Chinese reservoirs (Zheng et al., 2011).

average emissions of 1400 mg m-2 d-1 of CO2 and 20 mg m-2 d-1 of CH4. Their estimated emissions from tropical reservoirs (3500 mg m-2 d-1 of CO2 and 300 mg m-2 d-1 of CH4), though, were based on data from a very small number of systems (four) and might have been overestimated due to the inclusion of young reservoirs (1-2 year old) which have high emissions. After estimating the global area occupied by all reservoirs types, the authors calculated the global emissions of GHG to be 273 Tg of CO2 and 48 Tg of CH4 per year. Considering that CH4 global warming potential is 25 times higher than that of CO2 (IPCC, 2007), these emissions corresponded to 2,600 Tg of CO2-equivalents per year.

After 2000, there was an important increase in the amount of data on GHG emissions from reservoirs located in both temperate (e.g. Huttunen et al., 2002; Aberg et al., 2004; Bergstrom et al., 2004; Soumis et al., 2004; Tremblay et al., 2004; Duchemin et al., 2006) and tropical regions (e.g. Delmas et al., 2001; Fearnside, 2002; Rosa et al., 2004; Abril et al., 2005; Guerin et al., 2006 ; Guerin et al., 2007). Comparisons between emissions in different regions were applied as tools to understand the factors controlling emissions from reservoirs. For example, CO2 emissions from Swedish reservoirs were lower than those reported for other boreal regions, which was attributed to the fact that in Sweden often comparatively small areas with thin layers of organic soil are flooded for reservoir construction (Bergstrom et al., 2004). In 2011 a review of the achievements in 20 years of measurements of CH4 emission from tropical and equatorial reservoirs came out (Demarty & Bastien, 2011). The document claims that GHG emissions might have been underestimated in the tropics due to the neglect of CH4 emissions.

Finally, the latest global assessment of GHG emissions from hydroelectric reservoirs compiled data from 85 globally distributed systems which account for about 20% of the global area occupied by hydroelectric reservoirs (Barros et al., 2011). The authors estimated that hydroelectric reservoirs globally emit about 51 Tg of carbon per year (48 Tg per year as CO2 and 3 Tg per year as CH4 or 288 Tg of CO2-equivalents per year) which is low compared to the first global estimation (321 Tg of carbon per year, St Louis et al., 2000). This difference is argued to be caused (i) by the greater amount of data currently available and (ii) by the smaller area occupied by hydroelectric reservoirs (350.000 km2, Barros et al., 2011) when compared to the area occupied by all types of reservoirs (1.500.000 km2, St Louis et al., 2000). Furthermore, this latest assessment showed that GHG emissions are correlated to reservoirs age and latitude, with the highest emission rates occurring in the Amazon region.

## **3. Environmental effects of hydroelectric reservoirs and the consequences for GHG emissions**

By definition, hydropower is a renewable source of electricity in which power is derived from the energy of water moving from higher to lower elevations. The amount of energy generated depends both on the accumulated water volume and on the difference in height between the water inlet and the outflow. While dams perform an important function, their effect on landscapes is remarkable: a fragment of river and its adjacent terrestrial environment are transformed in a new freshwater system, the reservoir. According to recent global inventories, 10,800 km3 of water were impounded in reservoirs (all kinds of reservoirs, e.g. irrigation, water supply, flood control, and aquaculture) in the last half century, causing the sea level to reduce by approximately 30 millimeters (Chao et al., 2008). The construction of reservoirs clearly represents, thus, one of the major human impacts on the hydrological cycle. The effects of such transformation, however, surpass the hydrological level. Impounded areas undergo a cascade of changes which influence, directly or indirectly, the local GHG fluxes (**Figure 3**).

Fig. 3. Diagram illustrating some of the major impacts of damming rivers and their effects on GHG emissions. Direct hydrological impacts of damming rivers trigger a cascade of shifts in the physical and chemical environment, leading to indirect (through changes in metabolism) and direct changes in GHG fluxes. External factors affecting GHG emissions are mainly related to atmospheric conditions and drainage basin characteristics. OC = organic carbon; PP = primary production. Arrows with the (+) symbol represent positive effect; arrows with the (-) symbol represent negative effect.  Greenhouse Gas Emissions from Hydroelectric

Keil, 1995; Hedges et al., 1997).

subsequent release of CO

timescales (Dean & Gorham, 1998).

turbulent flow keeps oxygen concentrations high, CH

2 and CH

as on the quantity and quality of organic matter.

Reservoirs: What Knowledge Do We Have and What is Lacking? 63

Prior to dam construction, rivers generally have rapid water flow rates which vary with drainage basin size and topography and respond to the seasonality in the upstream precipitation regime. Precipitation regimes also control variations in river water level, which

terrestrial organic matter decomposition (Cole et al., 2007; Tranvik et al., 2009). Because the

2005) except in highly organic matter-rich rivers (e.g. Lima et al., 1998). In compensation,

photosynthesis tends to exceed respiration. Furthermore, rivers play a crucial role in removing carbon from the global cycle by carrying large amounts of terrestrial material to the ocean (Schlesinger & Melack, 1981) where it is partially permanently buried (Hedges &

The construction of dams implies the extension of flooded area. This implies that flooded terrestrial vegetation no longer performs photosynthesis. Instead, the organic matter stored in vegetation, as well as in flooded soil, becomes available for bacterial decomposition (with

being a net sink to a net source of carbon to the atmosphere. The amount of GHG emitted from decomposition of flooded vegetation depends on the size of the flooded area, as well

Moreover, the enlargement of flooded area with reservoirs formation comes with an increase in water volume and causes water residence time (or turnover time) at individual impoundments to increase from less than one day to several years in the case of large dams. The reduction of river water velocity in the headwater of reservoirs results in a decreased sediment-carrying capacity and in the retention of particulate matter transported by rivers. Reservoirs are usually constructed at the lower end of large drainage basins, where rivers usually carry terrestrial material derived from the entire upstream drainage basin. Despite being energetically favourable – it guarantees maximal water inflow and, thus, enhances the potential for energy generation – this strategy implies that high amounts of terrestrial material constantly flow into and accumulate in the reservoirs. Globally, it has been estimated that reservoirs may cause the amount of material delivered to the oceans to decrease by more than 50% (Vorosmarty et al., 2003). Organic matter settled at the bottom of the reservoir has a higher chance to be mineralized than when it would have settled at the ocean bottom (see section 4 for details). As a consequence the net global emission of GHG is enlarged. Nevertheless, it should also be remarked that part of the organic material deposited in the sediment of reservoirs is not mineralized and may accumulate for long

Due to the transition from a turbulent-shallow river to a relatively static-deep system, reservoirs tend to undergo thermal stratification, especially in warm regions. The process of thermal stratification is triggered by differences in water density and leads to the formation of water layers: epilimnion (top), metalimnion (intermediate) and hypolimnion (bottom). Once these layers are formed, there is no mixture between top and bottom layers unless some stress (e.g. wind, shift in temperature or increase in water inflow from rivers) breaks water column stability. The establishment of stratification has important effects on oxygen

dynamics and, thus, on GHG emissions from reservoirs (see section 4 for details).

2 due to

2 sink, since

4 emissions hardly occur (Abril et al.,

4). It means that the flooded terrestrial area shifts from

usually varies within a predictable range. Rivers naturally emit large amounts of CO

terrestrial vegetation surrounding a river course usually functions as a CO

Fig. 3. Diagram illustrating some of the major impacts of damming rivers and their effects on GHG emissions. Direct hydrological impacts of damming rivers trigger a cascade of shifts i

leading to indirect (through changes in metabolism) and direct changes in GHG fluxes. External factors affecting

GHG emissions are mainly related to atmospheric conditions and drainage basin characteristics. OC = organic

carbon; PP = primary production. Arrows with the (+) symbol represent positive effect; arrows with the (-) symbol

represent negative effect.

n the physical and chemical environment,

Prior to dam construction, rivers generally have rapid water flow rates which vary with drainage basin size and topography and respond to the seasonality in the upstream precipitation regime. Precipitation regimes also control variations in river water level, which usually varies within a predictable range. Rivers naturally emit large amounts of CO2 due to terrestrial organic matter decomposition (Cole et al., 2007; Tranvik et al., 2009). Because the turbulent flow keeps oxygen concentrations high, CH4 emissions hardly occur (Abril et al., 2005) except in highly organic matter-rich rivers (e.g. Lima et al., 1998). In compensation, terrestrial vegetation surrounding a river course usually functions as a CO2 sink, since photosynthesis tends to exceed respiration. Furthermore, rivers play a crucial role in removing carbon from the global cycle by carrying large amounts of terrestrial material to the ocean (Schlesinger & Melack, 1981) where it is partially permanently buried (Hedges & Keil, 1995; Hedges et al., 1997).

The construction of dams implies the extension of flooded area. This implies that flooded terrestrial vegetation no longer performs photosynthesis. Instead, the organic matter stored in vegetation, as well as in flooded soil, becomes available for bacterial decomposition (with subsequent release of CO2 and CH4). It means that the flooded terrestrial area shifts from being a net sink to a net source of carbon to the atmosphere. The amount of GHG emitted from decomposition of flooded vegetation depends on the size of the flooded area, as well as on the quantity and quality of organic matter.

Moreover, the enlargement of flooded area with reservoirs formation comes with an increase in water volume and causes water residence time (or turnover time) at individual impoundments to increase from less than one day to several years in the case of large dams. The reduction of river water velocity in the headwater of reservoirs results in a decreased sediment-carrying capacity and in the retention of particulate matter transported by rivers. Reservoirs are usually constructed at the lower end of large drainage basins, where rivers usually carry terrestrial material derived from the entire upstream drainage basin. Despite being energetically favourable – it guarantees maximal water inflow and, thus, enhances the potential for energy generation – this strategy implies that high amounts of terrestrial material constantly flow into and accumulate in the reservoirs. Globally, it has been estimated that reservoirs may cause the amount of material delivered to the oceans to decrease by more than 50% (Vorosmarty et al., 2003). Organic matter settled at the bottom of the reservoir has a higher chance to be mineralized than when it would have settled at the ocean bottom (see section 4 for details). As a consequence the net global emission of GHG is enlarged. Nevertheless, it should also be remarked that part of the organic material deposited in the sediment of reservoirs is not mineralized and may accumulate for long timescales (Dean & Gorham, 1998).

Due to the transition from a turbulent-shallow river to a relatively static-deep system, reservoirs tend to undergo thermal stratification, especially in warm regions. The process of thermal stratification is triggered by differences in water density and leads to the formation of water layers: epilimnion (top), metalimnion (intermediate) and hypolimnion (bottom). Once these layers are formed, there is no mixture between top and bottom layers unless some stress (e.g. wind, shift in temperature or increase in water inflow from rivers) breaks water column stability. The establishment of stratification has important effects on oxygen dynamics and, thus, on GHG emissions from reservoirs (see section 4 for details).

Greenhouse Gas Emissions from Hydroelectric

surface and groundwater runoff.

elevated (Roland et al., 2011).

compartments.

oxygenated water layer (Guerin & Abril, 2007).

Reservoirs: What Knowledge Do We Have and What is Lacking? 65

(mostly CO2 but also CH4). Moreover, dissolved CO2 can flow into reservoirs from both

A global assessment of lakes with a worldwide distribution showed that most of the lakes in the world are net sources of CO2 to the atmosphere (Cole et al., 1994; Sobek et al., 2005). The frequently high CO2 concentrations are argued to rely on the high influx of organic matter and the consequent excess respiration by heterotrophic bacteria. The high availability of terrestrial organic matter in hydroelectric reservoirs causes particularly high heterotrophic respiration in these systems. Especially in young reservoirs where the availability of decomposable organic matter is high, the respiration by aerobic planktonic bacteria is

Mineralization may also occur anaerobically in the water column of stratified reservoirs (Abril et al., 2005). With the establishment of stratification, the oxygen consumption due to organic matter mineralization in the bottom layers of reservoirs is no longer offset by water exchange with top oxygenated water layers (see section 3 for details). Thus, bottom water layers become gradually anoxic. Under anoxic conditions, methanogenic bacteria metabolize organic compounds, hydrogen and CO2 into CH4 (methanogenesis), leading to high CH4 concentrations in the bottom layer of reservoirs (Abril et al., 2005). Due to the persistent stratification of reservoirs located in warmer regions, methanogenesis is usually an important pathway of GHG production in those systems (Demarty & Bastien, 2011). However, most of the CH4 produced in the water column of reservoirs tend to be emitted to the atmosphere as CO2 due to the oxidation by metanotrophic bacteria at the top

The overall GHG production in the water column of reservoirs is small compared to the production in the sediments. Great amounts of particulate organic matter transported through rivers tend to sedimentate in reservoirs. Sedimentation is also the fate of organic matter produced by algal photosynthesis in the water column. Once in the sediment, organic matter (i) is mineralized by aerobic or anaerobic bacteria and released as CO2 and CH4 (among other compounds) to the water column, (ii) is re-suspended, or (iii) remains buried. The relative importance of each of these processes is system-specific and influences

Mineralization leads to a general chemical-metabolic gradient within undisturbed sediment profiles. Aerobic mineralization is restricted to the top sediment layer, which may have variable thicknesses depending (i) on the hypolimnetic oxygen concentration, (ii) on the oxygen diffusion coefficient, which is a function of sediment density, sediment porosity, temperature and turbulence at the sediment-water interface and (iii) on the rate of oxygen consumption during organic matter mineralization. Below this top oxygenated sediment layer, anaerobic processes dominate sediment metabolism. While CO2 is the only GHG gas produced as a result of aerobic decomposition, anaerobic processes produce CH4 in addition to CO2. These gases accumulate in sediment pore-water and tend to be released in the water column as a consequence of the difference between concentrations in these two

Diffusive CH4 tends to be oxidized into CO2 by metanotrophic bacteria at the oxygenated top sediment layer or in the water column. The extent of CH4 oxidation in the water column depends on oxygen availability and, thus, may be higher in deeper reservoirs (Lima, 2005).

the overall role of a reservoir as net sink or source of GHG to the atmosphere.

## **4. GHG dynamics in reservoirs – an ecological assessment**

The potential of any aquatic system as source or sink of GHG to the atmosphere is ultimately determined by (i) the concentration of GHG dissolved in the water as related to the atmospheric GHG concentration and (ii) the solubility of the gases in the water. The processes of production, consumption and emission of GHG in hydroelectric reservoirs are dependent on their hydrological characteristics, which make of them especial sites in terms of GHG dynamics when compared to other aquatic systems.

#### **4.1 Water column and sediment**

Most of the routes of production and consumption of CO2 and CH4 are controlled by the aquatic biota metabolism (**Figure 4**). In a nutshell: aerobic respiration produces CO2 which is consumed through photosynthesis; methanogenic bacteria produce CH4 which is consumed by methanotrophic bacteria. An exception to the general pattern of biota mediating CO2 production is the process of photo-oxidation, i.e. the break-down of dissolved organic molecules by solar radiation with the production of several compounds among which CO2 is the more abundant (Soumis et al., 2007; Bastien, 2005). Production and consumption of GHG within reservoirs, however, are not the only processes controlling GHG stocks in the water column. Tributary rivers flowing into reservoirs may carry variable amounts of GHG

Fig. 4. Main pathways of GHG production, consumption and emission to the atmosphere.

The potential of any aquatic system as source or sink of GHG to the atmosphere is ultimately determined by (i) the concentration of GHG dissolved in the water as related to the atmospheric GHG concentration and (ii) the solubility of the gases in the water. The processes of production, consumption and emission of GHG in hydroelectric reservoirs are dependent on their hydrological characteristics, which make of them especial sites in terms

Most of the routes of production and consumption of CO2 and CH4 are controlled by the aquatic biota metabolism (**Figure 4**). In a nutshell: aerobic respiration produces CO2 which is consumed through photosynthesis; methanogenic bacteria produce CH4 which is consumed by methanotrophic bacteria. An exception to the general pattern of biota mediating CO2 production is the process of photo-oxidation, i.e. the break-down of dissolved organic molecules by solar radiation with the production of several compounds among which CO2 is the more abundant (Soumis et al., 2007; Bastien, 2005). Production and consumption of GHG within reservoirs, however, are not the only processes controlling GHG stocks in the water column. Tributary rivers flowing into reservoirs may carry variable amounts of GHG

Fig. 4. Main pathways of GHG production, consumption and emission to the atmosphere.

**4. GHG dynamics in reservoirs – an ecological assessment** 

of GHG dynamics when compared to other aquatic systems.

**4.1 Water column and sediment** 

(mostly CO2 but also CH4). Moreover, dissolved CO2 can flow into reservoirs from both surface and groundwater runoff.

A global assessment of lakes with a worldwide distribution showed that most of the lakes in the world are net sources of CO2 to the atmosphere (Cole et al., 1994; Sobek et al., 2005). The frequently high CO2 concentrations are argued to rely on the high influx of organic matter and the consequent excess respiration by heterotrophic bacteria. The high availability of terrestrial organic matter in hydroelectric reservoirs causes particularly high heterotrophic respiration in these systems. Especially in young reservoirs where the availability of decomposable organic matter is high, the respiration by aerobic planktonic bacteria is elevated (Roland et al., 2011).

Mineralization may also occur anaerobically in the water column of stratified reservoirs (Abril et al., 2005). With the establishment of stratification, the oxygen consumption due to organic matter mineralization in the bottom layers of reservoirs is no longer offset by water exchange with top oxygenated water layers (see section 3 for details). Thus, bottom water layers become gradually anoxic. Under anoxic conditions, methanogenic bacteria metabolize organic compounds, hydrogen and CO2 into CH4 (methanogenesis), leading to high CH4 concentrations in the bottom layer of reservoirs (Abril et al., 2005). Due to the persistent stratification of reservoirs located in warmer regions, methanogenesis is usually an important pathway of GHG production in those systems (Demarty & Bastien, 2011). However, most of the CH4 produced in the water column of reservoirs tend to be emitted to the atmosphere as CO2 due to the oxidation by metanotrophic bacteria at the top oxygenated water layer (Guerin & Abril, 2007).

The overall GHG production in the water column of reservoirs is small compared to the production in the sediments. Great amounts of particulate organic matter transported through rivers tend to sedimentate in reservoirs. Sedimentation is also the fate of organic matter produced by algal photosynthesis in the water column. Once in the sediment, organic matter (i) is mineralized by aerobic or anaerobic bacteria and released as CO2 and CH4 (among other compounds) to the water column, (ii) is re-suspended, or (iii) remains buried. The relative importance of each of these processes is system-specific and influences the overall role of a reservoir as net sink or source of GHG to the atmosphere.

Mineralization leads to a general chemical-metabolic gradient within undisturbed sediment profiles. Aerobic mineralization is restricted to the top sediment layer, which may have variable thicknesses depending (i) on the hypolimnetic oxygen concentration, (ii) on the oxygen diffusion coefficient, which is a function of sediment density, sediment porosity, temperature and turbulence at the sediment-water interface and (iii) on the rate of oxygen consumption during organic matter mineralization. Below this top oxygenated sediment layer, anaerobic processes dominate sediment metabolism. While CO2 is the only GHG gas produced as a result of aerobic decomposition, anaerobic processes produce CH4 in addition to CO2. These gases accumulate in sediment pore-water and tend to be released in the water column as a consequence of the difference between concentrations in these two compartments.

Diffusive CH4 tends to be oxidized into CO2 by metanotrophic bacteria at the oxygenated top sediment layer or in the water column. The extent of CH4 oxidation in the water column depends on oxygen availability and, thus, may be higher in deeper reservoirs (Lima, 2005).

Greenhouse Gas Emissions from Hydroelectric

before reaching the surface in deep reservoirs.

of the dam (Guerin et al., 2006).

**atmosphere** 

**5.1 Wind** 

hydroelectric reservoirs.

Reservoirs: What Knowledge Do We Have and What is Lacking? 67

from 2 to 8 millimetres (Delsontro et al., 2010). The bubbles are usually formed in the sediment of reservoirs, under anoxic conditions. Most of the CH4 emissions in shallow reservoirs occur through bubbling whereas CH4 bubbles are usually dissolved in the water

The process of energy generation by the turbines of hydroelectric reservoirs leads to two pathways of GHG emission that do not occur in artificial reservoirs built for other purposes (e.g. irrigation, water supply, flood control, and aquaculture): turbulent degassing of water passing through turbines (energy generation unities) and degassing downstream of dams. The water inlet to generate energy is frequently located in medium or lower parts of the dam which means that water from deep layers of the reservoirs passes through the turbines. These deep water layers are usually CO2 and CH4-rich due to both high mineralization rates and high water pressure (i.e. high gas solubility). By passing through the turbines, the gases are exposed to a low pressure and high temperature condition which favors rapid emissions to the atmosphere (Kemenes et al., 2007). Despite the usually high GHG emissions at the turbines, high amounts of both CO2 and CH4 remain dissolved in the water downstream the dams. GHG produced in reservoirs may be encountered at sites as far as 40 km downstream

**5. External factors affecting the processes driving GHG emissions to the** 

The great variation in GHG emissions from hydroelectric reservoirs in different parts of the globe is explained by the effect of climate and drainage basin characteristics on GHG dynamics. Climate affects the inputs of carbon from land to the aquatic environment and the rates of GHG production and consumption. Moreover, wind, precipitation and temperature may have important effects on the rates of gas exchange in the water-atmosphere interface. This section brings an overview of some important processes and external factors influencing GHG emissions from freshwater systems with focus on the dynamics of

Wind is the major source of energy for water movement in lakes and reservoir. The wind stress at the water surface can results in small turbulence, horizontal currents, vertical currents, surface waves and internal waves (seiches). Where and how those processes will emerge and behave depends on a series of factors such as the size and the shape of water surface, fetch, wind direction, temperature, and surrounding terrain. The response of reservoirs to wind stress depends mainly on the morphometric characteristics of the basin and its location. Reservoirs are typically built in valleys, after confluence of several tributaries. It usually results in long and dendritic water bodies (Ford, 1990), contrasting to the generally circular- or bowl-shaped natural lakes. Therefore, depending on the surrounding terrain, wind can blow aligned with the main-long fetch, increasing its effect over water surfaces. This effect, however, is limited by the dendritic geometry or reservoir,

Wind effect strongly impacts on GHG emissions. Turbulence, for example, influences the velocity of gas transfer at the water-atmosphere interface (Wanninkhof, 1992). The role of

resulting in more complex circulation patterns in certain parts of reservoirs.

It has been shown, e.g., that oxidation in the water column of Petit Saut reservoir can reduce CH4 emissions at more than 85% (Guerin & Abril, 2007). However, in very organic-rich sediments, the high rates of mineralization combined with the hydrophobic characteristic of CH4 molecule, can cause CH4 to form bubbles which tend to be quickly emitted to the atmosphere (Bastviken et al., 2004; McGinnis et al., 2006). Even though the formation of CH4 bubbles is favoured by high temperatures, temperate reservoirs may also emit significant amounts of CH4 through bubbling (e.g. Delsontro et al., 2010).

Both aerobic and anaerobic mineralization of organic matter release large amounts of inorganic nutrients, in addition to CO2 and CH4, in the water column of reservoirs. The availability of inorganic nutrients, especially phosphorus, combined to the usually high light penetration depth in reservoirs (due to sedimentation of suspended material) favours CO2 uptake by photosynthesis. Photosynthesis in reservoirs, as in any other aquatic system, is performed by phytoplankton in the water column, attached algae (periphyton) and aquatic plants (macrophytes). However, due to the usually high depth (and high 'total volume/littoral volume' ratio) of reservoirs, the relative importance of periphyton and macrophyte tends to be low compared to the importance of phytoplankton. In eutrophic reservoirs (i.e. high nutrient concentrations and high phytoplankton production) photosynthesis can cause a strong reduction in CO2 concentrations turning the reservoirs into a sink of atmospheric CO2. This occurs in reservoirs worldwide, regardless of age or latitude (Barros et al., 2011). The number of reservoirs acting as carbon sinks is, however, low as compared to CO2 emitting reservoirs.

#### **4.2 GHG flux through the water-atmosphere interface**

The excess GHG in the water column of reservoirs tend to be emitted to the atmosphere. Emissions can occur through (i) molecular diffusion at the water-air interface (e.g. Roland et al., 2010; Teodoru et al., 2010), (ii) bubbling from the sediment (e.g. Abe et al., 2005; Lima, 2005) (iii) degassing from water passing at the turbines (e.g. Kemenes et al., 2007, 2011) or (iv) turbulent degassing in downstream rivers (e.g. Guerin et al., 2006) (**Figure 4**).

Diffusive fluxes of CO2 and CH4 at the water-atmosphere interface are dependent on the existence of a concentration gradient between these two compartments. If the water at the surface of a reservoir is supersaturated with CO2 or CH4 in relation to the atmosphere, gas fluxes occur towards the atmosphere. If, on the other side, surface water is under-saturated in relation to the atmosphere, gas fluxes are from the atmosphere to the water. In the latter case, the reservoir surface represents a sink of atmospheric carbon. The amount of GHG flowing through water-atmosphere interfaces depends on gas solubility in water. Gas solubility is negatively related to temperature and positively related to pressure, according to Le Chatalier's principle. Thus, GHG emissions through diffusion tend to be higher in reservoirs located in warmer regions and at lower altitudes. The surfaces of reservoirs are usually dominated by diffusive fluxes of CO2, even in cases in which bottom anoxia leads to high CH4 production (Barros et al., 2011). This is due to the intense oxidation of diffusive CH4 by methanotrophic bacteria above the interface between anoxic and oxygenated water layers.

On the other hand, GHG emissions through bubbling are dominated by CH4. This is due to the very low CH4 solubility in water, which permits the formation of bubbles varying in size from 2 to 8 millimetres (Delsontro et al., 2010). The bubbles are usually formed in the sediment of reservoirs, under anoxic conditions. Most of the CH4 emissions in shallow reservoirs occur through bubbling whereas CH4 bubbles are usually dissolved in the water before reaching the surface in deep reservoirs.

The process of energy generation by the turbines of hydroelectric reservoirs leads to two pathways of GHG emission that do not occur in artificial reservoirs built for other purposes (e.g. irrigation, water supply, flood control, and aquaculture): turbulent degassing of water passing through turbines (energy generation unities) and degassing downstream of dams. The water inlet to generate energy is frequently located in medium or lower parts of the dam which means that water from deep layers of the reservoirs passes through the turbines. These deep water layers are usually CO2 and CH4-rich due to both high mineralization rates and high water pressure (i.e. high gas solubility). By passing through the turbines, the gases are exposed to a low pressure and high temperature condition which favors rapid emissions to the atmosphere (Kemenes et al., 2007). Despite the usually high GHG emissions at the turbines, high amounts of both CO2 and CH4 remain dissolved in the water downstream the dams. GHG produced in reservoirs may be encountered at sites as far as 40 km downstream of the dam (Guerin et al., 2006).

## **5. External factors affecting the processes driving GHG emissions to the atmosphere**

The great variation in GHG emissions from hydroelectric reservoirs in different parts of the globe is explained by the effect of climate and drainage basin characteristics on GHG dynamics. Climate affects the inputs of carbon from land to the aquatic environment and the rates of GHG production and consumption. Moreover, wind, precipitation and temperature may have important effects on the rates of gas exchange in the water-atmosphere interface. This section brings an overview of some important processes and external factors influencing GHG emissions from freshwater systems with focus on the dynamics of hydroelectric reservoirs.

## **5.1 Wind**

66 Greenhouse Gases – Emission, Measurement and Management

It has been shown, e.g., that oxidation in the water column of Petit Saut reservoir can reduce CH4 emissions at more than 85% (Guerin & Abril, 2007). However, in very organic-rich sediments, the high rates of mineralization combined with the hydrophobic characteristic of CH4 molecule, can cause CH4 to form bubbles which tend to be quickly emitted to the atmosphere (Bastviken et al., 2004; McGinnis et al., 2006). Even though the formation of CH4 bubbles is favoured by high temperatures, temperate reservoirs may also emit significant

Both aerobic and anaerobic mineralization of organic matter release large amounts of inorganic nutrients, in addition to CO2 and CH4, in the water column of reservoirs. The availability of inorganic nutrients, especially phosphorus, combined to the usually high light penetration depth in reservoirs (due to sedimentation of suspended material) favours CO2 uptake by photosynthesis. Photosynthesis in reservoirs, as in any other aquatic system, is performed by phytoplankton in the water column, attached algae (periphyton) and aquatic plants (macrophytes). However, due to the usually high depth (and high 'total volume/littoral volume' ratio) of reservoirs, the relative importance of periphyton and macrophyte tends to be low compared to the importance of phytoplankton. In eutrophic reservoirs (i.e. high nutrient concentrations and high phytoplankton production) photosynthesis can cause a strong reduction in CO2 concentrations turning the reservoirs into a sink of atmospheric CO2. This occurs in reservoirs worldwide, regardless of age or latitude (Barros et al., 2011). The number of reservoirs acting as carbon sinks is, however,

The excess GHG in the water column of reservoirs tend to be emitted to the atmosphere. Emissions can occur through (i) molecular diffusion at the water-air interface (e.g. Roland et al., 2010; Teodoru et al., 2010), (ii) bubbling from the sediment (e.g. Abe et al., 2005; Lima, 2005) (iii) degassing from water passing at the turbines (e.g. Kemenes et al., 2007, 2011) or

Diffusive fluxes of CO2 and CH4 at the water-atmosphere interface are dependent on the existence of a concentration gradient between these two compartments. If the water at the surface of a reservoir is supersaturated with CO2 or CH4 in relation to the atmosphere, gas fluxes occur towards the atmosphere. If, on the other side, surface water is under-saturated in relation to the atmosphere, gas fluxes are from the atmosphere to the water. In the latter case, the reservoir surface represents a sink of atmospheric carbon. The amount of GHG flowing through water-atmosphere interfaces depends on gas solubility in water. Gas solubility is negatively related to temperature and positively related to pressure, according to Le Chatalier's principle. Thus, GHG emissions through diffusion tend to be higher in reservoirs located in warmer regions and at lower altitudes. The surfaces of reservoirs are usually dominated by diffusive fluxes of CO2, even in cases in which bottom anoxia leads to high CH4 production (Barros et al., 2011). This is due to the intense oxidation of diffusive CH4 by methanotrophic bacteria above the interface between anoxic and oxygenated water

On the other hand, GHG emissions through bubbling are dominated by CH4. This is due to the very low CH4 solubility in water, which permits the formation of bubbles varying in size

(iv) turbulent degassing in downstream rivers (e.g. Guerin et al., 2006) (**Figure 4**).

amounts of CH4 through bubbling (e.g. Delsontro et al., 2010).

low as compared to CO2 emitting reservoirs.

layers.

**4.2 GHG flux through the water-atmosphere interface** 

Wind is the major source of energy for water movement in lakes and reservoir. The wind stress at the water surface can results in small turbulence, horizontal currents, vertical currents, surface waves and internal waves (seiches). Where and how those processes will emerge and behave depends on a series of factors such as the size and the shape of water surface, fetch, wind direction, temperature, and surrounding terrain. The response of reservoirs to wind stress depends mainly on the morphometric characteristics of the basin and its location. Reservoirs are typically built in valleys, after confluence of several tributaries. It usually results in long and dendritic water bodies (Ford, 1990), contrasting to the generally circular- or bowl-shaped natural lakes. Therefore, depending on the surrounding terrain, wind can blow aligned with the main-long fetch, increasing its effect over water surfaces. This effect, however, is limited by the dendritic geometry or reservoir, resulting in more complex circulation patterns in certain parts of reservoirs.

Wind effect strongly impacts on GHG emissions. Turbulence, for example, influences the velocity of gas transfer at the water-atmosphere interface (Wanninkhof, 1992). The role of

Greenhouse Gas Emissions from Hydroelectric

**5.4 Drainage basin and river input** 

regions.

account.

**6. Future perspectives** 

the sediment of warmer systems (Gudasz et al., 2010).

Reservoirs: What Knowledge Do We Have and What is Lacking? 69

(Sand-Jensen et al., 2007), primary production in the water column (Flanagan et al., 2003) are strongly influenced by temperature. Although both processes tend increase with warming, it has been argued that respiration has usually a stronger response (Rivkin & Legendre, 2001; Biddanda & Cotner, 2002; Lopez-Urrutia et al., 2006; Sand-Jensen et al., 2007), leading to a net increase in CO2 production under warmer conditions (Kosten et al 2010). Moreover, evidences from lakes located in different climate regions have shown that more mineralization occurs (with further CO2 and CH4 production) and less carbon is buried in

Indeed, GHG emissions are higher in reservoirs located in warmer latitudes (Barros et al., 2011). The positive relationship between GHG emissions and temperature implies that environmental changes causing temperature to vary affects the role of hydroelectric reservoirs as GHG sources. Based on the predicted scenarios of global change in temperature (IPCC, 2007) GHG emissions from reservoir will increase, especially in cold

Large impoundments may show different zones in terms of CO2 emission because those fluxes are dependent on flooded biomass and watershed input of organic matter. Compared to natural lakes, reservoirs tend to have shorter water residence times and more complex heterogeneity due to the presence of one or more major water inlets, instead of multiple diffuse water sources characteristic for most natural lakes (Kennedy et al., 1985; Kennedy & Walker, 1990). Reservoirs are intrinsically linked to the rivers that feed them (Baxter, 1977), creating a river–reservoir continuum, in which water and sediment inputs are functions of the land use in the watershed (Kelly, 2001). Watershed land use is often highly correlated with algal cell and nutrients concentrations as has been shown for seven subtropical reservoirs (Burford et al., 2007). This clearly indicates that in order to understand and predict GHG emissions from reservoirs, features of the drainage basin have to be taken into

Another important factor to be considered is the difference between the densities of river and reservoir water. When a river reaches a reservoir the water plunges and can flow along the surface (overflow), intermediate (interflow) or deep (underflow) layer, depending on the difference between the water temperature and physical characteristics (e.g. total dissolved solids and suspended solids) (Martin & McCutcheon, 1999). The fate of organic matter influx largely depends on the way the incoming water flows into the reservoirs. If incorporated into deeper layers, anoxic conditions facilitate degradation of organic matter into CH4. The picture becomes more complicates when, due to the complexity of the system and hydrodynamic factors of the river entrance, waves arise in the interface between river/reservoir water. These waves cause transport of water from the nutrient-rich and

It is now clear that hydroelectric reservoirs represent a renewable but not a carbon-free source of electricity. Attempts to determine the carbon footprint of hydropower production have lead to significant scientific advances (see section 2 for details). However, it is still a

GHG-rich deep layer to the surface layer (Assireu et al., 2011).

wind on GHG emissions, however, is not restricted to causing surface turbulence. Wind can also cause deep water circulation and transport of dissolved compounds.

The overall picture that emerges is that reservoirs are highly temperature stratified in summer and only weakly forced by wind, while in spring/winter the mixing process is enhanced by high wind velocity. During high velocity wind events surface water moves toward the shore, where the water piles up and sinks in the process known as downwelling. At the same time at the opposite side, surface water is replaced by water that wells up from below (upwelling) (Monismith, 1985; Stevens & Imberger, 1996; Macintyre et al., 2002; Farrow & Stevens, 2003). However, wind direction and persistency are of great importance as well. Persistent wind aligned with the main reservoir body – resulting in a large fetch – is able to promote downwelling/upwelling (d/u) even at low velocities (Assireu et al., 2011). At the downwelling side of a reservoir, CH4 oxidation plays an important role, changing CH4 into CO2 during the downward dissolved oxygen transport and reducing the CH4 emission to the atmosphere. At the upwelling side of reservoir, the GHG flux to the atmosphere is enhanced due to the high CO2 and CH4 concentrations of emerging water. This clearly influences the spatial variability of GHG emissions within reservoirs (Roland et al., 2010). It was suggested that frequent d/u events in the tropical Manso reservoir could enhance CO2 emissions with approximately 12% (Assireu et al., 2005; Ometto et al., 2011).

### **5.2 Precipitation**

Precipitation can vary markedly in intensity and frequency from one year to another. Such variability can generate different kinds of seasonal patterns, modifying water turnover time and changing the intensity of environmental processes occurring in the water column of aquatic systems (Armengol et al. 1999). Furthermore, the external loading of organic matter and other compounds to the system often increases with intensive precipitation events. The intensity of these loadings depends on land use, vegetation cover and landscape patchiness (Rybak, 2001). Rain-induced high primary productivity has been observed in some African lakes (Lemoalle, 1975; Melack, 1979; Thomas et al., 2000). In tropical regions, the first rains after the start of the rainy season are expected to a lot of carbon into the reservoir that can be assimilated by bacterioplankton or be buried in the sediment. However, if the residence time of the water is very short, with high ushing rate, most of this carbon and nutrients will not remain long in the system (Amarasinghe & Vijverberg, 2002).

Precipitation also affects CO2 transfer at the air-water interface. Impinging raindrops causes turbulence, enhancing CO2 flux across the air-water interface (Takagaki & Komori, 2007). This means that, depending on the difference of gas concentrations between water and atmosphere, the emission/absorption of GHG in aquatic systems can be enhanced by increasing precipitation rates. The rainfall effects can be significant to the local CO2 budget between water and atmosphere during the rainy season but are not as important to the whole system budget as the wind share effect.

#### **5.3 Temperature**

Changes in temperature may affect, directly and indirectly, all processes involved in the production, consumption and emission of GHG in reservoirs. For example, respiration (Sand-Jensen et al., 2007), primary production in the water column (Flanagan et al., 2003) are strongly influenced by temperature. Although both processes tend increase with warming, it has been argued that respiration has usually a stronger response (Rivkin & Legendre, 2001; Biddanda & Cotner, 2002; Lopez-Urrutia et al., 2006; Sand-Jensen et al., 2007), leading to a net increase in CO2 production under warmer conditions (Kosten et al 2010). Moreover, evidences from lakes located in different climate regions have shown that more mineralization occurs (with further CO2 and CH4 production) and less carbon is buried in the sediment of warmer systems (Gudasz et al., 2010).

Indeed, GHG emissions are higher in reservoirs located in warmer latitudes (Barros et al., 2011). The positive relationship between GHG emissions and temperature implies that environmental changes causing temperature to vary affects the role of hydroelectric reservoirs as GHG sources. Based on the predicted scenarios of global change in temperature (IPCC, 2007) GHG emissions from reservoir will increase, especially in cold regions.

## **5.4 Drainage basin and river input**

68 Greenhouse Gases – Emission, Measurement and Management

wind on GHG emissions, however, is not restricted to causing surface turbulence. Wind can

The overall picture that emerges is that reservoirs are highly temperature stratified in summer and only weakly forced by wind, while in spring/winter the mixing process is enhanced by high wind velocity. During high velocity wind events surface water moves toward the shore, where the water piles up and sinks in the process known as downwelling. At the same time at the opposite side, surface water is replaced by water that wells up from below (upwelling) (Monismith, 1985; Stevens & Imberger, 1996; Macintyre et al., 2002; Farrow & Stevens, 2003). However, wind direction and persistency are of great importance as well. Persistent wind aligned with the main reservoir body – resulting in a large fetch – is able to promote downwelling/upwelling (d/u) even at low velocities (Assireu et al., 2011). At the downwelling side of a reservoir, CH4 oxidation plays an important role, changing CH4 into CO2 during the downward dissolved oxygen transport and reducing the CH4 emission to the atmosphere. At the upwelling side of reservoir, the GHG flux to the atmosphere is enhanced due to the high CO2 and CH4 concentrations of emerging water. This clearly influences the spatial variability of GHG emissions within reservoirs (Roland et al., 2010). It was suggested that frequent d/u events in the tropical Manso reservoir could enhance CO2 emissions with approximately

Precipitation can vary markedly in intensity and frequency from one year to another. Such variability can generate different kinds of seasonal patterns, modifying water turnover time and changing the intensity of environmental processes occurring in the water column of aquatic systems (Armengol et al. 1999). Furthermore, the external loading of organic matter and other compounds to the system often increases with intensive precipitation events. The intensity of these loadings depends on land use, vegetation cover and landscape patchiness (Rybak, 2001). Rain-induced high primary productivity has been observed in some African lakes (Lemoalle, 1975; Melack, 1979; Thomas et al., 2000). In tropical regions, the first rains after the start of the rainy season are expected to a lot of carbon into the reservoir that can be assimilated by bacterioplankton or be buried in the sediment. However, if the residence time of the water is very short, with high ushing rate, most of this carbon and nutrients will not

Precipitation also affects CO2 transfer at the air-water interface. Impinging raindrops causes turbulence, enhancing CO2 flux across the air-water interface (Takagaki & Komori, 2007). This means that, depending on the difference of gas concentrations between water and atmosphere, the emission/absorption of GHG in aquatic systems can be enhanced by increasing precipitation rates. The rainfall effects can be significant to the local CO2 budget between water and atmosphere during the rainy season but are not as important to the

Changes in temperature may affect, directly and indirectly, all processes involved in the production, consumption and emission of GHG in reservoirs. For example, respiration

also cause deep water circulation and transport of dissolved compounds.

12% (Assireu et al., 2005; Ometto et al., 2011).

remain long in the system (Amarasinghe & Vijverberg, 2002).

whole system budget as the wind share effect.

**5.2 Precipitation** 

**5.3 Temperature** 

Large impoundments may show different zones in terms of CO2 emission because those fluxes are dependent on flooded biomass and watershed input of organic matter. Compared to natural lakes, reservoirs tend to have shorter water residence times and more complex heterogeneity due to the presence of one or more major water inlets, instead of multiple diffuse water sources characteristic for most natural lakes (Kennedy et al., 1985; Kennedy & Walker, 1990). Reservoirs are intrinsically linked to the rivers that feed them (Baxter, 1977), creating a river–reservoir continuum, in which water and sediment inputs are functions of the land use in the watershed (Kelly, 2001). Watershed land use is often highly correlated with algal cell and nutrients concentrations as has been shown for seven subtropical reservoirs (Burford et al., 2007). This clearly indicates that in order to understand and predict GHG emissions from reservoirs, features of the drainage basin have to be taken into account.

Another important factor to be considered is the difference between the densities of river and reservoir water. When a river reaches a reservoir the water plunges and can flow along the surface (overflow), intermediate (interflow) or deep (underflow) layer, depending on the difference between the water temperature and physical characteristics (e.g. total dissolved solids and suspended solids) (Martin & McCutcheon, 1999). The fate of organic matter influx largely depends on the way the incoming water flows into the reservoirs. If incorporated into deeper layers, anoxic conditions facilitate degradation of organic matter into CH4. The picture becomes more complicates when, due to the complexity of the system and hydrodynamic factors of the river entrance, waves arise in the interface between river/reservoir water. These waves cause transport of water from the nutrient-rich and GHG-rich deep layer to the surface layer (Assireu et al., 2011).

## **6. Future perspectives**

It is now clear that hydroelectric reservoirs represent a renewable but not a carbon-free source of electricity. Attempts to determine the carbon footprint of hydropower production have lead to significant scientific advances (see section 2 for details). However, it is still a

Greenhouse Gas Emissions from Hydroelectric

reservoirs' role in the global carbon cycle.

for energy and environmental wellness in the future.

**7. References** 

(Roland et al., 2010).

Reservoirs: What Knowledge Do We Have and What is Lacking? 71

of GHG emissions from reservoirs should, thus, seek for a better understanding of the patterns of variation on GHG dynamics with time. Both long-term measurements (including several

Overcoming the issues related to variability of GHG emission in space and time becomes even more challenging due to extreme events (e.g. upwelling). Such events which are very hard to 'catch' may account to a major portion of the GHG emissions from a reservoir

Most of the GHG assessments in hydroelectric reservoirs have looked at emissions from decaying biomass after impoundment, as related to the previous emissions (see section 2 for details). Some studies have used a full life-cycle assessment, which considers the emissions during the construction phase of dams (e.g. Chamberland & Levesque, 1996; Gagnon & Van de Vate, 1997; Pacca & Horvath, 2002). However, little attention has been given to the fact that every hydropower plant has an intrinsic operation life-time after which its dam should be deactivated (Pacca, 2007). Moreover, dams may be deactivated and removed due to failures (e.g. error during construction, geological instability, poor maintenance, excess internal erosion etc.). It has been estimated that about 600 dams were removed in the last 100 years in the USA alone (Gleick et al., 2009). The removal of dams was the subject of a special issue published by

With dam removal, part of the sediment accumulated at the reservoirs will be flushed down river if not otherwise controlled (e.g. dredged). Sediment fate is an important factor determining the amount of carbon returning to the atmosphere – a greater fraction of dredged sediment becomes available for mineralization whereas organic carbon flushing downstream rivers may ultimately accumulate the ocean (Pacca, 2007). Despite the many uncertainties regarding the effects of dam removal, it is now clear that this issue should not be disregarded by future assessments. Answering questions as "how permanent is carbon accumulation in the sediment of reservoirs?" is mandatory to a better understanding of

With the continuous and integrated research focusing on revealing the actual role of hydroelectric reservoirs in the global carbon cycle, the gaps in knowledge tend to be reduced at a rapid pace. Recent advances have resulted in important tools for reducing GHG emissions from reservoirs through changes in reservoir management and, even more importantly, through the GHG-intelligent design and location of new hydropower plants. It is now known, for example, that future hydropower projects should seek for better GHG emission/MWh ratios by reducing flooded area to the maximum extent, removing terrestrial vegetation prior to impounding and preserving the vegetation in the drainage basin to avoid soil erosion. The research about hydroelectric projects has been promoting a new generation of scientist that are able to apply fundamental environment sciences – ecology, biology, chemistry, physics and mathematics – to issues representing global concerns. We believe that the scientific community is strongly prepared to discuss the needs

Abe, D.S.; Donald D. Adams; Galli, C.V.S.; Sikar, E. & Tundisi, J.G. (2005). Sediment

greenhouse gases (methane and carbon dioxide) in the Lobo-Broa Reservoir, São

years) and intensive measurements (in a more frequent basis, e.g. hours) are needed.

Bioscience in 2002 which did not mention the effects on GHG emission.

long way towards a full assessment of the role of hydroelectric reservoirs as sources of GHG to the atmosphere. The major gaps to be filled by future research are related to the quality of current carbon budget estimations. However, other issues need attention, e.g. the outcomes of removal of dams and the development of mitigation strategies.

The assessment of the net role of reservoirs as sinks or sources of GHG to the atmosphere requires more than measurements of carbon fluxes at the water-atmosphere interface of reservoirs. For example, the net change in the carbon cycle due to reservoir construction should consider pre-impoundment carbon sinks and sources from the original river course and the adjacent terrestrial landscape (Teodoru et al., 2010). Scenarios of pre- and postimpoundment fluxes, however, are neglected by most estimates of emissions. Even more importantly, there is a need to evaluate the net effect of reservoir as sink of atmospheric carbon (St Louis et al., 2000; Barros et al., 2011), by comparing carbon burial in reservoirs with carbon burial prior to impoundment (i.e. in the ocean).

The role of reservoirs sediment in the global carbon cycle is still poorly understood. Despite the importance of sediment as the main site for organic matter mineralization and GHG production (especially CH4) sediment carbon fluxes in reservoirs are restricted to few measurements (e.g. Aberg et al., 2004; Abe et al., 2005). Additionally, reservoirs accumulate large amounts of carbon in the sediment, which compensates at least part of the emissions. It has been estimated that approximately 400 gC m-2 y1 is buried in reservoirs (Dean & Gorham, 1998; Mulholland et al., 2001). This estimation, however, is based on nonstandardized methods which consider, e.g. loss in water storage capacity of reservoirs in the USA and constant values for carbon content and sediment density. There is an urgent need, thus, for research focusing on this potential carbon sink pathway.

Another important source of errors in carbon budgets relies on neglecting variability. Both the neglected variability among and within reservoirs as well as variability over time impacts the precision of carbon flux estimates. The persistent increase in the amount of data, by itself, may contribute to equal out the errors of local and global estimations. Nevertheless, addressing all possible sources of variability is an important task for future research on GHG emissions from reservoirs.

Regarding the variability among reservoirs, the limited amount of data from reservoirs located in the tropics was, until recently, severely limiting the precision of GHG emission estimations from reservoirs (e.g. St Louis et al., 2000). Within the last decade, many tropical reservoirs have become focus of study, permitting a better evaluation of the latitudinal effect on emissions (Barros et al., 2011). Nevertheless, there is still a great need for research in the tropics, since this is where most of the potential for new dams remains (Figure 2). Future research on tropical reservoirs should seek for better estimations of CH4 emissions, which seems to have been seriously underestimated in the last two decades (Demarty & Bastien, 2011; Wehrli, 2011) in spite of its greater warming potential (more than 20 times more) when compared to CO2.

Neglecting variability in GHG emissions over time may equally cause great errors in estimations. Especially because GHG emissions reduce as reservoirs age (Barros et al., 2011). Furthermore, fluctuations in carbon cycling may occur on a seasonal (Wang et al., 2011) and even on a daily basis (Roland et al., 2010). The changes in GHG emission rates with time makes the full assessment of the carbon budget in a reservoir is a difficult task. Future studies

long way towards a full assessment of the role of hydroelectric reservoirs as sources of GHG to the atmosphere. The major gaps to be filled by future research are related to the quality of current carbon budget estimations. However, other issues need attention, e.g. the outcomes

The assessment of the net role of reservoirs as sinks or sources of GHG to the atmosphere requires more than measurements of carbon fluxes at the water-atmosphere interface of reservoirs. For example, the net change in the carbon cycle due to reservoir construction should consider pre-impoundment carbon sinks and sources from the original river course and the adjacent terrestrial landscape (Teodoru et al., 2010). Scenarios of pre- and postimpoundment fluxes, however, are neglected by most estimates of emissions. Even more importantly, there is a need to evaluate the net effect of reservoir as sink of atmospheric carbon (St Louis et al., 2000; Barros et al., 2011), by comparing carbon burial in reservoirs

The role of reservoirs sediment in the global carbon cycle is still poorly understood. Despite the importance of sediment as the main site for organic matter mineralization and GHG production (especially CH4) sediment carbon fluxes in reservoirs are restricted to few measurements (e.g. Aberg et al., 2004; Abe et al., 2005). Additionally, reservoirs accumulate large amounts of carbon in the sediment, which compensates at least part of the emissions. It has been estimated that approximately 400 gC m-2 y1 is buried in reservoirs (Dean & Gorham, 1998; Mulholland et al., 2001). This estimation, however, is based on nonstandardized methods which consider, e.g. loss in water storage capacity of reservoirs in the USA and constant values for carbon content and sediment density. There is an urgent need,

Another important source of errors in carbon budgets relies on neglecting variability. Both the neglected variability among and within reservoirs as well as variability over time impacts the precision of carbon flux estimates. The persistent increase in the amount of data, by itself, may contribute to equal out the errors of local and global estimations. Nevertheless, addressing all possible sources of variability is an important task for future

Regarding the variability among reservoirs, the limited amount of data from reservoirs located in the tropics was, until recently, severely limiting the precision of GHG emission estimations from reservoirs (e.g. St Louis et al., 2000). Within the last decade, many tropical reservoirs have become focus of study, permitting a better evaluation of the latitudinal effect on emissions (Barros et al., 2011). Nevertheless, there is still a great need for research in the tropics, since this is where most of the potential for new dams remains (Figure 2). Future research on tropical reservoirs should seek for better estimations of CH4 emissions, which seems to have been seriously underestimated in the last two decades (Demarty & Bastien, 2011; Wehrli, 2011) in spite of its greater warming potential (more than 20 times more) when

Neglecting variability in GHG emissions over time may equally cause great errors in estimations. Especially because GHG emissions reduce as reservoirs age (Barros et al., 2011). Furthermore, fluctuations in carbon cycling may occur on a seasonal (Wang et al., 2011) and even on a daily basis (Roland et al., 2010). The changes in GHG emission rates with time makes the full assessment of the carbon budget in a reservoir is a difficult task. Future studies

of removal of dams and the development of mitigation strategies.

with carbon burial prior to impoundment (i.e. in the ocean).

thus, for research focusing on this potential carbon sink pathway.

research on GHG emissions from reservoirs.

compared to CO2.

of GHG emissions from reservoirs should, thus, seek for a better understanding of the patterns of variation on GHG dynamics with time. Both long-term measurements (including several years) and intensive measurements (in a more frequent basis, e.g. hours) are needed.

Overcoming the issues related to variability of GHG emission in space and time becomes even more challenging due to extreme events (e.g. upwelling). Such events which are very hard to 'catch' may account to a major portion of the GHG emissions from a reservoir (Roland et al., 2010).

Most of the GHG assessments in hydroelectric reservoirs have looked at emissions from decaying biomass after impoundment, as related to the previous emissions (see section 2 for details). Some studies have used a full life-cycle assessment, which considers the emissions during the construction phase of dams (e.g. Chamberland & Levesque, 1996; Gagnon & Van de Vate, 1997; Pacca & Horvath, 2002). However, little attention has been given to the fact that every hydropower plant has an intrinsic operation life-time after which its dam should be deactivated (Pacca, 2007). Moreover, dams may be deactivated and removed due to failures (e.g. error during construction, geological instability, poor maintenance, excess internal erosion etc.). It has been estimated that about 600 dams were removed in the last 100 years in the USA alone (Gleick et al., 2009). The removal of dams was the subject of a special issue published by Bioscience in 2002 which did not mention the effects on GHG emission.

With dam removal, part of the sediment accumulated at the reservoirs will be flushed down river if not otherwise controlled (e.g. dredged). Sediment fate is an important factor determining the amount of carbon returning to the atmosphere – a greater fraction of dredged sediment becomes available for mineralization whereas organic carbon flushing downstream rivers may ultimately accumulate the ocean (Pacca, 2007). Despite the many uncertainties regarding the effects of dam removal, it is now clear that this issue should not be disregarded by future assessments. Answering questions as "how permanent is carbon accumulation in the sediment of reservoirs?" is mandatory to a better understanding of reservoirs' role in the global carbon cycle.

With the continuous and integrated research focusing on revealing the actual role of hydroelectric reservoirs in the global carbon cycle, the gaps in knowledge tend to be reduced at a rapid pace. Recent advances have resulted in important tools for reducing GHG emissions from reservoirs through changes in reservoir management and, even more importantly, through the GHG-intelligent design and location of new hydropower plants. It is now known, for example, that future hydropower projects should seek for better GHG emission/MWh ratios by reducing flooded area to the maximum extent, removing terrestrial vegetation prior to impounding and preserving the vegetation in the drainage basin to avoid soil erosion. The research about hydroelectric projects has been promoting a new generation of scientist that are able to apply fundamental environment sciences – ecology, biology, chemistry, physics and mathematics – to issues representing global concerns. We believe that the scientific community is strongly prepared to discuss the needs for energy and environmental wellness in the future.

## **7. References**

Abe, D.S.; Donald D. Adams; Galli, C.V.S.; Sikar, E. & Tundisi, J.G. (2005). Sediment greenhouse gases (methane and carbon dioxide) in the Lobo-Broa Reservoir, São

Greenhouse Gas Emissions from Hydroelectric

431-445, 1432-9840

press. No.0, pp. 0048-9697

8075

Reservoirs: What Knowledge Do We Have and What is Lacking? 73

Biddanda, B.A. & Cotner, J.B. (2002). Love handles in aquatic ecosystems: The role of

Burford, M.A.; Johnson, S.A.; Cook, A.J.; Packer, T.V.; Taylor, B.M. & Townsley, E.R. (2007).

Chanudet, V.; Descloux, S.; Harby, A.; Sundt, H.; Hansen, B.H.; Brakstad, O.; Serça, D. &

Chao, B.F.; Wu, Y.H. & Li, Y.S. (2008). Impact of Artificial Reservoir Water Impoundment on Global Sea Level. *Science*, Vol.320. No.5873, April 11, 2008, pp. 212-214, Chen, H.; Wu, Y.Y.; Yuan, X.Z.; Gao, Y.H.; Wu, N. & Zhu, D. (2009). Methane emissions

Cole, J.J.; Prairie, Y.T.; Caraco, N.F.; Mcdowell, W.H.; Tranvik, L.J.; Striegl, R.G.; Duarte,

Cullenward, D. & Victor, D.G. (2006). The dam debate and its discontents. *Climatic Change*,

Dean, W.E. & Gorham, E. (1998). Magnitude and significance of carbon burial in lakes, reservoirs, and peatlands. *Geology*, Vol.26. No.6, Jun, pp. 535-538, 0091-7613 Delmas, R.; Galy-Lacaux, C. & Richard, S. (2001). Emissions of greenhouse gases from the

Delsontro, T.; McGinnis, D.F.; Sobek, S.; Ostrovsky, I. & Wehrli, B. (2010). Extreme Methane

Demarty, M. & Bastien, J. (2011). GHG emissions from hydroelectric reservoirs in tropical

Diem, T.; Koch, S.; Schwarzenbach, S.; Wehrli, B. & Schubert, C.J. (2007). Greenhouse-gas (CH4, N2O and CO2) emissions from hydroelectric reservoirs in Switzerland. Dos Santos, M.A.; Rosa, L.P.; Sikar, B.; Sikar, E. & Dos Santos, E.O. (2006). Gross greenhouse

Duchemin, E.; Lucotte, M.; Canuel, R. & Chamberland, A. (1995). Production of the

*Journal of Geophysical Research-Atmospheres*, Vol.114. Sep, pp. 0148-0227 Cole, J.J.; Caraco, N.F.; Kling, G.W. & Kratz, T.K. (1994). Carbon-Dioxide Supersaturation in

Vol.37. No.6-8, Jun-Aug, pp. 885-890, 0196-8904

*Ecosystems*, Vol.10. No.1, Feb, pp. 171-184, 1432-9840

*Policy*, Vol.39. No.7, Jul, pp. 4197-4206, 0301-4215

*Policy*, Vol.34. No.4, Mar, pp. 481-488, 0301-4215

*Global Biogeochem. Cycles*, Vol.9. No.4, pp. 529-540,

Vol.75. No.1-2, Mar, pp. 81-86, 0165-0009

pp. 993-1003, 0886-6236

0013-936X

dissolved organic carbon drawdown, resuspended sediments, and terrigenous inputs in the carbon balance of Lake Michigan. *Ecosystems*, Vol.5. No.5, Aug, pp.

Correlations between watershed and reservoir characteristics, and algal blooms in subtropical reservoirs. *Water Research*, Vol.41. No.18, Oct, pp. 4105-4114, 0043-1354 Chamberland, A. & Levesque, S. (1996). Hydroelectricity, an option to reduce greenhouse

gas emissions from thermal power plants. *Energy Conversion and Management*,

Guerin, F. (2011). Gross CO2 and CH4 emissions from the Nam Ngum and Nam Leuk sub-tropical reservoirs in Lao PDR. *Science of The Total Environment*, Vol.in

from newly created marshes in the drawdown area of the Three Gorges Reservoir.

the Surface Waters of Lakes. *Science*, Vol.265. No.5178, Sep 9, pp. 1568-1570, 0036-

C.M.; Kortelainen, P.; Downing, J.A.; Middelburg, J.J. & Melack, J. (2007). Plumbing the global carbon cycle: Integrating inland waters into the terrestrial carbon budget.

tropical hydroelectric reservoir of Petit Saut (French Guiana) compared with emissions from thermal alternatives. *Global Biogeochemical Cycles*, Vol.15. No.4, Dec,

Emissions from a Swiss Hydropower Reservoir: Contribution from Bubbling Sediments. *Environmental Science & Technology*, Vol.44. No.7, Apr 1, pp. 2419-2425,

and equatorial regions: Review of 20 years of CH4 emission measurements. *Energy* 

gas fluxes from hydro-power reservoir compared to thermo-power plants. *Energy* 

greenhouse gases CH4 and CO2 by hydroelectric reservoirs of the boreal region.

Paulo State, Brazil: Concentrations and diffuse emission fluxes for carbon budget considerations. *Lakes & Reservoirs: Research and Management*, Vol.10. No.4, pp. 201- 209, 1440-1770


Aberg, J.; Bergstrom, A.K.; Algesten, G.; Soderback, K. & Jansson, M. (2004). A comparison

Abril, G.; Guerin, F.; Richard, S.; Delmas, R.; Galy-Lacaux, C.; Gosse, P.; Tremblay, A.;

Armengol, J.; Garcia, J.C.; Comerma, M.; Romero, M.; Dolz, J.; Roura, M.; Han, B.H.; Vidal,

Assireu, A.T.; Alcântara, E.; Novo, E.M.L.M.; Roland, F.; Pacheco, F.S.; Stech, J.L. &

Assireu, A.T.; Stech, J.L.; Marinho, M.M.; Cesar, D.E.; Lorenzzentti, J.A.; Ferreira, R.M.;

Bambace, L.A.W.; Ramos, F.M.; Lima, I.B.T. & Rosa, R.R. (2007). Mitigation and recovery of

Barros, N.; Cole, J.J.; Tranvik, L.J.; Prairie, Y.T.; Bastviken, D.; Huszar, V.L.M.; Del Giorgio,

Bastien, J. (2005). Impacts of ultraviolet radiation on aquatic ecosystems: greenhouse gas

Bastviken, D.; Cole, J.; Pace, M. & Tranvik, L. (2004). Methane emissions from lakes:

Bayne, D.R.; Lawrence, J.M. & McGuire, J.A. (1983). Primary productivity studies during

Bergstrom, A.K.; Algesten, G.; Sobek, S.; Tranvik, L. & Jansson, M. (2004). Emission of CO2

estimate. *Global Biogeochemical Cycles*, Vol.18. No.4, Oct 20, 0886-6236 Baxter, R.M. (1977). Environmental Effects of Dams and Impoundments. *Annual Review of* 

*Ecology and Systematics*, Vol.8. pp. 255-283, 0066-4162

Lanka. *Hydrobiologia*, Vol.487. No.1, Nov, pp. 85-93, 0018-8158

Publishers, 585p. ISBN 85-87418-02-5, Barcelona

209, 1440-1770

538, 0043-1354

Vol.8. pp. 1-31

8576560593, São Carlos

1038-1046, 0360-5442

23455-5, New York

No.5, pp. 477-489, 0046-5070

No.1, Jan, pp. 25-42, 0003-9136

0894

Paulo State, Brazil: Concentrations and diffuse emission fluxes for carbon budget considerations. *Lakes & Reservoirs: Research and Management*, Vol.10. No.4, pp. 201-

of the carbon balances of a natural lake (L. Ortrasket) and a hydroelectric reservoir (L. Skinnmuddselet) in northern Sweden. *Water Research*, Vol.38. No.3, Feb, pp. 531-

Varfalvy, L.; Dos Santos, M.A. & Matvienko, B. (2005). Carbon dioxide and methane emissions and the carbon budget of a 10-year old tropical reservoir (Petit Saut, French Guiana). *Global Biogeochemical Cycles*, Vol.19. No.4, Oct, 0886-6236 Amarasinghe, P.B. & Vijverberg, J. (2002). Primary production in a tropical reservoir in Sri

A. & Simek, K. (1999). Longitudinal processes in canyon type reservoir: the case of Sau (N.E. SPAIN), pp. 313-345. In: *Theoretical reservoir ecology and its applications*, Tundisi, J.G. & Straskraba, M. (eds.), Brazilian Academy of Sciences and Backhuys

Lorenzzetti, J.A. (2011). Inside the hydro-physics processes at the plunge point location: an analysis by satellite and in situ data. *Hydrol. Earth Syst. Sci. Discuss.*,

Pacheco, F.S. & Roland, F. (2005). Princípios Físicos e Químicos a Serviço da Limnologia - Um Exercício. In: *Liçoes de Limnologia.*, Rima Editora, ISBN:

methane emissions from tropical hydroelectric dams. *Energy*, Vol.32. No.6, Jun, pp.

P. & Roland, F. (2011). Carbon emission from hydroelectric reservoirs linked to reservoir age and latitude. *Nature Geoscience*, Vol.4. No.9, Sep, pp. 593-596, 1752-

emissions and implications for hydroeletric reservoirs. In: *Greenhouse Gas Emissions - Fluxes and Process. Hydroelectric Reservoirs and Natural Environments*, Tremblay, A., Varfalvy, L., Roehm, C. & Garneau, M., 509-527, Springer-Verlag, ISBN 978-3-540-

Dependence of lake characteristics, two regional assessments, and a global

early years of West-Point reservoir, Alabama Georgia. *Freshwater Biology*, Vol.13.

from hydroelectric reservoirs in northern Sweden. *Archiv Fur Hydrobiologie*, Vol.159.


Greenhouse Gas Emissions from Hydroelectric

No.6705, Oct, pp. 881-884, 0028-0836

Vol.16. No.1, Mar, pp. -, 0886-6236

http://go.nature.com/6mAAWK

*Biogeosciences*, Vol.116. Jul, pp. 0148-0227

8276

4203

0148-0227

332,

8276

Dordrecht

New York

Reservoirs: What Knowledge Do We Have and What is Lacking? 75

Guerin, F.; Abril, G.; Richard, S.; Burban, B.; Reynouard, C.; Seyler, P. & Delmas, R. (2006).

Guerin, F.; Abril, G.; Serca, D.; Delon, C.; Richard, S.; Delmas, R.; Tremblay, A. & Varfalvy,

Hedges, J.I.; Keil, R.G. & Benner, R. (1997). What happens to terrestrial organic matter in the ocean? *Organic Geochemistry*, Vol.27. No.5-6, Oct-Nov, pp. 195-212, 0146-6380 Hoffert, M.I.; Caldeira, K.; Jain, A.K.; Haites, E.F.; Harvey, L.D.D.; Potter, S.D.; Schlesinger,

Huttunen, J.T.; Vaisanen, T.S.; Hellsten, S.K.; Heikkinen, M.; Nykanen, H.; Jungner, H.;

IEA (2008). International Energy Agency. Electricity/Heat in World in 2008. available via

Kelly, C.; Rudd, J.W.M.; St. Louis, V.L. & Moore, T. (1994). Turning attention to reservoir

Kelly, V.J. (2001). Influence of reservoirs on solute transport: a regional-scale approach.

Kemenes, A.; Forsberg, B.R. & Melack, J.M. (2007). Methane release below a tropical hydroelectric dam. *Geophysical Research Letters*, Vol.34. No.12, Jun, pp. 0094-8276 Kemenes, A.; Forsberg, B.R. & Melack, J.M. (2007 ). Methane release below a tropical

Kemenes, A.; Forsberg, B.R. & Melack, J.M. (2011). CO2 emissions from a tropical

Kennedy, R.H.; Thorton, K.W. & D.E., F. (1985). Characterization of the reservoir ecosystem.

Kennedy, R.H. & Walker, W.W. (1990). Reservoir nutrient dynamics. In: *Reservoir Limnology:* 

*Hydrological Processes*, Vol.15. No.7, May, pp. 1227-1249, 0885-6087

IPCC (2007). Intergovernmental Panel on Climate Change's Fourth Assessment Report Keller, M. & Stallard, R.F. (1994). Methane Emission by Bubbling from Gatun Lake, Panama.

Methane and carbon dioxide emissions from tropical reservoirs: Significance of downstream rivers. *Geophysical Research Letters*, Vol.33. No.21, Nov, pp. 0094-8276 Guerin, F.; Abril, G.; Richard, S.; Burban, B.; Reynouard, C.; Seyler, P. & Delmas, R. (2006 ).

Methane and carbon dioxide emissions from tropical reservoirs: Significance of downstream rivers. *Geophysical Research Letters*, Vol.33: L21407. Nov 14, pp. 0094-

L. (2007). Gas transfer velocities of CO2 and CH4 in a tropical reservoir and its river downstream. *Journal of Marine Systems*, Vol.66. No.1-4, Jun, pp. 161-172, 0924-7963 Hedges, J.I. & Keil, R.G. (1995). Sedimentary organic-matter preservation - an assessment

and speculative synthesis. *Marine Chemistry*, Vol.49. No.2-3, Apr, pp. 81-115, 0304-

M.E.; Schneider, S.H.; Watts, R.G.; Wigley, T.M.L. & Wuebbles, D.J. (1998). Energy implications of future stabilization of atmospheric CO2 content. *Nature*, Vol.395.

Niskanen, A.; Virtanen, M.O.; Lindqvist, O.V.; Nenonen, O.S. & Martikainen, P.J. (2002). Fluxes of CH4 ,CO2, and N2O in hydroelectric reservoirs Lokka and Porttipahta in the northern boreal zone in Finland. *Global Biogeochemical Cycles*,

*Journal of Geophysical Research-Atmospheres*, Vol.99. No.D4, Apr 20, pp. 8307-8319,

surfaces, a neglected area in greenhouse studies. *Eos Trans. AGU*, Vol.75. No.29, pp.

hydroelectric dam. *Geophysical Research Letters*, Vol.34: L1 2809. Jun 23, pp. 0094-

hydroelectric reservoir (Balbina, Brazil). *Journal of Geophysical Research-*

In: *Microbial Processes in Reservoirs.*, Gunnisson, D., Dr. W. Junk Publishers,

*Ecological Perspectives*, Thornton, K.W., B.L., K. & F.E., P., John Wiley and Sons,


Duchemin, E.; Lucotte, M.; Canuel, R. & Soumis, N. (2006). First assessment of methane and

Farrow, D.E. & Stevens, C.L. (2003). Numerical modelling of a surface-stress driven density-

Fearnside, P.M. (1995). Hydroelectric dams in the Brazilian amazon as sources of

Fearnside, P.M. (1997). Greenhouse-gas emissions from Amazonian hydroelectric reservoirs:

Fearnside, P.M. (2002). Greenhouse gas emissions from a hydroelectric reservoir (Brazil's

Fearnside, P.M. (2004). Greenhouse gas emissions from hydroelectric dams: Controversies

editorial comment. *Climatic Change*, Vol.66. No.1-2, Sep, pp. 1-8, 0165-0009 Fearnside, P.M. (2005). Do hydroelectric dams mitigate global warming? The case of Brazil's

Fearnside, P.M. (2006). Greenhouse gas emissions from hydroelectric dams: Reply to Rosa et

Flanagan, K.M.; McCauley, E.; Wrona, F. & Prowse, T. (2003). Climate change: The potential

Fisheries and Aquatic Sciences, Vol.60. No.6, Oct 20, pp. 635-639, 0706-652X Ford, D.E. (1990). Reservoir transport processes. In: *Reservoir limnology: ecological perspectives*, Thornton, K.W., Kimmel, B.L. & Payne, F.E., 256, JohnWiley & Sons, New York Gagnon, L. & Van de Vate, J.F. (1997). Greenhouse gas emissions from hydropower - The state of research in 1996. *Energy Policy*, Vol.25. No.1, Jan, pp. 7-13, 0301-4215 Galy-Lacaux, C.; Delmas, R.; Jambert, C.; Dumestre, J.F.; Labroue, L.; Richard, S. & Gosse, P.

Galy-Lacaux, C.; Delmas, R.; Kouadio, G.; Richard, S. & Gosse, P. (1999). Long-term

Gleick, P.H.; Cooley, H.; Cohen, M.J.; Marikawa, M.; Morrison, J. & Palanappn, M. (2009).

Development, Environment, and Security, Island Press, Washington, D.C. Gudasz, C.; Bastviken, D.; Steger, K.; Premke, K.; Sobek, S. & Tranvik, L.J. (2010).

Guerin, F. & Abril, G. (2007). Significance of pelagic aerobic methane oxidation in the

*Global Biogeochemical Cycles*, Vol.13. No.2, Jun, pp. 503-517, 0886-6236 Giles, J. (2006). Methane quashes green credentials of hydropower. *Nature*, Vol.444. No.7119,

al. *Climatic Change*, Vol.75. No.1-2, Mar, pp. 103-109, 0165-0009

*Environmental Conservation*, Vol.24. No.1, Mar, pp. 64-75, 0376-8929

Vol.133. No.1-4, Jan, pp. 69-96, 0049-6979

1440-1770

pp. 675-691,

0886-6236

Nov, pp. 524-525, 0028-0836

Vol.466. No.7305, Jul, pp. 478-U3, 0028-0836

*Biogeosciences*, Vol.112. No.G3, pp.

0833

carbon dioxide emissions from shallow and deep zones of boreal reservoirs upon ice break-up. *Lakes & Reservoirs: Research & Management*, Vol.11. No.1, pp. 9-19,

stratified fluid. *Journal of Engineering Mathematics*, Vol.47. No.1, Sep, pp. 1-16, 0022-

greenhouse gases. *Environmental Conservation*, Vol.22. No.1, Spr, pp. 7-19, 0376-8929

the example of Brazil's Tucurui Dam as compared to fossil fuel alternatives.

Tucurui Dam) and the energy policy implications. *Water Air and Soil Pollution*,

provide a springboard for rethinking a supposedly 'clean' energy source - An

Curuá-Una dam. *Mitigation and Adaptation Strategies for Global Change*, Vol.10. No.5,

for latitudinal effects on algal biomass in aquatic ecosystems, Canadian. Journal of

(1997). Gaseous emissions and oxygen consumption in hydroelectric dams: A case study in French Guyana. *Global Biogeochemical Cycles*, Vol.11. No.4, Dec, pp. 471-483,

greenhouse gas emissions from hydroelectric reservoirs in tropical forest regions.

Dams Removed or Decommissioned in the United States, 1912 to Present. In: *The World's Water 2008-2009*, Gleick, P.H., 239-264, Pacific Institute for Studies in

Temperature-controlled organic carbon mineralization in lake sediments. *Nature*,

methane and carbon budget of a tropical reservoir. *Journal of Geophysical Research-*


Greenhouse Gas Emissions from Hydroelectric

7447


Sep, pp. 0886-6236

Vol.111. No.D24, Dec, pp. 0148-0227

Vol.66. No.1-2, Sep, pp. 9-21, 0165-0009

*Ambio*, Vol.23. No.2, Mar, pp. 164-165, 0044-7447

Vol.52. No.12, Dec, pp. 2340-2353, 0046-5070

*Tellus*, Vol.33. No.2, pp. 172-187, 0040-2826

*Biogeochemistry*, Vol.86. pp. 123-135, s10533-007

*Bioscience*, Vol.50. No.9, Sep, pp. 766-775, 0006-3568

*Ecosystems*, Vol.14. No.1, Jan, pp. 28-46, 1432-9840

Reservoirs: What Knowledge Do We Have and What is Lacking? 77

Roehm, C. & Tremblay, A. (2006). Role of turbines in the carbon dioxide emissions from two

Roland, F.; Cimbleris, A.C.P.; Lobão, L.M. & Vidal, L.O. (2011). Bacterioplankton Metabolism in Hydroelectric Reservoirs. *Oecologia Australis*, Vol.15. pp. 605-617, Roland, F.; Vidal, L.O.; Pacheco, F.S.; Barros, N.O.; Assireu, A.; Ometto, J.; Cimbleris, A.C.P.

Rosa, L.P. & Schaeffer, R. (1994). Greenhouse-Gas Emissions from Hydroelectric Reservoirs.

Rudd, J.W.M.; Harris, R.; Kelly, C.A. & Hecky, R.E. (1993). Are hydroelectric reservoirs

Rybak, J. (2001). Long-term and seasonal dynamics of nutrient export rates from lake

Schlesinger, W.H. & Melack, J.M. (1981). Transport of organic-carbon in the worlds rivers.

Sobek, S.; Tranvik, L.J. & Cole, J.J. (2005). Temperature independence of carbon dioxide

Soumis, N.; Duchemin, E.; Canuel, R. & Lucotte, M. (2004). Greenhouse gas emissions from

Soumis, N.; Lucotte, M.; Larose, C.; Veillette, F. & Canuel, R. (2007). Photomineralization in

St Louis, V.L.; Kelly, C.A.; Duchemin, E.; Rudd, J.W.M. & Rosenberg, D.M. (2000). Reservoir

Stevens, C. & Imberger, J. (1996). The initial response of a stratified lake to a surface shear stress. *Journal of Fluid Mechanics*, Vol.312. Apr 10, pp. 39-66, 0022-1120 Takagaki, N. & Komori, S. (2007). Effects of rainfall on mass transfer across the air-water

Teodoru, C.R.; Prairie, Y.T. & Del Giorgio, P.A. (2010). Spatial Heterogeneity of Surface CO2

Thomas, S.; Cecchi, P.; Corbin, D. & Lemoalle, J. (2000). The different primary producers in a

interactions. *Freshwater Biology*, Vol.45. No.1, Sep, pp. 43-56, 0046-5070 Tranvik, L.J.; Downing, J.A.; Cotner, J.B.; Loiselle, S.A.; Striegl, R.G.; Ballatore, T.J.; Dillon,

boreal reservoirs, Quebec, Canada. *Journal of Geophysical Research-Atmospheres*,

& Cole, J.J. (2010). Variability of carbon dioxide flux from tropical (Cerrado) hydroelectric reservoirs. *Aquatic Sciences*, Vol.72. No.3, Jun, pp. 283-293, 1015-1621 Rosa, L.P.; Dos Santos, M.A.; Matvienko, B.; Dos Santos, E.O. & Sikar, E. (2004). Greenhouse

gas emissions from hydroelectric reservoirs in tropical regions. *Climatic Change*,

significant sources of greenhouse gases. *Ambio*, Vol.22. No.4, Jun, pp. 246-248, 0044-

watersheds of diversified land cover pattern. *International Association of Theoretical and Applied Limnology, Vol 27, Pt 5, Proceedings*, Vol.27. pp. 3132-3136, 0368-0770 Sand-Jensen, K.; Pedersen, N.L. & Sondergaard, M. (2007). Bacterial metabolism in small

temperate streams under contemporary and future climates. *Freshwater Biology*,

supersaturation in global lakes. *Global Biogeochemical Cycles*, Vol.19. No.2, Apr 5, pp.

reservoirs of the western United States. *Global Biogeochemical Cycles*, Vol.18. No.3,

a boreal hydroelectric reservoir: a comparison with natural aquatic ecosystems.

surfaces as sources of greenhouse gases to the atmosphere: A global estimate.

interface. *Journal of Geophysical Research-Oceans*, Vol.112. No.C6, Jun 8, pp. 0148-0227

Fluxes in a Newly Created Eastmain-1 Reservoir in Northern Quebec, Canada.

small African tropical reservoir during a drought: temporal changes and

P.; Finlay, K.; Fortino, K.; Knoll, L.B.; Kortelainen, P.L.; Kutser, T.; Larsen, S.; Laurion, I.; Leech, D.M.; McCallister, S.L.; McKnight, D.M.; Melack, J.M.; Overholt,


Kosten, K.; Roland, F.; Motta Marques, D.M.L.; Van Nes, E.H.; Mazzeo, N.; Sternberg, L.S.L.;

Kritzberg, E.S.; Cole, J.J.; Pace, M.M. & Graneli, W. (2005). Does autochthonous primary

Lemoalle, J. (1975). L'activité photosynthetic du phytoplankton en relation avec le niveau des eaux du Lac Tchad (Afrique). *Verh. int. Ver. Limnol.*, Vol.19. pp. 5, Lima, I.B.T. (2005). Blogeochemical distinction of methane releases from two Amazon hydroreservoirs. *Chemosphere*, Vol.59. No.11, Jun, pp. 1697-1702, 0045-6535 Lima, I.B.T.; Novo, E.M.L.M.; Ballester, M.V.R. & Ometto, J.P.H.B. (1998). Methane

Macintyre, S.; Romero, J.R. & Kling, G.W. (2002). Spatial-temporal variability in surface layer

McGinnis, D.F.; Greinert, J.; Artemov, Y.; Beaubien, S.E. & Wuest, A. (2006). Fate of rising

*Journal of Geophysical Research-Oceans*, Vol.111. No.C9, Sep, pp. 0148-0227 Melack, J.M. (1979). Photosynthetic Rates in 4 Tropical African Fresh Waters. *Freshwater* 

Monismith, S.G. (1985). Wind-Forced Motions in Stratified Lakes and Their Effect on Mixed-Layer Shear. *Limnology and Oceanography*, Vol.30. No.4, pp. 771-783, 0024-3590 Mulholland, P.J.; Fellows, C.S.; Tank, J.L.; Grimm, N.B.; Webster, J.R.; Hamilton, S.K.; Marti,

*Limnology and Oceanography*, Vol.47. No.3, May, pp. 656-671, 0024-3590 Martin, J.L. & Mccutcheon, S.C. (1999). Hydrodynamics and transport for water quality

Global Biogeochemical Cycles, Vol.24. May 8, 0886-6236

*States of America*, Vol.103. No.23, Jun, pp. 8739-8744, 0027-8424

103-111, 0948-3055

modeling. Boca Raton

*Biology*, Vol.9. No.6, pp. 555-571, 0046-5070

Scheffer, M. & Cole, J.J. (2010). Climate-dependent CO2 emissions from lakes.

production drive variability in bacterial metabolism and growth efficiency in lakes dominated by terrestrial C inputs? *Aquatic Microbial Ecology*, Vol.38. No.2, Feb, pp.

Production, Transport and Emission in Amazon Hydroelectric Plants. *Veranlundgen der Internationalen Vereinigung für Theoretische und Angewandte Limnologie.*, pp. Lopez-Urrutia, A.; San Martin, E.; Harris, R.P. & Irigoien, X. (2006). Scaling the metabolic

balance of the oceans. *Proceedings of the National Academy of Sciences of the United* 

deepening and lateral advection in an embayment of Lake Victoria, East Africa.

methane bubbles in stratified waters: How much methane reaches the atmosphere?

E.; Ashkenas, L.; Bowden, W.B.; Dodds, W.K.; Mcdowell, W.H.; Paul, M.J. & Peterson, B.J. (2001). Inter-biome comparison of factors controlling stream metabolism. *Freshwater Biology*, Vol.46. No.11, Nov, pp. 1503-1517, 0046-5070 Ometto, J.P.; Cimbleris, A.C.P.; Santos, M.A.; Rosa, L.P.; Abe, D.S.; Tundisi, J.G.; Stech, J.L.;

Barros, N.O. & Roland, F. Patterns of carbon emission as a function of energy generation in hydroelectric reservoirs. . *Submitted to Climatic Changes (2011)*, pp. Ometto, J.P.; Pacheco, F.S.; Cimbleris, A.C.P.; Stech, J.L.; Lorenzzetti, J.A.; Assireu, A.;

Santos, M.A.; Matvienko, B.; Rosa, L.P.; Galli, C.S.; Abe, D.S.; Tundisi, J.G.; Barros, N.; Mendonça, R.F. & Roland, F. (2011). Carbon dynamic and emissions in Brazilian hydropower reservoirs. In: *Energy Resources: Development, Distribution, and* 

electric power plants in the upper Colorado River Basin. *Environmental Science &* 

Pacca, S. (2007). Impacts from decommissioning of hydroelectric dams: a life cycle perspective. *Climatic Change*, Vol.84. No.3-4, Oct, pp. 281-294, 0165-0009 Pacca, S. & Horvath, A. (2002). Greenhouse gas emissions from building and operating

Rivkin, R.B. & Legendre, L. (2001). Biogenic carbon cycling in the upper ocean: Effects of microbial respiration. *Science*, Vol.291. No.5512, Mar, pp. 2398-2400, 0036-8075

*Exploitation*, Alcântara, E., Nova Science Publishers, New York

*Technology*, Vol.36. No.14, Jul, pp. 3194-3200, 0013-936X


**4**

**GHG Emissions Reduction Via Energy** 

There are three major sources of GHG; carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O). The world's CO2 emissions into the air have been increasing drastically over the past century. The industrial revolution and exploitation of natural resources such as coal and oil have greatly contributed to CO2 emissions. The world's CO2 emissions due to fuel utilization such as natural gas, liquid fuels such as oil, and coal has resulted in CO2 emissions due to energy usage of about 45 billion metric tons by the year 2030. In Japan, for instance, more than 80 % of its GHG emissions are due to energy-based sources resulted from fossil fuel consumption and a boost to the energy efficiency of a single refinery in

It is not a matter of indifference or/and skepticism any more, the majority in the world scientific communities do subscribe now to the fact that, the world's environment has been negatively affected by the global warming phenomenon which has caused the average temperature of the earth's surface to increase during the last century due to the irresponsible release of greenhouse gases (GHG) into the atmosphere. From GHG perspective, energy efficiency optimization is not only a fast track approach to reduce energy-based GHG/CO2 emissions but also a cost-effective option towards such endeavor. It does not need behavioral change, which is sometimes difficult to achieve on the short run. It does not also play on the people's level of romance regarding the health of our beloved universe. It touches the heart of mankind's old and new motivations and aspirations for better life. It is industrial people's own benefit. Energy efficiency optimization solution approach as a quick answer to energy-based GHG emissions reduction simply enables us attaining the "wateroil-impossible-mix" via mixing the useful, exhibited in saving money, with the beautiful of

At the equipment level, process equipment becomes inefficient when it uses higher energy than the designated one at the same feed and production rates. Extra energy consumption by equipment can be related to aging, part deterioration, process related causes, fouling and so on. Such reasons need to be scrutinized on timely basis in any industrial facility to avoid excessive energy consumption and consequently more GHG emissions. Equipment normally has its standard causes and characteristics for energy efficiency degradations and it can be captured through detailed analysis of its parameters change along the historical

Japan can result in a reduction of at least 50,000 ton CO2/year.

saving more than the money, the environment.

**1. Introduction** 

Faisal F. Al Musa, Ali H. Qahtani, Mana M. Owaidh, Meshabab S. Qahtani and Mahmoud Bahy Noureldin

**Efficiency Optimization** 

*Saudi Aramco, Dhahran* 

*Saudi Arabia* 

E.; Porter, J.A.; Prairie, Y.; Renwick, W.H.; Roland, F.; Sherman, B.S.; Schindler, D.W.; Sobek, S.; Tremblay, A.; Vanni, M.J.; Verschoor, A.M.; von Wachenfeldt, E.; Weyhenmeyer, G.A. (2009). Lakes and reservoirs as regulators of carbon cycling and climate. *Limnology and Oceanography*. Vol. 54. No.6, Nov 09, pp. 2298-2314, 0024-3590


## **GHG Emissions Reduction Via Energy Efficiency Optimization**

Faisal F. Al Musa, Ali H. Qahtani, Mana M. Owaidh, Meshabab S. Qahtani and Mahmoud Bahy Noureldin *Saudi Aramco, Dhahran Saudi Arabia* 

## **1. Introduction**

78 Greenhouse Gases – Emission, Measurement and Management

Tremblay, A.; Lambert, M. & Gagnon, L. (2004). Do hydroelectric reservoirs emit

Vidal, L.O.; Graneli, W.; Daniel, C.B.; Heiberg, L. & Roland, F. (2011). Carbon and

Vorosmarty, C.J.; Meybeck, M.; Fekete, B.; Sharma, K.; Green, P. & Syvitski, J.P.M. (2003).

Wang, F.; Wang, B.; Liu, C.Q.; Wang, Y.; Guan, J.; Liu, X. & Yu, Y. (2011). Carbon dioxide

Wanninkhof, R. (1992). Relationship between Wind-Speed and Gas-Exchange over the

WCD (2000). World Commission on Dams. Dams and Development: A New Framework for

Wehrli, B. (2011). Renewable but not carbon-free. *Nature Geoscience*, Vol.4. No.9, pp. 585-586,

Weisser, D. (2007). A guide to life-cycle greenhouse gas (GHG) emissions from electric supply technologies. *Energy*, Vol.32. No.9, Sep, pp. 1543-1559, 0360-5442 Zheng, H.; Zhao, X.J.; Zhao, T.Q.; Chen, F.L.; Xu, W.H.; Duan, X.N.; Wang, X.K. & Ouyang,

Decision-Making. Earthscan Publications. Available via

*Plankton Research*, Vol.33. No.11, pp. 1747-1756,

0024-3590

8181

3834, 1352-2310

http://go.nature.com/rnFEBI

Apr, pp. 1391-1396, 0885-6087

0148-0227

1752-0894

E.; Porter, J.A.; Prairie, Y.; Renwick, W.H.; Roland, F.; Sherman, B.S.; Schindler, D.W.; Sobek, S.; Tremblay, A.; Vanni, M.J.; Verschoor, A.M.; von Wachenfeldt, E.; Weyhenmeyer, G.A. (2009). Lakes and reservoirs as regulators of carbon cycling and climate. *Limnology and Oceanography*. Vol. 54. No.6, Nov 09, pp. 2298-2314,

greenhouse gases? *Environmental Management*, Vol.33. Jul, pp. S509-S517, 0364-152X

phosphorus regulating bacterial metabolism in oligotrophic boreal lakes. *Journal of* 

Anthropogenic sediment retention: major global impact from registered river impoundments. *Global and Planetary Change*, Vol.39. No.1-2, Oct, pp. 169-190, 0921-

emission from surface water in cascade reservoirs-river system on the Maotiao River, southwest of China. *Atmospheric Environment*, Vol.45. No.23, Jul, pp. 3827-

Ocean. *Journal of Geophysical Research-Oceans*, Vol.97. No.C5, May 15, pp. 7373-7382,

Z.Y. (2011). Spatial-temporal variations of methane emissions from the Ertan hydroelectric reservoir in southwest China. *Hydrological Processes*, Vol.25. No.9, There are three major sources of GHG; carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O). The world's CO2 emissions into the air have been increasing drastically over the past century. The industrial revolution and exploitation of natural resources such as coal and oil have greatly contributed to CO2 emissions. The world's CO2 emissions due to fuel utilization such as natural gas, liquid fuels such as oil, and coal has resulted in CO2 emissions due to energy usage of about 45 billion metric tons by the year 2030. In Japan, for instance, more than 80 % of its GHG emissions are due to energy-based sources resulted from fossil fuel consumption and a boost to the energy efficiency of a single refinery in Japan can result in a reduction of at least 50,000 ton CO2/year.

It is not a matter of indifference or/and skepticism any more, the majority in the world scientific communities do subscribe now to the fact that, the world's environment has been negatively affected by the global warming phenomenon which has caused the average temperature of the earth's surface to increase during the last century due to the irresponsible release of greenhouse gases (GHG) into the atmosphere. From GHG perspective, energy efficiency optimization is not only a fast track approach to reduce energy-based GHG/CO2 emissions but also a cost-effective option towards such endeavor. It does not need behavioral change, which is sometimes difficult to achieve on the short run. It does not also play on the people's level of romance regarding the health of our beloved universe. It touches the heart of mankind's old and new motivations and aspirations for better life. It is industrial people's own benefit. Energy efficiency optimization solution approach as a quick answer to energy-based GHG emissions reduction simply enables us attaining the "wateroil-impossible-mix" via mixing the useful, exhibited in saving money, with the beautiful of saving more than the money, the environment.

At the equipment level, process equipment becomes inefficient when it uses higher energy than the designated one at the same feed and production rates. Extra energy consumption by equipment can be related to aging, part deterioration, process related causes, fouling and so on. Such reasons need to be scrutinized on timely basis in any industrial facility to avoid excessive energy consumption and consequently more GHG emissions. Equipment normally has its standard causes and characteristics for energy efficiency degradations and it can be captured through detailed analysis of its parameters change along the historical

GHG Emissions Reduction Via Energy Efficiency Optimization 81

throughout the world and consequently generating GHG. The impact of oil refining industry on the environment is becoming more and more apparent nowadays. Industrial emissions contribute to the climate change by emitting green house gases (GHG) into atmosphere. Pollution prevention measures must take place in industrial sector to lower emissions especially in oil refineries which consume huge energy derived from fossil fuels. Pollution prevention is an adequate method for managing the environmental impact associated with industrial facilities. The right pollution prevention strategy aims to eliminate potential pollution from the process at the source before it is being emitted into the environment. Pollution prevention can be in form of energy demand reduction inside process plants. The reduction in energy demand will eventually lead to lower fuel consumption that will result in lower GHG emissions. Such reduction in energy-based GHG emissions will ultimately reduce the environmental impact of industrial facilities. The application of clean development mechanism (CDM) to an oil refinery presented here is confined to only one major unit in the oil refineries called Hydrocracking unit (HCU). The aim is to reduce energy generation-based CO2 emissions at the source via reducing the energy consumption of the HCU. The reduction in fuel usage can be achieved by better integration of the heat exchanger network of the unit to reduce fuel demand in the process by increasing the process-to-process heat exchange duties. Pinch technology is used in this context due to its applicability and practicality to process plants energy reduction. The reduction in energy demand, of a major oil refinery unit presented in this chapter, is an application that has potential implementation as a small scale CDM project. The CDM application/project introduced quantifies the potential CO2 emissions reduction resulted

The oil refinery used in this study is a medium size one, processes 120 000 bbl/d of crude oil. As usual the feed to the refinery is first stabilized by stripping off acid gases and other volatile compounds and then introduced to the atmospheric distillation column followed by vacuum distillation column. The distilled products are then introduced to different process units to produce a variety of petroleum products. The refinery is divided into several process areas. Those process areas are the stabilizer, the atmospheric crude distillation Unit, the vacuum distillation unit, the asphalt recovery, the amine gas treating, the gas concentration unit, the naphtha and kerosene Hydrotreating, demex, Platformer, hydrogen plant, hydrocracker and finally the utilities. The HCU is a major process unit in any refinery. It receives heavy oils and produces more valuable products such as diesel and fuel oil. The purpose of the HCU, presented in this chapter, is to process vacuum gas oil (VGO) from the Vacuum distillation unit (VDU) and De-metalized oil (DMO) from the asphalt oxidization unit. These two low value feed streams are converted into useful products such as liquefied petroleum gas (LPG), light naphtha, heavy naphtha, kerosene, light diesel oil (LDO), and heavy diesel oil (HDO) products. The HCU is divided into two main sections; the first

HCU consists of several process streams and not all of them are used in our case study calculation using pinch technology. The problem's streams data included in our study have, the stream supply temperature (Ts), target temperature (Tt), the heat transfer in term of thermal kW gained or released by the stream, and the heat transfer coefficient (HTC) for the heat transfer equipment. The process streams have been segregated into hot streams (any process stream that needs to be cooled) and cold streams (any process stream that need to be heated). Each stream has been assigned a name based on the content and direction into the

section is the reaction section and the second is the fractionation section.

from energy consumption reduction in the HCU.

sequence of events during its operation. The relation between GHG emissions and energy efficiency is subject to some factors, such as equipment and drivers types. Gas compressors are driven by electrical motors; steam turbines; and gas turbine drivers. Compressor's driver selection is a major contributor to gas emissions. Electric motors are the least emitters and gas turbines are the most among other drivers.

Industrial community is systematically practicing online equipment-based performance monitoring and diagnosis functions to measure, analyze, improve and control equipment's energy usage to keep its energy consumption under tight control. Although the motivator to steer the equipment energy consumption towards the design figure is to reduce the energy operating cost; that result is also contributing to the reduction of GHG emission. It is the industry practice to identify energy savings opportunities in equipment operation by taking proactive maintenance steps, applying advanced control strategies and real-time optimization. For example oil and gas equipment/units operation load management make reductions in energy usage and greenhouse emissions. Constantly invented/developed new materials, technologies, hardware and design software are also improving equipment's energy efficiency. For instance, materials scientists have helped steam power plant designers to design steam boilers for subcritical and super critical steam generation levels, which were not possible decades ago. Many successful efforts in capital allocation in many industrial facilities are also preventing energy inefficient equipment of being introduced to new industrial sites and/or phasing out inefficient ones in existing facilities before its designated retiring date, is also getting steam every day.

However, for decades ago energy efficiency optimization was merely addressing the energy efficiency of standalone process equipment. Since late seventies in the last century, the landscape has changed. It is not only energy efficiency for the standalone equipment/unit but also for subsystems, systems, complexes and even mega sites as well as industrial cities. This chapter is presenting the impact of adapting the state-of-art in energy efficiency optimization on the GHG emissions reduction. The chapter starts with intra-process integration and clean development mechanism application in refinery Hydrocracking unit (HCU), followed by a method and case study for inter-processes integration in oil refining and then is closed with brief industrial utility system retrofit case study in an oil plant using combined heat and power concept.

## **2. Intra-process integration and clean development mechanism (CDM)**

Oil refineries market is currently undergoing a significant reorientation, with demand moving away from the traditional strongholds of Europe and North America to other regions of the world. This transformation, which began before the recent economic recession, but was accelerated by its effects, both creates a number of opportunities and poses many number of threats to companies involved in the industry and to the environment. Refining operations are a vital aspect of any modern, industrialized economy, and hence much is being invested in trying to produce products that do not rely on refined fuels, we are a little away from a point at which these developments will significantly reduce the market for refined products. Refineries are as usual going to be influenced by environmental regulations, technological innovations, and other important factors and trends which are changing the dynamics of the industry. Oil refining business is likely to see significant investment over the next ten years, where growth will be unevenly spread

sequence of events during its operation. The relation between GHG emissions and energy efficiency is subject to some factors, such as equipment and drivers types. Gas compressors are driven by electrical motors; steam turbines; and gas turbine drivers. Compressor's driver selection is a major contributor to gas emissions. Electric motors are the least emitters and

Industrial community is systematically practicing online equipment-based performance monitoring and diagnosis functions to measure, analyze, improve and control equipment's energy usage to keep its energy consumption under tight control. Although the motivator to steer the equipment energy consumption towards the design figure is to reduce the energy operating cost; that result is also contributing to the reduction of GHG emission. It is the industry practice to identify energy savings opportunities in equipment operation by taking proactive maintenance steps, applying advanced control strategies and real-time optimization. For example oil and gas equipment/units operation load management make reductions in energy usage and greenhouse emissions. Constantly invented/developed new materials, technologies, hardware and design software are also improving equipment's energy efficiency. For instance, materials scientists have helped steam power plant designers to design steam boilers for subcritical and super critical steam generation levels, which were not possible decades ago. Many successful efforts in capital allocation in many industrial facilities are also preventing energy inefficient equipment of being introduced to new industrial sites and/or phasing out inefficient ones in existing facilities before its designated

However, for decades ago energy efficiency optimization was merely addressing the energy efficiency of standalone process equipment. Since late seventies in the last century, the landscape has changed. It is not only energy efficiency for the standalone equipment/unit but also for subsystems, systems, complexes and even mega sites as well as industrial cities. This chapter is presenting the impact of adapting the state-of-art in energy efficiency optimization on the GHG emissions reduction. The chapter starts with intra-process integration and clean development mechanism application in refinery Hydrocracking unit (HCU), followed by a method and case study for inter-processes integration in oil refining and then is closed with brief industrial utility system retrofit case study in an oil plant using

**2. Intra-process integration and clean development mechanism (CDM)** 

Oil refineries market is currently undergoing a significant reorientation, with demand moving away from the traditional strongholds of Europe and North America to other regions of the world. This transformation, which began before the recent economic recession, but was accelerated by its effects, both creates a number of opportunities and poses many number of threats to companies involved in the industry and to the environment. Refining operations are a vital aspect of any modern, industrialized economy, and hence much is being invested in trying to produce products that do not rely on refined fuels, we are a little away from a point at which these developments will significantly reduce the market for refined products. Refineries are as usual going to be influenced by environmental regulations, technological innovations, and other important factors and trends which are changing the dynamics of the industry. Oil refining business is likely to see significant investment over the next ten years, where growth will be unevenly spread

gas turbines are the most among other drivers.

retiring date, is also getting steam every day.

combined heat and power concept.

throughout the world and consequently generating GHG. The impact of oil refining industry on the environment is becoming more and more apparent nowadays. Industrial emissions contribute to the climate change by emitting green house gases (GHG) into atmosphere. Pollution prevention measures must take place in industrial sector to lower emissions especially in oil refineries which consume huge energy derived from fossil fuels. Pollution prevention is an adequate method for managing the environmental impact associated with industrial facilities. The right pollution prevention strategy aims to eliminate potential pollution from the process at the source before it is being emitted into the environment. Pollution prevention can be in form of energy demand reduction inside process plants. The reduction in energy demand will eventually lead to lower fuel consumption that will result in lower GHG emissions. Such reduction in energy-based GHG emissions will ultimately reduce the environmental impact of industrial facilities. The application of clean development mechanism (CDM) to an oil refinery presented here is confined to only one major unit in the oil refineries called Hydrocracking unit (HCU). The aim is to reduce energy generation-based CO2 emissions at the source via reducing the energy consumption of the HCU. The reduction in fuel usage can be achieved by better integration of the heat exchanger network of the unit to reduce fuel demand in the process by increasing the process-to-process heat exchange duties. Pinch technology is used in this context due to its applicability and practicality to process plants energy reduction. The reduction in energy demand, of a major oil refinery unit presented in this chapter, is an application that has potential implementation as a small scale CDM project. The CDM application/project introduced quantifies the potential CO2 emissions reduction resulted from energy consumption reduction in the HCU.

The oil refinery used in this study is a medium size one, processes 120 000 bbl/d of crude oil. As usual the feed to the refinery is first stabilized by stripping off acid gases and other volatile compounds and then introduced to the atmospheric distillation column followed by vacuum distillation column. The distilled products are then introduced to different process units to produce a variety of petroleum products. The refinery is divided into several process areas. Those process areas are the stabilizer, the atmospheric crude distillation Unit, the vacuum distillation unit, the asphalt recovery, the amine gas treating, the gas concentration unit, the naphtha and kerosene Hydrotreating, demex, Platformer, hydrogen plant, hydrocracker and finally the utilities. The HCU is a major process unit in any refinery. It receives heavy oils and produces more valuable products such as diesel and fuel oil. The purpose of the HCU, presented in this chapter, is to process vacuum gas oil (VGO) from the Vacuum distillation unit (VDU) and De-metalized oil (DMO) from the asphalt oxidization unit. These two low value feed streams are converted into useful products such as liquefied petroleum gas (LPG), light naphtha, heavy naphtha, kerosene, light diesel oil (LDO), and heavy diesel oil (HDO) products. The HCU is divided into two main sections; the first section is the reaction section and the second is the fractionation section.

HCU consists of several process streams and not all of them are used in our case study calculation using pinch technology. The problem's streams data included in our study have, the stream supply temperature (Ts), target temperature (Tt), the heat transfer in term of thermal kW gained or released by the stream, and the heat transfer coefficient (HTC) for the heat transfer equipment. The process streams have been segregated into hot streams (any process stream that needs to be cooled) and cold streams (any process stream that need to be heated). Each stream has been assigned a name based on the content and direction into the

GHG Emissions Reduction Via Energy Efficiency Optimization 83

HCU is estimated to be 38 kt of CO2 annually. The potential reduction of the other utilities demand such as cooling water and air cooing utilities is not identified here due to its low value. This potential reduction in CO2 emission as a result of the fuel gas reduction in the HCU due to energy efficiency enhancement can be adopted as a clean development

The HCU emission reduction potential can be adopted as a small-scale CDM project. Adopting the small-scale CDM project framework has several advantages such as; ability to combine identical project as one group of project; the project design document and methodologies are simplified for a small scale project and the baseline and monitoring

The establishment of the baseline emission is an important step in any CDM project evaluation. Therefore, we need to estimate the as is CO2 emissions without implementing the project. Fortunately, a real parameter data of fuel gas flow can be obtained because it is measured and the emission quantity is related to the fuel consumption. The fuel consumption rate can be measured for the past three years to evaluate the baseline emission situation. The current energy consumed by the HCU process heaters is 71 MWh which is equivalent to 256 GJ of fuel gas. As per given or sampled fuel composition, the CO2

PEFC,j,y = CO2 emission from fossil fuel combustion in process j during the year y

FCi,j,y = Quantity of fuel type I combusted in process j during the year y (mass or volume

For our HCU fuel gas reduction project, the above formula is used to calculate the baseline

The i value is assumed to be 1 because there is one fuel type (fuel gas) and the CO2 content

HCU exhibits, using pinch technology, an energy efficiency enhancement potential of 21 MWh in fuel. Such reduction in energy consumption can reduce GHG emissions in the HCU to 91,635 t CO2/y from the baseline calculated above if the project is implemented. Using the emission baseline figure, the potential reduction in GHG emissions after implementing

It is important to note here that beside the benefit obtained from \$ energy saving another \$ benefit can be attained due to carbon credit concept. The price of CO2 emission trading

COEFi,y = CO2 emission coefficient of fuel type I in year y (t CO2 /mass or volume unit)

emission factor in the fuel gas is calculated to be 0.217 kg CO2 per kWh.

The current baseline emission can be simply identified as follows:

PEFC,j,y = 71.1 MW x 0.217 kg CO2 /kWh x 8400 h = 12967 t CO2/y

per energy unit has been used instead of CO2 content per mass unit.

i = fuel types combusted in process j during year y.

the CDM project is estimated to be 38 035 t CO2/y.

mechanism (CDM) project.

procedures are reduced.

PEFC,j,y = ∑iFCi,j,y x COEFi,y

Where:

(tCO2/yr)

unit / year)

emission value.

process. Streams are identified based on their composition, if the stream passes through another heat transfer equipment and changes in temperature, then it will be another segment of the same stream; however, if the stream changes in composition in the case of new production from the separation column it becomes different stream.


Table 1. Problem Data for Hydrocracking Plant

The optimum ∆Tmin for HCU is calculated to estimate the minimum temperature difference in the heat exchanger network that will lead to the lowest possible utility demand in the process. The total annualized capital and energy cost is plotted vs. several, ∆Tmin values to estimate such optimum at the minimum annualized cost.

The optimum ∆Tmin for HCU was calculated to be 13 °C where the total annualized cost reaches the minimum. The total annualized cost consists of the annual energy cost and the annual heat exchanger capital cost. The annual energy cost is obtained simply by multiplying the utility cost by the expected operating hours in the year which is assumed to be 8,400 hours per year to incorporate maintenance and plant shut down. The annualized capital cost incorporates the capital cost multiplied by the money borrowing annualized factor.

The energy reduction as a result of applying the minimum ∆T in the HCU can be compared to the current HCU process energy demand and current GHG emissions. The potential savings in both energy consumption and CO2 emission are then identified. Using pinch techniques, the potential reduction in energy is about 20.1 MWh annually which is equivalent to fuel gas consumption of 631 000 GJ. The potential emission reduction in the HCU is estimated to be 38 kt of CO2 annually. The potential reduction of the other utilities demand such as cooling water and air cooing utilities is not identified here due to its low value. This potential reduction in CO2 emission as a result of the fuel gas reduction in the HCU due to energy efficiency enhancement can be adopted as a clean development mechanism (CDM) project.

The HCU emission reduction potential can be adopted as a small-scale CDM project. Adopting the small-scale CDM project framework has several advantages such as; ability to combine identical project as one group of project; the project design document and methodologies are simplified for a small scale project and the baseline and monitoring procedures are reduced.

The establishment of the baseline emission is an important step in any CDM project evaluation. Therefore, we need to estimate the as is CO2 emissions without implementing the project. Fortunately, a real parameter data of fuel gas flow can be obtained because it is measured and the emission quantity is related to the fuel consumption. The fuel consumption rate can be measured for the past three years to evaluate the baseline emission situation. The current energy consumed by the HCU process heaters is 71 MWh which is equivalent to 256 GJ of fuel gas. As per given or sampled fuel composition, the CO2 emission factor in the fuel gas is calculated to be 0.217 kg CO2 per kWh.

The current baseline emission can be simply identified as follows:

PEFC,j,y = ∑iFCi,j,y x COEFi,y

Where:

82 Greenhouse Gases – Emission, Measurement and Management

process. Streams are identified based on their composition, if the stream passes through another heat transfer equipment and changes in temperature, then it will be another segment of the same stream; however, if the stream changes in composition in the case of

**No Stream Name Ts, C Tt, C kW kW/C kW/m^2.C** COLD VGO/DMO Feed 105.00 393.30 20196 70.05 3.41 COLD 393.33 437.78 14200 319.49 56.76 COLD De-C4 col feed 56.67 170.56 18197 159.78 2.27 COLD De-4 Reb Duty 226.11 281.11 17589 319.81 56.76 COLD Frac Feed 266.11 377.22 27644 248.80 56.76 COLD HN Strip reb duty 160.00 171.11 1804 162.35 5.68 COLD Kero Strip Reb duty 226.67 235.00 732 87.85 5.68 COLD LN feed to Frac 60.00 109.44 1092 22.09 3.41 COLD 109.44 265.56 929 5.95 0.57 HOT DHC rxn outlet from V-1/2 468.33 385.00 16710 200.52 0.45 HOT HC rxn outlet from V-3/4 426.67 180.00 20196 81.87 1.70 HOT V6 overhead vapor (to V8) 165.00 60.00 37727 359.31 0.40 HOT 60.00 43.33 2115 126.88 1.70 HOT V-14 oh vapor 71.11 48.89 1172 52.74 4.55 HOT Fract OH Vapor 81.11 60.00 16353 774.62 5.68 HOT De-4 OH vapor 83.89 57.22 8246 309.22 5.68 HOT LiN product draw 60.00 37.78 375 16.87 2.84 HOT HN product draw 160.00 60.00 3262 32.62 2.27 HOT 60.00 37.78 719 32.35 1.70 HOT HN p/a loop 137.78 77.22 7918 130.76 1.70 HOT Kero Product to Storage 227.22 60.00 2580 15.43 2.27 HOT 60.00 48.89 129 11.60 1.70 HOT Kero p/a loop 210.00 77.22 5915 44.55 2.27 HOT LDO draw 298.89 252.22 929 19.90 2.27 HOT 252.22 196.11 1092 19.47 2.27 HOT 196.11 60.00 4310 31.67 2.27 HOT HDO p/a (combined) 321.11 170.56 4100 27.23 2.27 HOT HDO product draw 321.11 60.00 10442 39.99 2.27 HOT Frac btms recycle 362.78 121.11 18197 75.30 2.56 HOT Frac btms FO draw 362.78 79.44 3951 13.95 2.56

**DH CP HTC**

The optimum ∆Tmin for HCU is calculated to estimate the minimum temperature difference in the heat exchanger network that will lead to the lowest possible utility demand in the process. The total annualized capital and energy cost is plotted vs. several, ∆Tmin values to

The optimum ∆Tmin for HCU was calculated to be 13 °C where the total annualized cost reaches the minimum. The total annualized cost consists of the annual energy cost and the annual heat exchanger capital cost. The annual energy cost is obtained simply by multiplying the utility cost by the expected operating hours in the year which is assumed to be 8,400 hours per year to incorporate maintenance and plant shut down. The annualized capital cost

The energy reduction as a result of applying the minimum ∆T in the HCU can be compared to the current HCU process energy demand and current GHG emissions. The potential savings in both energy consumption and CO2 emission are then identified. Using pinch techniques, the potential reduction in energy is about 20.1 MWh annually which is equivalent to fuel gas consumption of 631 000 GJ. The potential emission reduction in the

incorporates the capital cost multiplied by the money borrowing annualized factor.

new production from the separation column it becomes different stream.

Table 1. Problem Data for Hydrocracking Plant

estimate such optimum at the minimum annualized cost.

PEFC,j,y = CO2 emission from fossil fuel combustion in process j during the year y (tCO2/yr)

FCi,j,y = Quantity of fuel type I combusted in process j during the year y (mass or volume unit / year)

COEFi,y = CO2 emission coefficient of fuel type I in year y (t CO2 /mass or volume unit)

i = fuel types combusted in process j during year y.

For our HCU fuel gas reduction project, the above formula is used to calculate the baseline emission value.

PEFC,j,y = 71.1 MW x 0.217 kg CO2 /kWh x 8400 h = 12967 t CO2/y

The i value is assumed to be 1 because there is one fuel type (fuel gas) and the CO2 content per energy unit has been used instead of CO2 content per mass unit.

HCU exhibits, using pinch technology, an energy efficiency enhancement potential of 21 MWh in fuel. Such reduction in energy consumption can reduce GHG emissions in the HCU to 91,635 t CO2/y from the baseline calculated above if the project is implemented. Using the emission baseline figure, the potential reduction in GHG emissions after implementing the CDM project is estimated to be 38 035 t CO2/y.

It is important to note here that beside the benefit obtained from \$ energy saving another \$ benefit can be attained due to carbon credit concept. The price of CO2 emission trading

GHG Emissions Reduction Via Energy Efficiency Optimization 85

soft constraints in a systematic way to find better energy targets. New method uses for each hot stream specific ∆T\_min and allows minor possible combinations of process conditions to be modified (e.g. 5 F ± in the supply and target temperatures) to customizing the waste heat recovery problem in a way that renders better reductions in energy consumption cost( energy quantity and/or quality/work optimization). The waste energy recovery problem using such new targeting method can now has several degrees of freedom for the waste energy recovery problem optimization, instead of the one parameter problem optimization

The above mentioned energy integration targeting methods while can be used at any scale is nowadays focused only on industrial facility via direct integration between the hot and cold streams of its process units. It is usually applied at the process level; and known as intraprocess integration. It has proved to be very successful to reducing both energy consumption and energy-based GHG emissions. Integration among many processes, in adjacent geographical locations, can bring in more degrees of freedom to optimize the waste energy recovery problem and consequently presents new horizon to the energy-based GHG emissions reduction to levels never thought of before. For many reasons, direct interprocesses integration is not widely practiced in industry. Many of the reasons hindering the application of inter-process integration among several processes for better energy consumption cost reduction and less energy-based GHG emissions are very valid and need to be addressed in a novel way to enable wider adaptation of direct inter-process integration in existing industrial facilities and naturally for mega facilities, zones and even cities in the

Since the emanation of the pinch technology and its evolution to pinch analysis technique for process synthesis, the direct inter-processes integration has been considered impractical. Many arguments, such as the processes are normally have different start up and shut down times; the processes can work at partial loads; the processes can have seasonal changes in its conditions; utility systems, heaters and HEN capital will not be reduced due to changes in processes schedule and operation philosophy; the disturbance in one process can propagate to another one if they are integrated; the distance-time/velocity lags affect the controllability of processes; the geographical distances among processes will cost us energy in pumping or compression and capital in piping, pumping and compression; safety might be impacted due to the travel of a fluid from one hazardous area to another; the fear of leakage and so on, are certainly valid. Therefore, direct inter-processes integration while is beneficial to energy conservation and the GHG emissions reduction, is still to date almost ignored in

Due to those concerns most of the current methods for inter-process integration are indirect and conducted using buffer systems. Buffer systems are either steam system or hot oil system or sometimes both. However, the industrial community does agree that, direct integration approach in inter-process integration (between several plants) is more efficient and can render more saving in energy consumption and energy-based greenhouse gas emissions [10:16]. In this chapter we demonstrate, that huge potential for energy consumption and GHG emissions reduction in oil refining (more than 5 % in-house) can be

It is instructive to note here that, while we agree with the validity of those arguments we believe that we can still find a room for improvement in energy efficiency enhancement and

currently used [7:9].

future plants.

industrial community.

attained through smart integration among processes.

nowadays can play role in determining the feasibility of HCU energy efficiency/emission reduction project. For instance, the annual potential revenue of the CDM project can be estimated to be \$ 950,000 per year using an average price of \$25 carbon credit per ton [1:3].

## **3. Inter-processes integration for enhanced energy efficiency and energybased GHG emissions reduction**

Pinch analysis is the technology that provides a systematic methodology for energy saving in processes and total sites. The methodology is based upon thermodynamic principles. Pinch Analysis/technology was first developed in the late 1970s as a technique for optimization of thermal heat recovery, and rapidly gained wide acceptance as a practical approach to the design of Heat Exchanger Networks (HENs). Since then, it has evolved into a general methodology for optimization, based on the principles of process integration.

The technique calculates thermodynamically attainable energy targets for a given process and identifies how to achieve them. A key insight is the pinch temperature, which is the most constrained point in the process. The most detailed explanation of the techniques is by Linnhoff et al. Other pinch analyses were developed for several applications such as massexchange networks (El-Halwagi and Manousiouthakis, 1989), water minimization (Wang and Smith, 1994), and material recycle (El-Halwagi et al., 2003).

Pinch analysis has been applied successfully not only to energy systems (heat recovery, pressure drop recovery, power generation), but also to fresh water conservation, wastewater minimization, production capacity de-bottlenecking, and management of chemical species in complex processes.

Applying Pinch technology in heat exchangers networks (HEN) synthesis and retrofit, enable the engineer to calculate the energy requirement for any process, and produce thermally efficient and practical designs. Energy savings are significant compared to previous best designs. Pinch technology is also applied to the optimization/integration of the supply-side, consisting of on-site utilities, such as boilers, furnaces, steam and gas turbines, cogeneration, heat pumps, and refrigeration systems [4:6].

Generally speaking, the objective of the heat exchangers network synthesis for waste heat recovery problem is to design a network that meets an economic criterion such as minimum total annualized cost. Sometimes minimum number of units and minimum energy consumption and GHG emissions in special applications become very legitimate objectives too.

The heat exchangers network synthesis is a multi-variable multi-dimensional optimization problem in which the total network driving distribution depends on each stream conditions and each hot stream minimum approach temperature for heat recovery. Such variables can contribute to number of units, shells, and both the heating and cooling utilities requirements as well as its mix. In pinch technology this multi-variable optimization problem has been reduced to a single variable optimization problem which is the global ∆T\_min of the problem that in pinch technology needs to be optimized.

Recent advances in the field of energy efficiency optimization advocate the need to conduct the waste heat recovery targeting phase using several ∆T\_min (minimum temperature approach between hot and cold composite curves) and/or using problem process conditions

nowadays can play role in determining the feasibility of HCU energy efficiency/emission reduction project. For instance, the annual potential revenue of the CDM project can be estimated to be \$ 950,000 per year using an average price of \$25 carbon credit per ton [1:3].

Pinch analysis is the technology that provides a systematic methodology for energy saving in processes and total sites. The methodology is based upon thermodynamic principles. Pinch Analysis/technology was first developed in the late 1970s as a technique for optimization of thermal heat recovery, and rapidly gained wide acceptance as a practical approach to the design of Heat Exchanger Networks (HENs). Since then, it has evolved into a general methodology for optimization, based on the principles of process integration.

The technique calculates thermodynamically attainable energy targets for a given process and identifies how to achieve them. A key insight is the pinch temperature, which is the most constrained point in the process. The most detailed explanation of the techniques is by Linnhoff et al. Other pinch analyses were developed for several applications such as massexchange networks (El-Halwagi and Manousiouthakis, 1989), water minimization (Wang

Pinch analysis has been applied successfully not only to energy systems (heat recovery, pressure drop recovery, power generation), but also to fresh water conservation, wastewater minimization, production capacity de-bottlenecking, and management of chemical species

Applying Pinch technology in heat exchangers networks (HEN) synthesis and retrofit, enable the engineer to calculate the energy requirement for any process, and produce thermally efficient and practical designs. Energy savings are significant compared to previous best designs. Pinch technology is also applied to the optimization/integration of the supply-side, consisting of on-site utilities, such as boilers, furnaces, steam and gas

Generally speaking, the objective of the heat exchangers network synthesis for waste heat recovery problem is to design a network that meets an economic criterion such as minimum total annualized cost. Sometimes minimum number of units and minimum energy consumption and GHG emissions in special applications become very legitimate objectives

The heat exchangers network synthesis is a multi-variable multi-dimensional optimization problem in which the total network driving distribution depends on each stream conditions and each hot stream minimum approach temperature for heat recovery. Such variables can contribute to number of units, shells, and both the heating and cooling utilities requirements as well as its mix. In pinch technology this multi-variable optimization problem has been reduced to a single variable optimization problem which is the global ∆T\_min of the

Recent advances in the field of energy efficiency optimization advocate the need to conduct the waste heat recovery targeting phase using several ∆T\_min (minimum temperature approach between hot and cold composite curves) and/or using problem process conditions

and Smith, 1994), and material recycle (El-Halwagi et al., 2003).

turbines, cogeneration, heat pumps, and refrigeration systems [4:6].

problem that in pinch technology needs to be optimized.

**3. Inter-processes integration for enhanced energy efficiency and energy-**

**based GHG emissions reduction** 

in complex processes.

too.

soft constraints in a systematic way to find better energy targets. New method uses for each hot stream specific ∆T\_min and allows minor possible combinations of process conditions to be modified (e.g. 5 F ± in the supply and target temperatures) to customizing the waste heat recovery problem in a way that renders better reductions in energy consumption cost( energy quantity and/or quality/work optimization). The waste energy recovery problem using such new targeting method can now has several degrees of freedom for the waste energy recovery problem optimization, instead of the one parameter problem optimization currently used [7:9].

The above mentioned energy integration targeting methods while can be used at any scale is nowadays focused only on industrial facility via direct integration between the hot and cold streams of its process units. It is usually applied at the process level; and known as intraprocess integration. It has proved to be very successful to reducing both energy consumption and energy-based GHG emissions. Integration among many processes, in adjacent geographical locations, can bring in more degrees of freedom to optimize the waste energy recovery problem and consequently presents new horizon to the energy-based GHG emissions reduction to levels never thought of before. For many reasons, direct interprocesses integration is not widely practiced in industry. Many of the reasons hindering the application of inter-process integration among several processes for better energy consumption cost reduction and less energy-based GHG emissions are very valid and need to be addressed in a novel way to enable wider adaptation of direct inter-process integration in existing industrial facilities and naturally for mega facilities, zones and even cities in the future plants.

Since the emanation of the pinch technology and its evolution to pinch analysis technique for process synthesis, the direct inter-processes integration has been considered impractical. Many arguments, such as the processes are normally have different start up and shut down times; the processes can work at partial loads; the processes can have seasonal changes in its conditions; utility systems, heaters and HEN capital will not be reduced due to changes in processes schedule and operation philosophy; the disturbance in one process can propagate to another one if they are integrated; the distance-time/velocity lags affect the controllability of processes; the geographical distances among processes will cost us energy in pumping or compression and capital in piping, pumping and compression; safety might be impacted due to the travel of a fluid from one hazardous area to another; the fear of leakage and so on, are certainly valid. Therefore, direct inter-processes integration while is beneficial to energy conservation and the GHG emissions reduction, is still to date almost ignored in industrial community.

Due to those concerns most of the current methods for inter-process integration are indirect and conducted using buffer systems. Buffer systems are either steam system or hot oil system or sometimes both. However, the industrial community does agree that, direct integration approach in inter-process integration (between several plants) is more efficient and can render more saving in energy consumption and energy-based greenhouse gas emissions [10:16]. In this chapter we demonstrate, that huge potential for energy consumption and GHG emissions reduction in oil refining (more than 5 % in-house) can be attained through smart integration among processes.

It is instructive to note here that, while we agree with the validity of those arguments we believe that we can still find a room for improvement in energy efficiency enhancement and

GHG Emissions Reduction Via Energy Efficiency Optimization 87

The details of each stream data as presented by Fraser and Gillespie [18], are shown in tables

**1- FCCU**  Type No Ts (oC) Tt (oC) Heat Capacity (KW/oC)

**2- CDU/VDU**  Type No Ts (oC) Tt (oC) Heat Capacity (KW/oC)

H 1 172.5 67.6 116.951 H 2 260.0 189.8 75.104 H 3 309.0 269.5 95.138 H 4 333.4 189.4 14.91 H 5 116.8 49.7 72.242 H 6 272.0 210.0 303.711 H 7 210.0 79.8 58.8 H 8 146.0 18.2 144.92 H 9 50.5 18.2 152.687 H 10 189.0 26.1 69.661 H 11 198.9 171.1 13.477 C 12 26.0 261.7 221.887 C 13 261.7 356.5 430.191 C 14 338.2 409.8 257.147 C 15 26.7 96.1 136.283

H 1 165.5 90.0 22.616 H 2 282.0 796.5 54.389 H 3 274.0 37.5 9.163 H 4 164.0 27.0 36.141 H 5 327.0 261.0 44.321 H 6 363.0 246.0 26.76 H 7 327.0 165.0 16.772 H 8 201.0 104.0 5.405 H 9 140.9 38.0 162.055 H 10 144.5 51.0 15.252 C 11 74.0 295.0 62.462 C 12 143.0 164.0 129.383 C 13 94.0 125.0 126.44

3, 4, 5 & 6.below.

Table 3. Stream data for the FCCU

Table 4. Stream data for the CDU/VDU

energy-based GHG emissions reduction via identifying at the energy integration targeting phase best possible scenarios for inter-processes integration and then trying to find cost effective solutions to the above mentioned concerns via smart plants "matching".

The general intuition regarding the integration among several processes that assumes, the more you integrate among the processes, the better you save in energy consumption and consequently in the energy-based GHG emissions sometimes are not always fully true.

## **3.1 Direct inter-processes integration case study data**

For N plants, there will be a cretin number of possible inter-process integration combinations. Starting from 1 for a single plant, 2 combinations for two plants, 5 combinations for three plants, 15 combinations for four plants (our example) up to 4.6386x10^18 possible combinations for 25 plants only. These possible combinations are identified through Bell's numbers. The Bell numbers (1, 1, 2, 5, 15, 52, 203, 877, 4140, 21147, 115975, 678570, 4213597, ...) describe the number of ways a set with n elements can be partitioned into disjoint, non-empty subsets [17].

For example, the set {1, 2, 3} can be partitioned in the following ways:


In this chapter, the data used for the demonstration of the possible energy consumption and GHG emissions reduction due to inter-processes integration method is a four existing plants in a typical oil refinery. It is used to illustrate the useful impact of inter- process integration between different plants on both energy consumption and energy-based GHG emissions reduction. These data extracted from literature are presented below. We selected the heat recovery approach temperature for the whole problem to be 15 C [18]. Our intention for such selection is to use reasonable value for our method testing and not to compare with results of other methods. For the purpose of exhibiting the method of selecting which units to integrate inside a whole facility and the potential impact of the optimal selection on both energy consumption and GHG emissions reduction we considered from the whole refinery, the big energy consumers (the elephants in energy consumption) such as Fluid Catalytic Cracking Unit (FCCU), Crude/Vacuum Distillation Unit (CDU/VDU), Visbreaker/thermal cracking Unit (VBU) and Platformer/reformer Unit (PLAT). The four plants selected for the inter-processes direct integration targeting study presented in this chapter are numbered as per the table below.


Table 2. The four plants and their referenced numbers

energy-based GHG emissions reduction via identifying at the energy integration targeting phase best possible scenarios for inter-processes integration and then trying to find cost

The general intuition regarding the integration among several processes that assumes, the more you integrate among the processes, the better you save in energy consumption and consequently in the energy-based GHG emissions sometimes are not always fully true.

For N plants, there will be a cretin number of possible inter-process integration combinations. Starting from 1 for a single plant, 2 combinations for two plants, 5 combinations for three plants, 15 combinations for four plants (our example) up to 4.6386x10^18 possible combinations for 25 plants only. These possible combinations are identified through Bell's numbers. The Bell numbers (1, 1, 2, 5, 15, 52, 203, 877, 4140, 21147, 115975, 678570, 4213597, ...) describe the number of ways a set with n elements can be

In this chapter, the data used for the demonstration of the possible energy consumption and GHG emissions reduction due to inter-processes integration method is a four existing plants in a typical oil refinery. It is used to illustrate the useful impact of inter- process integration between different plants on both energy consumption and energy-based GHG emissions reduction. These data extracted from literature are presented below. We selected the heat recovery approach temperature for the whole problem to be 15 C [18]. Our intention for such selection is to use reasonable value for our method testing and not to compare with results of other methods. For the purpose of exhibiting the method of selecting which units to integrate inside a whole facility and the potential impact of the optimal selection on both energy consumption and GHG emissions reduction we considered from the whole refinery, the big energy consumers (the elephants in energy consumption) such as Fluid Catalytic Cracking Unit (FCCU), Crude/Vacuum Distillation Unit (CDU/VDU), Visbreaker/thermal cracking Unit (VBU) and Platformer/reformer Unit (PLAT). The four plants selected for the inter-processes direct integration targeting study presented in this chapter are numbered as

**Plant** FCCU CDU/VDU VBU PLAT

**Number** 1 2 3 4

effective solutions to the above mentioned concerns via smart plants "matching".

**3.1 Direct inter-processes integration case study data** 

For example, the set {1, 2, 3} can be partitioned in the following ways:

partitioned into disjoint, non-empty subsets [17].

**Reference** 

Table 2. The four plants and their referenced numbers

 {{1}, {2}, {3}} {{1, 2}, {3}} {{1, 3}, {2}} {{1}, {2, 3}} {{1, 2, 3}}.

per the table below.

The details of each stream data as presented by Fraser and Gillespie [18], are shown in tables 3, 4, 5 & 6.below.


Table 3. Stream data for the FCCU


Table 4. Stream data for the CDU/VDU

GHG Emissions Reduction Via Energy Efficiency Optimization 89

As mentioned earlier, the four plants that were considered are FCCU (1), CDU/VDU (2), VBU (3) & PLAT(4). The 15 possible combinations which include 14 possible inter-process integration schemes (A to N) are listed in table 7 below. Combination labeled O is the no inter-process integration case where each plant is in a standalone intra-process integration

Sets **Combination Name** 

Before we calculate the energy targets for each combination above using pinch technology/process integration technique presented earlier in this chapter, let us try to test our intuition regarding the expected output regarding the combination that render best

Our first hypothesis/intuition is that, the energy consumption values due to inter-process integration among the four units, all as one unit, is going to be better than the intra-process integration/ standalone unit integration. If there is no benefit from inter-process integration, than standalone intra-process integration. Then, there is no point to spend capital for integration among processes; complicating the plant start-up and its abnormal situations management without saving in energy consumption and reduction in GHG emissions. This correct hypothesis means that set A in the table above, where all four units are treated as one unit is going to be the best set and set O in the same table above, where each unit is handled independently, is going to be the worst set from energy consumption point of view and naturally from energy-based GHG emissions too. In other words from energy consumption and energy-based GHG emissions, combination A is the upper limit of possible energy saving and GHG emissions reduction. The same logic that advocate the integration of all units to get best energy consumption saving, will lead us to a general crude intuition thinking that always, the more we integrate plants; the better reduction in both energy requirements and Greenhouse Gases emissions as well we get. This untested second

{1,2,3,4} A {1,2,3}{4} B {1,2,4}{3} C {1,2}{3,4} D {1,2}{3}{4} E {1,3,4}{2} F {1,3}{2,4} G {1,3}{2}{4} H {1,4}{2,3} I {1}{2,3,4} J {1}{2,3}{4} K {1,4}{2}{3} L {1}{2,4}{3} M {1}{2}{3,4} N {1}{2}{3}{4} O

Table 7. The 15 possible combinations for 4 plants

energy consumptions and the rank of each combination.

**3.2 Direct inter-processes integration case study results and discussion** 

status.


Table 5. Stream data for the VBU


Table 6. Stream data for the PLAT

**3- VBU**  Type No Ts (oC) Tt (oC) Heat Capacity (KW/oC)

**4- PLAT**  Type No Ts (oC) Tt (oC) Heat Capacity (KW/oC)

H 1 503.9 366.1 67.382 H 2 366.1 178.9 3.807 H 3 366.1 253.9 26.094 H 4 303.3 36.7 63.175 H 5 76.7 26.7 24.568 H 6 232.2 112.2 15.107 H 7 79.4 32.2 39.78 H 8 112.0 23.9 0.738 H 9 67.2 27.2 75.556 H 10 157.2 32.2 7.905 H 11 43.3 26.3 4.773 H 12 92.0 65.0 1.773 H 13 107.0 32.2 7.671 C 14 66.1 370.6 76.121 C 15 232.2 247.2 195.269 C 16 36.7 125.6 20.378 C 17 112.0 112.8 2506.929 C 18 157.2 163.9 106.929 C 19 92.0 97.2 38.98 C 20 370.6 495.6 94.091 C 21 452.8 497.2 106.519 C 22 480.6 496.1 114.065

H 1 135.6 30.0 19.506 H 2 255.0 176.1 9.887 H 3 353.3 198.9 43.165 H 4 198.9 171.1 14.477 H 5 171.1 75.0 19.279 C 6 327.8 457.8 54.685 C 7 158.3 160.0 221.49 C 8 126.7 176.7 15.599 C 9 126.7 176.7 133.329 C 10 126.7 146.7 21.364

Table 5. Stream data for the VBU

Table 6. Stream data for the PLAT

## **3.2 Direct inter-processes integration case study results and discussion**

As mentioned earlier, the four plants that were considered are FCCU (1), CDU/VDU (2), VBU (3) & PLAT(4). The 15 possible combinations which include 14 possible inter-process integration schemes (A to N) are listed in table 7 below. Combination labeled O is the no inter-process integration case where each plant is in a standalone intra-process integration status.


Table 7. The 15 possible combinations for 4 plants

Before we calculate the energy targets for each combination above using pinch technology/process integration technique presented earlier in this chapter, let us try to test our intuition regarding the expected output regarding the combination that render best energy consumptions and the rank of each combination.

Our first hypothesis/intuition is that, the energy consumption values due to inter-process integration among the four units, all as one unit, is going to be better than the intra-process integration/ standalone unit integration. If there is no benefit from inter-process integration, than standalone intra-process integration. Then, there is no point to spend capital for integration among processes; complicating the plant start-up and its abnormal situations management without saving in energy consumption and reduction in GHG emissions. This correct hypothesis means that set A in the table above, where all four units are treated as one unit is going to be the best set and set O in the same table above, where each unit is handled independently, is going to be the worst set from energy consumption point of view and naturally from energy-based GHG emissions too. In other words from energy consumption and energy-based GHG emissions, combination A is the upper limit of possible energy saving and GHG emissions reduction. The same logic that advocate the integration of all units to get best energy consumption saving, will lead us to a general crude intuition thinking that always, the more we integrate plants; the better reduction in both energy requirements and Greenhouse Gases emissions as well we get. This untested second

GHG Emissions Reduction Via Energy Efficiency Optimization 91

also integrated together. The following next best scenarios (6, 7, 8) are the reverse of this assumption. It is important to note here that the testing of the above hypothesis based upon the two suggested schemes A and B is going to be proved unsatisfactory even though the hypothesis looks logical and we cannot use this hypothesis as a rule for inter-process integration scenario selection. We will shortly see such conclusion after calculating the energy targets for each case, in the suggested schemes A and B, and discussing the findings from these calculations. As shown in table 9, the heating (Qh) and cooling (Qc) utilities targets, (at ∆T\_min=15 C) were calculated for each set of the 15 possible combinations identified for the four plants. Then, the minimum heating and cooling requirements for each set are calculated followed by the ranking of each set based on the total minimum energy requirements. The table below shows in the first column the sets list of all possible plants integration combinations, the second column is the combination label, the third column and the fourth column are the minimum heating and minimum cooling utilities requirement Qh1 & Qc1 respectively of the whole 4 plants together. It is useful to note that, the Qh1 & Qc1, Qh2 & Qc2, Qh3 & Qc3, and Qh4 & Qc4 are the minimum heating and minimum cooling requirements for the plants between brackets {}. For instance, in combination G, the Qh1 &Qc1 are the minimum heating and minimum cooling utilities requirements of plants {1,3} and the Qh2 &Qc2 are the minimum heating and minimum cooling utilities requirements of plants {2,4}. Qh total and Qc total required in G combination are the

summation of Qh1 & Qh2 and Qc1& Qc2, respectively.

Table 9. The 15 possible combinations for 4 plants

now that table 9 gives us a negative answer to such hypothesis.

Let us now have a deep look to table 9, "Rank" column. The table gives us the answer to our hypothesis test/question regarding what is the second best inter-process integration after the all together four plant inter-processes integration?. This question is important to us since full integration among all plants might be costly and impractical. Our hypothesis suggested adapting either one of two schemes (scheme A or scheme B) depicted in table 8. It is clear

hypothesis, is deciding the next best combination after the A combination (in which the four plants/units were all together integrated from energy consumption point of view). This hypothesis says, if you cannot integrate the four plants, rendering best energy saving, all together at least try to integrate three of them, and if you cannot integrate the three plants all together try to integrate two of them in pairs to get the best possible energy saving and energy-based GHG emissions reduction. That is the intuition obtained from combination A that shown the superiority of inter-processes integration on the standalone intra-process integration and can lead us to false conclusion, if we do not test this hypothesis.

To test this hypothesis we suggest two possible schemes. Our hypothesis is that, the next best integration among the processes may be either the one that has the highest number of integrated processes in a set or the one that has the highest number of integrations/pairings in the set. For example we need to test and find an answer to the following questions: Which combination is better, is it the B combination? where we have three units integrated together and one unit is in standalone intra-process integration status (i.e. one integration in the set) or is it the G combination? Where in this combination we have two integrations in the set (where unit 1 and unit 3 are integrated and unit 2 and unit 4 are integrated). The test procedure to the above second hypothesis will lead us to either one of the two schemes A and B shown in the table 8 below.


Table 8. The 15 possible combinations for 4 plants

The above two schemes says that the best combinations for the inter-processes integration is naturally the 4 plants all together. The next best (2, 3, 4, 5) are either the integration of three plants all together and the fourth is a standalone one as per scheme A or the two plants integrations in pairs as per scheme B, where for instance process plant/unit 1 and process plant/unit 2 are integrated together and process plant/unit 3 and process plant/unit 4 are

hypothesis, is deciding the next best combination after the A combination (in which the four plants/units were all together integrated from energy consumption point of view). This hypothesis says, if you cannot integrate the four plants, rendering best energy saving, all together at least try to integrate three of them, and if you cannot integrate the three plants all together try to integrate two of them in pairs to get the best possible energy saving and energy-based GHG emissions reduction. That is the intuition obtained from combination A that shown the superiority of inter-processes integration on the standalone intra-process

To test this hypothesis we suggest two possible schemes. Our hypothesis is that, the next best integration among the processes may be either the one that has the highest number of integrated processes in a set or the one that has the highest number of integrations/pairings in the set. For example we need to test and find an answer to the following questions: Which combination is better, is it the B combination? where we have three units integrated together and one unit is in standalone intra-process integration status (i.e. one integration in the set) or is it the G combination? Where in this combination we have two integrations in the set (where unit 1 and unit 3 are integrated and unit 2 and unit 4 are integrated). The test procedure to the above second hypothesis will lead us to either one of the two schemes A

**Scheme A Scheme B # Combinations Mix # Combinations Mix**  1 {1,2,3,4} All 1 {1,2,3,4} All

3 {1,3,4}{2} 3 {1,3}{2,4} 2-2 only

6 {1,3,4}{2}

9 {1,2}{3}{4}

1-3 or 3-1

2-1-1 only

2 {1}{2,3,4} 1-3 or 3-1 2 {1,2}{3,4}

4 {1,2,4}{3} 4 {1,4}{2,3} 5 {1,2,3}{4} 5 {1}{2,3,4}

7 {1,3}{2,4} 7 {1,2,4}{3} 8 {1,4}{2,3} 8 {1,2,3}{4}

10 {1,3}{2}{4} 10 {1,3}{2}{4} 11 {1,4}{2}{3} 11 {1,4}{2}{3} 12 {1}{2,3}{4} 12 {1}{2,3}{4} 13 {1}{2,4}{3} 13 {1}{2,4}{3} 14 {1}{2}{3,4} 14 {1}{2}{3,4}

Table 8. The 15 possible combinations for 4 plants

2-2 only

2-1-1 only

15 {1}{2}{3}{4} 1-1-1-1 15 {1}{2}{3}{4} 1-1-1-1

The above two schemes says that the best combinations for the inter-processes integration is naturally the 4 plants all together. The next best (2, 3, 4, 5) are either the integration of three plants all together and the fourth is a standalone one as per scheme A or the two plants integrations in pairs as per scheme B, where for instance process plant/unit 1 and process plant/unit 2 are integrated together and process plant/unit 3 and process plant/unit 4 are

integration and can lead us to false conclusion, if we do not test this hypothesis.

and B shown in the table 8 below.

6 {1,2}{3,4}

9 {1,2}{3}{4}

also integrated together. The following next best scenarios (6, 7, 8) are the reverse of this assumption. It is important to note here that the testing of the above hypothesis based upon the two suggested schemes A and B is going to be proved unsatisfactory even though the hypothesis looks logical and we cannot use this hypothesis as a rule for inter-process integration scenario selection. We will shortly see such conclusion after calculating the energy targets for each case, in the suggested schemes A and B, and discussing the findings from these calculations. As shown in table 9, the heating (Qh) and cooling (Qc) utilities targets, (at ∆T\_min=15 C) were calculated for each set of the 15 possible combinations identified for the four plants. Then, the minimum heating and cooling requirements for each set are calculated followed by the ranking of each set based on the total minimum energy requirements. The table below shows in the first column the sets list of all possible plants integration combinations, the second column is the combination label, the third column and the fourth column are the minimum heating and minimum cooling utilities requirement Qh1 & Qc1 respectively of the whole 4 plants together. It is useful to note that, the Qh1 & Qc1, Qh2 & Qc2, Qh3 & Qc3, and Qh4 & Qc4 are the minimum heating and minimum cooling requirements for the plants between brackets {}. For instance, in combination G, the Qh1 &Qc1 are the minimum heating and minimum cooling utilities requirements of plants {1,3} and the Qh2 &Qc2 are the minimum heating and minimum cooling utilities requirements of plants {2,4}. Qh total and Qc total required in G combination are the summation of Qh1 & Qh2 and Qc1& Qc2, respectively.


Table 9. The 15 possible combinations for 4 plants

Let us now have a deep look to table 9, "Rank" column. The table gives us the answer to our hypothesis test/question regarding what is the second best inter-process integration after the all together four plant inter-processes integration?. This question is important to us since full integration among all plants might be costly and impractical. Our hypothesis suggested adapting either one of two schemes (scheme A or scheme B) depicted in table 8. It is clear now that table 9 gives us a negative answer to such hypothesis.

GHG Emissions Reduction Via Energy Efficiency Optimization 93

GHG emissions calculation the relationship suggested by Smith [5]. For each MW of heat saved in a furnace using fuel gas and has about 90 % firing efficiency, the fuel gas saved is going to reduce the amount of CO2 emissions from that furnace by 300 kg. We are also

> **Ranking of Best Scenarios Rank Mix Combinations** 1 All {1,2,3,4} 2 1-3 or 3-1 {1,2,4}{3} 3 1-3 or 3-1 {1}{2,3,4} 4 1-3 or 3-1 {1,2,3}{4} 5 2-2 only {1,2}{3,4} 6 2-2 only {1,3}{2,4} 7 2-1-1 only {1}{2,4}{3} 8 2-2 only {1,4}{2,3} 9 2-1-1 only {1,2}{3}{4} 10 2-1-1 only {1}{2,3}{4} 11 1-3 or 3-1 {1,3,4}{2} 12 2-1-1 only {1,4}{2}{3} 13 2-1-1 only {1}{2}{3,4} 14 2-1-1 only {1,3}{2}{4} 15 1-1-1-1 {1}{2}{3}{4}

Table 10. The proposed best scenario for the 4 plants problem inter-process integration

Qh (KW) 29,751 55,801 6,944 18,885 **Qh total= 111,382**  Qc (KW) 17,550 24,800 3,344 9,245 **Qc total= 54,940** 

Qh (KW) 29,016 52,832 6,676 17,946 **Qh total= 106,470**  Qc (KW) 16,815 21,831 3,076 8,306 **Qc total= 50,028** 

∆T\_min=15 C

∆T\_min=5 C

**1 2 3 4** 

FCCU CDU/VDU VBU PLAT

**1 2 3 4** 

FCCU CDU/VDU VBU PLAT

Temp. 158 272 141.7 81.1

Temp. 148 266.7 131.7 79.4

Table 12. Minimum Heating/Cooling requirements at ∆T\_min= 5 C

Table 11. Minimum Heating/Cooling requirements at ∆T\_min=15 C

Pinch

Pinch

ignoring the GHG emissions saving due to the reduction in plant cooling utility.

Based on the calculations and numbers in the rank column in table 9, we generated table 10, to definitely answer the question of, What is the next best inter-process integration scenario? that comes after the four plants all-together inter-process integration. Table 10 shows few ranks different than our expectation. It is a fact that the optimum solution (ranked 1) happens when there is a full integration between all the four plants and the worst from energy saving point of view (rank 15) occurs when there is no integration between any of the four plants. In between these two scenarios, the sequence continues with the logic that says that better solutions are reached with the scenario that has higher number of integrated plants combinations ( the ones that has three plants all together and one standalone plant) such as ({1,2,4} rank 2, {2,3,4} rank 3, and finally {1,2,3}) rank 4. However, this is not always true since rank 11 ({1, 3, 4}, {2}) broke the expected priority/sequence and half of the six 2-1- 1 mix (ranks 7, 9 and 10) which contain one combination only( least possible inter-process integration), provides better energy consumption and less energy-based GHG emissions solution than it. These better combinations are as follows respectively: ({1}, {2, 4}, {3}) and ({1, 2}, {3}, {4}) and ({1}, {2, 3} , {4}). It says that for the specific refinery application data used in this chapter and the ∆T\_min=15 C used if you only integrate the CDU/VDU with the reformer/PLAT or the FCCU or the VBU you will have less process design complication and better energy consumption saving than integrating all the plants together without the ADU/VDU, left as a standalone. Another finding from table 10 calculations ranking is that, the 2-2 inter-processes integration scenarios in which every two plants are integrated together most of the 2-2 scenarios, two out of three possible combinations, are better than the 2-1-1 scenarios of integration (six possible combinations in which only two plants are integrated and the rest are standalone plants). This one combination which is not following the hypothesis now is the 2-2 mix (rank 8) that contains the combination ({1, 4} {2, 3}). Rank 7 combination which is only one process-to process matching is better not much but it is less complication in the process design and better energy consumption saving and consequently better reduction in energy-based GHG emissions. It means that complexity in process design is not always mandatory to save energy. In our refining application here the above finding tell us that integrating the ADU/VDU with PLAT is the best scenario (when we are only allowed one process to process integration to do). integration/matching ADU/VDU with the "wrong process" such as VBU can bring in negative impact. In summary, it is clear that all 3-1 mix sets are better than all the 2-1-1 mix sets and all the 2-2 mix sets except one 3-1 mix set (rank 11) where it has higher energy requirements than three 2-1-1 mix sets (rank 7, 9 and 10) and all 2-2 mix sets (rank 5, 6 and 8). The second best sets are all the 2-2 mix sets but the rank 8 set.

In order to evaluate the concept of inter-processes integration on the energy-based GHG emissions, we calculated the energy consumption of the standalone plants at different ∆T\_min and listed the results in tables 11, 12 and 13 as shown below. The minimum heating and cooling requirements are shown for each plant as a standalone assuming perfect intra-process integration and with no inter-process integration between plants at several minimum approach temperatures (∆T\_min) using ∆T\_min=15 , 5 and 1 C respectively. It is well known to the experienced in the field of energy efficiency optimization that when the ∆T\_min is reduced both the minimum heating and minimum cooling utilities requirement are reduced and the energy-based GHG emissions will be decreasing as well. It is also accepted to the experienced in the field that the heat exchangers network capital cost that achieves such saving in energy consumption, most of the time, will be increasing. For the sake of simplicity in calculating the GHG emissions reduction associated with energy saving, we are using for the energy-based

Based on the calculations and numbers in the rank column in table 9, we generated table 10, to definitely answer the question of, What is the next best inter-process integration scenario? that comes after the four plants all-together inter-process integration. Table 10 shows few ranks different than our expectation. It is a fact that the optimum solution (ranked 1) happens when there is a full integration between all the four plants and the worst from energy saving point of view (rank 15) occurs when there is no integration between any of the four plants. In between these two scenarios, the sequence continues with the logic that says that better solutions are reached with the scenario that has higher number of integrated plants combinations ( the ones that has three plants all together and one standalone plant) such as ({1,2,4} rank 2, {2,3,4} rank 3, and finally {1,2,3}) rank 4. However, this is not always true since rank 11 ({1, 3, 4}, {2}) broke the expected priority/sequence and half of the six 2-1- 1 mix (ranks 7, 9 and 10) which contain one combination only( least possible inter-process integration), provides better energy consumption and less energy-based GHG emissions solution than it. These better combinations are as follows respectively: ({1}, {2, 4}, {3}) and ({1, 2}, {3}, {4}) and ({1}, {2, 3} , {4}). It says that for the specific refinery application data used in this chapter and the ∆T\_min=15 C used if you only integrate the CDU/VDU with the reformer/PLAT or the FCCU or the VBU you will have less process design complication and better energy consumption saving than integrating all the plants together without the ADU/VDU, left as a standalone. Another finding from table 10 calculations ranking is that, the 2-2 inter-processes integration scenarios in which every two plants are integrated together most of the 2-2 scenarios, two out of three possible combinations, are better than the 2-1-1 scenarios of integration (six possible combinations in which only two plants are integrated and the rest are standalone plants). This one combination which is not following the hypothesis now is the 2-2 mix (rank 8) that contains the combination ({1, 4} {2, 3}). Rank 7 combination which is only one process-to process matching is better not much but it is less complication in the process design and better energy consumption saving and consequently better reduction in energy-based GHG emissions. It means that complexity in process design is not always mandatory to save energy. In our refining application here the above finding tell us that integrating the ADU/VDU with PLAT is the best scenario (when we are only allowed one process to process integration to do). integration/matching ADU/VDU with the "wrong process" such as VBU can bring in negative impact. In summary, it is clear that all 3-1 mix sets are better than all the 2-1-1 mix sets and all the 2-2 mix sets except one 3-1 mix set (rank 11) where it has higher energy requirements than three 2-1-1 mix sets (rank 7, 9 and 10) and all 2-2 mix sets (rank 5, 6 and 8). The second best sets are all the 2-2 mix sets

In order to evaluate the concept of inter-processes integration on the energy-based GHG emissions, we calculated the energy consumption of the standalone plants at different ∆T\_min and listed the results in tables 11, 12 and 13 as shown below. The minimum heating and cooling requirements are shown for each plant as a standalone assuming perfect intra-process integration and with no inter-process integration between plants at several minimum approach temperatures (∆T\_min) using ∆T\_min=15 , 5 and 1 C respectively. It is well known to the experienced in the field of energy efficiency optimization that when the ∆T\_min is reduced both the minimum heating and minimum cooling utilities requirement are reduced and the energy-based GHG emissions will be decreasing as well. It is also accepted to the experienced in the field that the heat exchangers network capital cost that achieves such saving in energy consumption, most of the time, will be increasing. For the sake of simplicity in calculating the GHG emissions reduction associated with energy saving, we are using for the energy-based

but the rank 8 set.

GHG emissions calculation the relationship suggested by Smith [5]. For each MW of heat saved in a furnace using fuel gas and has about 90 % firing efficiency, the fuel gas saved is going to reduce the amount of CO2 emissions from that furnace by 300 kg. We are also ignoring the GHG emissions saving due to the reduction in plant cooling utility.


Table 10. The proposed best scenario for the 4 plants problem inter-process integration


Table 11. Minimum Heating/Cooling requirements at ∆T\_min=15 C


Table 12. Minimum Heating/Cooling requirements at ∆T\_min= 5 C

GHG Emissions Reduction Via Energy Efficiency Optimization 95

 One of the famous options in industrial facilities to cut energy consumption cost and reduce emissions is the adaptation of the cogeneration technology. Co-generation or "CHP" is simply known as the production of two forms of useful energy from the same fuel source. Cogeneration systems are used to produce electricity, and use the excess (waste) heat for

There are several types of co-generation plants. One is the steam turbine- based cogeneration plant consisting of a steam turbine with the usual controlled steam extraction(s) for process steam supply. The other type is a gas-turbine- based cogeneration plant consisting of one or more gas turbines exhausting products of combustion through one or more heat-recovery steam generators (HRSGs), which produce steam for the heat supply [19:21]. The thermodynamic efficiency of a cogeneration plant is simply, (P + H)/Q1; where; P and H are the power and heat outputs of the cogeneration plant, respectively and Q1, is heat input .

The oil plant, presented here, is responsible for stabilizing and shipping crude oil and this normally needs large amount of energy mainly in form of steam and power. The plant receives crude oil from different wells and then separates the gas from the oil. The crude oil is stabilized, where light components are separated, and then pumped for shipment to users. The separated gas goes to natural gas liquid process (NGL) plants. Fuel is used by boilers and gas turbine generators. The steam generated from boilers goes to high pressure steam header at 625 psig. High pressure steam is used by back pressure steam turbines throughout the plant to drive oil shipper pumps. The steam coming out from the back pressure steam turbines is sent to the 60 psig header. The 60 psig steam is used to pre-heat

process steam generation, hot water heating, space heating, and other thermal needs.

**4. Combined heat and power plant (CHP) retrofit** 

the crude oil stream before entering crude stabilization column.

Fig. 1. Oil Stabilization Plant CHP Model


Table 13. Minimum Heating/Cooling requirements at ∆T\_min=1 C (impossible)

Comparing the total heating, cooling requirements and consequently the associate energybased GHG emissions is our next task. First we will do the comparison for the standalone intra- process integration at descending ∆T\_min to target for best GHG emission attainable using intra-process integration. Then we will compare the best possible target with the one that can be attained using inter- processes integration at ∆T\_min, equal to 15 C.

Using previously mentioned simple relationship between MW of heat saved and amount of CO2 reduction obtained, we can easily calculate the reduction in CO2 emissions from the refinery biggest energy consumers due to intra-process integration at descending ∆T\_min to be 31202 tone CO2/year and 62196 tone CO2/year respectively. It is 3 to 6 % maximum. Practically we can reach the 3 % but to get better than that we need extremely expensive HEN and the 6 %, at ∆T\_min equal to 1 C, is unattainable/impractical. To reach such 3% reduction in the GHG emissions using ∆T\_min equal 5 C will also need costly HEN design too. The results of direct inter-processes integration at ∆T\_min equal 15 C can render the same amount of GHG emission reduction of %3 and even better through 9 possible scenarios/combinations. Such combinations are listed and ranked from 1 to 9 in table 9. The unattainable 6 % reduction in the energy-based GHG emissions using intra-process integration can be reached and even a little better (up to 10 %) using direct inter-processes integration techniques. Four scenarios/combinations can exhibit such fact, taking the rank from 1 to 4 in table 9. These combinations are as follows: direct inter-process integration of ADU/VDU, FCCU, VBU and PLAT processes all together, FCCU, CDU/VDU and VBU together and PLAT as a standalone, FCCU, CDU/VDU and PLAT together and VBU as a standalone, CDU/VDU, VBU , and PLAT together and FCCU as standalone.

It can be noticed from this discussion that in order to attain more aggressive GHG emissions reduction targets using intra-process integration techniques, even if we tried to use ∆T\_min equal to 1 C which is currently impractical using the available state-of-art heat exchanger technology, we have no way but to use inter-process integration. There are 4 sets/combinations in the inter-process integration application at reasonable ∆T\_min equal to 15 C, which can defeat the minimum heating and cooling requirements and consequently the energy-based GHG emissions of the almost impossible ∆T\_min equal to 1 C in the intraprocess integration application. That means, for an oil refinery with excellent intra-process integration applications; to obtain in-house GHG emissions reduction target of 6 % or more, as per the case study presented in this chapter, we have to resort to inter-processes integration application. Nowadays, the industrial community perception is that the only way to do such inter-process integration is to make it indirectly via buffer system (steam or hot oil). Such approach is currently adapted in industry but it has its limitations. Such limitations combined with the need to push the envelope and reduce the refining business GHG emissions will lead towards using more of the direct inter-processes integration.

Qh (KW) 28,715 51,557 6,521 17,560 **Qh total= 104,353**  Qc (KW) 16,514 20,556 2,921 7,920 **Qc total= 47,911** 

Comparing the total heating, cooling requirements and consequently the associate energybased GHG emissions is our next task. First we will do the comparison for the standalone intra- process integration at descending ∆T\_min to target for best GHG emission attainable using intra-process integration. Then we will compare the best possible target with the one

Using previously mentioned simple relationship between MW of heat saved and amount of CO2 reduction obtained, we can easily calculate the reduction in CO2 emissions from the refinery biggest energy consumers due to intra-process integration at descending ∆T\_min to be 31202 tone CO2/year and 62196 tone CO2/year respectively. It is 3 to 6 % maximum. Practically we can reach the 3 % but to get better than that we need extremely expensive HEN and the 6 %, at ∆T\_min equal to 1 C, is unattainable/impractical. To reach such 3% reduction in the GHG emissions using ∆T\_min equal 5 C will also need costly HEN design too. The results of direct inter-processes integration at ∆T\_min equal 15 C can render the same amount of GHG emission reduction of %3 and even better through 9 possible scenarios/combinations. Such combinations are listed and ranked from 1 to 9 in table 9. The unattainable 6 % reduction in the energy-based GHG emissions using intra-process integration can be reached and even a little better (up to 10 %) using direct inter-processes integration techniques. Four scenarios/combinations can exhibit such fact, taking the rank from 1 to 4 in table 9. These combinations are as follows: direct inter-process integration of ADU/VDU, FCCU, VBU and PLAT processes all together, FCCU, CDU/VDU and VBU together and PLAT as a standalone, FCCU, CDU/VDU and PLAT together and VBU as a

<sup>∆</sup>T\_min=1 C (impossible) FCCU CDU/VDU VBU PLAT

**1 2 3 4** 

Table 13. Minimum Heating/Cooling requirements at ∆T\_min=1 C (impossible)

that can be attained using inter- processes integration at ∆T\_min, equal to 15 C.

standalone, CDU/VDU, VBU , and PLAT together and FCCU as standalone.

It can be noticed from this discussion that in order to attain more aggressive GHG emissions reduction targets using intra-process integration techniques, even if we tried to use ∆T\_min equal to 1 C which is currently impractical using the available state-of-art heat exchanger technology, we have no way but to use inter-process integration. There are 4 sets/combinations in the inter-process integration application at reasonable ∆T\_min equal to 15 C, which can defeat the minimum heating and cooling requirements and consequently the energy-based GHG emissions of the almost impossible ∆T\_min equal to 1 C in the intraprocess integration application. That means, for an oil refinery with excellent intra-process integration applications; to obtain in-house GHG emissions reduction target of 6 % or more, as per the case study presented in this chapter, we have to resort to inter-processes integration application. Nowadays, the industrial community perception is that the only way to do such inter-process integration is to make it indirectly via buffer system (steam or hot oil). Such approach is currently adapted in industry but it has its limitations. Such limitations combined with the need to push the envelope and reduce the refining business GHG emissions will lead towards using more of the direct inter-processes integration.

Temp. 144 262.7 137.7 79.4

Pinch

## **4. Combined heat and power plant (CHP) retrofit**

 One of the famous options in industrial facilities to cut energy consumption cost and reduce emissions is the adaptation of the cogeneration technology. Co-generation or "CHP" is simply known as the production of two forms of useful energy from the same fuel source. Cogeneration systems are used to produce electricity, and use the excess (waste) heat for process steam generation, hot water heating, space heating, and other thermal needs.

There are several types of co-generation plants. One is the steam turbine- based cogeneration plant consisting of a steam turbine with the usual controlled steam extraction(s) for process steam supply. The other type is a gas-turbine- based cogeneration plant consisting of one or more gas turbines exhausting products of combustion through one or more heat-recovery steam generators (HRSGs), which produce steam for the heat supply [19:21]. The thermodynamic efficiency of a cogeneration plant is simply, (P + H)/Q1; where; P and H are the power and heat outputs of the cogeneration plant, respectively and Q1, is heat input .

The oil plant, presented here, is responsible for stabilizing and shipping crude oil and this normally needs large amount of energy mainly in form of steam and power. The plant receives crude oil from different wells and then separates the gas from the oil. The crude oil is stabilized, where light components are separated, and then pumped for shipment to users. The separated gas goes to natural gas liquid process (NGL) plants. Fuel is used by boilers and gas turbine generators. The steam generated from boilers goes to high pressure steam header at 625 psig. High pressure steam is used by back pressure steam turbines throughout the plant to drive oil shipper pumps. The steam coming out from the back pressure steam turbines is sent to the 60 psig header. The 60 psig steam is used to pre-heat the crude oil stream before entering crude stabilization column.

Fig. 1. Oil Stabilization Plant CHP Model

GHG Emissions Reduction Via Energy Efficiency Optimization 97

Consider a power demand of 53 MW; the corresponding fuel input to simple cycle gas

At winter time the units have to operate at full load (40 MW each) and generate steam (340 Klb/h) of steam. The plant power need only 53 MW, so there will be an access power of 27

For calculating the useful heat energy, assuming 100% utilization. The steam Internal energy (Enthalpy) for 625 psig, 380 F steam header is 1370 Btu/lb and BFW enthalpy is about 270

= (1370 – 270) \* 340,000 = 374 MMBtu/hr

Fuel Input (Btu/hr) = 773,075 SCF/hr \* 1090 Btu/SCF = 842.6 MMBtu/hr.

Having the two co-generation units installed, and the two old boilers demolished. Let's see

 SO2 (g/s) = [0.126\*(16\*32\*2)/(0.7302\*527)/1000] \* (Fuel Gas Consumption) (2) Knowing, the average availability of the two boilers per year and their consumption and using the above formulas (1) & (2) gives a reduction on total emissions of 1.25 metric ton per day. Other benefit appears when converting to co-generation cycle is that the heat going to

In summary, the adaptation of co-generation technology in the oil plant exhibits economical and environmental benefits over the simple cycle power generation and stand alone steam boilers. Well designed and operated cogeneration plant will always improve energy efficiency for systems requiring steam and power in oil and gas facilities. The typical energy

NOx (g/s) = 0.32 \* 0.126 \* (Fuel Gas Consumption) (1)

ி௨ ൌ ௪ାௌ௧ ி௨

Power (MMBtu/hr) = 53,000 KW \* 3412.14 Btu/KWh = 180.8 MMBtu/hr.

Fuel Input (MMBtu/hr) = 532,860 SCF/hr \* 1090 Btu/SCF = 580.8 MMBtu/hr.

Co-generation cycle (µ) = ை௨௧௨௧

MW; let's assume that the excess power has no value. Calculating the efficiency:

Power (MMBtu/hr) = 53,000 KW \* 3412.14 Btu/KWh = 180.8 MMBtu/hr.

the ambient will be significantly reduced from 1000 F to about 300 F.

Steam heat energy = ∆H (Btu/lb) \* Steam Flow (Lb/hr)

Where,

Btu/lb.

Power = electrical power output in Btu/h

Fuel = Fuel energy input in Btu/h

turbines is 532.86 KSCF/hr.

Simple cycle (µ) = 31.13 %.

Co-generation cycle (µ) = 66%

the impact on the total emissions: NOx & CO2 Emissions calculation:

For the case of cogeneration system:

Fig.1 above gives an overview on the oil plants' utilities model. The oil plant has 10 boilers with total steam production of 4.5 Million bounds per hour (Mlb/h). The maximum steam demand can be found in winter interval where the plant is in need of almost the full boilers capacity. Oil plant historical data shows that the maximum steam demand, at winter time, is equal to 4.4 Mlb/h, and the minimum steam demand, at summer, is equal to 3.2 Mlb/h. The oil plant has two simple cycle gas turbine generators, each capable of generating 40 MW at 38 C. The plant power demand varies between 75 MW in summer and 53 MW in winter and the average power consumption per year is about 60 MW. The oil plant is tied up to the nation-wide electricity grid. Usually the plant internal power generation units are reliable but that is made to avoid any failure that can result in any disruption to the oil plant production.

Retrofitting the existing two simple cycle gas turbines with two heat recovery steam generators (HRSGs) can result in less energy consumption and less GHG emission. The steam generated from the two HRSGs would be equivalent to a production of two old small boilers producing 340 Klb/h of steam. Thus, the fuel used in the two old boilers will be saved in addition to eliminating their emissions.

In order to calculate the fuel saving, it is necessary to have a relation between the fuel consumption and the steam generation of the two old boilers. Such relation can be developed from plant's historical real time data via curve fitting techniques.


Table 14. Two small boilers fuel vs. steam equations

From plant's historical data, the average steam produced from the two small boilers is 340 (thousand bounds per hour) Klb/h; i.e. each producing 170 klb/h.

Fuel savings (avoided) = Boilers Fuel consumption

= (170 1.0917 + 6.4138) + (170 1.3983 - 15.38)

= 414.3338 KSCF/h (thousand standard cubic feet per hour)

Energy saving (MMBtu/h) = Fuel Btu Content \* Boilers fuel consumption

= 1.090 KBtu/SCF 414.3338 KSCF/h = 451.6 MMBtu/h

Note: Assuming HHV of fuel is 1090 Btu/SCF

Now, let's compare the efficiency of simple cycle with the expected co-generation efficiency, for simple cycle system:

$$\text{Simple cycle (\mu)} \frac{Output}{Fuel} = \frac{Power}{Fuel}$$

Where,

96 Greenhouse Gases – Emission, Measurement and Management

Fig.1 above gives an overview on the oil plants' utilities model. The oil plant has 10 boilers with total steam production of 4.5 Million bounds per hour (Mlb/h). The maximum steam demand can be found in winter interval where the plant is in need of almost the full boilers capacity. Oil plant historical data shows that the maximum steam demand, at winter time, is equal to 4.4 Mlb/h, and the minimum steam demand, at summer, is equal to 3.2 Mlb/h. The oil plant has two simple cycle gas turbine generators, each capable of generating 40 MW at 38 C. The plant power demand varies between 75 MW in summer and 53 MW in winter and the average power consumption per year is about 60 MW. The oil plant is tied up to the nation-wide electricity grid. Usually the plant internal power generation units are reliable but that is made to avoid any failure that can result in any disruption to the oil plant

Retrofitting the existing two simple cycle gas turbines with two heat recovery steam generators (HRSGs) can result in less energy consumption and less GHG emission. The steam generated from the two HRSGs would be equivalent to a production of two old small boilers producing 340 Klb/h of steam. Thus, the fuel used in the two old boilers will be

In order to calculate the fuel saving, it is necessary to have a relation between the fuel consumption and the steam generation of the two old boilers. Such relation can be

Boiler # Fuel VS. Steam

From plant's historical data, the average steam produced from the two small boilers is 340

= (170 1.0917 + 6.4138) + (170 1.3983 - 15.38)

= 414.3338 KSCF/h (thousand standard cubic feet per hour)

= 1.090 KBtu/SCF 414.3338 KSCF/h = 451.6 MMBtu/h

Now, let's compare the efficiency of simple cycle with the expected co-generation efficiency,

ி௨ ൌ ௪ ி௨

Simple cycle (µ) =ை௨௧௨௧

B-1 Fuel (MSCFh) = 1.0917 \* Stm (klb/h) + 6.4138 B-2 Fuel (MSCFh) = 1.3983 \* Stm (klb/h) – 15.38

developed from plant's historical real time data via curve fitting techniques.

production.

saved in addition to eliminating their emissions.

Table 14. Two small boilers fuel vs. steam equations

Fuel savings (avoided) = Boilers Fuel consumption

Note: Assuming HHV of fuel is 1090 Btu/SCF

for simple cycle system:

(thousand bounds per hour) Klb/h; i.e. each producing 170 klb/h.

Energy saving (MMBtu/h) = Fuel Btu Content \* Boilers fuel consumption

Power = electrical power output in Btu/h

Fuel = Fuel energy input in Btu/h

Consider a power demand of 53 MW; the corresponding fuel input to simple cycle gas turbines is 532.86 KSCF/hr.

Power (MMBtu/hr) = 53,000 KW \* 3412.14 Btu/KWh = 180.8 MMBtu/hr.

Fuel Input (MMBtu/hr) = 532,860 SCF/hr \* 1090 Btu/SCF = 580.8 MMBtu/hr.

Simple cycle (µ) = 31.13 %.

For the case of cogeneration system:

Co-generation cycle (µ) = ை௨௧௨௧ ி௨ ൌ ௪ାௌ௧ ி௨

At winter time the units have to operate at full load (40 MW each) and generate steam (340 Klb/h) of steam. The plant power need only 53 MW, so there will be an access power of 27 MW; let's assume that the excess power has no value. Calculating the efficiency:

Power (MMBtu/hr) = 53,000 KW \* 3412.14 Btu/KWh = 180.8 MMBtu/hr.

For calculating the useful heat energy, assuming 100% utilization. The steam Internal energy (Enthalpy) for 625 psig, 380 F steam header is 1370 Btu/lb and BFW enthalpy is about 270 Btu/lb.

Steam heat energy = ∆H (Btu/lb) \* Steam Flow (Lb/hr)

= (1370 – 270) \* 340,000 = 374 MMBtu/hr

Fuel Input (Btu/hr) = 773,075 SCF/hr \* 1090 Btu/SCF = 842.6 MMBtu/hr.

Co-generation cycle (µ) = 66%

Having the two co-generation units installed, and the two old boilers demolished. Let's see the impact on the total emissions:

NOx & CO2 Emissions calculation:

$$\text{NOx (g/s)} = 0.32 \, ^\circ 0.126 \, ^\circ \text{ (Fuel Gas Consumption)}\tag{1}$$

$$\text{SO2 (g/s)} = \left[ 0.126^\* (16^\* 32^\* 2) / (0.7302^\* 527) / 1000 \right] \text{\* (Fuel Gas Consumption)} \tag{2}$$

Knowing, the average availability of the two boilers per year and their consumption and using the above formulas (1) & (2) gives a reduction on total emissions of 1.25 metric ton per day. Other benefit appears when converting to co-generation cycle is that the heat going to the ambient will be significantly reduced from 1000 F to about 300 F.

In summary, the adaptation of co-generation technology in the oil plant exhibits economical and environmental benefits over the simple cycle power generation and stand alone steam boilers. Well designed and operated cogeneration plant will always improve energy efficiency for systems requiring steam and power in oil and gas facilities. The typical energy

GHG Emissions Reduction Via Energy Efficiency Optimization 99

[2] Hayashi, D & Michaelowa, A 2007, 'Lessons from Submission and Approval Process of

[3] UNFCCC 2009a, *Efficient Utilization of Waster Energy in Industrial Facility*, New York,

[4] Linnhoff,B.(1993). Pinch analysis-state of the art overview. *Trans.Inst.Chem.Eng.* ,Part A5,

[7] Noureldin, M.B., and A. Hasan,(2006), Global energy targets and optimal operating

[8] Noureldin, M.B and Aseeri, A.S.(2010), System, method and program product for

[9] Noureldin, M.B and Aseeri, A.S.,(2011), Advanced method and software for targeting

[10] Ahmad, S., and Hui, C.W. (1991), Heat recovery between areas of integrity, Computers

[11] Hui, C.W. and Ahmad, S.,(1994), Total site heat integration using the utility system,

[12] Dhole, V.R.; Linhoff B. (1992), Total site targets for fuel, cogeneration, emissions and

[13] Amidpour, M., and Polley, G.T. (1997), Application of problem decomposition in

[14] Rodera, H., and M.J. Bagajewicz,(1999a), Targeting procedures for energy savings by

[15] Rodera, H., and M.J. Bagajewicz,(2001), Multipurpose Heat exchanger networks for heat

[16] Bagajewicz, M.I and Rodera, H..,(2002), Multiple plant heat integration in a total site",

[18] Fraser, D.M., and Gillespie, N.E , (1992), The application of pinch technology to retrofit

[19] Mavromatis, S.P. and Kokossis, A.C., (1998), Conceptual optimization of utility

[20] Varbanov, P.,S., Doyle, S., and Smith, R. (2004), modeling and optimization of utility systems, Trans IChemE, Part A, chem. Eng Res Des, 82(A5), 561-578

energy integration of an entire oil refinery , Transactions of Institute of Chemical

networks for operational variations-I. Targets and level optimization, Che.Eng. Sci,

process integration. Transactions of the Institute of Chemical engineering, 75, 53-

cooling, computers and Chemical engineering, 17, S101-S109

heat integration across plants", *AIChE* Journal, 45, 1721

integration across plants"Ind. Eng.Chem.Res., 40, 5585

conditions for waste-energy recovery in Bisphenol-A plant, Applied Thermal

targeting and optimal driving force distribution in energy recovery systems" US

and optimal driving force distribution in energy recovery systems", US Patent

[5] Smith, R (2005), *Chemical Process Design and Integration*, 2nd edition, Manchester: Wiley [6] El-Halwagi, M..M.,(1997),Pollution prevention through process integration, Academic

Austria, 19-22 March 2007.

USA

503-522.

Press, San Diego

Patent 7,698,002

7,873,443

63

Engineering, 26, 374-381.

and Chemical engineering, 12, 809-832

*AIChE* Journal, 48,(10), 2255-2270 [17] http://mathforum.org/advanced/robertd/bell.html

Engineering, 70, 395-406

53(8), 1585-1608

Computers and Chemical engineering, 18, 729

Large-Scale Energy Efficiency CDM Methodologies', paper presented to UNIDO/CTI/UK Trade and Investment Seminar on Energy Efficiency, Vienna,

efficiency of a co-generation system, normally between 70-90%, simply means significant reduction in CO2, SO2, NOx emissions compared to other stand alone generation of power and heat in oil and gas industry.

## **5. Concluding remarks**

GHG emission reduction can be addressed successfully using energy efficiency optimization techniques in design and operation of industrial facilities. Energy efficiency optimization in process plants, on the equipment level (compressors, boilers, furnaces and so on), subsystem level (combined heating and power CHP), complete process level (crude oil fluid catalytic cracking) and site-wide level (refinery and refinery-chemical integrated facility and even mega industrial complex) is a fast and cost effective way to reduce GHG emissions at the source.

While the industry's already adapted intra-process integration and indirect inter-processes integration bring value to the task of energy consumption and energy-based GHG emissions reduction, aggressive direct inter-processes integration can also be adapted on large scale in industrial community to boost the efforts for energy conservation and GHG emissions reduction. It can enable us stretch the envelope beyond GHG emissions targets and reach better ones in many industrial sites such as in-house oil refining GHG emissions reduction target.

It is instructive to mention that direct inter-process integration that can render us better results beyond the attainable from perfect intra process integration and indirect interprocesses integration do not necessarily have to be done using excessive connections among the plants to reach the desired targets for GHG emissions. From our industrial experience we can surly tell that in many cases two or three connections at most between plants are enough to reach reasonable level of the best desired energy consumption and GHG emissions reduction targets. The details of such finding will be presented in future work.

While we are addressing in this chapter the role of energy efficiency optimization in GHG emissions reduction, we also believe that increasing the use of renewable non-hydrocarbon based energy sources are very efficient way to tackle the GHG emissions problem even on the process plant level. Solar energy in industrial applications especially in remote areas for oil, water pumping and gas compression, and administration areas lighting and cooling can be a very viable. The utilization of solar power for water heating, steam generation and large scale air-conditioning, as solar cooling, is very valid greenhouse gas emissions reduction option. We believe that adapting both energy efficiency optimization and flexible customization of renewable as a clean source of energy at plant's level will be fast and costeffective approach to attain desired GHG emissions reduction targets.

## **6. References**

[1] Birkeland, H & Energi, N 2007, 'Provisions of Small Scale Projects', paper presented to CDM Capacity Building in Servian Institutions, Belgrade, 26 September 2007.

efficiency of a co-generation system, normally between 70-90%, simply means significant reduction in CO2, SO2, NOx emissions compared to other stand alone generation of power

GHG emission reduction can be addressed successfully using energy efficiency optimization techniques in design and operation of industrial facilities. Energy efficiency optimization in process plants, on the equipment level (compressors, boilers, furnaces and so on), subsystem level (combined heating and power CHP), complete process level (crude oil fluid catalytic cracking) and site-wide level (refinery and refinery-chemical integrated facility and even mega industrial complex) is a fast and cost effective way to reduce GHG emissions at

While the industry's already adapted intra-process integration and indirect inter-processes integration bring value to the task of energy consumption and energy-based GHG emissions reduction, aggressive direct inter-processes integration can also be adapted on large scale in industrial community to boost the efforts for energy conservation and GHG emissions reduction. It can enable us stretch the envelope beyond GHG emissions targets and reach better ones in many industrial sites such as in-house oil refining GHG emissions reduction

It is instructive to mention that direct inter-process integration that can render us better results beyond the attainable from perfect intra process integration and indirect interprocesses integration do not necessarily have to be done using excessive connections among the plants to reach the desired targets for GHG emissions. From our industrial experience we can surly tell that in many cases two or three connections at most between plants are enough to reach reasonable level of the best desired energy consumption and GHG emissions reduction targets. The details of such finding will be presented in future

While we are addressing in this chapter the role of energy efficiency optimization in GHG emissions reduction, we also believe that increasing the use of renewable non-hydrocarbon based energy sources are very efficient way to tackle the GHG emissions problem even on the process plant level. Solar energy in industrial applications especially in remote areas for oil, water pumping and gas compression, and administration areas lighting and cooling can be a very viable. The utilization of solar power for water heating, steam generation and large scale air-conditioning, as solar cooling, is very valid greenhouse gas emissions reduction option. We believe that adapting both energy efficiency optimization and flexible customization of renewable as a clean source of energy at plant's level will be fast and cost-

[1] Birkeland, H & Energi, N 2007, 'Provisions of Small Scale Projects', paper presented to CDM Capacity Building in Servian Institutions, Belgrade, 26 September 2007.

effective approach to attain desired GHG emissions reduction targets.

and heat in oil and gas industry.

**5. Concluding remarks** 

the source.

target.

work.

**6. References** 


**5** 

*Canada* 

**Greenhouse Gas Emissions from** 

S. Fournel1,2, S.P. Lemay1,2 and J.H. Palacios1

*University of Alberta, Edmonton,* 

*3Department of Agricultural, Food and Nutritional Science,* 

**Non-Cattle Confinement Buildings:** 

**Monitoring, Emission Factors and Mitigation** 

S. Godbout1,2, F. Pelletier1, J.P. Larouche1, M. Belzile1, J.J.R. Feddes1,3,

*1Research and Development Institute for the Agri-Environment, Québec City, QC,* 

*2Department of Soil Science and Agri-Food Engineering, Université Laval, Québec City, QC,* 

Worldwide environmental issues are dominated by climate change, especially by the increase in greenhouse gas (GHG) emissions (UNDP, 2007). The rise in of GHG concentrations in the atmosphere has become a major environmental concern as revealed in the Kyoto Protocol (AAFC, 2000). Besides contributing to global warming by absorbing infrared radiation, carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O) have been declared the most harmful gases for ecosystems, apart from ammonia (NH3) (Pain, 1998;

Agricultural practices account for 10 to 12% of world total GHG emissions, however, it could reach between 17 and 32% (8,5-16,5 Pg CO2-eq) by including all agriculture-related emission sources (Bellarby et al., 2008). Agricultural GHG emissions can be divided into three main groups: a) CH4 emissions from cattle enteric fermentation; b) CH4 and N2O emissions due to manure management practices; and c) N2O emissions from cultivated fields, including direct emissions from crop land and pasture and indirect emissions

Manure management alone is responsible for 13% of GHG emissions from the agricultural sector with CH4 and N2O accounting for 33 and 67% of CO2-eq, respectively (Steinfield et al., 2006). Current trends suggest that this level will substantially increase over the coming decades as the intensification of livestock activities continues. On the other hand, CH4 and N2O have a global warming potential of 21 and 310 times over hundred years greater than CO2, respectively, based on their ability to contribute to climate change (Houghton et al. 1995). Hence, the environmental impact of livestock operations can not be considered negligible.

Currently, many countries have to use internationally agreed values to evaluate their GHG emissions. By describing the GHG emission sources and presenting the emission factors from non-cattle production, this chapter will improve the knowledge of scientists and

resulting from the use of nitrogen fertilizer in agriculture.

**1. Introduction** 

Copeland, 2009).

[21] Varbanov, P.,S., Perry, S., Makwana, Y., Zhu, X., and Smith, R. (2004), Top-level analysis of site utility system, Transactions of Institute of Chemical Engineering, 82(A6), 784-795

## **Greenhouse Gas Emissions from Non-Cattle Confinement Buildings: Monitoring, Emission Factors and Mitigation**

S. Godbout1,2, F. Pelletier1, J.P. Larouche1, M. Belzile1, J.J.R. Feddes1,3, S. Fournel1,2, S.P. Lemay1,2 and J.H. Palacios1 *1Research and Development Institute for the Agri-Environment, Québec City, QC, 2Department of Soil Science and Agri-Food Engineering, Université Laval, Québec City, QC, 3Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Canada* 

## **1. Introduction**

100 Greenhouse Gases – Emission, Measurement and Management

[21] Varbanov, P.,S., Perry, S., Makwana, Y., Zhu, X., and Smith, R. (2004), Top-level analysis

784-795

of site utility system, Transactions of Institute of Chemical Engineering, 82(A6),

Worldwide environmental issues are dominated by climate change, especially by the increase in greenhouse gas (GHG) emissions (UNDP, 2007). The rise in of GHG concentrations in the atmosphere has become a major environmental concern as revealed in the Kyoto Protocol (AAFC, 2000). Besides contributing to global warming by absorbing infrared radiation, carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O) have been declared the most harmful gases for ecosystems, apart from ammonia (NH3) (Pain, 1998; Copeland, 2009).

Agricultural practices account for 10 to 12% of world total GHG emissions, however, it could reach between 17 and 32% (8,5-16,5 Pg CO2-eq) by including all agriculture-related emission sources (Bellarby et al., 2008). Agricultural GHG emissions can be divided into three main groups: a) CH4 emissions from cattle enteric fermentation; b) CH4 and N2O emissions due to manure management practices; and c) N2O emissions from cultivated fields, including direct emissions from crop land and pasture and indirect emissions resulting from the use of nitrogen fertilizer in agriculture.

Manure management alone is responsible for 13% of GHG emissions from the agricultural sector with CH4 and N2O accounting for 33 and 67% of CO2-eq, respectively (Steinfield et al., 2006). Current trends suggest that this level will substantially increase over the coming decades as the intensification of livestock activities continues. On the other hand, CH4 and N2O have a global warming potential of 21 and 310 times over hundred years greater than CO2, respectively, based on their ability to contribute to climate change (Houghton et al. 1995). Hence, the environmental impact of livestock operations can not be considered negligible.

Currently, many countries have to use internationally agreed values to evaluate their GHG emissions. By describing the GHG emission sources and presenting the emission factors from non-cattle production, this chapter will improve the knowledge of scientists and

Greenhouse Gas Emissions from Non-Cattle

is a conversion factor (mg min-1 to g yr-1).

Amines Volatile fatty acids (VFA) NH3, CO2 Nitrogen heterocycles Sulfur compounds Phenolic

> Nitrogen compounds

Cu, P, K, Zn, Mn, Ca, Co, Fe

Mineral fraction

inside temperature differential.

Confinement Buildings: Monitoring, Emission Factors and Mitigation 103

where *EGHG* represents CO2, CH4 or N2O emissions for one animal space during one sampling event (g yr-1 animal-1), *Cou*t is the GHG exhaust concentration from the animal space (ppmv), *Cin* is the incoming GHG concentration to the animal space (ppmv), *Q* is the

number of animals in the room, *Patm* and *Pv* are respectively the atmospheric pressure at sea level and the vapour pressure (Pa), *T* corresponds to the temperature (K), *MGHG* characterize the molar masses of CO2 (44 g mol-1), CH4 (16 g mol-1), or N2O (44 g mol-1), *Mair* signifies the molar mass of air (29 g mol-1), 287 is the thermodynamic constant of air (J kg-1 K-1) and 525,6

> Aldehydes Alcohols VFA CH4 CO2 H2O Thiols

Manure

Fig. 1. Emissions from anaerobic decomposition of liquid manure (adapted from de la Farge,

The majority of the pig and poultry operations are mechanically ventilated whereas confined cattle facilities are primarily naturally ventilated. The methodology to estimate gas emissions depends on wind speed, direction, building opening orientation and outside-

The inlet and outlet gas concentrations play an important role. Monitoring methodologies differ among countries. The air exchange rate of an animal housing facility must be

1978; IPT, 1998; Taiganides, 1987; UGPVB, 1996; O'Neill & Phillips, 1992)

Carbon compounds

> Organic fraction

Peptides Glycerol Sugars Fibers

Proteins Lipids Carbohydrates

Amino acids Fatty acids Alcohols

VFA CH4 CO2 H2O air min-1), *Nanimals* is the

H2S and other sulfur compounds

Cellulose lignin Esters CH4 CO2 H2O

> Sulfur compounds

average room air exchange rate during the sampling event (m3

politicians about the contribution of agriculture to global warming. Moreover, this chapter provides quick information on monitoring and mitigation of GHG emissions.

## **2. Gas emissions from animal confinement buildings**

## **2.1 The source of contaminants**

Contaminants exhausted from animal confinement buildings include various gases, dust particles, micro-organisms and odours. The most important gases are CO2, NH3, hydrogen sulphide (H2S), CH4, N2O and some trace gases (aldehydes, amines, aromatics, organic acids, sulphur compounds, etc.). NH3, CH4 and N2O are produced from manure decomposition while CO2 is primarily a product of animal metabolism (Hartung & Phillips, 1994).

In most confinement buildings, manure is stored as a liquid or semi-solid beneath the animals for a short or long period of time. Both the manure attached to the flooring material and the manure stored under the animals produce these gases.

Manure decomposition begins in the stomach of the animal where the consumed feed undergoes early anaerobic decomposition by the intestinal flora at a temperature between 38 and 40 °C. Once the manure is excreted and exposed to air, a new type of bacteria grows according to the manure management method practiced. Depending on ambient temperature, this change takes 12 to 24 h (Barrington, 1999).

In the case of solid manure to which straw is added, animal manure is decomposed by aerobic bacteria. These bacteria break down organic matter and stabilize the manure. Once stabilized, almost no gas or odorous compounds will be produced (Barrington, 1999).

In liquid manure, a population of facultative bacteria (aerobic or anaerobic) grows rapidly. These bacteria decompose organic matter and produce gases and odorous compounds. The emission of these contaminants is thus carried throughout storage (Barrington, 1999). Figure 1 presents a schematic view of the emission mechanisms from anaerobic decomposition of liquid manure.

The production of N2O during storage and treatment of animal manure occurs during nitrification and denitrification of nitrogen contained in the manure. Nitrification is the oxidation of ammonium (NH4+) to nitrate (NO3-), and denitrification is the reduction of NO3 to N2O or atmospheric nitrogen (N2). Generally, as the degree of aeration of the waste increases, so does the amount of N2O produced (Olsen et al., 2003).

#### **2.2 Emission calculation**

Two parameters are very important in determining gas emission rates, namely the gas concentration (inlet and outlet) and the air exchange rate (Fig. 2). Gas emissions are calculated by multiplying the difference in concentration by the mass flow of gas, which is calculated from the mass flow of air. The GHG emissions can be calculated for each sampling period using Equation 1. In this equation, the specific volume of air (*v* = (*Patm* – *Pv*)/(287 x *T*); ASABE, 2010) is used to obtain the mass flow of air from the volumetric flow rate.

$$E\_{GHG} = (\mathbf{C}\_{out} - \mathbf{C}\_{in}) \times \frac{Q}{N\_{amimals}} \times \frac{P\_{atm} - P\_v}{287 \times T} \times \frac{M\_{GHG}}{M\_{air}} \times 525,6 \tag{1}$$

politicians about the contribution of agriculture to global warming. Moreover, this chapter

Contaminants exhausted from animal confinement buildings include various gases, dust particles, micro-organisms and odours. The most important gases are CO2, NH3, hydrogen sulphide (H2S), CH4, N2O and some trace gases (aldehydes, amines, aromatics, organic acids, sulphur compounds, etc.). NH3, CH4 and N2O are produced from manure decomposition

In most confinement buildings, manure is stored as a liquid or semi-solid beneath the animals for a short or long period of time. Both the manure attached to the flooring material

Manure decomposition begins in the stomach of the animal where the consumed feed undergoes early anaerobic decomposition by the intestinal flora at a temperature between 38 and 40 °C. Once the manure is excreted and exposed to air, a new type of bacteria grows according to the manure management method practiced. Depending on ambient

In the case of solid manure to which straw is added, animal manure is decomposed by aerobic bacteria. These bacteria break down organic matter and stabilize the manure. Once stabilized, almost no gas or odorous compounds will be produced (Barrington, 1999).

In liquid manure, a population of facultative bacteria (aerobic or anaerobic) grows rapidly. These bacteria decompose organic matter and produce gases and odorous compounds. The emission of these contaminants is thus carried throughout storage (Barrington, 1999). Figure 1 presents a schematic view of the emission mechanisms from anaerobic decomposition of

The production of N2O during storage and treatment of animal manure occurs during nitrification and denitrification of nitrogen contained in the manure. Nitrification is the oxidation of ammonium (NH4+) to nitrate (NO3-), and denitrification is the reduction of NO3 to N2O or atmospheric nitrogen (N2). Generally, as the degree of aeration of the waste

Two parameters are very important in determining gas emission rates, namely the gas concentration (inlet and outlet) and the air exchange rate (Fig. 2). Gas emissions are calculated by multiplying the difference in concentration by the mass flow of gas, which is calculated from the mass flow of air. The GHG emissions can be calculated for each sampling period using Equation 1. In this equation, the specific volume of air (*v* = (*Patm* – *Pv*)/(287 x *T*); ASABE,

( ) 525,6 <sup>287</sup>

*N TM*

*animals air*

*atm v GHG*

(1)

provides quick information on monitoring and mitigation of GHG emissions.

while CO2 is primarily a product of animal metabolism (Hartung & Phillips, 1994).

**2. Gas emissions from animal confinement buildings** 

and the manure stored under the animals produce these gases.

temperature, this change takes 12 to 24 h (Barrington, 1999).

increases, so does the amount of N2O produced (Olsen et al., 2003).

2010) is used to obtain the mass flow of air from the volumetric flow rate.

*<sup>Q</sup> P PM E CC*

*GHG out in*

**2.1 The source of contaminants** 

liquid manure.

**2.2 Emission calculation** 

where *EGHG* represents CO2, CH4 or N2O emissions for one animal space during one sampling event (g yr-1 animal-1), *Cou*t is the GHG exhaust concentration from the animal space (ppmv), *Cin* is the incoming GHG concentration to the animal space (ppmv), *Q* is the average room air exchange rate during the sampling event (m3 air min-1), *Nanimals* is the number of animals in the room, *Patm* and *Pv* are respectively the atmospheric pressure at sea level and the vapour pressure (Pa), *T* corresponds to the temperature (K), *MGHG* characterize the molar masses of CO2 (44 g mol-1), CH4 (16 g mol-1), or N2O (44 g mol-1), *Mair* signifies the molar mass of air (29 g mol-1), 287 is the thermodynamic constant of air (J kg-1 K-1) and 525,6 is a conversion factor (mg min-1 to g yr-1).

Fig. 1. Emissions from anaerobic decomposition of liquid manure (adapted from de la Farge, 1978; IPT, 1998; Taiganides, 1987; UGPVB, 1996; O'Neill & Phillips, 1992)

The majority of the pig and poultry operations are mechanically ventilated whereas confined cattle facilities are primarily naturally ventilated. The methodology to estimate gas emissions depends on wind speed, direction, building opening orientation and outsideinside temperature differential.

The inlet and outlet gas concentrations play an important role. Monitoring methodologies differ among countries. The air exchange rate of an animal housing facility must be

Greenhouse Gas Emissions from Non-Cattle

1983).

low cost technique with easy apparatus implementation.

Confinement Buildings: Monitoring, Emission Factors and Mitigation 105

The air is mainly composed of N2, oxygen (O2) and argon (Ar) with several others gases in trace concentrations like CO2, CH4 and N2O. These components can be separated by chromatography and detected by different detectors more or less specific to the target gas. The technique is simple, proven and allows the simultaneous quantification of CO2, CH4 and N2O in the gaseous effluents discharged to the atmosphere. Compared to other techniques having the required sensitivity, like most modern spectroscopic techniques, the chromatography is known to produce reliable results and can be envisaged as a moderate to

These three GHG are easily separated at low temperature on a column filled with porous polymers Porapak Q or Chromosorb 102 (Cowper & DeRose, 1983). However, the analysis strategy depends on the detectors used and additional gases to be separated and quantified in the sample. Methane can be precisely measured by a flame ionization detector (FID) on a wide range of concentrations ranging from parts per million to volume percent (Cowper & DeRose, 1983) which is suitable for measuring emissions from a livestock building where CH4 concentrations will range from atmospheric pressure of 1,7 ppmv (Brasseur et al., 1999) to less than 5 000 ppmv for most of the time. To measure atmospheric concentrations of CO2, a particular approach should be implemented. The approach involves the reduction of CO2 to CH4 with hydrogen over a nickel catalyst and detection by the FID (Cowper & DeRose,

In order to obtain the required sensitivity for the quantification of N2O at concentrations found in ambient air, the electron capture detector (ECD) is commonly used to measure atmospheric concentrations. Some work has been done with N2 alone as a carrier gas (Jiang et al., 2007; Arnold et al., 2001; Loftfield et al., 1997) while others were performed with an Ar/CH4 mix (95/5) as carrier gas or as make-up gas to the detector (Jiang et al., 2007; Heinemeyer & Kaiser, 1996; Sitaula et al., 1992; Weiss, 1981; Mosier & Mack, 1980). It was also noted that impurities in a carrier gas can strongly influence the response of the ECD (Phillips et al., 1979) and that the addition of O2 in the carrier gas increases the sensitivity of ECD to allow the determination of atmospheric concentrations of CO2 (Cowper & DeRose, 1983). Even if the dynamic range of the ECD is limited, the range of concentration for CO2 and N2O encountered for most confinement buildings are limited. The typical measured concentrations range from atmospheric concentration (360 ppmv for CO2 and 0,31 ppmv for

N2O; Brasseur et al., 1999) up to 5 000 ppmv for CO2 and up to 10 ppmv for N2O.

The system developed for the quantification of gas emissions from the agricultural sector has two main functions, first the collection and management of the sample and second the analysis of the sample. Samples are taken sequentially from several sampling points and continuously transported to the analysis system. CO2, CH4 and N2O are analyzed with a gas chromatograph (GC). An example of the gas sample collection system is shown in Fig. 3.

For each sample location, gases are pumped through a membrane filter made of polytetrafluoroethylene (50 mm diameter, 0,2 μm pore size) and routed in a Teflon tube (6,4 mm OD, 0.8 mm wall) of variable length depending on the distance between the sampling location and the analysis system. The Teflon tubes are connected to a rotary valve allowing the sequential sampling and analysis of up to 16 locations. A purge diaphragm pump allows

**3.2 Sampling and management of gas samples** 

measured accurately. This would apply to both mechanically ventilated (MV) buildings and naturally ventilated (NV) facilities or a combination of the two ventilation systems. Buildings with combined systems are commonly referred to as hybrid ventilation systems (HV). Many methods have been developed to measure ventilation rates from animal housing facilities.

The following sections will address these topics with a complete description of an experimental setup used by the authors for sampling and analysis of GHG emitted by a number of non-cattle confinement buildings. Also best methods for measuring the air flow rate in both MV and NV barns are suggested.

Fig. 2. View of an emission measurement set up on a naturally ventilated barn.

## **3. Concentration measurements: gas sampling and analysis**

### **3.1 Atmospheric concentration gas analysis**

Since the ambient air is used for ventilation of animal buildings, the determination of gas emissions from agricultural activities initially requires a measurement of the ambient air concentration. The contribution of farming systems under study may affect the GHG ambient concentrations near the facility. The measurement of atmospheric concentration need to use equipment having great sensitivity and selectivity like those utilizing optical properties of gas such as Fourier transform infrared spectroscopy (FTIR), photoacoustic spectroscopy (PAS) and non dispersive infrared analyser (NDIR) or separation techniques like chromatography with selective detectors (Neftel et al., 2006).

Since agricultural and forest soils are involved in gas exchange with the atmosphere and many agricultural activities like animal husbandry are significant sources of CH4 and N2O, several studies were conducted to quantify emissions from soils and livestock buildings. In those projects, a number of researchers primarily interested in the characterization of emissions from livestock buildings often work with PAS in the infrared (Blanes-Vidal et al., 2008; Cabaraux et al., 2008; Philippe et al., 2007). While for other authors interested in analytical development or atmospheric flux measurements from soils, separation by chromatography seems to be the preferred means of detection and quantification of trace gases in ambient air (Loftfield et al., 1997; Sitaula et al., 1992; Weiss, 1981; Blackmer & Bremner, 1977).

measured accurately. This would apply to both mechanically ventilated (MV) buildings and naturally ventilated (NV) facilities or a combination of the two ventilation systems. Buildings with combined systems are commonly referred to as hybrid ventilation systems (HV). Many methods have been developed to measure ventilation rates from animal

The following sections will address these topics with a complete description of an experimental setup used by the authors for sampling and analysis of GHG emitted by a number of non-cattle confinement buildings. Also best methods for measuring the air flow

Fig. 2. View of an emission measurement set up on a naturally ventilated barn.

Since the ambient air is used for ventilation of animal buildings, the determination of gas emissions from agricultural activities initially requires a measurement of the ambient air concentration. The contribution of farming systems under study may affect the GHG ambient concentrations near the facility. The measurement of atmospheric concentration need to use equipment having great sensitivity and selectivity like those utilizing optical properties of gas such as Fourier transform infrared spectroscopy (FTIR), photoacoustic spectroscopy (PAS) and non dispersive infrared analyser (NDIR) or separation techniques

Since agricultural and forest soils are involved in gas exchange with the atmosphere and many agricultural activities like animal husbandry are significant sources of CH4 and N2O, several studies were conducted to quantify emissions from soils and livestock buildings. In those projects, a number of researchers primarily interested in the characterization of emissions from livestock buildings often work with PAS in the infrared (Blanes-Vidal et al., 2008; Cabaraux et al., 2008; Philippe et al., 2007). While for other authors interested in analytical development or atmospheric flux measurements from soils, separation by chromatography seems to be the preferred means of detection and quantification of trace gases in ambient air (Loftfield et al., 1997; Sitaula et al., 1992; Weiss, 1981; Blackmer &

**3. Concentration measurements: gas sampling and analysis** 

like chromatography with selective detectors (Neftel et al., 2006).

Wind direction

*Tout, RH, Patm Nanimals, Tin*

*Cin Cout*

housing facilities.

Bremner, 1977).

rate in both MV and NV barns are suggested.

**3.1 Atmospheric concentration gas analysis** 

The air is mainly composed of N2, oxygen (O2) and argon (Ar) with several others gases in trace concentrations like CO2, CH4 and N2O. These components can be separated by chromatography and detected by different detectors more or less specific to the target gas. The technique is simple, proven and allows the simultaneous quantification of CO2, CH4 and N2O in the gaseous effluents discharged to the atmosphere. Compared to other techniques having the required sensitivity, like most modern spectroscopic techniques, the chromatography is known to produce reliable results and can be envisaged as a moderate to low cost technique with easy apparatus implementation.

These three GHG are easily separated at low temperature on a column filled with porous polymers Porapak Q or Chromosorb 102 (Cowper & DeRose, 1983). However, the analysis strategy depends on the detectors used and additional gases to be separated and quantified in the sample. Methane can be precisely measured by a flame ionization detector (FID) on a wide range of concentrations ranging from parts per million to volume percent (Cowper & DeRose, 1983) which is suitable for measuring emissions from a livestock building where CH4 concentrations will range from atmospheric pressure of 1,7 ppmv (Brasseur et al., 1999) to less than 5 000 ppmv for most of the time. To measure atmospheric concentrations of CO2, a particular approach should be implemented. The approach involves the reduction of CO2 to CH4 with hydrogen over a nickel catalyst and detection by the FID (Cowper & DeRose, 1983).

In order to obtain the required sensitivity for the quantification of N2O at concentrations found in ambient air, the electron capture detector (ECD) is commonly used to measure atmospheric concentrations. Some work has been done with N2 alone as a carrier gas (Jiang et al., 2007; Arnold et al., 2001; Loftfield et al., 1997) while others were performed with an Ar/CH4 mix (95/5) as carrier gas or as make-up gas to the detector (Jiang et al., 2007; Heinemeyer & Kaiser, 1996; Sitaula et al., 1992; Weiss, 1981; Mosier & Mack, 1980). It was also noted that impurities in a carrier gas can strongly influence the response of the ECD (Phillips et al., 1979) and that the addition of O2 in the carrier gas increases the sensitivity of ECD to allow the determination of atmospheric concentrations of CO2 (Cowper & DeRose, 1983). Even if the dynamic range of the ECD is limited, the range of concentration for CO2 and N2O encountered for most confinement buildings are limited. The typical measured concentrations range from atmospheric concentration (360 ppmv for CO2 and 0,31 ppmv for N2O; Brasseur et al., 1999) up to 5 000 ppmv for CO2 and up to 10 ppmv for N2O.

## **3.2 Sampling and management of gas samples**

The system developed for the quantification of gas emissions from the agricultural sector has two main functions, first the collection and management of the sample and second the analysis of the sample. Samples are taken sequentially from several sampling points and continuously transported to the analysis system. CO2, CH4 and N2O are analyzed with a gas chromatograph (GC). An example of the gas sample collection system is shown in Fig. 3.

For each sample location, gases are pumped through a membrane filter made of polytetrafluoroethylene (50 mm diameter, 0,2 μm pore size) and routed in a Teflon tube (6,4 mm OD, 0.8 mm wall) of variable length depending on the distance between the sampling location and the analysis system. The Teflon tubes are connected to a rotary valve allowing the sequential sampling and analysis of up to 16 locations. A purge diaphragm pump allows

Greenhouse Gas Emissions from Non-Cattle

**3.3 Chromatographic analysis of greenhouse gases** 

Confinement Buildings: Monitoring, Emission Factors and Mitigation 107

controlled mobile laboratory. T-type thermocouples are used to monitor the temperature in the heated ducts and the temperature inside and outside the mobile lab. A data logger controlled by the computer of the GC can acquire and archive various parameters measured during periods of analysis. It also controls the electric actuator of the 16-position rotary valve and the various solenoid activated valves of the gas collection system of the samples.

The strategy for the chromatographic analysis is the separation of the three gases on packed columns filled with the porous polymer Porapak Q 80/100 mesh. A pre-column (3,2 mm OD, 1 m long) connected in series before the analytical column (3,2 mm OD, 3 m long) removes some substances that may be present in the gas samples. These substances, which are retained longer than N2O in the pre-column, can include water, NH3 and some sulphur compounds that may have adverse effects on the detectors or on the columns.

CH4 is quantified with a FID, while CO2 and N2O are measured with an ECD. However following the initial set-up of the chromatographic analysis, CO2 was quantified with the FID after reduction with hydrogen over a nickel catalyst. Subsequently, following the chance observation of the detection of CO2 by the ECD, the quantification of CO2 is transferred to the ECD. This minimises the gradual loss of the effectiveness of the nickel catalyst and also allows a better separation of CO2 and N2O which is not affected by the

Figure 4 shows a schematic of the tubing configuration of the GC used for the separation and quantification of the greenhouse gases and Fig. 5 shows a picture of the column

The oven of the GC is maintained at 60 °C for the duration of the analysis. The 10-port injection valve and the 6-port detector valve are mounted on top of the GC. The zero grade nitrogen is used as carrier gas and is introduced at the three entry points of the pneumatic system of the GC at an equal flow rate of 25 ml min-1. The two detectors are maintained at 325 °C and the FID is supplied with 30 ml min-1of hydrogen UHP and with 300 ml min-1of zero grade air produced by a commercial generator. All the necessary tubing, fittings and

The sequence of the analysis begins when the injection valve is actuated to allow the carrier gas to flow through the sample loop and thus transfer the sample gas into the pre-column and the analytical column. After elution of the N2O from the pre-column, the injection valve returns to its original position to allow the back flush of the pre-column and further elution of the analytical column. After detection of CH4 on the FID, the detector valve is actuated to allow quantification of the CO2 and N2O on the ECD. The retention times are 2,9 min for CO2, 1,6 min for CH4 and 3,8 min for N2O. Examples of chromatograms obtained for analysis of standard calibration gases and ambient air are presented in Figs. 6 and 7, respectively. With an analysis time of 5 min for the chromatographic analysis and a turnover of less than 7 min between analyses, the chromatographic system allows continuous

Normal operation of the GC is provided in part by the control software for the electrical parameters and also by periodic checks of the pressure on low pressure gauges on the

disruption of the baseline caused by the actuation of the detector valve.

arrangement inside the GC oven and a picture of the outside of the GC.

acquisition of sufficient data to adequately describe agricultural process.

valves installed in the GC are made of stainless steel.

Fig. 3. Gas sample collection system

for back flowing of ambient air through the Teflon gas lines that are not under analysis to minimize stagnation of sample in the tubes.

The gas flow from the source to the analyser is provided by the main diaphragm pump which delivers the gas to a stainless steel tee fitting. A small filtering sleeve made of sintered stainless steel (7 μm pore size) placed upstream of the main pump provides extra equipment protection against fine dust. The stainless steel tee allows the diaphragm pump to draw a fraction of the sample to flow continuously through the 1 000 µL sample loop of the GC. Two solenoid valves are used to isolate the sample loop in order to balance the pressure of the sample with atmospheric pressure before injection. The sample excess is exhausted to the atmosphere.

The solenoid valve placed between the 16-position rotary valve and the main pump allow for selecting the sample gas analysed from the rotary valve or a selection of calibration gases including ambient air and pressurized gas cylinders controlled by solenoid valves.

All components of the sample collection system that are in contact with the sample gas are either Teflon (ex.: sample tubes), Teflon coated (ex. : pump diaphragms) or stainless steel 316 (ex.: rotary valve, fittings, etc.).

The Teflon sampling tubes from the sampling locations to the analyzer are placed inside a series of ducts maintained by circulating air at approximately 35 °C. The complete system including the 16-position rotary valve, the GC and the pumps are installed in a temperature

for back flowing of ambient air through the Teflon gas lines that are not under analysis to

The gas flow from the source to the analyser is provided by the main diaphragm pump which delivers the gas to a stainless steel tee fitting. A small filtering sleeve made of sintered stainless steel (7 μm pore size) placed upstream of the main pump provides extra equipment protection against fine dust. The stainless steel tee allows the diaphragm pump to draw a fraction of the sample to flow continuously through the 1 000 µL sample loop of the GC. Two solenoid valves are used to isolate the sample loop in order to balance the pressure of the sample with atmospheric pressure before injection. The sample excess is exhausted to

The solenoid valve placed between the 16-position rotary valve and the main pump allow for selecting the sample gas analysed from the rotary valve or a selection of calibration gases

All components of the sample collection system that are in contact with the sample gas are either Teflon (ex.: sample tubes), Teflon coated (ex. : pump diaphragms) or stainless steel

The Teflon sampling tubes from the sampling locations to the analyzer are placed inside a series of ducts maintained by circulating air at approximately 35 °C. The complete system including the 16-position rotary valve, the GC and the pumps are installed in a temperature

including ambient air and pressurized gas cylinders controlled by solenoid valves.

Fig. 3. Gas sample collection system

316 (ex.: rotary valve, fittings, etc.).

the atmosphere.

minimize stagnation of sample in the tubes.

controlled mobile laboratory. T-type thermocouples are used to monitor the temperature in the heated ducts and the temperature inside and outside the mobile lab. A data logger controlled by the computer of the GC can acquire and archive various parameters measured during periods of analysis. It also controls the electric actuator of the 16-position rotary valve and the various solenoid activated valves of the gas collection system of the samples.

## **3.3 Chromatographic analysis of greenhouse gases**

The strategy for the chromatographic analysis is the separation of the three gases on packed columns filled with the porous polymer Porapak Q 80/100 mesh. A pre-column (3,2 mm OD, 1 m long) connected in series before the analytical column (3,2 mm OD, 3 m long) removes some substances that may be present in the gas samples. These substances, which are retained longer than N2O in the pre-column, can include water, NH3 and some sulphur compounds that may have adverse effects on the detectors or on the columns.

CH4 is quantified with a FID, while CO2 and N2O are measured with an ECD. However following the initial set-up of the chromatographic analysis, CO2 was quantified with the FID after reduction with hydrogen over a nickel catalyst. Subsequently, following the chance observation of the detection of CO2 by the ECD, the quantification of CO2 is transferred to the ECD. This minimises the gradual loss of the effectiveness of the nickel catalyst and also allows a better separation of CO2 and N2O which is not affected by the disruption of the baseline caused by the actuation of the detector valve.

Figure 4 shows a schematic of the tubing configuration of the GC used for the separation and quantification of the greenhouse gases and Fig. 5 shows a picture of the column arrangement inside the GC oven and a picture of the outside of the GC.

The oven of the GC is maintained at 60 °C for the duration of the analysis. The 10-port injection valve and the 6-port detector valve are mounted on top of the GC. The zero grade nitrogen is used as carrier gas and is introduced at the three entry points of the pneumatic system of the GC at an equal flow rate of 25 ml min-1. The two detectors are maintained at 325 °C and the FID is supplied with 30 ml min-1of hydrogen UHP and with 300 ml min-1of zero grade air produced by a commercial generator. All the necessary tubing, fittings and valves installed in the GC are made of stainless steel.

The sequence of the analysis begins when the injection valve is actuated to allow the carrier gas to flow through the sample loop and thus transfer the sample gas into the pre-column and the analytical column. After elution of the N2O from the pre-column, the injection valve returns to its original position to allow the back flush of the pre-column and further elution of the analytical column. After detection of CH4 on the FID, the detector valve is actuated to allow quantification of the CO2 and N2O on the ECD. The retention times are 2,9 min for CO2, 1,6 min for CH4 and 3,8 min for N2O. Examples of chromatograms obtained for analysis of standard calibration gases and ambient air are presented in Figs. 6 and 7, respectively. With an analysis time of 5 min for the chromatographic analysis and a turnover of less than 7 min between analyses, the chromatographic system allows continuous acquisition of sufficient data to adequately describe agricultural process.

Normal operation of the GC is provided in part by the control software for the electrical parameters and also by periodic checks of the pressure on low pressure gauges on the

Greenhouse Gas Emissions from Non-Cattle

Confinement Buildings: Monitoring, Emission Factors and Mitigation 109

Fig. 6. Chromatograms of the greenhouse standard calibration gases

Fig. 7. Chromatograms of an ambient air sample

cylinder gas regulators and on the pressure gauges mounted on three controls of the carrier gas admission in the GC which operate near 234 kPa.

Fig. 4. Tubing configuration of the chromatograph

Fig. 5. Standard gas chromatograph and oven

cylinder gas regulators and on the pressure gauges mounted on three controls of the carrier

gas admission in the GC which operate near 234 kPa.

Fig. 4. Tubing configuration of the chromatograph

Fig. 5. Standard gas chromatograph and oven

Fig. 6. Chromatograms of the greenhouse standard calibration gases

Fig. 7. Chromatograms of an ambient air sample

Greenhouse Gas Emissions from Non-Cattle

supplier information (Belzile et al., 2006).

**4.3 Natural and hybrid ventilated barn** 

suggested by CIGR (2002).

wind and thermal buoyancy.

the fans.

**5.1 Swine production** 

temperatures inside and outside the building.

Confinement Buildings: Monitoring, Emission Factors and Mitigation 111

difference between the interior of the room and the outside of the building as well as the fan rotational speed of each ventilation stage should be continuously measured during the trials. During measurements within the duct, conditions occurring during each trial are recreated (static pressure and rotation speeds of each stage of ventilation). Collected data make it possible to calculate regression equations for each fan predicting air flow rate based on room static pressure and fan rotation speed. Regression equations are calculated from the

For MV buildings, the ventilation related data collected are: daily number of animals and their mass, hourly mean exhaust fan rotational speeds, building static pressures and hourly

Ventilation rates can be predicted using a CO2 balance. Using data monitored in that barn, the animal CO2 production values can be calculated from published data like, e.g., values

Determining the ventilation rate of NV or HV buildings remains challenging. With ventilation controls becoming more sophisticated, side curtain and ridge vent openings can be controlled more precisely based on anticipatory logic used to determine upcoming temperatures and wind conditions. Traditionally, NV buildings have only been used for larger animals where changes in the thermal environment are not as critical. With improved controller logic, NV buildings will soon be used more extensively for housing smaller animals. Consequently, the ability to predict airflow in these facilities is very important in

A number of researchers have proposed methods to calculate the air exchange rates in NV buildings. These airflow rates are dependent on wind speed at the opening and effectiveness of the opening. Morsing et al. (2002) and Choinière et al. (1988) reported coefficients to be used for agricultural buildings to calculate the wind speed at the opening. Nääs et al. (1998) and Choinière et al. (1988) described an algorithm to determine an opening effectiveness relative to the wind direction. This methodology is described as the ventilation rate due to

For HV buildings, the ventilation related data collected included: daily number of animals and their mass, hourly wind speed and direction, curtain opening areas, operating status of the exhaust fans and hourly temperatures inside and outside the building. As for MV buildings, ventilation rates from NV and HV barns can be predicted by establishing a CO2 balance. In order to be representative, the CO2 concentrations should be measured at least at three locations: on each side of the barn close to the curtains and in the center of the barn at the animal level. For HV barns, the CO2 concentrations should also be measured close to

The literature identifies emission factors from the three types of swine confinement buildings (Table 2). The swine production system begins with the maternity stage

controlling the thermal and non-thermal well-being of the confined animals.

**5. Emission factors from swine and poultry confined buildings** 

### **3.4 Quality control of the analyses**

Chromatographic instrumentation analysis is usually calibrated at the beginning of each period of analysis using standard calibration gases qualified as "Certified Standards". The three GHG are available mixed and diluted with N2 in a single cylinder and at concentrations similar to those expected in real samples. The response factors of the chromatographic analysis are calculated with a single point calibration for each analyzed gas since the response of the detectors used are linear over the concentration ranges encountered.

To document the long-term performance of the overall system and to control the quality of the data obtained, standard analysis are performed automatically at a specified frequency. The samples used are the ambient air in the mobile laboratory and the standard gases from cylinders used for calibration of the GC. The curves showing the responses of the system over time are used to observe and confirm the periods of normal operation of the system and other statistical calculations on the results are used to estimate the overall accuracy of analysis including management and quantification of the sample. Table 1 shows typical examples of results for standard analysis.


Table 1. Typical concentrations measured for standard analysis

## **4. Ventilation rate measurement**

#### **4.1 Basic principle**

To determine the gas emission rates of an animal housing facility, the air exchange rate of the facility must be measured accurately. In northern climates, HV buildings have become more common since buildings relying on NV during cold weather conditions develop numerous problems. These problems include poor inlet air distribution into the building, poor control of the exchange air within the building, uneven temperature and air velocity distribution throughout the building. Therefore, HV buildings rely on MV during cold weather conditions and on NV during spring, summer and fall weather conditions.

#### **4.2 Mechanically ventilated barn**

Many methods have been developed to measure ventilation rates from animal housing facilities. These have included airborne tracer techniques (Leonard et al., 1984), diffusion of animal-produced CO2 (Feddes et al., 1984) or heat (Barber et al., 1994), vane anemometers (Heber et al., 2000), orifice plates (Godbout et al., 2005), multi-port averaging pitot tubes (Clark et al., 2008) and thermal (e.g., hot-wire) anemometers (Feddes & McQuitty, 1980). Each of these methods, however, has limitations.

Fan airflow rates can be evaluated using a standardized ventilation conduit developed using the standard ANSI/ASHRAE 41.2-1987 (RA 92) (ASHRAE, 1992). The static pressure difference between the interior of the room and the outside of the building as well as the fan rotational speed of each ventilation stage should be continuously measured during the trials. During measurements within the duct, conditions occurring during each trial are recreated (static pressure and rotation speeds of each stage of ventilation). Collected data make it possible to calculate regression equations for each fan predicting air flow rate based on room static pressure and fan rotation speed. Regression equations are calculated from the supplier information (Belzile et al., 2006).

For MV buildings, the ventilation related data collected are: daily number of animals and their mass, hourly mean exhaust fan rotational speeds, building static pressures and hourly temperatures inside and outside the building.

Ventilation rates can be predicted using a CO2 balance. Using data monitored in that barn, the animal CO2 production values can be calculated from published data like, e.g., values suggested by CIGR (2002).

## **4.3 Natural and hybrid ventilated barn**

110 Greenhouse Gases – Emission, Measurement and Management

Chromatographic instrumentation analysis is usually calibrated at the beginning of each period of analysis using standard calibration gases qualified as "Certified Standards". The three GHG are available mixed and diluted with N2 in a single cylinder and at concentrations similar to those expected in real samples. The response factors of the chromatographic analysis are calculated with a single point calibration for each analyzed gas since the response of the

To document the long-term performance of the overall system and to control the quality of the data obtained, standard analysis are performed automatically at a specified frequency. The samples used are the ambient air in the mobile laboratory and the standard gases from cylinders used for calibration of the GC. The curves showing the responses of the system over time are used to observe and confirm the periods of normal operation of the system and other statistical calculations on the results are used to estimate the overall accuracy of analysis including management and quantification of the sample. Table 1 shows typical

Mean value (ppmv) 2,0 595 0,32 20,5 1 510 2,1 Precision (%) 5,8 4,6 16,0 1,0 3,8 5,8 Maximum value (ppmv) 2,4 638 0,41 20,7 1 589 2,3 Minimum value (ppmv) 1,9 541 0,21 20,1 1 403 1,9

To determine the gas emission rates of an animal housing facility, the air exchange rate of the facility must be measured accurately. In northern climates, HV buildings have become more common since buildings relying on NV during cold weather conditions develop numerous problems. These problems include poor inlet air distribution into the building, poor control of the exchange air within the building, uneven temperature and air velocity distribution throughout the building. Therefore, HV buildings rely on MV during cold

Many methods have been developed to measure ventilation rates from animal housing facilities. These have included airborne tracer techniques (Leonard et al., 1984), diffusion of animal-produced CO2 (Feddes et al., 1984) or heat (Barber et al., 1994), vane anemometers (Heber et al., 2000), orifice plates (Godbout et al., 2005), multi-port averaging pitot tubes (Clark et al., 2008) and thermal (e.g., hot-wire) anemometers (Feddes & McQuitty, 1980).

Fan airflow rates can be evaluated using a standardized ventilation conduit developed using the standard ANSI/ASHRAE 41.2-1987 (RA 92) (ASHRAE, 1992). The static pressure

weather conditions and on NV during spring, summer and fall weather conditions.

Ambient air Standard calibration gases CH4 CO2 N2O CH4 CO2 N2O

detectors used are linear over the concentration ranges encountered.

Table 1. Typical concentrations measured for standard analysis

**3.4 Quality control of the analyses** 

examples of results for standard analysis.

**4. Ventilation rate measurement** 

**4.2 Mechanically ventilated barn** 

Each of these methods, however, has limitations.

**4.1 Basic principle** 

Determining the ventilation rate of NV or HV buildings remains challenging. With ventilation controls becoming more sophisticated, side curtain and ridge vent openings can be controlled more precisely based on anticipatory logic used to determine upcoming temperatures and wind conditions. Traditionally, NV buildings have only been used for larger animals where changes in the thermal environment are not as critical. With improved controller logic, NV buildings will soon be used more extensively for housing smaller animals. Consequently, the ability to predict airflow in these facilities is very important in controlling the thermal and non-thermal well-being of the confined animals.

A number of researchers have proposed methods to calculate the air exchange rates in NV buildings. These airflow rates are dependent on wind speed at the opening and effectiveness of the opening. Morsing et al. (2002) and Choinière et al. (1988) reported coefficients to be used for agricultural buildings to calculate the wind speed at the opening. Nääs et al. (1998) and Choinière et al. (1988) described an algorithm to determine an opening effectiveness relative to the wind direction. This methodology is described as the ventilation rate due to wind and thermal buoyancy.

For HV buildings, the ventilation related data collected included: daily number of animals and their mass, hourly wind speed and direction, curtain opening areas, operating status of the exhaust fans and hourly temperatures inside and outside the building. As for MV buildings, ventilation rates from NV and HV barns can be predicted by establishing a CO2 balance. In order to be representative, the CO2 concentrations should be measured at least at three locations: on each side of the barn close to the curtains and in the center of the barn at the animal level. For HV barns, the CO2 concentrations should also be measured close to the fans.

## **5. Emission factors from swine and poultry confined buildings**

## **5.1 Swine production**

The literature identifies emission factors from the three types of swine confinement buildings (Table 2). The swine production system begins with the maternity stage

Greenhouse Gas Emissions from Non-Cattle

production confinement buildings.

Gas Units Mean

GHG \* kg CO2-eq. yr-1

GHG \* kg CO2-eq. yr-1

and the global warming potential of CH4 (21) and N2O (310).

(17,6 g yr-1 head -1 vs. 10,9 g yr-1 head -1).

**6. Mitigation techniques 6.1 In animal production** 

**5.2 Poultry production** 

Production type

Broiler

Layer

Confinement Buildings: Monitoring, Emission Factors and Mitigation 113

Broilers are mainly reared on a floor surface covered with bedding. Laying hens can be reared in multiple-deck battery cages, aviary systems, high-rise systems or percheries. In the first case, there is a possibility to dry manure directly under the cages with different manure drying systems. Table 3 presents emission factors from these two types of poultry

emissions \*\*

\* Total greenhouse gas emissions calculated on a CO2-equivalent basis considering the mean emissions

\*\* References: Chadwick et al., 1999; EPA, 2001; Fabbri et al., 2007; Fournel, 2011; Groot Koerkamp & Uenk, 1997; Hörnig et al., 2001; Monteny et al., 2001; Neser et al., 1997, as cited in Jungbluth et al., 2001;

Broiler and layer productions emit similar quantities of CO2 to the atmosphere (31,5 and 28,2 kg yr-1 head-1, respectively). However, the different layer systems using liquid manure management generate a greater emission factor for CH4 (44,7 g yr-1 head -1) comparatively to broiler systems with litter (12,3 g yr-1 head -1). On the other hand, litter increases the succession of nitrification and denitrification phases which result in greater N2O emissions

Several technologies have been developed to reduce gas and odour emissions from livestock housing. Several of these techniques, including reduction efficiencies for each technology, were inventoried by Godbout et al. (2010) from an exhaustive literature review. The inventory revealed that progress has been made in reducing odour and ammonia emissions, while less concern has been placed on GHG emissions. Three distinct techniques have been

Separating urine and feces beneath the slats and removing both the solid and liquid fractions frequently is a reliable manure management technique to reduce gas and odour

recognized, namely under slat separation, air cleaning and nutrient management.

**6.2 In-barn manure management: under-slat separation** 

Sneath et al., 1996, as cited in Jungbluth et al., 2001; Wathes et al., 1997; Wu-Haan et al., 2007.

Table 3. GHG emission factors for broiler and layer confinement buildings

CO2 kg yr-1 bird-1 31,5 31,5 31,5 - CH4 g yr-1 bird-1 12,3 8,3 20,0 6,7 N2O g yr-1 bird-1 17,6 4,0 34,2 12,6

CO2 kg yr-1 hen-1 28,2 12,6 37,8 2,87 CH4 g yr-1 hen-1 44,7 4,0 80,0 18,7 N2O g yr-1 hen-1 10,9 0,63 30,0 11,3

Minimum value

bird-1 5,71 - - -

hen-1 4,32 - - -

Maximum value

Standard deviation

comprising both of gestating sows and farrowing sows with their piglets. Gestating sows are usually reared either in individual stalls or in group pens. Farrowing sows and their piglets are mainly kept in farrowing crates. The swine nursery or post-weaning building rears the weaning piglets brought from the maternity. Piglets remain there until they reach a certain weight, generally between 20 and 25 kg. Feed and ambient conditions are adjusted with pig growth. Then, pigs are transferred to grower/finisher facilities until they leave for slaughter.


\* Total greenhouse gas emissions calculated on a CO2-equivalent basis considering the mean emissions and the global warming potential of CH4 (21) and N2O (310).

\*\* References: Gallman et al., 2003; Gallmann & Hartung, 2000; Godbout et al., 2003, 2006; Groot Koerkamp & Uenk, 1997; Guarino et al., 2003; Guimont et al., 2007; Hinz & Linke, 1998; Lemay et al., 2007; Sharpe et al., 2001; Zhang et al., 2007.

Table 2. GHG emission factors for swine confined buildings (adapted from Hamelin et al., 2009)

Overall, sows produce more CO2 on an animal basis (5,29 kg d-1 animal-1) than weanling piglets (0,55 kg d-1 animal-1) or grower/finisher pigs (1,92 kg d-1 animal-1) since the CO2 production increases as the animal weight grows. In fact, the emission ratios between two growing stages correspond approximately to the animal unit ratios.

In the same way, the greater amount of urine and feces excreted by sows favours the establishment of anaerobic conditions and the CH4 emission (30,1 g d-1 animal-1) in comparison with the offspring (2,77 and 5,54 g d-1 animal-1, respectively for weanling piglets and grower/finisher pigs).

N2O emissions from maternity and nursery were relatively close to zero as found in several studies. Grower/finisher pigs emit 0,66 g N2O d-1 animal-1. The non frequent change in protein requirements during the growth stage leads to more N being excreted.

## **5.2 Poultry production**

112 Greenhouse Gases – Emission, Measurement and Management

comprising both of gestating sows and farrowing sows with their piglets. Gestating sows are usually reared either in individual stalls or in group pens. Farrowing sows and their piglets are mainly kept in farrowing crates. The swine nursery or post-weaning building rears the weaning piglets brought from the maternity. Piglets remain there until they reach a certain weight, generally between 20 and 25 kg. Feed and ambient conditions are adjusted with pig growth. Then, pigs are transferred to grower/finisher facilities until they leave for

emissions\*\*

CO2 kg d-1 sow-1 5,29 1,83 9,35 2,26 CH4 g d-1 sow-1 30,1 13,3 119,7 25,3 N2O g d-1 sow-1 0,00 0,00 0,00 0,00

CO2 kg d-1 piglet-1 0,55 0,49 0,59 0,04 CH4 g d-1 piglet-1 2,77 0,32 10,7 4,11 N2O g d-1 piglet-1 0,007 0,000 0,010 0,005

CO2 kg d-1 pig-1 1,92 0,30 5,00 1,05 CH4 g d-1 pig-1 5,54 1,16 17,5 4,72 N2O g d-1 pig-1 0,66 0,00 3,50 1,39

\* Total greenhouse gas emissions calculated on a CO2-equivalent basis considering the mean emissions

Table 2. GHG emission factors for swine confined buildings (adapted from Hamelin et al.,

Overall, sows produce more CO2 on an animal basis (5,29 kg d-1 animal-1) than weanling piglets (0,55 kg d-1 animal-1) or grower/finisher pigs (1,92 kg d-1 animal-1) since the CO2 production increases as the animal weight grows. In fact, the emission ratios between two

In the same way, the greater amount of urine and feces excreted by sows favours the establishment of anaerobic conditions and the CH4 emission (30,1 g d-1 animal-1) in comparison with the offspring (2,77 and 5,54 g d-1 animal-1, respectively for weanling piglets

N2O emissions from maternity and nursery were relatively close to zero as found in several studies. Grower/finisher pigs emit 0,66 g N2O d-1 animal-1. The non frequent change in

protein requirements during the growth stage leads to more N being excreted.

\*\* References: Gallman et al., 2003; Gallmann & Hartung, 2000; Godbout et al., 2003, 2006; Groot Koerkamp & Uenk, 1997; Guarino et al., 2003; Guimont et al., 2007; Hinz & Linke, 1998; Lemay et al.,

Minimum value

sow-1 632 - - -

piglet-1 60,3 - - -

pig-1 321 - - -

Maximum value

Standard deviation

slaughter.

Growing

Maternity

Nursery

Grower/ finisher

2009)

phase Gas Units Mean

GHG \* g CO2-eq. d-1

GHG \* g CO2-eq. d-1

GHG \* g CO2-eq. d-1

and the global warming potential of CH4 (21) and N2O (310).

growing stages correspond approximately to the animal unit ratios.

2007; Sharpe et al., 2001; Zhang et al., 2007.

and grower/finisher pigs).

Broilers are mainly reared on a floor surface covered with bedding. Laying hens can be reared in multiple-deck battery cages, aviary systems, high-rise systems or percheries. In the first case, there is a possibility to dry manure directly under the cages with different manure drying systems. Table 3 presents emission factors from these two types of poultry production confinement buildings.


\* Total greenhouse gas emissions calculated on a CO2-equivalent basis considering the mean emissions and the global warming potential of CH4 (21) and N2O (310).

\*\* References: Chadwick et al., 1999; EPA, 2001; Fabbri et al., 2007; Fournel, 2011; Groot Koerkamp & Uenk, 1997; Hörnig et al., 2001; Monteny et al., 2001; Neser et al., 1997, as cited in Jungbluth et al., 2001; Sneath et al., 1996, as cited in Jungbluth et al., 2001; Wathes et al., 1997; Wu-Haan et al., 2007.

Table 3. GHG emission factors for broiler and layer confinement buildings

Broiler and layer productions emit similar quantities of CO2 to the atmosphere (31,5 and 28,2 kg yr-1 head-1, respectively). However, the different layer systems using liquid manure management generate a greater emission factor for CH4 (44,7 g yr-1 head -1) comparatively to broiler systems with litter (12,3 g yr-1 head -1). On the other hand, litter increases the succession of nitrification and denitrification phases which result in greater N2O emissions (17,6 g yr-1 head -1 vs. 10,9 g yr-1 head -1).

## **6. Mitigation techniques**

## **6.1 In animal production**

Several technologies have been developed to reduce gas and odour emissions from livestock housing. Several of these techniques, including reduction efficiencies for each technology, were inventoried by Godbout et al. (2010) from an exhaustive literature review. The inventory revealed that progress has been made in reducing odour and ammonia emissions, while less concern has been placed on GHG emissions. Three distinct techniques have been recognized, namely under slat separation, air cleaning and nutrient management.

#### **6.2 In-barn manure management: under-slat separation**

Separating urine and feces beneath the slats and removing both the solid and liquid fractions frequently is a reliable manure management technique to reduce gas and odour

Greenhouse Gas Emissions from Non-Cattle

they use the same principle.

**6.3 Air cleaning** 

Confinement Buildings: Monitoring, Emission Factors and Mitigation 115

In V-shaped scrapers (Fig. 8b), the feces stay on inclined gutter walls while urine is gathered into the bottom of the gutter and continuously drains out of the room by gravity. The solid fraction remaining on the inclined walls is scraped at a certain frequency using mechanically driven scrapers. Godbout et al. (2010) measured CO2 and CH4 emissions from a V-shaped scraper and compared these to a pull-plug system (emptied every week). However, even if

Conveyor belts (Fig. 8c) were adapted from poultry to swine production, placing the belt at an angle under the slatted portion of the pens. Its lower edge feeds into a pipe that collects the urine and transports it to the end of the building, thus allowing the separate collection of urine and feces within the hog house (van Kempen et al., 2003). Koger et al. (2003) founded that CH4 emission from a belt-based housing were reduced between 52 and 83% throughout

Most of studies evaluating these techniques have been mainly conducted in order to measure the reduction of NH3 emissions (Voermans and van Asseldonk, 1990; Voermans and van Poppel, 1993; Hendriks and van de Weerdhof, 1999). The study carried out by Belzile et al. (2006) found 13% and 19% CO2 and CH4 emission reduction, respectively, from a conveyor net compared to a drainage system without separation (emptied once a week). The same gas reduction values could be expected from the other separation techniques since

Air cleaning techniques for livestock buildings have the potential to improve air quality. However, efforts have been focused for improving performances on the reduction of dust and the abatement of NH3 and H2S. The air cleaning techniques are classified into two broad categories, physicochemical treatment and biological treatment (Godbout et al., 2010).

The physical - chemical absorption (scrubbing) is the physicochemical method most widely used for the treatment of air. This is a technology developed for many industrial applications. Gases are absorbed when the air from the barn is in contact with a liquid in which gas become soluble within the solution. The mass transfer from gas to the liquid is achieved by using a filter material within the filter unit (Devinny et al., 1999). The filter material usually has a large porosity, or void volume, and a large specific area (Melse & Ogink, 2005). Water is often used as the liquid solvent and its pH can be adjusted (basic or acid) depending on the pollutant to increase the solubility of gases. Contaminated air is introduced, either horizontally (crosscurrent) or upwards (counter-current), resulting in good contact between air and water, and enabling mass transfer from gas to liquid phase. A fraction of the trickling water is continuously recirculated; another fraction is discharged

On the other hand, a biological treatment of air is based on the capacity of microorganisms to transform organic and inorganic pollutants into non-toxic compounds and odour free (Devinny et al., 1999; Hartung et al., 2001; Revah & Morgan-Segastume, 2005). Three main types of bioreactors are currently used: biofilters, biotrickling filters and bioscrubbers (Fig. 10). The basic mechanism is the same for all biological treatment systems; the difference is

emission reductions were observed, there were no statistical differences.

the grower period studied comparatively to conventional pig houses.

and replaced by fresh water (Fig. 9) (Melse & Ogink, 2005).

emissions from buildings (Andersson, 1995; Arogo et al., 2001; Bernard et al., 2003; Jongebreur, 1981). The aim of this technique is to separate solid and liquid phases of the manure immediately after it falls through the slats to reduce the contact time between both phases. Three major under-slat manure separation systems have been studied over the years: the conveyor net, the V-shape scraper and the conveyor belt.

The conveyor net is composed of a mesh, tensioned under the slats, through which urine can flow while feces are collected. When this separation system is mechanized, the conveyor scrapes the feces to the end of the building while the urine stays in a gutter that is sloped toward a conventional storage pit (Fig. 8a).

Fig. 8. a) Conveyor net (Lemay et al., 2007); b) V-shape scraper (Lemay et al., 2007); c) Diagram of the conveyor belt (adapted from Lemay et al., 2007)

In V-shaped scrapers (Fig. 8b), the feces stay on inclined gutter walls while urine is gathered into the bottom of the gutter and continuously drains out of the room by gravity. The solid fraction remaining on the inclined walls is scraped at a certain frequency using mechanically driven scrapers. Godbout et al. (2010) measured CO2 and CH4 emissions from a V-shaped scraper and compared these to a pull-plug system (emptied every week). However, even if emission reductions were observed, there were no statistical differences.

Conveyor belts (Fig. 8c) were adapted from poultry to swine production, placing the belt at an angle under the slatted portion of the pens. Its lower edge feeds into a pipe that collects the urine and transports it to the end of the building, thus allowing the separate collection of urine and feces within the hog house (van Kempen et al., 2003). Koger et al. (2003) founded that CH4 emission from a belt-based housing were reduced between 52 and 83% throughout the grower period studied comparatively to conventional pig houses.

Most of studies evaluating these techniques have been mainly conducted in order to measure the reduction of NH3 emissions (Voermans and van Asseldonk, 1990; Voermans and van Poppel, 1993; Hendriks and van de Weerdhof, 1999). The study carried out by Belzile et al. (2006) found 13% and 19% CO2 and CH4 emission reduction, respectively, from a conveyor net compared to a drainage system without separation (emptied once a week). The same gas reduction values could be expected from the other separation techniques since they use the same principle.

## **6.3 Air cleaning**

114 Greenhouse Gases – Emission, Measurement and Management

emissions from buildings (Andersson, 1995; Arogo et al., 2001; Bernard et al., 2003; Jongebreur, 1981). The aim of this technique is to separate solid and liquid phases of the manure immediately after it falls through the slats to reduce the contact time between both phases. Three major under-slat manure separation systems have been studied over the

The conveyor net is composed of a mesh, tensioned under the slats, through which urine can flow while feces are collected. When this separation system is mechanized, the conveyor scrapes the feces to the end of the building while the urine stays in a gutter that is sloped

(a) (b)

(c)

Fig. 8. a) Conveyor net (Lemay et al., 2007); b) V-shape scraper (Lemay et al., 2007); c)

Diagram of the conveyor belt (adapted from Lemay et al., 2007)

years: the conveyor net, the V-shape scraper and the conveyor belt.

toward a conventional storage pit (Fig. 8a).

Air cleaning techniques for livestock buildings have the potential to improve air quality. However, efforts have been focused for improving performances on the reduction of dust and the abatement of NH3 and H2S. The air cleaning techniques are classified into two broad categories, physicochemical treatment and biological treatment (Godbout et al., 2010).

The physical - chemical absorption (scrubbing) is the physicochemical method most widely used for the treatment of air. This is a technology developed for many industrial applications. Gases are absorbed when the air from the barn is in contact with a liquid in which gas become soluble within the solution. The mass transfer from gas to the liquid is achieved by using a filter material within the filter unit (Devinny et al., 1999). The filter material usually has a large porosity, or void volume, and a large specific area (Melse & Ogink, 2005). Water is often used as the liquid solvent and its pH can be adjusted (basic or acid) depending on the pollutant to increase the solubility of gases. Contaminated air is introduced, either horizontally (crosscurrent) or upwards (counter-current), resulting in good contact between air and water, and enabling mass transfer from gas to liquid phase. A fraction of the trickling water is continuously recirculated; another fraction is discharged and replaced by fresh water (Fig. 9) (Melse & Ogink, 2005).

On the other hand, a biological treatment of air is based on the capacity of microorganisms to transform organic and inorganic pollutants into non-toxic compounds and odour free (Devinny et al., 1999; Hartung et al., 2001; Revah & Morgan-Segastume, 2005). Three main types of bioreactors are currently used: biofilters, biotrickling filters and bioscrubbers (Fig. 10). The basic mechanism is the same for all biological treatment systems; the difference is

Greenhouse Gas Emissions from Non-Cattle

Confinement Buildings: Monitoring, Emission Factors and Mitigation 117

The discharged water from a scrubber might be used as nitrogen fertilizer for crops; sometimes the water is added to the liquid manure storage (Melse et al., 2009). The discharge water from a biotrickling filter might be treated in a denitrification process in

(a) (b)

(c)

Fig. 10. a) Diagram of a closed biofilter system (adapted from Devinny et al., 1999); b) Diagram of a biotrickling filter (adapted from Revah & Morgan-Segastume, 2005); c) Diagram of a bioscrubber (adapted from Revah & Morgan-Segastume, 2005).

order to decrease the nitrogen content (Melse et al., 2009; Sakuma et al., 2008).

Reactor Microorganisms Liquid phase

Biofilter Fixed Stationary

Biotrickling filter Fixed Flowing Bioscrubber Suspended Flowing

Table 4. Classification of biological reactors for air treatment

due to the equipment configuration to carry out the transfer between the gas and the liquid, and on the pollutant biodegradation process (Table 4) (Devinny et al. 1999; Revah & Morgan-Segastume, 2005). Removal efficiency for NH3 and H2S emissions with a biological treatment can range from 6 to 100% and 3 to 99%, respectively (Nicolai & Janni 2001; Armeen, 2008; Iranpour et al., 2005). The reduction of odour emission is also widely variable, going from 29 to 100% depending of the operation conditions (Luo, 2001). A first bioreactor prototype developed by Belzile et al. (2010) found that NH3 emissions from small-scale swine chambers were reduced by 62 to 91% and H2S emissions were decreased by 24 to 66% by the biological treatment compared to a drainage system without separation (emptied once a week). However no significant reduction was obtained for CO2 and CH4 emissions at this first stage.

due to the equipment configuration to carry out the transfer between the gas and the liquid, and on the pollutant biodegradation process (Table 4) (Devinny et al. 1999; Revah & Morgan-Segastume, 2005). Removal efficiency for NH3 and H2S emissions with a biological treatment can range from 6 to 100% and 3 to 99%, respectively (Nicolai & Janni 2001; Armeen, 2008; Iranpour et al., 2005). The reduction of odour emission is also widely variable, going from 29 to 100% depending of the operation conditions (Luo, 2001). A first bioreactor prototype developed by Belzile et al. (2010) found that NH3 emissions from small-scale swine chambers were reduced by 62 to 91% and H2S emissions were decreased by 24 to 66% by the biological treatment compared to a drainage system without separation (emptied once a week). However no significant reduction was obtained for CO2 and CH4

Fig. 9. A counter-current air scrubber (adapted from Melse & Ogink, 2005)

emissions at this first stage.

The discharged water from a scrubber might be used as nitrogen fertilizer for crops; sometimes the water is added to the liquid manure storage (Melse et al., 2009). The discharge water from a biotrickling filter might be treated in a denitrification process in order to decrease the nitrogen content (Melse et al., 2009; Sakuma et al., 2008).


Table 4. Classification of biological reactors for air treatment

Fig. 10. a) Diagram of a closed biofilter system (adapted from Devinny et al., 1999); b) Diagram of a biotrickling filter (adapted from Revah & Morgan-Segastume, 2005); c) Diagram of a bioscrubber (adapted from Revah & Morgan-Segastume, 2005).

Greenhouse Gas Emissions from Non-Cattle

typical emissions from building.

effect on GHG is not clearly shown.

No.6, pp. 21-28

**8. Acknowledgments** 

chapter.

**9. References** 

Confinement Buildings: Monitoring, Emission Factors and Mitigation 119

In the case where is not possible to measure emissions, values from literature can be used for swine, broiler and layer productions. The typical CO2 emissions for sows, weanling piglets and grower/finisher pigs are 5,29, 0,55 and 1,92 kg d-1 animal-1, respectively. The greater amount of urine and faeces excreted by sows favours the establishment of anaerobic conditions and the CH4 emissions (30,1 g d-1 animal-1) in comparison with the offspring (2,77 and 5,54 g d-1 animal-1, respectively for weanling piglets and grower/finisher pigs). N2O emissions from maternity and nursery were relatively close to zero as found in several studies. Grower / finisher pigs emit 0,66 g N2O d-1 animal-1. Broiler and layer productions emit similar quantities of CO2 to the atmosphere (31,5 and 28,2 kg yr-1 animal-1, respectively). However, the different layer systems using liquid manure management generate a greater emission factor for CH4 (44,7 g yr-1 animal-1) comparatively to broiler systems with litter (12,3 g yr-1 animal-1). The N2O emissions range from 10,9 to 17,6 g yr-1 head-1. However, since the gas emissions are influenced by many factors, on-site measurement should be privileged instead of typical values from literature to determine the

Several technologies have been developed to reduce odour and gas emissions from swine housing. Two in-barn approaches are encouraging: the under slat separation system and diet manipulation while air cleaning systems have been developed for the exhaust air outlet. In agreement with the literature, the under slat separation system can reduce around 13% and 19% of CO2 and CH4 emissions, respectively. No studies have reported GHG emission reductions using diet manipulation. Many types of air cleaning systems already exist and they have been developed mainly for odour and ammonia emission reductions and the

The authors wish to acknowledge P. Brassard and M. Girard for their contributions to this

Andersson, M. (1995). *The Effect of Different Manuring Systems on Ammonia Emissions in Pig* 

Armeen, A.; Feddes, J. J. R.; Leonard, J. J., & Coleman, R. N. (2008). Biofilters to Treat Swine

Arnold, S. L.; Parkin, T. B.; Doran, J. W.; Eghball, B. & Mosier, A. R. (2001). Automated Gas

Arogo, J.; Westerman, P.W.; Heber, A.J.; Robarge, W.P. & Classen, J.J. (2001). Ammonia in

Department of Agricultural Biosystems and technology

*Science and Plant Analysis*, Vol.32, No.17&18, pp. 2795–2807

4089, Sacramento, California, USA, July 30-August 1, 2001 ASABE. (2010). *Psychrometric Data*, American Society of Agricultural and Biological

Engineers,ASAE Standard D271.2 APR1976 (R2010)

*Buildings. A Comparing Study between Scrapers Under Slatted Floors and Rubber Mats Under Slatted Floors*, Report 100, Lund, Swedish University of Agricultural Sciences,

Facility Air: Part 1. Nitrogen Mass Balance. *Canadian Biosystems Engineering*, Vol.50,

Sampling System for Laboratory Analysis of CH4 and N2O. *Communications in Soil* 

Animal Production –A Review, *ASAE Annual International Meeting*, Paper No 01-

### **6.4 Nutrient management**

An additional method to reduce emissions caused by excess nitrogen is the alteration of the ratio of nitrogen excretion in urine versus feces by nutrient management (Mroz et al., 1993). Reduction of dietary protein combined with supplementation of synthetic amino acids in pig diets might reduce total nitrogen excretion by 25 to 40% (Hartung & Phillips, 1994; Kay & Lee, 1997). Additionally, the inclusion of fermentable carbohydrates or non-starch polysaccharides into diets stimulates bacterial fermentation in the hindgut and reduced urinary versus fecal nitrogen ratio by 68% (Canh et al., 1997a).

However, GHG emission reduction is generally not measured or documented when using nutrient management. Principally, studies target reducing NH3 and other odorant compound emissions (Garry et al., 2007; Le et al., 2006; Lyngbye et al., 2006). GHG emissions measurements (CO2, CH4 and N2O) were carried out by Godbout et al. (2010) when protein content is reduced and lysine is increased in the diet. As a result, such diet treatment presented no impact on CO2 and N2O emissions while CH4 emissions increased by 58% compared to a commercial diet. Therefore, a more thorough analysis should be carried out for a better understanding of dietary management in GHG emission reduction.

## **7. Summary and conclusions**

Contaminants exhausted from confined animal buildings include various gases, dust particles, micro-organisms and odours. The most important gases are CO2, NH3, H2S, CH4, N2O and some trace gases (aldehydes, amines, aromatics, organic acids, sulphur compounds, etc.). The main GHG emitted from livestock building are CH4, N2O (from manure decomposition) and CO2 (from animal metabolism). Generally, CH4 emissions are more present in liquid manure management while N2O is produced under solid manure management.

The emission is the product of the gas concentration and the air exchange rate. An accurate measurement of these values is very important and is still a challenge today for emissions from agricultural sources. Since the agricultural emissions are very low, the concentration measurement requires equipment having great sensitivity and selectivity like those utilizing optical properties of gas such as FTIR, PAS and NDIR or separation techniques like chromatography with selective detectors. Most of the gases in air which are mainly composed of N2, O2 and Ar with several others gases in trace concentrations like CO2, CH4 and N2O can be separated by chromatography and detected by different detectors more or less specific to the target gas. The technique is simple, proven and allows the simultaneous quantification of CO2, CH4 and N2O. Compared to other techniques having the required sensitivity, like modern spectroscopic techniques, gas chromatography is known to produce reliable results and can be envisaged as moderate to low cost techniques with easy apparatus implementation.

The air flow measurement is very important and often, a lot of uncertainty is related to this value bringing an error in the emission determination. The measurement techniques are function of the ventilation system. Three main systems exist: MV, NV and HV buildings. Various methods have been developed to measure ventilation rate from animal housing facilities including airborne tracer techniques, diffusion of animal-produced CO2 or heat, vane anemometers and orifice plates. Each of these methods, however, has limitations.

In the case where is not possible to measure emissions, values from literature can be used for swine, broiler and layer productions. The typical CO2 emissions for sows, weanling piglets and grower/finisher pigs are 5,29, 0,55 and 1,92 kg d-1 animal-1, respectively. The greater amount of urine and faeces excreted by sows favours the establishment of anaerobic conditions and the CH4 emissions (30,1 g d-1 animal-1) in comparison with the offspring (2,77 and 5,54 g d-1 animal-1, respectively for weanling piglets and grower/finisher pigs). N2O emissions from maternity and nursery were relatively close to zero as found in several studies. Grower / finisher pigs emit 0,66 g N2O d-1 animal-1. Broiler and layer productions emit similar quantities of CO2 to the atmosphere (31,5 and 28,2 kg yr-1 animal-1, respectively). However, the different layer systems using liquid manure management generate a greater emission factor for CH4 (44,7 g yr-1 animal-1) comparatively to broiler systems with litter (12,3 g yr-1 animal-1). The N2O emissions range from 10,9 to 17,6 g yr-1 head-1. However, since the gas emissions are influenced by many factors, on-site measurement should be privileged instead of typical values from literature to determine the typical emissions from building.

Several technologies have been developed to reduce odour and gas emissions from swine housing. Two in-barn approaches are encouraging: the under slat separation system and diet manipulation while air cleaning systems have been developed for the exhaust air outlet. In agreement with the literature, the under slat separation system can reduce around 13% and 19% of CO2 and CH4 emissions, respectively. No studies have reported GHG emission reductions using diet manipulation. Many types of air cleaning systems already exist and they have been developed mainly for odour and ammonia emission reductions and the effect on GHG is not clearly shown.

## **8. Acknowledgments**

The authors wish to acknowledge P. Brassard and M. Girard for their contributions to this chapter.

## **9. References**

118 Greenhouse Gases – Emission, Measurement and Management

An additional method to reduce emissions caused by excess nitrogen is the alteration of the ratio of nitrogen excretion in urine versus feces by nutrient management (Mroz et al., 1993). Reduction of dietary protein combined with supplementation of synthetic amino acids in pig diets might reduce total nitrogen excretion by 25 to 40% (Hartung & Phillips, 1994; Kay & Lee, 1997). Additionally, the inclusion of fermentable carbohydrates or non-starch polysaccharides into diets stimulates bacterial fermentation in the hindgut and reduced

However, GHG emission reduction is generally not measured or documented when using nutrient management. Principally, studies target reducing NH3 and other odorant compound emissions (Garry et al., 2007; Le et al., 2006; Lyngbye et al., 2006). GHG emissions measurements (CO2, CH4 and N2O) were carried out by Godbout et al. (2010) when protein content is reduced and lysine is increased in the diet. As a result, such diet treatment presented no impact on CO2 and N2O emissions while CH4 emissions increased by 58% compared to a commercial diet. Therefore, a more thorough analysis should be carried out for a better understanding of dietary management in GHG emission reduction.

Contaminants exhausted from confined animal buildings include various gases, dust particles, micro-organisms and odours. The most important gases are CO2, NH3, H2S, CH4, N2O and some trace gases (aldehydes, amines, aromatics, organic acids, sulphur compounds, etc.). The main GHG emitted from livestock building are CH4, N2O (from manure decomposition) and CO2 (from animal metabolism). Generally, CH4 emissions are more present in liquid manure management while N2O is produced under solid manure

The emission is the product of the gas concentration and the air exchange rate. An accurate measurement of these values is very important and is still a challenge today for emissions from agricultural sources. Since the agricultural emissions are very low, the concentration measurement requires equipment having great sensitivity and selectivity like those utilizing optical properties of gas such as FTIR, PAS and NDIR or separation techniques like chromatography with selective detectors. Most of the gases in air which are mainly composed of N2, O2 and Ar with several others gases in trace concentrations like CO2, CH4 and N2O can be separated by chromatography and detected by different detectors more or less specific to the target gas. The technique is simple, proven and allows the simultaneous quantification of CO2, CH4 and N2O. Compared to other techniques having the required sensitivity, like modern spectroscopic techniques, gas chromatography is known to produce reliable results and can be envisaged as moderate to low cost techniques with easy

The air flow measurement is very important and often, a lot of uncertainty is related to this value bringing an error in the emission determination. The measurement techniques are function of the ventilation system. Three main systems exist: MV, NV and HV buildings. Various methods have been developed to measure ventilation rate from animal housing facilities including airborne tracer techniques, diffusion of animal-produced CO2 or heat, vane anemometers and orifice plates. Each of these methods, however, has limitations.

urinary versus fecal nitrogen ratio by 68% (Canh et al., 1997a).

**6.4 Nutrient management** 

**7. Summary and conclusions** 

management.

apparatus implementation.


Greenhouse Gas Emissions from Non-Cattle

Vol.50, pp. 5.1-5.7

Sabatier, Toulouse, France

Vol.26, No.1, p. 53-56

Denmark, June 1-4, 2003

USA

Lewis Publishers, Washington, DC, USA

293-297

Confinement Buildings: Monitoring, Emission Factors and Mitigation 121

Choiniere, Y.; Blais, F. & Munroe, J.A. (1988). A Wind Tunnel Study of Airflow Patterns in a

CIGR. (2002). *4th Report of Working Group on Climatization of Animal Houses – Heat and* 

Clark, O.G.; Segura, J.C.; Feddes, J.J.R. & Ouellette, C. (2008). Multiport Averaging Pitot

Cowper, C.J. & DeRose, A.J. (1983). The Analysis of Gases by Chromatography. Pergamon Series in Analytical Chemistry, Volume 7, Pergamon Press Ltd., United Kingdom de la Farge, B. (1978). Les déchets de l'élevage porcin : traitement – valorisation [Swine

Devinny, J. S.; Deshusses, M. A. & Webster, T. S. (1999). *Biofiltration for Air Pollution Control*,

EPA. (2001). *Emissions from Animal Feeding Operations*. Environmental Protection Agency,

Fabbri, C.; Valli, L., Guarino, M., Costa, A. & Mazzotta, V. (2007). Ammonia, Methane,

Feddes, J.J.R. & J.B. McQuitty. (1980). Design of an Air-Speed Sensor System. *Canadian* 

Feddes, J.J.R.; Leonard, J.J. & McQuitty, J.B. (1984). Carbon Dioxide Concentration As a

Fournel, S. (2011). *Étude comparative des émissions de gaz et d'odeurs issues de différents systèmes* 

Production], M.Sc. thesis, Université Laval, Québec City, Québec, Canada Gallmann, E. & Hartung, E. (2000). Evaluation of the Emission Rates of Ammonia and

Gallmann, E.; Hartung, E. & Jungbluth, T. (2003). Long-Term Study Regarding the Emission

Garry, B.P.; Fogarty, M.; Curran, T.P. & O'Doherty, J.V. (2007). The Effect of Cereal Type

Standards Division, Office of air quality planning and standards

Laying Hens. *Biosystems Engineering*, Vol.97, pp. 441-455

*Agricultural Engineering*, Vol.22, No.1, pp. 97-99

Agricultural Engineering, Section II, Horsens, Denmark

Naturally Ventilated Building. *Canadian Agricultural Engineering*, Vol.30, No.2, pp.

*Moisture Production at Animal and House Levels,* International Commission of

Tube to Measure Airflow Rates from Exhaust Fans. *Canadian Biosystems Engineering*,

Production Wastes: Treatment – Valorization]. Ph.D thesis, Université Paul

Draft Report No. 68-D6-0011. Research Triangle Park, North Carolina: Emission

Nitrous Oxide and Particulate Matter Emissions from Two Different Buildings for

Measure of Air Exchange in Animal Housing. *Canadian Agricultural Engineering*,

*de gestion des déjections en production d'œufs de consommation* [Comparative Study of Gas and Odour Emissions from Different Manure Management Systems in Egg

Greenhouse Gases from Swine Housings, *Proceedings of the 2nd International Conference on Air Pollution from Agricultural Operations*, pp. 92-99, Des Moines, Iowa,

Rates of Ammonia and Greenhouse Gases from Different Housing Systems for Fattening Pigs – Final Results, *Proceedings of the International Symposium on Gaseous and odour emissions from animal production facilities*, pp. 122-130, Horsens,

and Exogenous Enzyme Supplementation in Pig Diets on Performance, Odour and Ammonia Emissions from Finisher Pigs. *Livestock Science*, Vol.109, pp. 212–215


Heating, Refrigerating and Air-Conditioning Engineers, inc., ANSI/ASHRAE 41.2-

nauséabonds? [Why Do Gases from Liquid Manure Stink So Much?]. *Le producteur* 

(2010). Reduction of Gas and Odour Emissions from a Swine Building Using a Biotrickling Filter, *XVIIth World Congress of the International Commission of* 

de la séparation fèces-urine sous caillebotis sur la qualité de l'air ambiant en porcherie [Impact of the Separation of Feces and Urine Under the Slats on the Quality of Ambient Air in Swine Buildings]. *Journées Recherche Porcine*, Vol.38, pp.

Emissions in Milk-Fed Calf Buildings, *CSBE/SCGAB 2006 Annual Conference*, July

de l'Institut de recherche et de développement en agroenvironnement inc. [Report of the Research and Development Institute for the Agri-Environment inc.]. In: *Commission sur le développement durable de la production porcine au Québec*, April 16,

Methane and Nitrous Oxide from Pig Houses and Slurry: Effects of Rooting Material, Animal Activity and Ventilation Flow. *Agriculture, Ecosystems and* 

*Change*. Oxford University Press, New York. In: Cabaraux, J.F.; Philippe, F.X.; Laitat, M.; Canart, B.; Vandenheede, M. & Nicks, B. (2008). Gaseous Emissions from Weaned Pigs Raised on Different Floor Systems. *Agriculture, Ecosystems and* 

Factors on Nitrogen Partitioning and Compositions of Urine and Feces of Fattening

Oxide Emissions from Farmed Livestock. *Atmospheric Environment*, Vol.33, No.20,

ASHRAE. (1992). *Standard Methods for Laboratory Airflow Measurement*, American Society of

Barrington, S. (1999). Comment se fait-il que les gaz des fumiers liquides soient tellement

Barber, E.M. ; Strom, J.S.; Bantle, M.R.L.; Morsing, S. & Feddes, J.J.R. (1994). Ventilation Performance Standards – Applying Danish Experience in Canada. *Applied* 

Belzile, M.; Lemay, S.P.; Zegan, D.; Feddes, J.J.R.; Godbout, S.; Larouche, J.-P.; & Martel, M.

*Agricultural Engineering*, Québec City, Québec, Canada, June 13-17, 2010 Belzile, M.; Godbout, S. ; Lemay, S.P. ; Lavoie, J. ; Lachance, I. & Pouliot, F. (2006). Impact

Belzile, M.; Lemay, S.P.; Godbout, S.; Côté, C. & Lavoie, J. (2006). Gas Concentrations and

Bernard, C. ; Côté, D. ; Giroux, M. ; Grégoire, R. ; Joncas, R. & Martin, D.-Y. (2003). Mémoire

Blackmer, A.M. & Bremner, J.M. (1977). Gas Chromatographic Analysis of Soil Atmospheres. *Soil Science Society of American Journal*, Vol.41, pp. 908-912 Blanes-Vidal, V.; Hansen, M.N.; Pedersen, S. & Rom, H.B. (2008). Emissions of Ammonia,

Brasseur, G.P.; Orlando, J.J. & Tyndall, G.S., (Ed.). (1999). *Atmospheric Chemistry and Global* 

Canh, T.T.; Verstegen, M.W.A.; Aarnink, A.J.A. & Schrama, J.W. (1997). Influence of Dietary

Chadwick, D.R.; Sneath, R.W., Phillips, V.R. & Pain, B.F. (1999). A UK Inventory of Nitrous

1987 (RA 92), ISSN 1041-2336, Atlanta, Georgia, USA

*Engineering in Agriculture*, Vol.10, No.3, pp. 395-401

*de lait québécois*, Vol.20, No.3, pp. 41-45

21-26.

2003

16-19, 2006

*Environment*, Vol.124, pp. 237-244

*Environment*, Vol.130, pp. 86-92

pp. 3345-3354

Pigs. *Journal of Animal Science*, Vol.75, pp. 700-706


Greenhouse Gas Emissions from Non-Cattle

Confinement Buildings: Monitoring, Emission Factors and Mitigation 123

Heber, A.; Lim, T.; Ni, J.; Grant, R. & Sutton, A. (2000). *Development of a Site-Specific Odor* 

Heinemeyer, O. & Kaiser, E.A. (1996). Automated Gas Injector System for Gas

Hendriks, H.J.M & van de Weerdhof, A.M. (1999). *Dutch Notes on Bat for Pig- and Poultry* 

Hinz, T. & Linke, S. (1998). A Comprehensive Experimental Study of Aerial Pollutants in

Hörnig, G.; Brunsch, R., Stollberg, U., Jelínek, A., Plíva, P. & Češpiva, M. (2001). Ammonia,

Houghton, J.T.; Meira Filho, L.G.; Callander, B.A.; Harris, N.; Kattenberg, A. & Maskel, K.,

Iranpour, R.; Cox, H.H.J.; Deshusses M.A. & Schroeder, E.D. (2005). Literature Review of Air

Jongebreur, A.A. (1981). *Housing System and Their Influence on the Environment. Environmental Aspects of Housing for Animal Production*, pp. 431-436, Butterworths, London, UK Jungbluth, T.; Hartung, E. & Brose, G. (2001). Greenhouse Gas Emissions from Animal Houses and Manure Stores. *Nutrient Cycling in Agroecosystems*, Vol.60, pp. 133-145 Kay, R.M. & Lee, P.A. (1997). Ammonia Emission from Pig Buildings and Characteristics of

Koger, J.B.; Kaspers, B.A.; Burnette, R.P.; van Kempen, M.H.J.G. & van Kempen, T.A.T.G.

Compound Removal. *Environmental Progress*, Vol.24, No.3, pp. 254-267 ITP. (1998). Odeurs et environnement – Cas de la production porcine [Odours and Environment – Swine Production Case]. Institut technique du porc, Paris, France Jiang, X.; Ku, W.L.; Shia, R.L.; Li, Q.; Elkins, J.W.; Prinn, R.G. & Yung, Y.L. (2007). Seasonal

Washington, DC: National Pork Producers Council

*American Journal*, Vol.60, pp. 808-811

*Engineering Research*, Vol.70, pp. 119-129

and Fisheries, The Netherlands

Cambridge University Press

doi:10.1029/2006GB002691

October, 2001

Netherlands

USA, 21 May, 2003

*Impact Distance Guideline for Swine Production Systems*, Report No. 98-131.

Chromatography: Atmospheric Nitrous Oxide Analysis. *Soil Science Society of* 

*Intensive Livestock Farming*. Draft published by Ministery of Housing, Spatial Planning and the Environment et Ministery of Agriculture, Nature Management

and Emissions from Livestock Buildings – Part 2: Results. *Journal of Agricultural* 

Methane and Carbon Dioxide Emissions from Laying Hens Kept in Battery Cages and Aviary Systems, *Proceedings of the 2nd Agricultural Engineering Conference of Central and East European Countries*, pp. 37-43, Prague, Czech Republic, 23-24

(Ed). (1995). *IPCC Second Assessment Report: Climate Change 1995*, 572 pp.,

Pollution Control Biofilters and Biotrickling Filters for Odor and Volatile Organic

Cycle of N2O: Analysis of Data. *Global Biogeochemical Cycles*, Vol.21,

Slurry Produced by Pigs Offered Low Crude Protein Diets, *Proceedings of the International Symposium on Ammonia and Odour Control from Animal Production Facilities*, pp. 253-260, CIGR and EurAgEng publication, Rosmalen, The

(2003). *Manure Belts for Harvesting Urine and Feces Separately and Improving Air Quality in Swine Facilities*. Report to the Smithfield Panel, Raleigh, North Carolina,


Guarino, M.; Fabbri, C., Navarotto, P., Valli, L., Moscatelli, G., Rossetti, M. & Mazotta, V.

Godbout, S.; Lemay, S.P.; Pelletier, F.; Belzile, M.; Larouche, J.P.; Tamini, L.D.; Palacios, J. H.

Godbout, S.; Belzile, M., Lachance, I., Lemay, S.P., Turgeon, M.J., Dufour, V., Pouliot, F. &

Godbout, S.; Lavoie, J.; Lemay, S.P.; Lachance, I.; Pouliot, F. & Belzile, M. (2005). Impact of

*animal production facilities*, pp. 426-443, Horsens, Denmark, June 1-4, 2003 Groot Koerkamp, P.W.G. & Uenk, G.H. (1997). Climatic Conditions and Aerial Pollutants

Hamelin, L.; Pelletier, F. & Godbout, S. (2007). Impacts environnementaux, économiques et

Hartung, J. & Phillips, V.R. (1994). Control of Gaseous Emissions from Livestock Buildings

Research and Development Institute for the Agri-Environment inc.

*Annual International Meeting*, Paper No. 055018, Tampa, Florida, USA Godbout, S.; Laguë, C., Lemay, S.P., Marquis, A. & Fonstad, T.A. (2003). Greenhouse Gas

*facilities*, pp. 140-149, Horsens, Denmark, June 1-4, 2003

Institute for the Agri-Environment inc.

Institute for the Agri-Environment inc.

Coop fédérée & AAC

173-189

(2003). Ammonia, Methane and Nitrous Oxide Emissions and Particulate Matter Concentrations in Two Different Buildings for Fattening Pigs, *Proceedings of the International Symposium on Gaseous and odour emissions from animal production* 

& Zegan, D. (2010). *Réduction des émissions gazeuses et odorantes aux bâtiments porcins : techniques simples et efficaces applicables à la ferme* [Gas and Odour Emission Reductions in Swine Barns: Simple and Effective Techniques Applicable at the Farm], Final Report, Québec City, Québec, Canada: Research and Development

Marquis, A. (2006). *Évaluation technico-économique d'un système de séparation solide/liquide des déjections à la source dans un bâtiment porcin et les impacts sur l'environnement – Volet II* [Technico-Economical Assessment of an In-Barn Solid/Liquid Separation System for Pig Slurry and Impacts on the Environment – Phase II], Final Report, Québec City, Québec, Canada: Research and Development

In-Barn Manure Separation on the Biological Air Quality of Swine Buildings, *ASAE* 

and Odour from Swine Operations Under Liquid Manure Management in Canada, *Proceedings of the International Symposium on Gaseous and odour emissions from* 

in and Emissions from Commercial Animal Production Systems in the Netherlands, *Proceedings of the International Symposium of Ammonia and Odour Control from Animal Production Facilities*, pp. 139-144, AB Rosmalen : Netherlands Guimont, H.; Dufour, V. ; Pelletier, F. ; Coulibaly, A. ; Giguère, C. ; Godbout, S. ; Lemay,

S.P. ; Massé, D.I. ; Pouliot, F. & Fortin, N. (2007). Évaluation technique et économique d'un système d'isolement des fèces avec grattes en « V » dans un engraissement commercial. [Technical and Economical Assessment of a « V » Scraper System for Isolation of Feaces in a Commercial Fattening Barn], Rapport final [Final Report], Québec City, Québec, Canada : CDPQ, IRDA, La

sociaux des différentes chaînes de gestion des déjections des poules pondeuses [Environmental, Economical and Social Impacts of Different Manure Management Systems in Layer Production], Final Report, Québec City, Québec, Canada:

and Manure Stores. *Journal of Agricultural Engineering Research*, Vol.57, No.3, pp.


Greenhouse Gas Emissions from Non-Cattle

Vol.51, pp. 1819-1825

496

Jacobson.

Springer, Verlag, Berlin, Heidelberg

Elsevier, Amsterdam, 324 pp.

des producteurs de viande de Bretagne, 67 pp.

Engineering, University of Minnesota, USA

Gas Division, Environment Canada, 205 pp.

*Agricultural Engineering Research*, Vol.53, No.1, pp. 23-50

Combination System. *Water research*, Vol.42, pp. 4507–4513

Confinement Buildings: Monitoring, Emission Factors and Mitigation 125

Nicolai, R.E. & Janni, K.A. (2001). Biofilter Media Mixture Ratio of Wood Chips and

Olsen, K.; Wellisch, M.; Boileau, P.; Blain, D.; Ha, C.; Henderson, L.; Liang, C.; McCarthy, J.

O'Neill, D.H. & Phillips, V.R. (1992). A Review of the Control of Odour Nuisance from

Philippe, F.X.; Laitat, M.; Canart, B.; Vandenheede, M. & Nicks, B. (2007). Comparison of

Revah, S. & Morgan-Sagastume, J.M. (2005). Methods of Odor and VOC Control. In:

Sakuma, T.; Jinsiriwanit, S.; Hattori, T. & Deshusses, M.A. (2008). Removal of Ammonia

Sharpe, R.R.; Harper, L.A. & Simmons, J.D. (2001). Methane Emissions from Swine Houses

Sitaula, B.K.; Luo, J. & Bakken, L.R. (1992). Rapid Analysis of Climate Gases by Wide Bore

Taiganides, E. P. (1987). Animal Waste Management and Wastewater Treatment. In: *World* 

UGPVB. (1996). Les gaz en élevage porcin: une problématique à dominance ammoniac

van Kempen, T.; Kaspers, B.; Burnette, P.; van Kempen, M. & Koger, J.B. (2003). Swine

Voermans, J.A.M. & van Asseldonk, M.M.L. (1990). Separation of Pig Manure Under a Slatted Floor. *Proefverslag-Proefstation-voor-de-Varkenshouderij*, No P1.151, 32 pp. Voermans, J.A.M. & van Poppel, F. (1993). Scraper Systems in Pig Houses. *Proceedings of* 

*Livestock environment IV*, pp. 650-656, Coventry, UK, July 6-9, 1993

in North Carolina. *Chemosphere – Global change science*, Vol.3, pp. 1-6

Compost Treating Swine Odors, Department of Biosystems and Agricultural

& McKibbon, S. (2003). *Canada's Greenhouse Gas Inventory: 1999-2001*, Greenhouse

Livestock Buildings. Part 3. Properties of the Odorous Substances, Which Have Been Identified in Livestock Wastes or in the Air Around Them. *Journal of* 

Ammonia and Greenhouse Gas Emissions Furing the Fattening of Pigs, Kept Either on Fully Slatted Floor or on Deep Litter. *Livestock Science*, Vol.111, pp. 144-152 Phillips, M.P.; Sievers, R.E.; Goldan, P.D.; Kuster, W.C. & Fehsenfeld, F.C. (1979).

Enhancement of Electron Capture Detector Sensitivity to Nonelectron Attaching Compounds by Addition of Nitrous Oxide to the Carrier Gas. *Analytical Chemistry*,

*Biotechnology for Odour and Air Pollution Control*, Shareefdeen, Z. & Singh, A., (Ed.),

from Contaminated Air in a Biotrickling Flter – Denitrifying Bioreactor

Capillary Gas Chromatography. *Journal of Environmental Quality*, Vol.21, pp. 493-

*Animal Science B6, Animal Production and Environmental Health*, Strauch (Ed.),

[Gases in Swine Production: Ammonia Is a Major Case], Union des groupements

Housing With A Belt For Separating Urine And Feces; Key To Flexibility. *Swine Housings II Proceedings Conference*, pp. 159-165, Research Triangle Park, North Carolina, USA, 12-15 October, 2003. ASAE Publication Number 701P1303, ed. L.D.


Le, P.D.; Aarnink, A.J.A.; Jongbloed, A.W.; van der Peet Schwering, C.M.C.; Ogink, N.W.M.

Leonard, J.J.; Feddes, J.J.R. & McQuitty, J.B. (1984). Measurement of VR Using a Tracer Gas.

Loftfield, N.; Flessa, H.; Augustin, J. & Beese, F. (1997). Automated Gas Chromatographic

Lyngbye, M.; Hansen, M.J.; Riis, A.L.; Jensen, T.L. & Sørensen, G. (2006). 1000 Olfactometry

Melse, R.W.; Ogink, N.W.M. & Rulkens, W.H. (2009). Air Treatment Techniques for

Monteny, G.J.; Groenestein, C.M. & Hilhorst, M.A. (2001). Interactions and Coupling

Morsing, S.; Ikeguchi, A.; Bennetsen, J.C.;. Strøm, J.S; Ravn, P. & Okushima, L. (2002). Wind

Mosier, A. R. & Mack, L. (1980). Gas Chromatographic System for Precise, Rapid Analysis of Nitrous Oxide. *Soil Science Society of American Journal*, Vol.44, pp. 1121-1123 Mroz, Z.; Jongbloed, A.W.; Beers, S.; Kemme, P.A.; de Jonge, L.; van Berkum, A.K. & van der

Nääs, I.A.; Moura, D.J.; Bucklin, R.A. & Fialho, F.B. (1998). An Algortithm for Determining

Neftel, A.; Fischer, C. & Flechard, C. (2006). Measurements of Greenhouse Gas Fluxes from

System for Rapid Analysis of the Atmospheric Trace Gases Methane, Carbon Dioxide, and Nitrous Oxide*. Journal of Environmental Quality*, Vol.26 pp. 560-564 Luo, J. (2001). A Pilot-Scale Study on Biofilters for Controlling Animal Rendering Process

Analyses and 100 TD-GC/MS Analyses to Evaluate Methods for Reducing Odour from Finishing Units in Denmark, *Workshop on Agricultural Air Quality*,

Abatement of Emissions from Intensive Livestock Production. *The Open Agriculture* 

between Emissions of Methane and Nitrous Oxide from Animal Husbandry.

Induced Isothermal Airflow Patterns in a Scale Model of a Naturally Ventilated Swine Barn with Cathedral Ceiling. *Applied Engineering in Agriculture*, Vol.18, No.1,

Lee, R.A. (1993). Preliminary Studies on Excretory Patterns of Nitrogen and Anaerobic Deterioration of Faecal Protein from Pigs Fed Various Carbohydrates. In: *Nitrogen Flow in Pig Production and Environmental Consequences*, Verstegen, M.W.A.; den Hartog, L.A.; van Kempen, G.J.M. & Metz, J.H.M., (Ed.), 247-252,

Opening Effectiveness in Natural Ventilation by Wind. *Transactions of the ASAE*,

Development Institute for the Agri-Environment inc.

Washington, D.C., USA

*Journal*, Vol.3, pp. 6-12

Vol.41, No.3, pp. 767-771

pp. 97-101

*Canadian Agricultural Engineering*, Vol.26, No.1, pp. 49-51

*Nutrient Cycling in Agroecosystems*, Vol.60, pp. 123–132

Pudoc Scientific Publishers, Wageningen, The Netherlands

Agriculture. *International Congress Series*, vol.1293, pp. 3-12

Odours. *Water Science and Technology*, Vol.44, No.9, pp. 277-285

& Verstegen, M.W.A. (2006). Effects of Crystalline Amino Acid Supplementation to the Diet on Odor from Pig Manure. *Journal Animal Science*, Vol.85, pp. 791–801 Lemay, S.P.; Godbout, S., Bergeron, R., Belzile, M., Predicala, B., Laguë,C., Marquis, A.,

Pouliot, F. & Rondeau, F. (2007). *Développement d'un nouveau concept d'enclos pour élevages porcins "sans lattes" permettant de séparer les déjections et de réduire les émissions gazeuses et odorants* [Development of a New Pen Concept Without Slats Allowing to Separate Slurry Phases and Reduce Gaseous and Odorant Emissions for Pig Production], Final Report, Québec City, Québec, Canada: Research and


**6**

*USA* 

Risa Kumazawa *Duquesne University* 

**The Effect of Organic Farms on Global** 

The emission of greenhouse gases (GHGs) is a great concern for mankind as it believed that when GHGs such as carbon dioxide (CO2), nitrous oxide (N2O), and methane (CH4), trap energy from the sun, they contribute to global warming. Even though greenhouse gases occur naturally in the atmosphere, there are noteworthy man-made contributions through agricultural and industrial production. Agricultural land itself contributes to 12% of global GHG emissions while emissions from all sectors related to agricultural production contribute to an estimated 25-30% of all GHG emissions (International Trade Centre, 2007). The Kyoto Protocol was established in Kyoto, Japan in December, 1997 as an amendment to the United Nations Framework Convention on Climate Change (UNFCCC) to regulate global GHG emissions. It remains the most comprehensive international agreement to date through the setting of target reductions by industrialized nations in North America, Europe, Asia and Australia, commonly known as "Annex I" countries, shown in Table A-1 of the Appendix.

The emissions of methane, projected to have approximately 20 times the global warming potential of carbon dioxide in trapping heat in the atmosphere (Morgenstern, 1991; Shih et al., 2006; International Trade Centre, 2007; EPA, 2011a) are caused by enteric fermentation and manure management, both of which are strongly correlated with livestock numbers (Vermont & De Cara, 2010), in addition to rice cultivation (McCarl & Schneider, 2000; Nalley et al., 2011). Enteric fermentation is a digestive process where microbes in ruminant animals such as cows, goat and sheep break down food, producing methane as a byproduct. Anaerobic decomposition (without oxygen) of organic matter in livestock manure leads to methane emissions. The U.S. Environmental Protection Agency's website also mentions natural sources of methane such as wetlands, oceans, wildfires as well as man-made sources

The emissions of nitrous oxide, projected to be far more potent at about 300 times the global warming potential of carbon dioxide (International Trade Centre, 2007; EPA, 2011b), stem from livestock management (nitrogen content from animal feeds), soil disturbance (both soil loss and degradation), fertilizer/ other chemical use and the burning of agricultural residues (Ruttan, 2002; Flugge & Schilizzi, 2005). The combustion of fossil fuels and microbes in

Carbon dioxide emissions, the least potent GHG in agriculture stem from fuel use (diesel and petrol) to operate heavy machinery and equipment (Flugge & Schilizzi, 2005) and soils

tropical forests are also discussed as sources on the EPA's website (EPA, 2011b).

**1. Introduction** 

including landfills (EPA, 2011a).

**Greenhouse Gas Emissions** 


## **The Effect of Organic Farms on Global Greenhouse Gas Emissions**

Risa Kumazawa *Duquesne University USA* 

## **1. Introduction**

126 Greenhouse Gases – Emission, Measurement and Management

Wathes, C.M.; Holden, M.R., Sneath, R.W., White, R.P. & Phillips, V.R. (1997).

Weiss, R.F. (1981). Determinations of Carbon Dioxide and Methane by Dual Catalyst Flame

Zhang, Q.; Zhou, X.J., Cicek, N. & Tenuta, M. (2007). Measurement of Odour and

Chromatography. *Journal of Chromatographic Science*, Vol.19, p. 611-616 Wu-Haan, W.; Powers, W.J., Angel, C.R., Hale III, C.E. & Applegate, T.J. (2007). Effect of an

*Poultry Science*, Vol.38, No.1, pp. 14-28

*Biosystems Engineering*, Vol.49, pp. 6.13-6.20

Concentrations and Emissions Rates of Aerial Ammonia, Nitrous Oxide, Methane, Carbon Dioxide, Dust, and Endotoxin in U.K. Broiler and Layer Houses. *British* 

Ionization Chromatography and Nitrous Oxide by Electron Capture

Acidifying Diet Combined with Zeolite and Slight Protein Reduction on Air Emissions from Laying Hens of Different Ages. *Poultry Science*, Vol.86, pp. 182-190

Greenhouse Gas Emissions in Two Swine Farrowing Operations. *Canadian* 

The emission of greenhouse gases (GHGs) is a great concern for mankind as it believed that when GHGs such as carbon dioxide (CO2), nitrous oxide (N2O), and methane (CH4), trap energy from the sun, they contribute to global warming. Even though greenhouse gases occur naturally in the atmosphere, there are noteworthy man-made contributions through agricultural and industrial production. Agricultural land itself contributes to 12% of global GHG emissions while emissions from all sectors related to agricultural production contribute to an estimated 25-30% of all GHG emissions (International Trade Centre, 2007). The Kyoto Protocol was established in Kyoto, Japan in December, 1997 as an amendment to the United Nations Framework Convention on Climate Change (UNFCCC) to regulate global GHG emissions. It remains the most comprehensive international agreement to date through the setting of target reductions by industrialized nations in North America, Europe, Asia and Australia, commonly known as "Annex I" countries, shown in Table A-1 of the Appendix.

The emissions of methane, projected to have approximately 20 times the global warming potential of carbon dioxide in trapping heat in the atmosphere (Morgenstern, 1991; Shih et al., 2006; International Trade Centre, 2007; EPA, 2011a) are caused by enteric fermentation and manure management, both of which are strongly correlated with livestock numbers (Vermont & De Cara, 2010), in addition to rice cultivation (McCarl & Schneider, 2000; Nalley et al., 2011). Enteric fermentation is a digestive process where microbes in ruminant animals such as cows, goat and sheep break down food, producing methane as a byproduct. Anaerobic decomposition (without oxygen) of organic matter in livestock manure leads to methane emissions. The U.S. Environmental Protection Agency's website also mentions natural sources of methane such as wetlands, oceans, wildfires as well as man-made sources including landfills (EPA, 2011a).

The emissions of nitrous oxide, projected to be far more potent at about 300 times the global warming potential of carbon dioxide (International Trade Centre, 2007; EPA, 2011b), stem from livestock management (nitrogen content from animal feeds), soil disturbance (both soil loss and degradation), fertilizer/ other chemical use and the burning of agricultural residues (Ruttan, 2002; Flugge & Schilizzi, 2005). The combustion of fossil fuels and microbes in tropical forests are also discussed as sources on the EPA's website (EPA, 2011b).

Carbon dioxide emissions, the least potent GHG in agriculture stem from fuel use (diesel and petrol) to operate heavy machinery and equipment (Flugge & Schilizzi, 2005) and soils

The Effect of Organic Farms on Global Greenhouse Gas Emissions 129

are not subject to the country-specific reductions outlined in the Kyoto Protocol. The pattern is not clear for agricultural CO2 emissions which may appear in several of the existing categories shown in Table 1. In addition, the total CO2 emissions exclude the supplementary data on CO2 emissions from land-use change and forestry which is only available for 14% of the countries (4 Annex I countries and 22 non-Annex I countries). As a result, I do not analyze agricultural carbon dioxide emissions in this chapter and focus on just agricultural

Figure 1 and Figure 2 show agricultural emissions of both methane and nitrous oxide (the left panels are for the total levels and the right panels are for the per capita levels) averaged across Annex I and non-Annex I countries. They are measured in thousand metric tons CO2 equivalent (MtCO2e) and MtCO2e respectively. This data is compiled from World Bank's World Development Indicators (WDI) from 1990, available every five years until 2005.

Fig. 1. (a) Agricultural CH4 Emissions and (b) Agricultural CH4 Emissions per Capita

methane and nitrous oxide emissions.

from changes in land use, including deforestation (International Trade Centre, 2007). However, the amount of CO2 emissions from agriculture alone is trivial (McCarl & Schneider, 2000)1 even though the total emission of this GHG is the largest compared to the five others (Grunewald & Martinez-Zarozos, 2009).

Table 1 shows the most recent GHG emissions by Annex I status and sector for 2005. These numbers are compiled using World Resources Institute's Climate Analysis Indicators Tool (CAIT, version 8.0) for 39 Annex I countries (including the European Union) and 146 non-Annex I countries. In furnished tables, Malta is erroneously labeled as an Annex I country. It is not certain whether Malta is also counted as an Annex I country in the tool that was used to generate this table. It became a European Union member (with Cyprus) after the protocol was signed but has not changed its non-Annex I status (UNFCCC, 2008).


Table 1. Sources of 2005 GHG Emissions by Sector

While data is available for methane and nitrous oxide emissions solely in agriculture, there is no comparable carbon dioxide emission data specific to the industry. As expected, agricultural CH4 and N2O emissions contribute more for non-Annex I countries, those that

<sup>1</sup> The EPA's estimate of agricultural CO2 emissions for the United States in 1996 was less than 1% of the U.S. total emissions of this GHG.

from changes in land use, including deforestation (International Trade Centre, 2007). However, the amount of CO2 emissions from agriculture alone is trivial (McCarl & Schneider, 2000)1 even though the total emission of this GHG is the largest compared to the

Table 1 shows the most recent GHG emissions by Annex I status and sector for 2005. These numbers are compiled using World Resources Institute's Climate Analysis Indicators Tool (CAIT, version 8.0) for 39 Annex I countries (including the European Union) and 146 non-Annex I countries. In furnished tables, Malta is erroneously labeled as an Annex I country. It is not certain whether Malta is also counted as an Annex I country in the tool that was used to generate this table. It became a European Union member (with Cyprus) after the protocol

**Countries** 

Electricity & Heat 44.9% 45.1% 44.2% Manufacturing & Construction 18.9% 14.5% 24.0% Transportation 19.5% 24.7% 13.5% Fugitive Emissions 0.7% 0.2% 1.2% Industrial Processes 4.3% 1.9% 7.0%

Other Fuel Combustion 3.9% 2.4% 4.6% Fugitive Emissions 24.2% 38.7% 19.5% Industrial Processes 0.1% 0.3% 0.0% Agriculture 51.4% 36.6% 57.7% Waste 20.4% 22.1% 18.2%

Other Fuel Combustion 7.1% 13.6% 4.3% Industrial Processes 4.8% 9.3% 3.2% Agriculture 84.7% 72.4% 89.5% Waste 3.4% 4.7% 3.0%

While data is available for methane and nitrous oxide emissions solely in agriculture, there is no comparable carbon dioxide emission data specific to the industry. As expected, agricultural CH4 and N2O emissions contribute more for non-Annex I countries, those that

1 The EPA's estimate of agricultural CO2 emissions for the United States in 1996 was less than 1% of the

**Annex I Countries**  **Non-Annex I Countries** 

was signed but has not changed its non-Annex I status (UNFCCC, 2008).

five others (Grunewald & Martinez-Zarozos, 2009).

**Sources of GHG Emissions All** 

**1) Carbon Dioxide (CO2) Emissions:**

**2) Methane (CH4) Emissions:**

**3) Nitrous Oxide (N2O) Emissions:**

Table 1. Sources of 2005 GHG Emissions by Sector

U.S. total emissions of this GHG.

Energy:

are not subject to the country-specific reductions outlined in the Kyoto Protocol. The pattern is not clear for agricultural CO2 emissions which may appear in several of the existing categories shown in Table 1. In addition, the total CO2 emissions exclude the supplementary data on CO2 emissions from land-use change and forestry which is only available for 14% of the countries (4 Annex I countries and 22 non-Annex I countries). As a result, I do not analyze agricultural carbon dioxide emissions in this chapter and focus on just agricultural methane and nitrous oxide emissions.

Figure 1 and Figure 2 show agricultural emissions of both methane and nitrous oxide (the left panels are for the total levels and the right panels are for the per capita levels) averaged across Annex I and non-Annex I countries. They are measured in thousand metric tons CO2 equivalent (MtCO2e) and MtCO2e respectively. This data is compiled from World Bank's World Development Indicators (WDI) from 1990, available every five years until 2005.

Fig. 1. (a) Agricultural CH4 Emissions and (b) Agricultural CH4 Emissions per Capita

The Effect of Organic Farms on Global Greenhouse Gas Emissions 131

without the use of synthetic pesticides, herbicides, chemical fertilizers, growth hormones, antibiotics or gene manipulation" (Chen, 2009). As the Kyoto Protocol is not aimed at reducing the emissions of developing countries, I argue that organic farming practices may have accounted for the reduction of GHGs in only Annex I countries. These are also countries that can afford organic production as such farms tend to operate on a smaller scale and may not employ the least-cost production techniques due to the tradeoffs between quality and quantity. Furthermore, agriculture in developing countries tends to be large-scale and more likely non-organic since it serves as the primary export industry that has the competitive edge in low-cost production in the world. With frequent use of fertilizers, pesticides and herbicides to maximize agricultural output, such practices are predicted to have increased the greenhouse

This research makes a contribution to the literature that explores a direct link between organic farming practices and the emissions of nitrous oxide and methane across countries. The chapter continues as follows. After a cross-disciplinary and detailed literature review in various areas, I provide the economic model to be estimated. The chapter concludes after a

Below, a review of the literature is provided in three key areas: (1) the Kyoto Protocol and its effectiveness; (2) previous models of GHG emissions using the Environmental Kuznets Curve (EKC) model and (3) the mitigating effects of organic farms on GHG emissions.

The effectiveness of the Kyoto Protocol has continued to be contemplated by researchers across disciplines. The international agreement calls for the reduction in combined emissions of six of the main greenhouse gases, carbon dioxide, nitrous oxide, methane, hydrofluorocarbons, perfluorocarbons and sulfur hexafluoride (McCarl & Schneider, 2000; UNFCCC, 2008). There are specific emission targets listed for each of the 38 industrialized nations plus the European Union ("Annex I countries"), outlined in a section called "Annex B" in the protocol. These reduction targets in GHG emissions which are listed in Table A-1 of the Appendix will account for a 5.3 percent collective reduction of the 1990 emission levels by the first commitment period of 2008-2012 (Sathiendrakumar, 2003; Finus, 2008; Grunewald & Martinez-Zarzoso, 2009). Belarus and Turkey were not parties to the UNFCCC and so they are Annex I countries that have no target reductions. The United States remains the sole Annex I country that has not ratified the Kyoto Protocol despite being an integral member (Kumazawa & Callaghan, 2010). Most countries use the 1990 baseline emissions but Japan and the former

While setting rigid country-specific target reductions, the protocol allows for flexibility in meeting these targets through emissions trading, clean development mechanism (CDM) or joint implementation between countries as long as eligibility requirements are met (Rollings-Magnusson & Magnusson, 2000; McKibbon & Wilcoxen, 2002; Finus, 2008). The Kyoto Protocol recognizes the uniqueness of the agricultural industry in the emissions reduction process. In another section of the protocol called "Annex A," the agricultural sources of emissions listed include enteric fermentation, manure management and deforestation while sinks to offset the emissions include afforestation and reforestation

gas emissions in non-Annex I countries relative to Annex I countries.

discussion of the empirical results and conclusions.

**2.1 The Kyoto protocol and Its pros and cons** 

members of the Soviet Union have argued for flexible baselines.

(McCarl & Schneider, 2000; UNFCCC, 2008).

**2. Review of the literature** 

Fig. 2. (a) Agricultural N2O Emissions and (b) Agricultural N2O Emissions per Capita

Without controlling for population, both agricultural CH4 and N2O emissions increased for just non-Annex I countries. This is not surprising given that one of the world's most heavily populated countries with substantial emissions, China, is in this category. Since the signing of the Kyoto Protocol, Annex I countries have been experiencing declines in the total levels. On a per capita basis, both GHG emissions declined at about the same rate after 1995 even though non-Annex I countries are not subject to target reductions. Roca et al. (2001) argue that the environmental pressure to assimilate is an important consideration for countries.

In recent years, there has been a shift toward organic farm production in Western societies, primarily for promoting better health and for implementing sustainable business practices (Ruiz de Maya et al., 2011). It has been argued that organic farms provide an unintended side benefit of reducing the greenhouse gas emissions (International Trade Centre, 2007). Organic products are defined as, "goods that respect the environment and that are manufactured without the use of synthetic pesticides, herbicides, chemical fertilizers, growth hormones, antibiotics or gene manipulation" (Chen, 2009). As the Kyoto Protocol is not aimed at reducing the emissions of developing countries, I argue that organic farming practices may have accounted for the reduction of GHGs in only Annex I countries. These are also countries that can afford organic production as such farms tend to operate on a smaller scale and may not employ the least-cost production techniques due to the tradeoffs between quality and quantity. Furthermore, agriculture in developing countries tends to be large-scale and more likely non-organic since it serves as the primary export industry that has the competitive edge in low-cost production in the world. With frequent use of fertilizers, pesticides and herbicides to maximize agricultural output, such practices are predicted to have increased the greenhouse gas emissions in non-Annex I countries relative to Annex I countries.

This research makes a contribution to the literature that explores a direct link between organic farming practices and the emissions of nitrous oxide and methane across countries. The chapter continues as follows. After a cross-disciplinary and detailed literature review in various areas, I provide the economic model to be estimated. The chapter concludes after a discussion of the empirical results and conclusions.

## **2. Review of the literature**

130 Greenhouse Gases – Emission, Measurement and Management

Fig. 2. (a) Agricultural N2O Emissions and (b) Agricultural N2O Emissions per Capita

Without controlling for population, both agricultural CH4 and N2O emissions increased for just non-Annex I countries. This is not surprising given that one of the world's most heavily populated countries with substantial emissions, China, is in this category. Since the signing of the Kyoto Protocol, Annex I countries have been experiencing declines in the total levels. On a per capita basis, both GHG emissions declined at about the same rate after 1995 even though non-Annex I countries are not subject to target reductions. Roca et al. (2001) argue that the environmental pressure to assimilate is an important consideration for countries. In recent years, there has been a shift toward organic farm production in Western societies, primarily for promoting better health and for implementing sustainable business practices (Ruiz de Maya et al., 2011). It has been argued that organic farms provide an unintended side benefit of reducing the greenhouse gas emissions (International Trade Centre, 2007). Organic products are defined as, "goods that respect the environment and that are manufactured Below, a review of the literature is provided in three key areas: (1) the Kyoto Protocol and its effectiveness; (2) previous models of GHG emissions using the Environmental Kuznets Curve (EKC) model and (3) the mitigating effects of organic farms on GHG emissions.

## **2.1 The Kyoto protocol and Its pros and cons**

The effectiveness of the Kyoto Protocol has continued to be contemplated by researchers across disciplines. The international agreement calls for the reduction in combined emissions of six of the main greenhouse gases, carbon dioxide, nitrous oxide, methane, hydrofluorocarbons, perfluorocarbons and sulfur hexafluoride (McCarl & Schneider, 2000; UNFCCC, 2008). There are specific emission targets listed for each of the 38 industrialized nations plus the European Union ("Annex I countries"), outlined in a section called "Annex B" in the protocol. These reduction targets in GHG emissions which are listed in Table A-1 of the Appendix will account for a 5.3 percent collective reduction of the 1990 emission levels by the first commitment period of 2008-2012 (Sathiendrakumar, 2003; Finus, 2008; Grunewald & Martinez-Zarzoso, 2009). Belarus and Turkey were not parties to the UNFCCC and so they are Annex I countries that have no target reductions. The United States remains the sole Annex I country that has not ratified the Kyoto Protocol despite being an integral member (Kumazawa & Callaghan, 2010). Most countries use the 1990 baseline emissions but Japan and the former members of the Soviet Union have argued for flexible baselines.

While setting rigid country-specific target reductions, the protocol allows for flexibility in meeting these targets through emissions trading, clean development mechanism (CDM) or joint implementation between countries as long as eligibility requirements are met (Rollings-Magnusson & Magnusson, 2000; McKibbon & Wilcoxen, 2002; Finus, 2008). The Kyoto Protocol recognizes the uniqueness of the agricultural industry in the emissions reduction process. In another section of the protocol called "Annex A," the agricultural sources of emissions listed include enteric fermentation, manure management and deforestation while sinks to offset the emissions include afforestation and reforestation (McCarl & Schneider, 2000; UNFCCC, 2008).

The Effect of Organic Farms on Global Greenhouse Gas Emissions 133

the conventional EKC model where the subscript i denotes country and the subscript t denotes year in a cross-sectional time series data set. stands for a matrix of year dummy variables that control for the time trend. The time-invariant heterogeneity (i) and the idiosyncratic time-varying error term (it) make up the composite error term. Past research indicates that political institutions play a role (Congleton, 1992). This may be an example of the unobservable heterogeneity, in addition to country land mass and country temperature/ climate which stay relatively constant over time. The betas are the regression coefficients,

 ln(E)it=0 +1ln(Yit)+2[ln(Yit)2]+t +i+it (1) Grossman & Krueger (1995) and Schmalensee et al. (1998) find empirical evidence for the EKC model that developed countries experience the "inverted-U" EKC in their environmental emissions. Grossman and Krueger (1995) find the turning point to be before a country reaches a per capita income of \$8,000. Notable exceptions are Shafik (1994) and Dijkgraaf & Vollebergh (2005) who do not find the same shape for their EKC. The role of industrial production has been incorporated in recent studies of CO2 emissions. Aldy (2007) finds that emission-intensive industrial production grows faster when there are fewer regulations. Kumazawa & Callaghan (2010) estimate an "augmented EKC model" (the conventional EKC model with additional independent variables) controlling for industrial production and test for structural breaks in the pre- and post- agreement years of the Kyoto Protocol. The authors find that carbon dioxide emissions, especially for industrialized nations (Annex I countries) show a decline since the signing. The model used in this chapter will be a variation of this augmented EKC model, specifically for agricultural production. Roca et al. (2001) find that only sulfur dioxide emissions exhibits the classic inverted Ushape of the EKC for Spain. Both methane and nitrous oxide emissions have increased but

Fig. 3. Environmental Kuznets Curve (EKC)

the intercept and 1 and 2the slopes.

not decreased with economic growth for the country.

**2.3 The mitigating effects of GHG emissions through organic farming** 

Mitigating the effects of non-CO2 greenhouse gases is not an entire new idea in agriculture. Collecting methane from livestock manure through anaerobic digesters and using the captured gas, known as "biogas" to generate electricity reduces overall GHG emissions on farms (Shih et al., 2006; Lazarus et al., 2011). McCarl & Schneider (2000) advocate the

One of the shortcomings of the Kyoto Protocol is that it did not go into force until 2005, eight years after it was adopted in 1997. It was only after Russia's ratification that 55 countries which had emitted at least 55% of the greenhouse gases had ratified the protocol (Finus, 2008; Grunewald & Martinez-Zarzoso, 2009). The UNFCCC's principle of "common but differentiated responsibilities" means that the responsibilities for reduction efforts fall disproportionately on the industrialized nations because they had contributed more toward GHG emissions than others. Another criticism of the protocol is that it does not address each GHG and sector individually as separate but linked agreements (Barrett, 2008).

The most problematic areas in the international agreement have been compliance and participation (Bohringer, 2003). The United States is still the only Annex I country which has yet to ratify the protocol and its target reduction of 7 percent is not binding even though a substantial 25% of the world's emissions originate in the US (Sathiendrakumar, 2003). The cost of compliance is not cheap and is estimated to be as much as 2.6 percent of the US GDP (Jaffe et al., 1995), a costly price to pay for a country which is in a chronic trade deficit. Australia and Russia initially followed in the footsteps of the United States but they ratified the protocol in 2007 and 2004, respectively, with provisions.

## **2.2 Previous research using the EKC Model**

Agricultural production has long been considered to be an essential condition for a country's overall economic growth (Ruttan, 2002) in the development phase. However, few paid attention to how this growth was achieved until recent decades when it became evident from scientific data that global warming should be a cause for major concern. As countries experience economic development, they typically transition away from traditional and labor-intensive small-scale farming toward capital-intensive and even chemicallyintensive large-scale farming (Ruttan, 2002; Goodstein, 2008). The economies of scale from the latter farms contribute to lower prices, creating comparative advantages in the global market. It is only when countries become richer that people can afford to demand more control over pollution and environmental degradation (Goodstein, 2008; Dasgupta et al., 2002), a reason why more regulations for pollutants exist in the environmentally-conscious and affluent Western societies.

The Environmental Kuznets Curve (EKC) hypothesis states that there is an inverted Ushaped relationship between economic development and environmental damages. Figure 3 shows that until a country reaches a turning point, there is continuous environmental degradation, at which point the country reverses the path and starts to experience an environmental improvement (Grossman and Krueger, 1995).

Past empirical research that uses the EKC model analyzes the impact of carbon dioxide (Shafik, 1994; Schmalensee et al., 1998; Dijkgraaf & Vollebergh, 2005; Aldy, 2007; Grunewald & Martinez-Zarzoso, 2009; Kumazawa & Callaghan, 2010) and common air pollutants including sulfur dioxide (Grossman & Krueger, 1995; Roca et al., 2001; Harbaugh et al., 2002). One of the few studies which investigated the EKC model for nitrous oxide and methane is the research by Roca et al. (2001).

Empirically, the Environmental Kuznets Curve is shown by the effect of the GDP per capita (Y) and GDP per capita squared on the emissions per capita (E), all in logged form for a panel data regression analysis. The square of the logged income per capita is included to allow for nonlinearity in the parameter. The log transformation of the variables Y and E

One of the shortcomings of the Kyoto Protocol is that it did not go into force until 2005, eight years after it was adopted in 1997. It was only after Russia's ratification that 55 countries which had emitted at least 55% of the greenhouse gases had ratified the protocol (Finus, 2008; Grunewald & Martinez-Zarzoso, 2009). The UNFCCC's principle of "common but differentiated responsibilities" means that the responsibilities for reduction efforts fall disproportionately on the industrialized nations because they had contributed more toward GHG emissions than others. Another criticism of the protocol is that it does not address each

The most problematic areas in the international agreement have been compliance and participation (Bohringer, 2003). The United States is still the only Annex I country which has yet to ratify the protocol and its target reduction of 7 percent is not binding even though a substantial 25% of the world's emissions originate in the US (Sathiendrakumar, 2003). The cost of compliance is not cheap and is estimated to be as much as 2.6 percent of the US GDP (Jaffe et al., 1995), a costly price to pay for a country which is in a chronic trade deficit. Australia and Russia initially followed in the footsteps of the United States but they ratified

Agricultural production has long been considered to be an essential condition for a country's overall economic growth (Ruttan, 2002) in the development phase. However, few paid attention to how this growth was achieved until recent decades when it became evident from scientific data that global warming should be a cause for major concern. As countries experience economic development, they typically transition away from traditional and labor-intensive small-scale farming toward capital-intensive and even chemicallyintensive large-scale farming (Ruttan, 2002; Goodstein, 2008). The economies of scale from the latter farms contribute to lower prices, creating comparative advantages in the global market. It is only when countries become richer that people can afford to demand more control over pollution and environmental degradation (Goodstein, 2008; Dasgupta et al., 2002), a reason why more regulations for pollutants exist in the environmentally-conscious

The Environmental Kuznets Curve (EKC) hypothesis states that there is an inverted Ushaped relationship between economic development and environmental damages. Figure 3 shows that until a country reaches a turning point, there is continuous environmental degradation, at which point the country reverses the path and starts to experience an

Past empirical research that uses the EKC model analyzes the impact of carbon dioxide (Shafik, 1994; Schmalensee et al., 1998; Dijkgraaf & Vollebergh, 2005; Aldy, 2007; Grunewald & Martinez-Zarzoso, 2009; Kumazawa & Callaghan, 2010) and common air pollutants including sulfur dioxide (Grossman & Krueger, 1995; Roca et al., 2001; Harbaugh et al., 2002). One of the few studies which investigated the EKC model for nitrous oxide and

Empirically, the Environmental Kuznets Curve is shown by the effect of the GDP per capita (Y) and GDP per capita squared on the emissions per capita (E), all in logged form for a panel data regression analysis. The square of the logged income per capita is included to allow for nonlinearity in the parameter. The log transformation of the variables Y and E

GHG and sector individually as separate but linked agreements (Barrett, 2008).

the protocol in 2007 and 2004, respectively, with provisions.

environmental improvement (Grossman and Krueger, 1995).

methane is the research by Roca et al. (2001).

**2.2 Previous research using the EKC Model** 

and affluent Western societies.

Fig. 3. Environmental Kuznets Curve (EKC)

the conventional EKC model where the subscript i denotes country and the subscript t denotes year in a cross-sectional time series data set. stands for a matrix of year dummy variables that control for the time trend. The time-invariant heterogeneity (i) and the idiosyncratic time-varying error term (it) make up the composite error term. Past research indicates that political institutions play a role (Congleton, 1992). This may be an example of the unobservable heterogeneity, in addition to country land mass and country temperature/ climate which stay relatively constant over time. The betas are the regression coefficients, the intercept and 1 and 2the slopes.

$$\ln(\text{E})\_{\text{it}} = \text{\text{\textquotedblleft}}\_{\text{t}} + \text{\text{\textquotedblright}}\_{\text{l}} \ln(\text{Y}\_{\text{it}}) + \text{\textquotedblleft}\_{\text{Z}} [\ln(\text{Y}\_{\text{it}}) \text{\textquotedblright}] + \tau\_{\text{t}} + \alpha\_{\text{i}} + \varepsilon\_{\text{it}} \tag{1}$$

Grossman & Krueger (1995) and Schmalensee et al. (1998) find empirical evidence for the EKC model that developed countries experience the "inverted-U" EKC in their environmental emissions. Grossman and Krueger (1995) find the turning point to be before a country reaches a per capita income of \$8,000. Notable exceptions are Shafik (1994) and Dijkgraaf & Vollebergh (2005) who do not find the same shape for their EKC. The role of industrial production has been incorporated in recent studies of CO2 emissions. Aldy (2007) finds that emission-intensive industrial production grows faster when there are fewer regulations. Kumazawa & Callaghan (2010) estimate an "augmented EKC model" (the conventional EKC model with additional independent variables) controlling for industrial production and test for structural breaks in the pre- and post- agreement years of the Kyoto Protocol. The authors find that carbon dioxide emissions, especially for industrialized nations (Annex I countries) show a decline since the signing. The model used in this chapter will be a variation of this augmented EKC model, specifically for agricultural production. Roca et al. (2001) find that only sulfur dioxide emissions exhibits the classic inverted Ushape of the EKC for Spain. Both methane and nitrous oxide emissions have increased but not decreased with economic growth for the country.

#### **2.3 The mitigating effects of GHG emissions through organic farming**

Mitigating the effects of non-CO2 greenhouse gases is not an entire new idea in agriculture. Collecting methane from livestock manure through anaerobic digesters and using the captured gas, known as "biogas" to generate electricity reduces overall GHG emissions on farms (Shih et al., 2006; Lazarus et al., 2011). McCarl & Schneider (2000) advocate the

The Effect of Organic Farms on Global Greenhouse Gas Emissions 135

arable land in hectares is obtained for all countries using FAOSTAT, a data extraction tool

There are further problems in obtaining additional control variables for the estimation of the augmented EKC model for CH4 and N2O emissions. The most problematic is the key information on organic farming. Both the WDI and FAOSTAT have variables on organic land relative to agricultural or arable land but there are more missing values than observations for most countries until the late 2000s. Euromonitor International's World Environment Factbook (2008) has comprehensive data on the number of organic farms, average size of organic farms (in hectares), land used in organic farming (in hectares) and the share of organic land relative to total land for 2001 to 2007 for 71 countries of the world. Even though the use of these variables would reduce the sample size to less than 50%, as

I obtained data on consumption of pesticides (thousand tons), herbicides (thousand tons), livestock numbers (in 1000 heads) and production of cereals (all grains including rice in thousand tons) from the World Environment Factbook (2008) as well. Each of these variables can also be obtained using FAOSTAT in finer detail (for example, gross production value of rice and consumption of nitrogen from fertilizers) but again, there were substantial missing values for most countries which made the data unusable. Table A-3 in the Appendix outlines

The final sample of 55 countries for the augmented EKC model consists disproportionately of industrialized nations due to the availability of data. Annex I countries are overrepresented (there is 87% of them) while non-Annex I countries are underrepresented (there is only 14% of them). These countries make up approximately 60% of the world's agricultural emissions of methane and 50% of the world's agricultural emissions of nitrous oxide. China which emits 15% of the world's emission of agricultural methane and 22% of the world's agricultural nitrous oxide is excluded in the analysis because it does not have the key organic farming information. Japan, Philippines, Singapore and South Korea are

Since the data for emissions of agricultural methane and nitrous oxide are available in 1990, 1995, 2000 and 2005, the country data is stacked for the four years in panel format, allowing for an unbalanced panel. In the conventional EKC model, a panel regression is run for the logged per capita emissions on logged per capita income and the squared logged per capita income, with three dummy variables for the years. This is done for each GHG separately for all 131 countries and then by Annex I status. A positive sign on ln (GDP per capita) and a negative sign on ln (GDP)2 presence of an inverted U-shaped EKC. The results are presented in Table 2 and Table 3. The Hausman Specification Test is used to test between the fixed-

The EKC model, augmented with organic farm practices and other sources of the GHGs, has some data constraints. The additional variables are collected for 2001 to 2007 and 2005 is the only where all variables are available at the same time. For livestock, pesticide, herbicide and cereal data, I approximate the 2000 data with 2001 data, as there is not enough data in the time series to be extrapolating the missing values reliably for each country. The resulting

all variables collected, the data sources and availability of countries and years.

excluded due to the same reason, for the lack of organic farm data.

effect model and the random-effect model, where i is randomly distributed.

panel data is for two years, for a subset of 55 countries.

**4. Empirical estimation** 

for the Food and Agriculture Organization of the United Nations.

there are no alternatives, I do so without much hesitation.

reduction in the use of nitrogen fertilizers and animal feed with nitrogen content to reduce nitrous oxide emissions.

Mitigating the effects of N2O and CH4 through organic farm practices follow along similar lines. The only study to date on this topic is the one conducted by the International Trade Centre (2007). In it, the following arguments are made. Organic farms engage in ley-farming where the field is left fallow after a few years of growing cash crops. The grasses and other plants grown on these fields during the fallow phase increase the nitrogen content of soils naturally without relying on nitrogen fertilizers which is seen as a direct cause of nitrous oxide emissions. In 2005 alone, global consumption of nitrogen fertilizers amounted to a staggering 91 million tons (International Trade Centre, 2007). Manure from livestock can also be recycled as a natural fertilizer, instead of being treated as waste material which, if not treated properly also contributes to emissions.

Organic agriculture also reduces the number of livestock per hectare, which is directly proportional to the emissions of both nitrous oxide and methane. The concentration of nitrous oxide can be more manageable with fewer livestock. In addition, the effects of methane from enteric fermentation and anaerobic decomposition can be reduced by having fewer livestock. Anaerobic digestion of manure and the production of biogas, while not a completely organic production method, have been at the forefront especially on organic farms. These are methods used on organic farms that directly reduce both methane and nitrous oxide emissions in addition to providing chemical and fertilizer-free products for the health-conscious consumers.

## **3. Data**

Even though the idea of investigating the relationship between greenhouse gases and agriculture specifically to do with organic farming practices is very simple, compiling the data set proved to be extremely difficult. While carbon dioxide emission data is readily available for most countries annually over several decades, the same cannot be said of nitrous oxide and methane emission data.

At first glance, the World Resources Institute's Climate Analysis Indicators Tool (CAIT, version 8.0) seems to provide the most comprehensive data for 185 countries in 1990, 1995, 2000 and 2005 for both GHGs. However, upon careful analysis, there seem to be errors in reporting. Sometimes, countries that have small but positive total emissions per capita of either N20 or CH4 (in units of MtCO2e) have zero emissions due to rounding to only one decimal place (for instance, 0.04 rounded to one decimal place becomes 0.0). This makes the distinction between a very small number and an actual zero to be indistinguishable. The agricultural emissions of both of these gases also suffer from the same problem.

The World Bank's World Development Indicators (WDI) also reports agricultural emissions of both N2O and CH4 for 133 countries (excluding Hong Kong, Taiwan and other island nations) in the same four years. These numbers do not always match those from the World Resources Institute but are measured in thousands of MtCO2e, with several decimal places. I obtain these numbers so that the emissions per capita can be calculated with ease by multiplying the emissions by a thousand and dividing by the population. The population, real GDP per capita (in 2000 \$US) and fertilizer consumption information are also extracted from WDI for the same countries that have the two GHG emission data. The size of the

reduction in the use of nitrogen fertilizers and animal feed with nitrogen content to reduce

Mitigating the effects of N2O and CH4 through organic farm practices follow along similar lines. The only study to date on this topic is the one conducted by the International Trade Centre (2007). In it, the following arguments are made. Organic farms engage in ley-farming where the field is left fallow after a few years of growing cash crops. The grasses and other plants grown on these fields during the fallow phase increase the nitrogen content of soils naturally without relying on nitrogen fertilizers which is seen as a direct cause of nitrous oxide emissions. In 2005 alone, global consumption of nitrogen fertilizers amounted to a staggering 91 million tons (International Trade Centre, 2007). Manure from livestock can also be recycled as a natural fertilizer, instead of being treated as waste material which, if

Organic agriculture also reduces the number of livestock per hectare, which is directly proportional to the emissions of both nitrous oxide and methane. The concentration of nitrous oxide can be more manageable with fewer livestock. In addition, the effects of methane from enteric fermentation and anaerobic decomposition can be reduced by having fewer livestock. Anaerobic digestion of manure and the production of biogas, while not a completely organic production method, have been at the forefront especially on organic farms. These are methods used on organic farms that directly reduce both methane and nitrous oxide emissions in addition to providing chemical and fertilizer-free products for the

Even though the idea of investigating the relationship between greenhouse gases and agriculture specifically to do with organic farming practices is very simple, compiling the data set proved to be extremely difficult. While carbon dioxide emission data is readily available for most countries annually over several decades, the same cannot be said of

At first glance, the World Resources Institute's Climate Analysis Indicators Tool (CAIT, version 8.0) seems to provide the most comprehensive data for 185 countries in 1990, 1995, 2000 and 2005 for both GHGs. However, upon careful analysis, there seem to be errors in reporting. Sometimes, countries that have small but positive total emissions per capita of either N20 or CH4 (in units of MtCO2e) have zero emissions due to rounding to only one decimal place (for instance, 0.04 rounded to one decimal place becomes 0.0). This makes the distinction between a very small number and an actual zero to be indistinguishable. The

The World Bank's World Development Indicators (WDI) also reports agricultural emissions of both N2O and CH4 for 133 countries (excluding Hong Kong, Taiwan and other island nations) in the same four years. These numbers do not always match those from the World Resources Institute but are measured in thousands of MtCO2e, with several decimal places. I obtain these numbers so that the emissions per capita can be calculated with ease by multiplying the emissions by a thousand and dividing by the population. The population, real GDP per capita (in 2000 \$US) and fertilizer consumption information are also extracted from WDI for the same countries that have the two GHG emission data. The size of the

agricultural emissions of both of these gases also suffer from the same problem.

nitrous oxide emissions.

health-conscious consumers.

nitrous oxide and methane emission data.

**3. Data** 

not treated properly also contributes to emissions.

arable land in hectares is obtained for all countries using FAOSTAT, a data extraction tool for the Food and Agriculture Organization of the United Nations.

There are further problems in obtaining additional control variables for the estimation of the augmented EKC model for CH4 and N2O emissions. The most problematic is the key information on organic farming. Both the WDI and FAOSTAT have variables on organic land relative to agricultural or arable land but there are more missing values than observations for most countries until the late 2000s. Euromonitor International's World Environment Factbook (2008) has comprehensive data on the number of organic farms, average size of organic farms (in hectares), land used in organic farming (in hectares) and the share of organic land relative to total land for 2001 to 2007 for 71 countries of the world. Even though the use of these variables would reduce the sample size to less than 50%, as there are no alternatives, I do so without much hesitation.

I obtained data on consumption of pesticides (thousand tons), herbicides (thousand tons), livestock numbers (in 1000 heads) and production of cereals (all grains including rice in thousand tons) from the World Environment Factbook (2008) as well. Each of these variables can also be obtained using FAOSTAT in finer detail (for example, gross production value of rice and consumption of nitrogen from fertilizers) but again, there were substantial missing values for most countries which made the data unusable. Table A-3 in the Appendix outlines all variables collected, the data sources and availability of countries and years.

The final sample of 55 countries for the augmented EKC model consists disproportionately of industrialized nations due to the availability of data. Annex I countries are overrepresented (there is 87% of them) while non-Annex I countries are underrepresented (there is only 14% of them). These countries make up approximately 60% of the world's agricultural emissions of methane and 50% of the world's agricultural emissions of nitrous oxide. China which emits 15% of the world's emission of agricultural methane and 22% of the world's agricultural nitrous oxide is excluded in the analysis because it does not have the key organic farming information. Japan, Philippines, Singapore and South Korea are excluded due to the same reason, for the lack of organic farm data.

## **4. Empirical estimation**

Since the data for emissions of agricultural methane and nitrous oxide are available in 1990, 1995, 2000 and 2005, the country data is stacked for the four years in panel format, allowing for an unbalanced panel. In the conventional EKC model, a panel regression is run for the logged per capita emissions on logged per capita income and the squared logged per capita income, with three dummy variables for the years. This is done for each GHG separately for all 131 countries and then by Annex I status. A positive sign on ln (GDP per capita) and a negative sign on ln (GDP)2 presence of an inverted U-shaped EKC. The results are presented in Table 2 and Table 3. The Hausman Specification Test is used to test between the fixedeffect model and the random-effect model, where i is randomly distributed.

The EKC model, augmented with organic farm practices and other sources of the GHGs, has some data constraints. The additional variables are collected for 2001 to 2007 and 2005 is the only where all variables are available at the same time. For livestock, pesticide, herbicide and cereal data, I approximate the 2000 data with 2001 data, as there is not enough data in the time series to be extrapolating the missing values reliably for each country. The resulting panel data is for two years, for a subset of 55 countries.

The Effect of Organic Farms on Global Greenhouse Gas Emissions 137

In Table 2, the coefficients on ln (GDP per capita) and ln (GDP per Capita)2 indicate that Annex I countries face an unexpected U-shaped EKC while non-Annex I countries face the expected inverted U-shaped EKC. However, the result for Annex I countries may be due to the relative short span of emission data which is only available every five years. Most developed countries would have already achieved economic growth by 1990. For Annex I countries, on average, a one percent increase in income per capita decreases per capita emissions of methane by 2 percent. For non-Annex I countries, on average, a one percent increase in income per capita raises emission per capita by 1 percent. All year dummy variables have positive and statistically significant coefficients. The Hausman Specification Tests across all three columns indicate that the fixed-effect model is the appropriate model which takes into consideration the time-invariant characteristics of

In Table 3, the EKC has the expected inverted-U shape for non-Annex I countries and the unexpected U-shape for Annex I countries again. This time, the coefficient on ln (GDP per capita) is not statistically significant, meaning that income per capita has no effect on emissions per capita of nitrous oxide. The coefficient for non-Annex I countries is slightly higher compared to the previous table. All year dummy variables, while unreported in the table, have positive and statistically significant coefficients. The Hausman Specification Tests across all three columns indicate that the fixed-effect model is the appropriate model

**Countries** 

(0.223)

(0.016)

(0.025)

(0.025)

(0.021)

(0.801)

R2 0.375 0.681 0.313 Number of Observations 513 152 357 Number of Countries 131 38 92 Average Number of Years 3.9 4.0 3.9 Hausman Specification Test (2) 36.50\*\*\* 53.31\*\*\* 108.19\*\*\*

Table 3. Conventional EKC Model for Agricultural Nitrous Oxide Emissions

**Annex I Countries** 

> -0.347 (0.516)

0.068\*\* (0.031)

0.564\*\*\* (0.043)

0.355\*\*\* (0.044)

0.166\*\*\* (0.032)


> 1.214\*\*\* (0.304)


0.244\*\*\* (0.030)

0.145\*\*\* (0.029)

0.066\*\* (0.026)


**Variable All** 

ln (GDP per capita) 0.714\*\*\*

ln (GDP per capita)2 -0.024

Dummy Variable for 1990 0.307\*\*\*

Dummy Variable for 1995 0.168\*\*\*

Dummy Variable for 2000 0.080\*\*\*

Constant -5.381

Notes: (1) Standard errors in parentheses below coefficients. (2) Statistical significance at 1% (\*\*\*) , 5% (\*\*) and 10% (\*).

each country.

again.

For agricultural emissions of CH4, the conventional EKC model is augmented with arable land size, the number of organic farms and livestock, consumption of pesticides and production of cereals. These independent variables are chosen based on the literature. They are converted to logs so that elasticities are estimated. Methane stems from both enteric fermentation and anaerobic decomposition, manure management of livestock and rice cultivation. All variable, except the number of organic farms, are expected to increase methane emissions. The number of organic farms is expected to reduce methane emissions through organic practices used in agricultural production. The results are presented in Table 4.

For agricultural emissions of N2O, the conventional EKC model is augmented with arable land size, the number of organic farms and livestock and consumption of fertilizers, pesticides and herbicides, all in logged form. These variables are also chosen based on the literature. Nitrous oxide stems from chemical usage on farms and livestock management. All variables, except the number of organic farms, are expected to increase nitrous oxide emissions. The results are presented in Table 5.

## **5. Results**

The conventional EKC models for agricultural methane and nitrous oxide emissions are presented in Tables 2 and 3. A full list of countries used is provided in Table A-2 of the Appendix.


Notes: (1) Standard errors in parentheses below coefficients.

(2) Statistical significance at 1% (\*\*\*) , 5% (\*\*) and 10% (\*).

Table 2. Conventional EKC Model for Agricultural Methane Emissions

For agricultural emissions of CH4, the conventional EKC model is augmented with arable land size, the number of organic farms and livestock, consumption of pesticides and production of cereals. These independent variables are chosen based on the literature. They are converted to logs so that elasticities are estimated. Methane stems from both enteric fermentation and anaerobic decomposition, manure management of livestock and rice cultivation. All variable, except the number of organic farms, are expected to increase methane emissions. The number of organic farms is expected to reduce methane emissions through organic practices used in agricultural production. The results are

For agricultural emissions of N2O, the conventional EKC model is augmented with arable land size, the number of organic farms and livestock and consumption of fertilizers, pesticides and herbicides, all in logged form. These variables are also chosen based on the literature. Nitrous oxide stems from chemical usage on farms and livestock management. All variables, except the number of organic farms, are expected to increase nitrous oxide

The conventional EKC models for agricultural methane and nitrous oxide emissions are presented in Tables 2 and 3. A full list of countries used is provided in Table A-2 of the

**Countries** 

(0.219)

(0.015)

(0.025)

(0.024)

(0.021)

(0.787)

R2 0.410 0.682 0.360 Number of Observations 513 152 357 Number of Countries 131 38 92 Average Number of Years 3.9 4.0 3.9 Hausman Specification Test (2) 16.52\*\* 33.07\*\*\* 30.25\*\*\*

**Annex I Countries** 


0.169\*\*\* (0.035)

0.670\*\*\* (0.049)

0.454\*\*\* (0.049)

0.185\*\*\* (0.036)

4.110 (2.443) **Non-Annex I Countries** 

> 1.057\*\*\* (0.277)


0.249\*\*\* (0.027)

0.138\*\*\* (0.026)

0.049\*\* (0.024)


presented in Table 4.

**5. Results** 

Appendix.

emissions. The results are presented in Table 5.

**Variable All** 

ln (GDP per capita) 0.411\*

ln (GDP per capita)2 -0.013

Dummy Variable for 1990 0.330\*\*\*

Dummy Variable for 1995 0.184\*\*\*

Dummy Variable for 2000 0.072\*\*\*

Constant -3.360

Notes: (1) Standard errors in parentheses below coefficients. (2) Statistical significance at 1% (\*\*\*) , 5% (\*\*) and 10% (\*).

Table 2. Conventional EKC Model for Agricultural Methane Emissions

In Table 2, the coefficients on ln (GDP per capita) and ln (GDP per Capita)2 indicate that Annex I countries face an unexpected U-shaped EKC while non-Annex I countries face the expected inverted U-shaped EKC. However, the result for Annex I countries may be due to the relative short span of emission data which is only available every five years. Most developed countries would have already achieved economic growth by 1990. For Annex I countries, on average, a one percent increase in income per capita decreases per capita emissions of methane by 2 percent. For non-Annex I countries, on average, a one percent increase in income per capita raises emission per capita by 1 percent. All year dummy variables have positive and statistically significant coefficients. The Hausman Specification Tests across all three columns indicate that the fixed-effect model is the appropriate model which takes into consideration the time-invariant characteristics of each country.

In Table 3, the EKC has the expected inverted-U shape for non-Annex I countries and the unexpected U-shape for Annex I countries again. This time, the coefficient on ln (GDP per capita) is not statistically significant, meaning that income per capita has no effect on emissions per capita of nitrous oxide. The coefficient for non-Annex I countries is slightly higher compared to the previous table. All year dummy variables, while unreported in the table, have positive and statistically significant coefficients. The Hausman Specification Tests across all three columns indicate that the fixed-effect model is the appropriate model again.


Notes: (1) Standard errors in parentheses below coefficients.

Table 3. Conventional EKC Model for Agricultural Nitrous Oxide Emissions

 <sup>(2)</sup> Statistical significance at 1% (\*\*\*) , 5% (\*\*) and 10% (\*).

The Effect of Organic Farms on Global Greenhouse Gas Emissions 139

effects on methane emissions. Perhaps, it signifies the shortcoming of what these variables actually measure. There is no distinction on whether the organic farms are dairy farms or not. If there are disproportionate shares of dairy farms among organic farms, it is entirely possible to expect the sign on the variable to be positive through enteric fermentation and

It is noteworthy that for all columns except Annex I countries, the Hausman Specification Test could not reject the random-effect model. A shortcoming of this model may be that the natural sources of methane such as the number of landfills and wetlands were not

The augmented EKC model for nitrous oxide is presented in Table 5. Again, the results differ from those of the conventional EKC model. The income variables are not statistically significant except for the case where all countries are considered together and show the

**Countries**

(0.551)

(0.038)

(0.131)

(0.029)

(0.123)

(0.155)

(0.153)

(0.061)

(0.026)

(3.384)

R2 0.506 0.725 0.936 Number of Observations 94 60 34 Number of Countries 52 32 20 Average Number of Years 1.8 1.9 1.7 Hausman Specification Test (2) 40.6\*\*\* 53.18\*\*\* 63.48\*\*\*

**Annex I Countries**

> 0.444 (0.674)


0.100 (0.126)

0.081\*\* (0.035)

0.038 (0.160)


0.122 (0.180)

0.099 (0.062)

0.090\*\*\* (0.026)


> -0.623 (0.822)

> 0.072 (0.059)

0.586\*\*\* (0.145)



1.045\*\*\* (0.201)


0.460\*\*\* (0.108)

0.004 (0.058)


anaerobic decomposition.

inverted-U shape.

controlled for due to the lack of available data.

**Variable All** 

ln (GDP per capita) 1.536\*\*\*

ln (GDP per capita)2 -0.086\*\*

ln (Arable Land in Hectares) 0.233\*

ln (Number of Organic Farms) 0.022

ln (Number of Livestock) 0.213\*

ln (Consumption of Pesticides) 0.367\*\*

ln (Consumption of Herbicides) -0.250

ln (Fertilizer Consumption) 0.155\*\*

Dummy Variable for 2000 0.031

Constant -16.996

Notes: (1) Standard errors in parentheses below coefficients. (2) Statistical significance at 1% (\*\*\*) , 5% (\*\*) and 10% (\*).

Table 5. Augmented EKC Model for Agricultural Nitrous Oxide Emissions

The augmented EKC model for methane is presented in Table 4. Compared to the conventional EKC model, the results are dissimilar for the different groups of countries due to the smaller cross-sections of countries and fewer years used. First, the inclusion of additional independent variables makes the EKC seemingly less robust, an observation made by Harbaugh et al. (2002) as both income variables are no longer statistically significant. For Annex I countries, livestock plays a significant role in increasing methane emissions. On average, a one percent increase in livestock numbers increases the agricultural methane emissions per capita by 0.8 percent, ceteris paribus. However, for non-Annex I countries, it is the consumption of fertilizers that matters. On average, a one percent increase in fertilizer consumption increases the agricultural methane emissions per capita by 0.3 percent, ceteris paribus. The production of cereal shows no significant impact on methane. This may be due to the fact that this variable measures the production of other grains, too.


Notes: (1) Standard errors in parentheses below coefficients.

(2) Statistical significance at 1% (\*\*\*) , 5% (\*\*) and 10% (\*).

Table 4. Augmented EKC Model for Agricultural Methane Emissions

The result that stands out is the effect of the number of organic farms for Annex I countries. It has an unexpected positive sign and is statistically significant. While not reported, other variables for organic farm production (average size of organic farms, land used in organic farming and the share of organic land relative to total land) also exhibit positive but smaller

The augmented EKC model for methane is presented in Table 4. Compared to the conventional EKC model, the results are dissimilar for the different groups of countries due to the smaller cross-sections of countries and fewer years used. First, the inclusion of additional independent variables makes the EKC seemingly less robust, an observation made by Harbaugh et al. (2002) as both income variables are no longer statistically significant. For Annex I countries, livestock plays a significant role in increasing methane emissions. On average, a one percent increase in livestock numbers increases the agricultural methane emissions per capita by 0.8 percent, ceteris paribus. However, for non-Annex I countries, it is the consumption of fertilizers that matters. On average, a one percent increase in fertilizer consumption increases the agricultural methane emissions per capita by 0.3 percent, ceteris paribus. The production of cereal shows no significant impact on methane. This may be due to

**Countries**

(0.382)

(0.024)

(0.072)

(0.021)

(0.072)

(0.050)

(0.058)

(0.015)

(1.799)

R2 0.545 0.901 0.505 Number of Observations 97 62 35 Number of Countries 54 33 21 Average Number of Years 1.8 1.9 1.7 Hausman Specification Test (2) 6.27 40.41\*\*\* 6.50

The result that stands out is the effect of the number of organic farms for Annex I countries. It has an unexpected positive sign and is statistically significant. While not reported, other variables for organic farm production (average size of organic farms, land used in organic farming and the share of organic land relative to total land) also exhibit positive but smaller

**Annex I Countries**

> -0.083 (0.414)

> 0.012 (0.027)

> -0.027 (0.083)

0.055\*\*\* (0.019)

0.783\*\*\* (0.088)

0.036 (0.044)

0.021 (0.054)

0.062\*\*\* (0.017)


**Non-Annex I Countries** 

> 0.122 (0.970)

0.0002 (0.064)

0.017 (0.138)


0.171 (0.144)

0.306\*\*\* (0.097)

0.037 (0.107)

0.036 (0.042)


the fact that this variable measures the production of other grains, too.

**Variable All** 

ln (GDP per capita) 0.053

ln (GDP per capita)2 0.008

ln (Arable Land in Hectares) -0.155\*\*

ln (Number of Organic Farms) -0.004

ln (Number of Livestock) 0.405\*\*\*

ln (Consumption of Pesticides) 0.059

ln (Production of Cereals) -0.012

Dummy Variable for 2000 0.065\*\*\*

Constant -6.404

Notes: (1) Standard errors in parentheses below coefficients. (2) Statistical significance at 1% (\*\*\*) , 5% (\*\*) and 10% (\*).

Table 4. Augmented EKC Model for Agricultural Methane Emissions

effects on methane emissions. Perhaps, it signifies the shortcoming of what these variables actually measure. There is no distinction on whether the organic farms are dairy farms or not. If there are disproportionate shares of dairy farms among organic farms, it is entirely possible to expect the sign on the variable to be positive through enteric fermentation and anaerobic decomposition.

It is noteworthy that for all columns except Annex I countries, the Hausman Specification Test could not reject the random-effect model. A shortcoming of this model may be that the natural sources of methane such as the number of landfills and wetlands were not controlled for due to the lack of available data.

The augmented EKC model for nitrous oxide is presented in Table 5. Again, the results differ from those of the conventional EKC model. The income variables are not statistically significant except for the case where all countries are considered together and show the inverted-U shape.


Notes: (1) Standard errors in parentheses below coefficients.

(2) Statistical significance at 1% (\*\*\*) , 5% (\*\*) and 10% (\*).

Table 5. Augmented EKC Model for Agricultural Nitrous Oxide Emissions

The Effect of Organic Farms on Global Greenhouse Gas Emissions 141

that data is not readily available for all countries for various aspects of agricultural

The results of the augmented EKC model are mixed. For example, livestock numbers only contribute to methane emissions in Annex I countries. The number of organic farms has a positive and statistically significant impact on emissions of both CH4 and N2O, only in industrialized nations. This variable may not be measuring the organic practices of such farms and needs further analysis. Nitrous oxide emissions are contributed primarily

The implication of this research is that there is hope for organic agricultural practices which will help to mitigate the emissions of methane and nitrous oxide, which are more harmful and prevalent in agriculture compared to carbon dioxide. Engaging in sustainable agricultural practices will not only be healthy for consumers but will be more productive for farmers if they creatively choose methods that reduce greenhouse emissions and are environmentally friendly. The Kyoto Protocol's effectiveness in the future also lies in incorporating the role of developing nations in the reduction process. Finding the balance between these will help

**baseline emission levels)**

Australia +8% 2007 Austria -8% 2002 Belarus -- 2004 Belgium -8% 2002 Bulgaria -8% 2002 Canada -6% 2002 Croatia -5% 2007 Czech Republic -8% 2001 Denmark -8% 2002 Estonia -8% 2002 European Union -8% 2002 Finland -8% 2002 France -8% 2002 Germany -8% 2002 Greece -8% 2002 Hungary -6% 2002 Iceland +10% 2002 Ireland -8% 2002 Italy -8% 2002 Japan -6% 2002 Latvia -8% 2002 Liechtenstein -8% 2004 Lithuania -8% 2003

**Ratification Date** 

production and for all known sources of methane and nitrous oxide.

through fertilizer and pesticide use of non-Annex I countries.

reduce the emissions of both methane and nitrous oxide in agriculture.

**Country 2008-2012 Target (% of 1990**

**7. Appendix**

Across all columns, the Hausman Specification Tests show the rejection of the random-effect model in favor of the fixed-effect model. Furthermore, the natural sources of N2O are not controlled for in the regressions due to the lack of available data.

For Annex I countries, only the number of organic farms matter and it is again, statistically significant with the wrong sign. For non-Annex I countries, the use of chemicals in agricultural production increase N2O emissions. On average, a one percent increase in pesticide consumption increases agricultural emissions of nitrous oxide per capita by one percent, all else equal. Similarly, a one percent increase in fertilizer consumption increases the same GHG emission per capita by 0.5 percent, all else equal. Curiously, the consumption of herbicides decreases emissions per capita. Perhaps, this is indicative of how the combination of chemicals used in agricultural production affects nitrous oxide emissions.

## **6. Conclusions**

This chapter is one of the first kinds to investigate the Environmental Kuznets Curve hypothesis for nitrous oxide and methane, specifically in agriculture. Using a sample of 131 countries, the conventional EKC model shows the classic inverted-U EKC for non-Annex I countries of the Kyoto Protocol for both GHGs. These are countries which are not subject to the target emission reductions in Annex B of the international agreement. As these countries experienced economic development, per capita agricultural emissions of both methane and income per capita increased until the turning points were reached and started to decline. Controlling for income, the dummy variables for years indicate higher emissions.

On the other hand, the Annex I countries, those that have committed to reductions in emissions of the six greenhouse gases by ratifying the Kyoto Protocol, exhibit a U-shaped EKC for both GHGs. This means that emissions have been decreasing with income but have increased again after a critical point. However, most Annex I countries had already experienced much of their economic development before 1990. Without additional data, it is difficult to ascertain what the shape of this EKC implies for Annex I countries. While the cross-section of countries used was large, the time series only spanned four years as the emission data is only collected every 5 years in 1990, 1995, 2000 and 2005. Even though the protocol went into force, the true test of the effectiveness lies in comparing the numbers in 2010 which reflects the first commitment period of 2008-2012.

The estimation of the augmented EKC model proved to be rather difficult. While the inclusion of additional independent variables makes the EKC seemingly less robust (Harbaugh et al., 2002), it is not reasonable to claim that the income variables pick up all variations in environmental emissions using the conventional EKC model. Omitted variable bias is a serious problem that cannot be ignored. In light of the study by Kumazawa & Callaghan (2010) which focused on augmenting the conventional EKC to show the impact of industrial production on the emissions of carbon dioxide, I estimated the augmented EKC for agricultural production on the emissions of methane and nitrous oxide in the industry.

Additional variables included in the augmented EKC model are the number of livestock, chemicals (fertilizers, pesticides, herbicides) used and the number of organic farms, along with a few others. However, these variables are not available for numerous countries and in the same years as the emission data. The resulting sub-sample used was for 55 countries of the original 131 countries, which are disproportionately Annex I countries and excludes a high emission non-Annex I country, China. This is one of the shortcomings of this research, that data is not readily available for all countries for various aspects of agricultural production and for all known sources of methane and nitrous oxide.

The results of the augmented EKC model are mixed. For example, livestock numbers only contribute to methane emissions in Annex I countries. The number of organic farms has a positive and statistically significant impact on emissions of both CH4 and N2O, only in industrialized nations. This variable may not be measuring the organic practices of such farms and needs further analysis. Nitrous oxide emissions are contributed primarily through fertilizer and pesticide use of non-Annex I countries.

The implication of this research is that there is hope for organic agricultural practices which will help to mitigate the emissions of methane and nitrous oxide, which are more harmful and prevalent in agriculture compared to carbon dioxide. Engaging in sustainable agricultural practices will not only be healthy for consumers but will be more productive for farmers if they creatively choose methods that reduce greenhouse emissions and are environmentally friendly. The Kyoto Protocol's effectiveness in the future also lies in incorporating the role of developing nations in the reduction process. Finding the balance between these will help reduce the emissions of both methane and nitrous oxide in agriculture.


## **7. Appendix**

140 Greenhouse Gases – Emission, Measurement and Management

Across all columns, the Hausman Specification Tests show the rejection of the random-effect model in favor of the fixed-effect model. Furthermore, the natural sources of N2O are not

For Annex I countries, only the number of organic farms matter and it is again, statistically significant with the wrong sign. For non-Annex I countries, the use of chemicals in agricultural production increase N2O emissions. On average, a one percent increase in pesticide consumption increases agricultural emissions of nitrous oxide per capita by one percent, all else equal. Similarly, a one percent increase in fertilizer consumption increases the same GHG emission per capita by 0.5 percent, all else equal. Curiously, the consumption of herbicides decreases emissions per capita. Perhaps, this is indicative of how the combination of chemicals

This chapter is one of the first kinds to investigate the Environmental Kuznets Curve hypothesis for nitrous oxide and methane, specifically in agriculture. Using a sample of 131 countries, the conventional EKC model shows the classic inverted-U EKC for non-Annex I countries of the Kyoto Protocol for both GHGs. These are countries which are not subject to the target emission reductions in Annex B of the international agreement. As these countries experienced economic development, per capita agricultural emissions of both methane and income per capita increased until the turning points were reached and started to decline.

On the other hand, the Annex I countries, those that have committed to reductions in emissions of the six greenhouse gases by ratifying the Kyoto Protocol, exhibit a U-shaped EKC for both GHGs. This means that emissions have been decreasing with income but have increased again after a critical point. However, most Annex I countries had already experienced much of their economic development before 1990. Without additional data, it is difficult to ascertain what the shape of this EKC implies for Annex I countries. While the cross-section of countries used was large, the time series only spanned four years as the emission data is only collected every 5 years in 1990, 1995, 2000 and 2005. Even though the protocol went into force, the true test of the effectiveness lies in comparing the numbers in

The estimation of the augmented EKC model proved to be rather difficult. While the inclusion of additional independent variables makes the EKC seemingly less robust (Harbaugh et al., 2002), it is not reasonable to claim that the income variables pick up all variations in environmental emissions using the conventional EKC model. Omitted variable bias is a serious problem that cannot be ignored. In light of the study by Kumazawa & Callaghan (2010) which focused on augmenting the conventional EKC to show the impact of industrial production on the emissions of carbon dioxide, I estimated the augmented EKC for agricultural production on the emissions of methane and nitrous oxide in the industry. Additional variables included in the augmented EKC model are the number of livestock, chemicals (fertilizers, pesticides, herbicides) used and the number of organic farms, along with a few others. However, these variables are not available for numerous countries and in the same years as the emission data. The resulting sub-sample used was for 55 countries of the original 131 countries, which are disproportionately Annex I countries and excludes a high emission non-Annex I country, China. This is one of the shortcomings of this research,

Controlling for income, the dummy variables for years indicate higher emissions.

controlled for in the regressions due to the lack of available data.

used in agricultural production affects nitrous oxide emissions.

2010 which reflects the first commitment period of 2008-2012.

**6. Conclusions** 

The Effect of Organic Farms on Global Greenhouse Gas Emissions 143

Table A-2. Countries with Emissions Data (Used in EKC Model in Tables 2 and 3) and Subset of Italicized Countries (Used in Augmented EKC Model in Tables 4 and 5)

> World Development Indicator (Word Bank)

> World Development Indicator (Word Bank)

World Development

World Development

FAOSTAT (Food and Agriculture Organization of the United Nations)

World Environmental Factbook (Euromonitor

World Environmental Factbook (Euromonitor

World Environmental Factbook (Euromonitor

World Environmental Factbook (Euromonitor

World Environmental Factbook (Euromonitor

International)

International)

International)

International)

International)

**Variable Data Source Year(s)** 

Population World Development

Dummy Variable for Annex I Status Kumazawa & Callaghan

Mexico Mongolia *Morocco*  Mozambique Myanmar Namibia Nepal *Netherlands New Zealand*  Nicaragua Nigeria *Norway*  Oman

*Turkey*  Turkmenistan *Ukraine* 

Uruguay Uzbekistan *Venezuela Vietnam*  Yemen Zambia Zimbabwe

**Available** 

1990, 1995, 2000, 2005 <sup>133</sup>

1990, 1995, 2000, 2005 <sup>133</sup>

1961-2009 234

2001-2007 71

2001-2007 71

2001-2007 71

2001-2007 71

2001-2001 71

Indicator (Word Bank) 1960-2009 215

Indicator (Word Bank) 1960-2009 215

Indicator (Word Bank) 1990-2009 215

(2010) 1980-2009 210

United Arab Emirates *United Kingdom United States of America* 

> **Countries Available**

*Czech Republic* 

Dominican Republic

*Denmark* 

*Ecuador Egypt*  El Salvatore Eritrea *Estonia*  Ethiopia *Finland France*

Democratic People's Republic of Korea Democratic Republic of Congo

Agricultural Methane Emissions per

Agricultural Nitrous Oxide Emissions Per Capita (thousand MtCO2e)

Gross Domestic Product per Capita

Fertilizer Consumption (kg per hectare of Arable Land)

Capita (thousand MtCO2e)

(constant 2000 \$US)

Arable Land (Hectares)

Livestock (1000 Heads)

Number of Organic Farms

Pesticide Consumption (1000 tons)

Herbicide Consumption (1000 tons)

Production of Cereals (1000 tons)

Table A-3. Variables and their Sources


Table A-1. Annex I Countries with Emission Targets in Annex B of the Kyoto Protocol


Gabon Georgia *Germany*  Ghana *Greece*  Guatemala Haiti Honduras *Hungary*  Iceland *India Indonesia*  Iran Iraq *Ireland Israel Italy*  Jamaica Luxembourg Japan *Jordan Kazakhstan*  Kenya Kuwait Kyrgyzstan *Latvia*  Lebanon Libya *Lithuania Malaysia*  Malta

*Pakistan* Panama Paraguay *Peru*  Philippines *Poland Portugal*  Qatar

*Romania* 

Republic of Korea Republic of Macedonia Republic of Moldova

*Russian Federation Saudi Arabia*  Senegal Singapore *Slovakia Slovenia South Africa Spain*  Sri Lanka Sudan *Sweden Switzerland* 

Syrian Arab Republic

Trinidad & Tobago

Tajikistan Tanzania *Thailand*  Togo

*Tunisia* 

Luxembourg -8% 2002 Monaco -8% 2006 Netherlands -8% 2002 New Zealand 0% 2002 Norway +1% 2002 Portugal -8% 2002 Romania -8% 2001 Russian Federation 0% 2004 Slovakia -8% 2002 Slovenia -8% 2002 Spain -8% 2002 Switzerland -8% 2003 Turkey -- 2009 Ukraine 0% 2004 United Kingdom -8% 2002 United States -7% Not Yet Ratified Table A-1. Annex I Countries with Emission Targets in Annex B of the Kyoto Protocol

Albania Algeria Angola *Argentina*  Armenia *Australia Austria*  Azerbaijan Bahrain Bangladesh Belarus *Belgium*  Benin Bolivia

Bosnia & Herzegovina

Brunei Darussalam

Botswana *Brazil* 

Bulgaria Cambodia Cameroon *Canada Chile*  China *Colombia*  Congo Costa Rica Côte d'Ivoire *Croatia*  Cuba Cyprus


Table A-2. Countries with Emissions Data (Used in EKC Model in Tables 2 and 3) and Subset of Italicized Countries (Used in Augmented EKC Model in Tables 4 and 5)


Table A-3. Variables and their Sources

The Effect of Organic Farms on Global Greenhouse Gas Emissions 145

International Trade Centre [UNCTAD/WTO] (2007). Organic Farming and Climate Change,

from https://www.fibl-shop.org/shop/pdf/mb-1500-climate-change.pdf Kumazawa, R. & Callaghan, M.S. (2010). The Effect of the Kyoto Protocol on Carbon Dioxide

Lazarus, W.F., Goodkind, A., Gallagher, P. & Conway, R. (2011). Carbon Prices Required to

McKibbin, W.J. & Wilcoxen, P.J. (2002). The Role of Economics in Climate Change Policy.

Morgenstern, R.D. (1991). Comprehensive Approach to Global Climate Change. *Papers and*

Nalley, L., Popp, M. & Fortin, C. (2011). The Impact of Reducing Greenhouse Gas Emissions

Roca, J., Padilla, E., Farre´, M. & Galletto, V. (2001). Economic Growth and Atmospheric

Rollings-Magnusson, S. & Magnusson, R.C. (2000). The Kyoto Protocol: Implications of a

Ruiz de Maya, S., Lopez-Lopez, I. & Munuera, J.L. (2011). Organic Food Consumption in

Ruttan, V.W. (2002). Productivity Growth in Agriculture: Sources and Constraints. *The* 

Sathiendrakumar, R. (2003). Greenhouse Emission Reduction and Sustainable Development.

Schmalensee, R., Stoker, T.M. & Judson, R.A. (1998). World Carbon Dioxide Emissions.

Shafik, N. (1994). Economic Development and Environmental Quality: An Econometric

Shih, J.S., Burtraw, D., Palmer, K. & Siikamaki, J. (2006). Air Emissions of Ammonia and

Retrieved from http://www.rff.org/documents/RFF-DP-06-11.pdf. United Nations Framework Convention on Climate Change (2008). Kyoto Protocol reference

Methane from Livestock Operations, *RFF DP 06-11*, Resources for the Future.

Manual on Accounting of Emissions and Assigned Amounts. Retrieved from http://unfccc.int/resource/docs/publications/08\_unfccc\_kp\_ref\_manual.pdf.

*Proceedings of the American Economic Association*, 81, 2, pp. 140-145.

http://ageconsearch.umn.edu/bitstream/98628/2/p11-01revised.pdf. McCarl, B.A. & Schneider, U.A. (2000). U.S. Agriculture's Role in a Greenhouse Gas

*Environment*, 20, 2, pp. 138-149.

10.1007/s12197-010-9164-5.

*Economics*, 22, 1, pp. 134-159.

*Journal of Economic Perspectives*, 16, 2, pp. 107-129.

*Journal of Economic Perspectives*, 16, 4, pp. 161-184.

*Review of Economics and Statistics*, 80, 1, pp. 15-27.

*International Journal of Social Economics*, 30, 12, pp. 1233-1248.

Analysis. *Oxford Economic Papers* 46, October, pp. 757-773.

*Resource Economic Review*, 40, 1, pp.63-80.

*Ecological Economics*, 39, 3, pp.85-99.

*Economics*, 70, 10, pp.1767-1775.

359.

Water Pollution in Less-Developed Countries, 1975 to 2000. *Organization and* 

*Doc.No. MDS-08-152.E*. Research Institute for Organic Agriculture (FiBL). Retrieved

Emissions. *Journal of Economics and Finance*, online edition from Springer, DOI:

Make Digesters Profitable on U.S. Dairy Farms of Different Sizes*. Department of Applied Economics Staff Paper P11-1*, University of Minnesota. Retrieved from

Emission Mitigation World: An Economic Perspective. *Review of Agricultural* 

in Crop Agriculture: A Spatial- and Production-Level Analysis. *Agricultural and* 

Pollution in Spain: Discussing the Environmental Kuznets Curve Hypothesis.

Flawed but Important Environmental Policy. *Canadian Public Policy*, 26, 3, pp. 347-

Europe: International Segmentation Based on Value System Differences. *Ecological* 

#### **8. References**


Aldy, J.E. (2007). Divergence in State-Level per Capita Carbon Dioxide Emissions. *Land* 

Barrett, S. (2008). Climate Treaties and the Imperative Enforcement. *Oxford Review of* 

Bohringer, C. (2003). The Kyoto Protocol: A Review and Perspectives. *Oxford Review of* 

Chen, M.F. (2009). Attitude toward Organic Foods among Taiwanese as Related to Health

Congleton, R.D. (1992). Political Institutions and Pollution Control. *Review of Economics and* 

Dasgupta, S., Laplanta, B., Wang, H. & Wheeler, D. (2002). Confronting the Environmental

Dijkgraaf, E. & Vollebergh, H.R. (2005). A Test for Parameter Homogeneity in CO2 Panel EKC Estimations. *Environment & Resource Economics* 32, 2, pp. 229-239. Environmental Protection Agency (2011) Sources and Emissions (Methane). Retrieved from

Environmental Protection Agency (2011) Sources and Emissions (Nitrous Oxide). Retrieved

Euromonitor International (2008). *World Environmental Databook 2008/2009* (2e), Euromonitor

Finus, M. (2008). The Enforcement Mechanism of the Kyoto Protocol: Flawed or Promising

Flugge, F. & Schilizzi, S. (2005). Greenhouse Gas Abatement Policies and the Value of

Food and Agriculture Organization of the United Nations (2011). FAOSTAT. Data compiled

Goodstein, E.S. (2008). *Economics and the Environment* (5e), John Wiley and Sons, Inc., ISBN

Grossman, G.M. & Krueger, A.B. (1995). Economic Growth and the Environment. *Quarterly* 

Grunewald, N. & Martinez-Zarzoso, I. (2009). Driving Factors of Carbon Dioxide Emissions

Helm, D. (2003). The Assessment: Climate Change Policy. *Oxford Review of Economic Policy*

Jaffe, A., S.R. Peterson, Portney, P.R. & Stavins, R.N. (1995) Environmental Regulation and

Jorgenson, A.K. (2007). Does Foreign Investment Harm the Air We Breathe and the Water

and the Impact from Kyoto Protocol. CESifo Working Paper No. 2758. Retrieved from http://www.cesifo-group.de/portal/pls/portal/docs/1/1186086.PDF. Harbaugh, W.T., Levinson, A. & Wilson, D.M. (2002). Reexamining the Empirical Evidence

for an Environmental Kuznets Curve. *Review of Economics and Statistics*, 84, 3, pp.

the Competitiveness of U.S. Manufacturing: What Does the Evidence Tell Us?

We Drink? A Cross-National Study of Carbon Dioxide Emissions and Organic

Carbon Sinks: Do Grazing and Cropping Systems Have Different Destinies?

Kuznets Curve. *Journal of Economic Perspectives*, 16, 1, pp. 147-168.

Consciousness, Environmental Attitudes, and the Mediating Effects of a Healthy

**8. References** 

*Economics,* 83, 3, pp. 353-369.

*Statistics*, 74, 3, pp. 412 – 421.

*Economic Policy* , 24, 2, pp. 239-258.

*Economic Policy*, 19, 3, pp. 451-466.

Lifestyle. *British Food Journal*, 111, 2, pp.165-178.

http://www.epa.gov/methane/sources.html.

*Ecological Economics*, 55, 4, pp. 584-598.

978-0-471-76309-3, Hoboken, NJ, USA .

*Journal of Economics*, 110, 2: pp. 353-377.

*Journal of Economic Literature* 33, 1, pp. 132-163.

541-551.

19, 3, pp. 349-361.

from http://www.epa.gov/nitrousoxide/sources.html.

International Plc, ISBN: 978-1-84264-471-3, London, UK.

from http://faostat.fao.org/site/291/default.aspx.

Concepts? *Letters in Spatial and Resource Sciences*, 1, 1, pp. 13-25.

Water Pollution in Less-Developed Countries, 1975 to 2000. *Organization and Environment*, 20, 2, pp. 138-149.


**7**

Judith Patterson *Concordia University* 

*Canada* 

**Exploitation of Unconventional Fossil Fuels:**

Carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), and ozone (O3) are the principal naturally occurring radiatively active gases in the earth's atmosphere (Ruddiman, 2001). These gases are responsible for trapping outgoing infrared radiation in the earth's atmosphere, heating it, and thus keeping the earth's average temperature at approximately 15°C. These gases act as the panes of glass in a greenhouse, trapping in heat from sunlight,

The concentration of greenhouse gases throughout earth's history has fluctuated naturally, with plate tectonics being the primary, first order control. However, since the Industrial Revolution rapid rises in combustion-related GHG's, principally CO2, have resulted from related increases in the extraction and burning of the fossil fuels oil, natural gas, and coal. Concern has existed over rising GHG levels and potential climate change since the early 1970's (reviewed in Patterson and Perl, 2007). It is now the consensus of the majority of the scientific community that increases in atmospheric GHG are responsible for measured

The onset of industrialization in the 18th Century was enabled by the widespread use of coal, followed by oil and natural gas in the 19th and 20th centuries. The first resources of all three fossil fuels used were those easily accessible on land. With depletion of these sources, exploration and development has expanded to offshore fields and also to those fossil carbon

The increase in the annual amounts of fossil fuel combustion in terms of barrels of oil, cubic metres of natural gas, and tonnes of coal is estimated to account for an increasingly larger component of the annual flux of CO2 to the atmosphere. While there are natural sources of CO2 emissions to the atmosphere (e.g. volcanic eruptions, decay of vegetation), the global increase in CO2 concentration is attributable primarily to fossil fuel burning (Intergovernmental Panel on Climate Change [IPCC], 2007). Similarly, extraction and distribution related CH4 emissions and combustion related N2O are increasing as well. Following the 2008-09 recession, world primary energy consumption increased 5.6% (British Petroleum [BP], 2011). In the 21st century, fossil fuels now account for 81% of the global

primary energy mix (Table 1; International Energy Agency [IEA], 2011a).

and consequently are also referred to as greenhouse gases (GHG).

increases in global average air and sea surface temperature.

accumulations inaccessible by conventional means.

**1. Introduction** 

**Enhanced Greenhouse Gas Emissions** 


## **Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions**

Judith Patterson *Concordia University Canada* 

## **1. Introduction**

146 Greenhouse Gases – Emission, Measurement and Management

Vermont, B. & De Cara, S. (2010). How Costly is Mitigation of Non-CO2 Greenhouse Gases from Agriculture? A Meta-Analysis. *Ecological Economics*, 69, 7, pp.1373-1386. Winkler, H. (2008). Measurable, Reportable and Verifiable: the Keys to Mitigation in the

World Resources Institute (2011). Climate Analysis Indicators Tool (CAIT) Version 8.0., Washington, DC. Data compiled from http://cait.wri.org/cait.php?page=yearly.

Copenhagen Deal. *Climate Policy*, 8, pp. 534-547.

Carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), and ozone (O3) are the principal naturally occurring radiatively active gases in the earth's atmosphere (Ruddiman, 2001). These gases are responsible for trapping outgoing infrared radiation in the earth's atmosphere, heating it, and thus keeping the earth's average temperature at approximately 15°C. These gases act as the panes of glass in a greenhouse, trapping in heat from sunlight, and consequently are also referred to as greenhouse gases (GHG).

The concentration of greenhouse gases throughout earth's history has fluctuated naturally, with plate tectonics being the primary, first order control. However, since the Industrial Revolution rapid rises in combustion-related GHG's, principally CO2, have resulted from related increases in the extraction and burning of the fossil fuels oil, natural gas, and coal. Concern has existed over rising GHG levels and potential climate change since the early 1970's (reviewed in Patterson and Perl, 2007). It is now the consensus of the majority of the scientific community that increases in atmospheric GHG are responsible for measured increases in global average air and sea surface temperature.

The onset of industrialization in the 18th Century was enabled by the widespread use of coal, followed by oil and natural gas in the 19th and 20th centuries. The first resources of all three fossil fuels used were those easily accessible on land. With depletion of these sources, exploration and development has expanded to offshore fields and also to those fossil carbon accumulations inaccessible by conventional means.

The increase in the annual amounts of fossil fuel combustion in terms of barrels of oil, cubic metres of natural gas, and tonnes of coal is estimated to account for an increasingly larger component of the annual flux of CO2 to the atmosphere. While there are natural sources of CO2 emissions to the atmosphere (e.g. volcanic eruptions, decay of vegetation), the global increase in CO2 concentration is attributable primarily to fossil fuel burning (Intergovernmental Panel on Climate Change [IPCC], 2007). Similarly, extraction and distribution related CH4 emissions and combustion related N2O are increasing as well. Following the 2008-09 recession, world primary energy consumption increased 5.6% (British Petroleum [BP], 2011). In the 21st century, fossil fuels now account for 81% of the global primary energy mix (Table 1; International Energy Agency [IEA], 2011a).

Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions 149

The concentration of CO2 in the atmosphere has varied throughout geologic history. For ancient sequences, millions of years old, atmospheric concentrations are inferred by the use of proxies. However, using ice cores drilled into continental glaciers, measured gas values from air trapped in the ice have yielded the values of atmospheric CO2 concentrations for the past 800,000 years of the Pleistocene and Holocene periods. This research on ice cores has been one of the most significant and valuable contributions made to the understanding of greenhouse gases and climate in the recent geological record. In 1958, a continuous monitoring programme of atmospheric CO2 values was undertaken on the island of Mauna

Variations in the concentration of atmospheric CO2 through the past 800,000 years are illustrated in Figure 1. CO2 concentrations changed, as did temperature throughout the past eight glacial cycles, with low levels of CO2 during globally cold intervals with ice sheet expansion and high amounts of CO2 in warm, interglacial periods. As seen in Figure 1, CO2 values fluctuated between fairly fixed boundaries, ranging from 170 to 290 ppmv, until the beginning of the Industrial Revolution. Minimum CO2 values were recorded during glacial maximums, and maximum CO2 values in interglacial periods. The last 10,000 years, known as the Holocene epoch, have been remarkably stable in temperature and CO2 values. However, beginning with the Industrial Revolution, CO2 levels began to increase and at the present value of 390 ppmv are now higher than at anytime in the last 800,000 years. This increase is

Fig. 1. Record of atmospheric CO2 in ppmv 800,000 years ago to the present. Ice core data from Supplementary materials in Luthi et al, 2008, including Monnin et al, 2001; Pepin et al, 2001; Petit et al, 1999; Raynaud et al, 2005; Siegenthaler et al, 2005. Data from 1958 CE

onwards, Keeling et al, 2001, and Keeling et al, 2011

Loa in Hawaii (Keeling et al, 2001; Keeling et al, 2011).

Added to this now is a new component. The initially accessed oil and gas reservoirs were shallow, and of high quality (National Energy Technology Laboratory, [NETL], 2011). With the depletion of easily accessed conventional deposits of oil and natural gas, the unconventional deposits of these fuels are increasingly being exploited. However with these resources a large energy commitment, which often involves carbon combustion, is required to extract the fuel and render it viable for refining and ultimately consumable in existing equipment. Therefore, there is not just the conventional input of energy and materials to produce fuel, but there are also additional combustion and extra resources involved in the extraction and/or upgrading of the unconventional fuel before it can be treated as a conventional product. This gives the final refined product from an unconventional source a higher carbon intensity (amount of carbon per unit of fuel burned). This creates an additive greenhouse effect to every unit of unconventional oil or gas consumed.


Table 1. Global Primary Energy Mix, 2008. Oil and gas values include fuel from both conventional and unconventional sources. Data from IEA (2011a).

The purpose of this paper is to review the role of extraction and processing of unconventional oil and natural gas and the impact on greenhouse gas production. The peak of production of conventional oil has not resulted in diminished use of oil; rather, it has resulted in increased production of oil from unconventional sources. Similarly, natural gas from unconventional sources is assuming an increasing role in the global gas market (IEA, 2011b). Production of fossil fuels from unconventional sources creates more greenhouse gases than from conventional sources. Previous work by Brandt and Ferrell (2007), using IPCC scenarios, found significant potential impacts on GHG's from the substitution of unconventional oil products for conventional oil. This work examines the increased reliance on unconventional oil and natural gas to meet growing energy demands, and the impact on GHG emissions if increasing demand for fuel is met by unconventional sources with higher carbon intensity.

## **2. Greenhouse gases in the atmosphere**

Radiatively active gases are produced by many different means in the earth system. The carbon cycle is governed by complex mechanisms involving plate tectonics, the biosphere, and solid earth-atmosphere interactions (Press and Siever, 1986). Carbon dioxide (CO2) is introduced to the atmosphere by several processes, including volcanism, combustion, decay of organic material, and respiration. CO2 is also removed from the atmosphere by several processes, including photosynthesis, dissolution in the oceans, geological burial of organic material, and ultimately plate subduction at tectonic boundaries (Press and Siever, 1986).

Added to this now is a new component. The initially accessed oil and gas reservoirs were shallow, and of high quality (National Energy Technology Laboratory, [NETL], 2011). With the depletion of easily accessed conventional deposits of oil and natural gas, the unconventional deposits of these fuels are increasingly being exploited. However with these resources a large energy commitment, which often involves carbon combustion, is required to extract the fuel and render it viable for refining and ultimately consumable in existing equipment. Therefore, there is not just the conventional input of energy and materials to produce fuel, but there are also additional combustion and extra resources involved in the extraction and/or upgrading of the unconventional fuel before it can be treated as a conventional product. This gives the final refined product from an unconventional source a higher carbon intensity (amount of carbon per unit of fuel burned). This creates an additive greenhouse effect to every unit of

> **Fuel Global Use %**  Oil 33 Coal 27 Natural gas 21 Nuclear 6

Biomass burning 10 Hydro electricity 2 Others 1

The purpose of this paper is to review the role of extraction and processing of unconventional oil and natural gas and the impact on greenhouse gas production. The peak of production of conventional oil has not resulted in diminished use of oil; rather, it has resulted in increased production of oil from unconventional sources. Similarly, natural gas from unconventional sources is assuming an increasing role in the global gas market (IEA, 2011b). Production of fossil fuels from unconventional sources creates more greenhouse gases than from conventional sources. Previous work by Brandt and Ferrell (2007), using IPCC scenarios, found significant potential impacts on GHG's from the substitution of unconventional oil products for conventional oil. This work examines the increased reliance on unconventional oil and natural gas to meet growing energy demands, and the impact on GHG emissions if increasing demand for fuel is met by unconventional sources with higher

Radiatively active gases are produced by many different means in the earth system. The carbon cycle is governed by complex mechanisms involving plate tectonics, the biosphere, and solid earth-atmosphere interactions (Press and Siever, 1986). Carbon dioxide (CO2) is introduced to the atmosphere by several processes, including volcanism, combustion, decay of organic material, and respiration. CO2 is also removed from the atmosphere by several processes, including photosynthesis, dissolution in the oceans, geological burial of organic material, and ultimately plate subduction at tectonic boundaries (Press and Siever, 1986).

Table 1. Global Primary Energy Mix, 2008. Oil and gas values include fuel from both

Renewables

conventional and unconventional sources. Data from IEA (2011a).

unconventional oil or gas consumed.

carbon intensity.

**2. Greenhouse gases in the atmosphere** 

The concentration of CO2 in the atmosphere has varied throughout geologic history. For ancient sequences, millions of years old, atmospheric concentrations are inferred by the use of proxies. However, using ice cores drilled into continental glaciers, measured gas values from air trapped in the ice have yielded the values of atmospheric CO2 concentrations for the past 800,000 years of the Pleistocene and Holocene periods. This research on ice cores has been one of the most significant and valuable contributions made to the understanding of greenhouse gases and climate in the recent geological record. In 1958, a continuous monitoring programme of atmospheric CO2 values was undertaken on the island of Mauna Loa in Hawaii (Keeling et al, 2001; Keeling et al, 2011).

Variations in the concentration of atmospheric CO2 through the past 800,000 years are illustrated in Figure 1. CO2 concentrations changed, as did temperature throughout the past eight glacial cycles, with low levels of CO2 during globally cold intervals with ice sheet expansion and high amounts of CO2 in warm, interglacial periods. As seen in Figure 1, CO2 values fluctuated between fairly fixed boundaries, ranging from 170 to 290 ppmv, until the beginning of the Industrial Revolution. Minimum CO2 values were recorded during glacial maximums, and maximum CO2 values in interglacial periods. The last 10,000 years, known as the Holocene epoch, have been remarkably stable in temperature and CO2 values. However, beginning with the Industrial Revolution, CO2 levels began to increase and at the present value of 390 ppmv are now higher than at anytime in the last 800,000 years. This increase is

Fig. 1. Record of atmospheric CO2 in ppmv 800,000 years ago to the present. Ice core data from Supplementary materials in Luthi et al, 2008, including Monnin et al, 2001; Pepin et al, 2001; Petit et al, 1999; Raynaud et al, 2005; Siegenthaler et al, 2005. Data from 1958 CE onwards, Keeling et al, 2001, and Keeling et al, 2011

Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions 151

oil, and natural gas extraction and transportation are also a factor in the increase in atmospheric methane. Presently, agriculture and fossil fuel exploitation account for twothirds of annual anthropogenically derived CH4 emissions (Montzka et al, 2011). The main contributor is agriculture (dominated by ruminant emissions and rice cultivation). Extraction and transportation of natural gas and oil, followed by coal, are the dominant energy related emission sources of CH4 (Montzka et al, 2011), and contribute 18% of anthropogenic CH4 emissions globally (U.S. Environmental Protection Agency [EPA], 2011). It should be noted however, that the United Nations Framework on Climate Change (UNFCC) data on gas production and emissions are only reported for Annex-I parties (IEA,

The distribution of anthropogenically produced greenhouse gas emissions, expressed in CO2 equivalents, is shown in Table 2 below. This does not include greenhouse gas emissions from natural sources (e.g. volcanoes, wetlands). The unit of measure, CO2 eq, is defined as "the amount of CO2 emission that would cause the same time-integrated radiative forcing, over a given time horizon, as an emitted amount of a long-lived GHG or a mixture of GHGs" (IPCC, 2007, p. 14). This value is obtained by multiplying the amount of a GHG by its Global Warming Potential (GWP) over a given time horizon (IPCC, 2007). The GWP is a means of comparing the relative climatic impacts of the different gases (Montzka

CO2 Fossil fuel use 56.6 CO2 Deforestation, decay of biomass 17.3 CO2 Cement production and natural gas flaring 2.8 CH4 Agriculture and fossil fuel production 14.3 N2O Combustion, agriculture 7.9

perfluorocarbons (PFC's) and sulphur

Table 2. Distribution of anthropogenically generated GHG by sector in CO2eq. From IPCC

Fugitive emissions are all GHG emissions from the oil and gas industry except for emissions from fossil fuel combustion (IPCC, 2006). The sources of fugitive emissions include leaks, evaporation, venting, flaring, pipeline breaks or dig-ins, and well blowouts. Some are well characterized, e.g. tank evaporation, process vents, and flare systems (IPCC, 2006), but

The industrial use of fossil fuels began in the 18th C with coal, and oil followed in the late 19th C. Oil is now the dominant fuel in the global primary energy mix, accounting for 33% of energy used in 2008 (IEA, 2010). In 2010 there was a 3.1% increase in global oil consumption from 2009 (British Petroleum [BP], 2011). This was in part driven by the emerging economies

F Hydrofluorocarbons (HFC's),

hexafluoride (SF6)

others are accidental and unpredictable.

Source Percentage of Annual

1.1

Anthropogenic Global Emissions in CO2 eq

2011a) and thus actual emissions may be higher.

et al, 2011).

(2007)

**3. Oil** 

Greenhouse Gas

attributed to the intensified use of fossil fuels with the onset of industrialization and, to a lesser degree, land use changes (IPCC, 2007). The earth's geological systems are unable to remove CO2 at the same rate as it is input, and thus excess CO2 is accumulating in the atmosphere.

Methane (CH4) is produced due to anaerobic decay of organic material. Methane is also the principal constituent of natural gas. The cycling of methane (CH4) is complexly tied to geological and biological activity. Research has shown that large-scale releases of methane to the atmosphere have occurred due to geological activity, including bursts from methane clathrates (e.g. Nisbet, 2002) and degassing of methane-rich sedimentary rocks following intrusion of magmas (e.g. Storey et al, 2007). Atmospheric concentrations of CH4 in the last 800,000 years are shown in Figure 2. Ice core records show the same pattern of CH4 as that seen with CO2 through the glacial cycles, with low levels of CH4 when it is cold and high levels during warm interglacial intervals. Throughout the past 800,000 years, methane also fluctuated between upper and lower boundary conditions, as measured from ice cores, and since the Industrial Revolution has increased dramatically to the present value of 1800 ppbv.

Fig. 2. Record of atmospheric CH4 in ppbv over the past 800,000 years to the present. Ice Core data from online Supplementary materials in Loulergue et al, 2008, including Delmotte et al, 2004; Spahni et al, 2005; Lisiecki and Raymo, 2005; recent data from Dlugokencky, E., et al, 2009.

The increase of atmospheric methane since the Industrial Revolution is attributed to increased anthropogenic activities including land use changes, increased rice cultivation, and growth in the number of ruminant livestock. Fugitive emissions of methane from coal, oil, and natural gas extraction and transportation are also a factor in the increase in atmospheric methane. Presently, agriculture and fossil fuel exploitation account for twothirds of annual anthropogenically derived CH4 emissions (Montzka et al, 2011). The main contributor is agriculture (dominated by ruminant emissions and rice cultivation). Extraction and transportation of natural gas and oil, followed by coal, are the dominant energy related emission sources of CH4 (Montzka et al, 2011), and contribute 18% of anthropogenic CH4 emissions globally (U.S. Environmental Protection Agency [EPA], 2011). It should be noted however, that the United Nations Framework on Climate Change (UNFCC) data on gas production and emissions are only reported for Annex-I parties (IEA, 2011a) and thus actual emissions may be higher.

The distribution of anthropogenically produced greenhouse gas emissions, expressed in CO2 equivalents, is shown in Table 2 below. This does not include greenhouse gas emissions from natural sources (e.g. volcanoes, wetlands). The unit of measure, CO2 eq, is defined as "the amount of CO2 emission that would cause the same time-integrated radiative forcing, over a given time horizon, as an emitted amount of a long-lived GHG or a mixture of GHGs" (IPCC, 2007, p. 14). This value is obtained by multiplying the amount of a GHG by its Global Warming Potential (GWP) over a given time horizon (IPCC, 2007). The GWP is a means of comparing the relative climatic impacts of the different gases (Montzka et al, 2011).


Table 2. Distribution of anthropogenically generated GHG by sector in CO2eq. From IPCC (2007)

Fugitive emissions are all GHG emissions from the oil and gas industry except for emissions from fossil fuel combustion (IPCC, 2006). The sources of fugitive emissions include leaks, evaporation, venting, flaring, pipeline breaks or dig-ins, and well blowouts. Some are well characterized, e.g. tank evaporation, process vents, and flare systems (IPCC, 2006), but others are accidental and unpredictable.

## **3. Oil**

150 Greenhouse Gases – Emission, Measurement and Management

attributed to the intensified use of fossil fuels with the onset of industrialization and, to a lesser degree, land use changes (IPCC, 2007). The earth's geological systems are unable to remove CO2 at the same rate as it is input, and thus excess CO2 is accumulating in the

Methane (CH4) is produced due to anaerobic decay of organic material. Methane is also the principal constituent of natural gas. The cycling of methane (CH4) is complexly tied to geological and biological activity. Research has shown that large-scale releases of methane to the atmosphere have occurred due to geological activity, including bursts from methane clathrates (e.g. Nisbet, 2002) and degassing of methane-rich sedimentary rocks following intrusion of magmas (e.g. Storey et al, 2007). Atmospheric concentrations of CH4 in the last 800,000 years are shown in Figure 2. Ice core records show the same pattern of CH4 as that seen with CO2 through the glacial cycles, with low levels of CH4 when it is cold and high levels during warm interglacial intervals. Throughout the past 800,000 years, methane also fluctuated between upper and lower boundary conditions, as measured from ice cores, and since the Industrial Revolution has increased dramatically to the present value of 1800 ppbv.

Fig. 2. Record of atmospheric CH4 in ppbv over the past 800,000 years to the present. Ice Core data from online Supplementary materials in Loulergue et al, 2008, including Delmotte et al, 2004; Spahni et al, 2005; Lisiecki and Raymo, 2005; recent data from Dlugokencky, E., et

The increase of atmospheric methane since the Industrial Revolution is attributed to increased anthropogenic activities including land use changes, increased rice cultivation, and growth in the number of ruminant livestock. Fugitive emissions of methane from coal,

atmosphere.

al, 2009.

The industrial use of fossil fuels began in the 18th C with coal, and oil followed in the late 19th C. Oil is now the dominant fuel in the global primary energy mix, accounting for 33% of energy used in 2008 (IEA, 2010). In 2010 there was a 3.1% increase in global oil consumption from 2009 (British Petroleum [BP], 2011). This was in part driven by the emerging economies

Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions 153

embargo in the early 1970's, the Iranian Revolution of 1979 and ensuing global economic downturn (Patterson and Perl, 2007), and most recently the 2008-09 recession, but overall the

Oil is a finite resource. M. King Hubbert forecast the peak in production in U.S. oil, based on his observations in Texas oil fields (Hubbert, 1949). Production from an individual well and aggregates of wells that make up a field, rises steadily at first and then, when approximately half the reserve has been extracted, the yield peaks and declines in the form of a normal distribution. Based on these observations, Hubbert (1971) went on to predict an estimate for the peak in production of global oil. This concept was revisited in the now-classic Campbell

1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

Currently, production of conventional crude is at 68-69 mb/d, with unconventional oil, NGL's, and condensates making up the balance. In 2009, world production of oil was 81.0 mb/d (IEA, 2010). Of this, 67.9 mb/d of crude were produced; natural gas liquids (10.8 mb/d) and unconventional oil (2.3 mb/d) made up the balance (IEA, 2010). It appears that the peak of production of conventional crude oil was at 70 mb/d in 2006 (IEA, 2010). The International Energy Agency believes that if governments put in place the energy and climate policies that they have currently committed to, then the peak of conventional crude

There is tremendous debate about the size of remaining conventional reserves (Sorrel et al, 2010). For example, with the decline of Arctic sea ice in the summer months, new fields are available for exploration. Offshore discoveries, including deepwater, have contributed

Fig. 3. Global oil production and consumption 1965-2010. Data from BP (2011).

Production Consumption

trend has grown inexorably upwards.

0

oil has indeed probably passed (IEA, 2010).

10000

20000

30000

40000

50000

**thousands of barrels a day**

60000

70000

80000

90000

100000

and Laherrère paper "The End of Cheap Oil" (1998).

of China and India, which will increasingly shape global energy demand (IEA, 2010). "China is currently the most important country in shaping future energy markets" (IEA, 2011a, p. 15), and it has surpassed the U.S. as the world's largest energy consumer (BP, 2011). Transportation is expected to continue to be the main driver for the increased demand for oil (IEA, 2010).

Oil resources are classified as conventional and unconventional. Conventional oil consists of crude, natural gas liquids (NGL), and condensates (IEA, 2010; Sorrel et al, 2010). Crude oil is a mixture of hydrocarbons that exists as a liquid under surface conditions (IEA, 2010). It initially flows easily from a well and is one of the least expensive and easiest to process of the petroleum products. The bulk of the oil that has been used to date on the planet was the easily accessible crude oil. NGL's are light hydrocarbons that are produced within a natural gas stream in a hydrocarbon reservoir (IEA, 2010). Condensates are light liquid hydrocarbons recovered from gas reservoirs, and are classed as NGL's (IEA, 2010).

Unconventional oil is derived from sources that include extra-heavy oil, bitumen1 (tar/oil sands), oil shale, coal (coal-to-liquid CTL), and natural gas (gas-to-liquid GTL); (IEA, 2010; Sorrel et al, 2010). Primary fields of unconventional oil are the bitumen deposits in Alberta, Canada and the heavy oil in the Orinoco Belt of Venezuela. Unconventional oil requires the addition of resources – natural gas, energy and water – to extract and transform the material into oil that can then be processed at conventional refineries. Biofuels and coal-to-liquid (CTL) are not included in this paper.

The variation between conventional and unconventional oil is illustrated by API values (Table 3). API gravity is a measure of the density of a petroleum liquid relative to water. An API >10 means lighter than water, <10 is heavier than water (Stratton et al, 2010).


Table 3. API gravity values for the different oils (IEA, 2010; Martinez-Palou et al, 2011).

#### **3.1 The peak of conventional and the rise of unconventional oil**

The annual global production and consumption of oil, from all sources (crude, bitumen, shale oil, and NGL's), is shown in Figure 32. Production has risen from 31.8 mb/d in 1965 to 82.1 mb/d (million barrels a day) in 2010 (BP, 2011). There have been dips, due to the oil

<sup>1</sup> In this paper, the term bitumen is used. It is an objective descriptor of the extra-heavy petroleum

found in Northern Alberta, Canada, and avoids "taking sides" in the oil vs. tar sands debate. 2 "Differences between these world consumption figures and world production statistics are accounted for by stock changes, consumption of non-petroleum additives and substitute fuels, and unavoidable disparities in the definition, measurement or conversion of oil supply and demand data." BP, 2011, p. 9.

of China and India, which will increasingly shape global energy demand (IEA, 2010). "China is currently the most important country in shaping future energy markets" (IEA, 2011a, p. 15), and it has surpassed the U.S. as the world's largest energy consumer (BP, 2011). Transportation is expected to continue to be the main driver for the increased demand

Oil resources are classified as conventional and unconventional. Conventional oil consists of crude, natural gas liquids (NGL), and condensates (IEA, 2010; Sorrel et al, 2010). Crude oil is a mixture of hydrocarbons that exists as a liquid under surface conditions (IEA, 2010). It initially flows easily from a well and is one of the least expensive and easiest to process of the petroleum products. The bulk of the oil that has been used to date on the planet was the easily accessible crude oil. NGL's are light hydrocarbons that are produced within a natural gas stream in a hydrocarbon reservoir (IEA, 2010). Condensates are light liquid

Unconventional oil is derived from sources that include extra-heavy oil, bitumen1 (tar/oil sands), oil shale, coal (coal-to-liquid CTL), and natural gas (gas-to-liquid GTL); (IEA, 2010; Sorrel et al, 2010). Primary fields of unconventional oil are the bitumen deposits in Alberta, Canada and the heavy oil in the Orinoco Belt of Venezuela. Unconventional oil requires the addition of resources – natural gas, energy and water – to extract and transform the material into oil that can then be processed at conventional refineries. Biofuels and coal-to-liquid

The variation between conventional and unconventional oil is illustrated by API values (Table 3). API gravity is a measure of the density of a petroleum liquid relative to water. An

<10

API >10 means lighter than water, <10 is heavier than water (Stratton et al, 2010).

Oil API°

Light crude >35 Medium 26-35 Heavy <20

Table 3. API gravity values for the different oils (IEA, 2010; Martinez-Palou et al, 2011).

1 In this paper, the term bitumen is used. It is an objective descriptor of the extra-heavy petroleum found in Northern Alberta, Canada, and avoids "taking sides" in the oil vs. tar sands debate. 2 "Differences between these world consumption figures and world production statistics are accounted for by stock changes, consumption of non-petroleum additives and substitute fuels, and unavoidable disparities in the definition, measurement or conversion of oil supply and demand data." BP, 2011, p. 9.

The annual global production and consumption of oil, from all sources (crude, bitumen, shale oil, and NGL's), is shown in Figure 32. Production has risen from 31.8 mb/d in 1965 to 82.1 mb/d (million barrels a day) in 2010 (BP, 2011). There have been dips, due to the oil

Extra-heavy and

**3.1 The peak of conventional and the rise of unconventional oil** 

Bitumen

hydrocarbons recovered from gas reservoirs, and are classed as NGL's (IEA, 2010).

for oil (IEA, 2010).

(CTL) are not included in this paper.

embargo in the early 1970's, the Iranian Revolution of 1979 and ensuing global economic downturn (Patterson and Perl, 2007), and most recently the 2008-09 recession, but overall the trend has grown inexorably upwards.

Oil is a finite resource. M. King Hubbert forecast the peak in production in U.S. oil, based on his observations in Texas oil fields (Hubbert, 1949). Production from an individual well and aggregates of wells that make up a field, rises steadily at first and then, when approximately half the reserve has been extracted, the yield peaks and declines in the form of a normal distribution. Based on these observations, Hubbert (1971) went on to predict an estimate for the peak in production of global oil. This concept was revisited in the now-classic Campbell and Laherrère paper "The End of Cheap Oil" (1998).

Fig. 3. Global oil production and consumption 1965-2010. Data from BP (2011).

Currently, production of conventional crude is at 68-69 mb/d, with unconventional oil, NGL's, and condensates making up the balance. In 2009, world production of oil was 81.0 mb/d (IEA, 2010). Of this, 67.9 mb/d of crude were produced; natural gas liquids (10.8 mb/d) and unconventional oil (2.3 mb/d) made up the balance (IEA, 2010). It appears that the peak of production of conventional crude oil was at 70 mb/d in 2006 (IEA, 2010). The International Energy Agency believes that if governments put in place the energy and climate policies that they have currently committed to, then the peak of conventional crude oil has indeed probably passed (IEA, 2010).

There is tremendous debate about the size of remaining conventional reserves (Sorrel et al, 2010). For example, with the decline of Arctic sea ice in the summer months, new fields are available for exploration. Offshore discoveries, including deepwater, have contributed

Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions 155

Oil from unconventional sources is predicted to play an increasing role in the world oil supply (IEA, 2010). Over 50% of the world's petroleum reserves (Head et al, 2003) consist of degraded oil, preserved either as heavy oil or bitumen. Heavy oil and extra heavy oil production is currently rising around the world (Martinez-Palou et al, 2010), and Venezuela is expected to increase output from the Orinoco belt (Watkins, 2010). Some estimates, reported in Baynard (2011), suggest that oil exploration and production activities in the

The Canadian bitumen deposits of Alberta will also play a significant role. Estimates for future production from the Canadian bitumen sand are varied and conflicting. Rates of production of 3.7 mb/d are forecast for 2025, (Canadian Association of Petroleum Producers [CAPP], 2011), while predictions for production in 2030 range from 5 mb/d (Soderburgh et

Evidence for the predicted growth in production from Alberta bitumen deposits can also be seen in plans for pipeline development. As of this writing the contentious Keystone XL pipeline, from northern Alberta to Texas, has passed Canadian and U.S. Environmental Protection Agency (EPA) approval and is contingent upon approval by the White House. In Canada, the Northern Gateway Pipeline has been proposed to run from northern Alberta to the deep-water port of Kitimat B.C. for export overseas, presumably to the Chinese market. These plans are meeting fierce opposition (see http://www.huffingtonpost.ca/

http://www.tankersnothankers.ca). Also, Enbridge Pipelines Inc. has requested permission from the National Energy Board of Canada to proceed with the Trailbreaker project. This would initially change flow in a portion of the pipeline in Ontario. Ultimately, a reversal in pipeline flow between Portland, Maine, (U.S.A.) and Montreal, Canada would allow Alberta bitumen products to travel to the east coast of the U.S., for export overseas and to southern U.S. ports and refineries (Vanderklippe, 2011a). Finally, both Canadian railways, Canadian National, and Canadian Pacific, are exploring increased oil movement by rail as business

The production of conventional oil involves exploration, drilling, establishment of the wellhead, and construction of pipelines to move the oil to refineries. In later stages, as the well becomes depleted, secondary recovery with pumping, or water or CO2 injection, is used to extract the oil. The enhanced recovery techniques add to the energy costs of the production of conventional crude oil. Greenhouse gas emissions associated with the production of conventional crude oil are combustion related CO2 from construction, operation, and transportation. Methane emissions from field operations include venting, oil storage tanks, and well venting and flaring (IPCC, 2006; EPA, 2011). Natural gas in remote

In the case of conventional crude oil, the product that flows out of the well needs little modification before entering the pipeline. However, this is not the case with the unconventional oil. Any existing technology for the conversion of unconventional fossil fuel

Venezuelan heavy oil belt may increase by 600% in the next two decades.

al, 2007) to 6 mb/d (Cambrian Energy Research Associates [CERA], 2009).

**3.2 Production of unconventional oil; energy and resources required** 

locations is often disposed of by flaring (Martinez-Palou et al, 2011).

2011/09/02/keystone-xl-protest-naomi-klein\_n\_947117.html;

options (Vanderklippe, 2011b).

significantly to conventional reserves since the 1990's. Since 2000, more than half the oil discovered is in deep water (IEA, 2010). However, the average size of new fields has continued to fall (IEA, 2010), and it is thought unlikely that new giant fields of conventional oil will be found (Sorrel et al, 2010). Rather than a peak of production of conventional oil, Sorrel et al (2010) suggest that the production values from conventional fields may form a "bumpy plateau".

Patterson and Perl (2007) discussed the possibility of two paths for world energy consumption, with the peak of production of conventional crude oil. Replacements for conventional crude are all more costly, hence it was postulated that the increased price in oil could have the effect of reducing oil consumption that CO2 climate mitigation policies (e.g. Kyoto Protocol) had failed to achieve. The other path was increased utilization of unconventional resources and rising global oil combustion, with concomitant emissions, regardless of price.

The record of global average oil prices is shown in Figure 4. Although there have been significant fluctuations, the average annual price has remained above US\$60 for six years now (BP, 2011). Given the apparent unresponsiveness of consumption to high prices, there seems to be no monetary throttle on the continued rise in oil production and consumption.

Fig. 4. Oil Prices 1965-2010 adjusted to US\$2010. Data from BP (2011).

The decline in production from producing conventional fields means that new production must come from other sources if demand is to be met. In 2010, this means an extra 3 mb/d must be added each year either from enhanced recovery of existing conventional fields, discovery of new conventional fields, or through the exploitation of unconventional resources (Sorrel et al, 2010).

significantly to conventional reserves since the 1990's. Since 2000, more than half the oil discovered is in deep water (IEA, 2010). However, the average size of new fields has continued to fall (IEA, 2010), and it is thought unlikely that new giant fields of conventional oil will be found (Sorrel et al, 2010). Rather than a peak of production of conventional oil, Sorrel et al (2010) suggest that the production values from conventional fields may form a

Patterson and Perl (2007) discussed the possibility of two paths for world energy consumption, with the peak of production of conventional crude oil. Replacements for conventional crude are all more costly, hence it was postulated that the increased price in oil could have the effect of reducing oil consumption that CO2 climate mitigation policies (e.g. Kyoto Protocol) had failed to achieve. The other path was increased utilization of unconventional resources and rising global oil combustion, with concomitant emissions,

The record of global average oil prices is shown in Figure 4. Although there have been significant fluctuations, the average annual price has remained above US\$60 for six years now (BP, 2011). Given the apparent unresponsiveness of consumption to high prices, there seems to be no monetary throttle on the continued rise in oil production and consumption.

1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

The decline in production from producing conventional fields means that new production must come from other sources if demand is to be met. In 2010, this means an extra 3 mb/d must be added each year either from enhanced recovery of existing conventional fields, discovery of new conventional fields, or through the exploitation of unconventional

Fig. 4. Oil Prices 1965-2010 adjusted to US\$2010. Data from BP (2011).

"bumpy plateau".

regardless of price.

0.00

resources (Sorrel et al, 2010).

20.00

40.00

60.00

**US\$2010**

80.00

100.00

120.00

Oil from unconventional sources is predicted to play an increasing role in the world oil supply (IEA, 2010). Over 50% of the world's petroleum reserves (Head et al, 2003) consist of degraded oil, preserved either as heavy oil or bitumen. Heavy oil and extra heavy oil production is currently rising around the world (Martinez-Palou et al, 2010), and Venezuela is expected to increase output from the Orinoco belt (Watkins, 2010). Some estimates, reported in Baynard (2011), suggest that oil exploration and production activities in the Venezuelan heavy oil belt may increase by 600% in the next two decades.

The Canadian bitumen deposits of Alberta will also play a significant role. Estimates for future production from the Canadian bitumen sand are varied and conflicting. Rates of production of 3.7 mb/d are forecast for 2025, (Canadian Association of Petroleum Producers [CAPP], 2011), while predictions for production in 2030 range from 5 mb/d (Soderburgh et al, 2007) to 6 mb/d (Cambrian Energy Research Associates [CERA], 2009).

Evidence for the predicted growth in production from Alberta bitumen deposits can also be seen in plans for pipeline development. As of this writing the contentious Keystone XL pipeline, from northern Alberta to Texas, has passed Canadian and U.S. Environmental Protection Agency (EPA) approval and is contingent upon approval by the White House. In Canada, the Northern Gateway Pipeline has been proposed to run from northern Alberta to the deep-water port of Kitimat B.C. for export overseas, presumably to the Chinese market. These plans are meeting fierce opposition (see http://www.huffingtonpost.ca/ 2011/09/02/keystone-xl-protest-naomi-klein\_n\_947117.html;

http://www.tankersnothankers.ca). Also, Enbridge Pipelines Inc. has requested permission from the National Energy Board of Canada to proceed with the Trailbreaker project. This would initially change flow in a portion of the pipeline in Ontario. Ultimately, a reversal in pipeline flow between Portland, Maine, (U.S.A.) and Montreal, Canada would allow Alberta bitumen products to travel to the east coast of the U.S., for export overseas and to southern U.S. ports and refineries (Vanderklippe, 2011a). Finally, both Canadian railways, Canadian National, and Canadian Pacific, are exploring increased oil movement by rail as business options (Vanderklippe, 2011b).

#### **3.2 Production of unconventional oil; energy and resources required**

The production of conventional oil involves exploration, drilling, establishment of the wellhead, and construction of pipelines to move the oil to refineries. In later stages, as the well becomes depleted, secondary recovery with pumping, or water or CO2 injection, is used to extract the oil. The enhanced recovery techniques add to the energy costs of the production of conventional crude oil. Greenhouse gas emissions associated with the production of conventional crude oil are combustion related CO2 from construction, operation, and transportation. Methane emissions from field operations include venting, oil storage tanks, and well venting and flaring (IPCC, 2006; EPA, 2011). Natural gas in remote locations is often disposed of by flaring (Martinez-Palou et al, 2011).

In the case of conventional crude oil, the product that flows out of the well needs little modification before entering the pipeline. However, this is not the case with the unconventional oil. Any existing technology for the conversion of unconventional fossil fuel

Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions 157

an in situ operation, seismic lines are cleared, drilling sites are constructed, roads are built,

Steam assisted gravity drainage (SAGD) is a forefront technology for in situ extraction, and is the most economically attractive method for the deep deposits of the Athabasca field (Chow et al, 2008). Advances in horizontal drilling have aided the development of this technology. Two horizontal wells are drilled, one above the other; steam at 250°C is injected into the top well, and the loosened bitumen is collected in the bottom production well (Isaacs, 2007). A new technology involves the use of expanding solvents. In this technique, expanding solvent steam assisted gravity drainage (ES-SAGD) combines steam and solvent

The Cyclic Steam Stimulation (CSS) process involves the injection of high temperature steam at high pressure into the bitumen deposit underground (Chow et al, 2008). The highpressure steam fractures the sediment, allowing the steam to spread and heat the bitumen,

Once the bitumen is extracted from underground, the majority of the product is treated with diluent so that it will flow and goes by pipeline to refineries, while the remainder is upgraded in northern Alberta (Alberta Chamber of Commerce, 2004). The main source of the diluent is natural gas. Blending with traditional condensate diluent requires a 70:30 ratio of bitumen to condensate (CAPP, 2009). Upgraded light crude can also be used and is blended in a 50:50 ratio with the bitumen for pipeline transport to refineries (CAPP, 2009).

The transportation of bitumen, heavy or extra-heavy oils that have not yet been upgraded from the fields to the energy markets, pose technological challenges because of high density, viscosity, low API gravity, and salt and heavy metal content (Martinez-Palou et al, 2011). While sending oil by pipeline is the most efficient and convenient method of transportation, the low mobility of these oils, and wax and ashphaltene deposits on pipeline walls, create

To overcome these problems, pipeline transportation of heavy oil can be effected by reducing the viscosity of the oil, minimizing wall drag in the pipe, and in-situ upgrading of the oil prior to transport (Martinez-Palou et al, 2011). Viscosity can be reduced by dilution with natural gas condensates, partial upgrading, formation of oil/water emulsions, reconfiguration of internal shear, and heating the oil and pipelines. Adding heat to the oil and pipeline and re-heating through directed fire heaters at pumping stations is the second most common method for reducing the viscosity of the heavy oil. (Martinez-Palou et al, 2011). In offshore settings, subsea pipelines must be heated when transporting heavy or extra heavy oil. All these techniques have added energy costs, and consequently greenhouse gas implications, particularly the

The production of heavy and extra-heavy oils face similar production challenges.

and pipelines are constructed (Johnson and Miyanishi, 2008; Schneider et al, 2003).

in the injection process (Chow et al, 2008).

technical problems (Martinez-Palou et al, 2011).

**conventional and unconventional petroleum** 

direct heating of the heavy oil along the length of the pipeline.

**3.3 Quantification of Life cycle Greenhouse gas emissions associated with** 

The greenhouse gas emissions associated with both conventional and non-conventional liquid petroleum fuels have been investigated as these industries have grown, and concern

reducing its viscosity and allowing it to flow (Chow et al, 2008).

to liquid hydrocarbon suitable for refining as conventional petroleum carries significant environmental burdens. The unconventional resource must be extracted, separated from the host rock and, depending upon the composition, upgraded before being sent to a refinery. All of this requires energy, water, and feedstock for upgrading. In the section below the processes used to produce oil from bitumen are used as an outline to illustrate the energy and resources required.

Bitumen is a complex hydrocarbon that requires enhanced extraction techniques and upgrading before the liquid product can be sent to a conventional liquid petroleum refinery. Bitumen has a high viscosity and will not flow unaltered. There are two methods by which the bitumen is produced; surface mining followed by extraction of the bitumen from the sand, and in-situ separation of the bitumen and removal from the ground. In both cases, upgrading of the bitumen to a lighter hydrocarbon fluid resembling conventional petroleum is required before refining of the product can take place. Currently most of the bitumen that is removed by surface mining is upgraded as part of the overall on-site process, and these are known as integrated operations (CAPP, 2009). Only some of the in situ operations upgrade the product before transportation to refineries; the bitumen from other in situ plants is blended with a diluent to enable the non-upgraded material to flow in pipelines to southern upgrading and refining facilities (CAPP, 2009).

In the open-pit mining sites, truck and shovel operations are the main method of extracting and transporting the oil sand from the ground (Isaacs, 2007). Ore is transported to crushers by heavy truck (Alberta Chamber of Resources, 2004) with energy supplied by diesel fuel. At the crushers, the ore is broken down (Alberta Chamber of Resources, 2004) and then the crushed ore undergoes slurrying and hydrotransport in pipelines (Isaacs, 2007) to the separation facility (Alberta Chamber of Resources, 2004). In this process, called conditioning, the crushed oil sand is mixed with steam and water. In the dynamic movement, the separation of oil from sand begins (Isaacs, 2007; Chow et al, 2008). Most separation of mined bitumen is done by using the Clark method (Hyndman and Luhning, 1991), involving hot water, NaOH, and steam (Holowenko et al, 2000); first developed by Dr. Karl Clark in 1929 (Chow et al, 2008). The separation of the bitumen from the sand is an iterative process in which as much bitumen as possible is extracted from the sand slurry before being sent to the tailings pond (Alberta Chamber of Resources, 2004; Chow et al, 2008).

Following extraction of the bitumen it is sent for upgrading, the process whereby the extracted bitumen is transformed into a synthetic crude oil that can be sent by conventional pipeline to refineries where it can be used as a feedstock, similar to conventional petroleum (Hyndman and Luhning, 1991). The upgrading step is a complex petrochemical engineering process that reduces the viscosity of the hydrocarbon product and decreases the sulphur, nitrogen, and metals content (Hyndman and Luhning, 1991; Isaacs, 2007; Humphries, 2008). Natural gas is the principle feedstock for hydrogen in the upgrading process (Soderburgh et al, 2007; Humphries, 2008).

Extraction of bitumen from the host sandstone at depths too great for economically viable surface mining is accomplished by the use of techniques that reduce the viscosity of the bitumen and allow it to be pumped to the surface (Chow et al, 2008). In the development of

to liquid hydrocarbon suitable for refining as conventional petroleum carries significant environmental burdens. The unconventional resource must be extracted, separated from the host rock and, depending upon the composition, upgraded before being sent to a refinery. All of this requires energy, water, and feedstock for upgrading. In the section below the processes used to produce oil from bitumen are used as an outline to illustrate the energy

Bitumen is a complex hydrocarbon that requires enhanced extraction techniques and upgrading before the liquid product can be sent to a conventional liquid petroleum refinery. Bitumen has a high viscosity and will not flow unaltered. There are two methods by which the bitumen is produced; surface mining followed by extraction of the bitumen from the sand, and in-situ separation of the bitumen and removal from the ground. In both cases, upgrading of the bitumen to a lighter hydrocarbon fluid resembling conventional petroleum is required before refining of the product can take place. Currently most of the bitumen that is removed by surface mining is upgraded as part of the overall on-site process, and these are known as integrated operations (CAPP, 2009). Only some of the in situ operations upgrade the product before transportation to refineries; the bitumen from other in situ plants is blended with a diluent to enable the non-upgraded material to flow in pipelines to

In the open-pit mining sites, truck and shovel operations are the main method of extracting and transporting the oil sand from the ground (Isaacs, 2007). Ore is transported to crushers by heavy truck (Alberta Chamber of Resources, 2004) with energy supplied by diesel fuel. At the crushers, the ore is broken down (Alberta Chamber of Resources, 2004) and then the crushed ore undergoes slurrying and hydrotransport in pipelines (Isaacs, 2007) to the separation facility (Alberta Chamber of Resources, 2004). In this process, called conditioning, the crushed oil sand is mixed with steam and water. In the dynamic movement, the separation of oil from sand begins (Isaacs, 2007; Chow et al, 2008). Most separation of mined bitumen is done by using the Clark method (Hyndman and Luhning, 1991), involving hot water, NaOH, and steam (Holowenko et al, 2000); first developed by Dr. Karl Clark in 1929 (Chow et al, 2008). The separation of the bitumen from the sand is an iterative process in which as much bitumen as possible is extracted from the sand slurry before being sent to the tailings pond (Alberta Chamber of Resources, 2004; Chow et al,

Following extraction of the bitumen it is sent for upgrading, the process whereby the extracted bitumen is transformed into a synthetic crude oil that can be sent by conventional pipeline to refineries where it can be used as a feedstock, similar to conventional petroleum (Hyndman and Luhning, 1991). The upgrading step is a complex petrochemical engineering process that reduces the viscosity of the hydrocarbon product and decreases the sulphur, nitrogen, and metals content (Hyndman and Luhning, 1991; Isaacs, 2007; Humphries, 2008). Natural gas is the principle feedstock for hydrogen in the upgrading process (Soderburgh et

Extraction of bitumen from the host sandstone at depths too great for economically viable surface mining is accomplished by the use of techniques that reduce the viscosity of the bitumen and allow it to be pumped to the surface (Chow et al, 2008). In the development of

and resources required.

2008).

al, 2007; Humphries, 2008).

southern upgrading and refining facilities (CAPP, 2009).

an in situ operation, seismic lines are cleared, drilling sites are constructed, roads are built, and pipelines are constructed (Johnson and Miyanishi, 2008; Schneider et al, 2003).

Steam assisted gravity drainage (SAGD) is a forefront technology for in situ extraction, and is the most economically attractive method for the deep deposits of the Athabasca field (Chow et al, 2008). Advances in horizontal drilling have aided the development of this technology. Two horizontal wells are drilled, one above the other; steam at 250°C is injected into the top well, and the loosened bitumen is collected in the bottom production well (Isaacs, 2007). A new technology involves the use of expanding solvents. In this technique, expanding solvent steam assisted gravity drainage (ES-SAGD) combines steam and solvent in the injection process (Chow et al, 2008).

The Cyclic Steam Stimulation (CSS) process involves the injection of high temperature steam at high pressure into the bitumen deposit underground (Chow et al, 2008). The highpressure steam fractures the sediment, allowing the steam to spread and heat the bitumen, reducing its viscosity and allowing it to flow (Chow et al, 2008).

Once the bitumen is extracted from underground, the majority of the product is treated with diluent so that it will flow and goes by pipeline to refineries, while the remainder is upgraded in northern Alberta (Alberta Chamber of Commerce, 2004). The main source of the diluent is natural gas. Blending with traditional condensate diluent requires a 70:30 ratio of bitumen to condensate (CAPP, 2009). Upgraded light crude can also be used and is blended in a 50:50 ratio with the bitumen for pipeline transport to refineries (CAPP, 2009). The production of heavy and extra-heavy oils face similar production challenges.

The transportation of bitumen, heavy or extra-heavy oils that have not yet been upgraded from the fields to the energy markets, pose technological challenges because of high density, viscosity, low API gravity, and salt and heavy metal content (Martinez-Palou et al, 2011). While sending oil by pipeline is the most efficient and convenient method of transportation, the low mobility of these oils, and wax and ashphaltene deposits on pipeline walls, create technical problems (Martinez-Palou et al, 2011).

To overcome these problems, pipeline transportation of heavy oil can be effected by reducing the viscosity of the oil, minimizing wall drag in the pipe, and in-situ upgrading of the oil prior to transport (Martinez-Palou et al, 2011). Viscosity can be reduced by dilution with natural gas condensates, partial upgrading, formation of oil/water emulsions, reconfiguration of internal shear, and heating the oil and pipelines. Adding heat to the oil and pipeline and re-heating through directed fire heaters at pumping stations is the second most common method for reducing the viscosity of the heavy oil. (Martinez-Palou et al, 2011). In offshore settings, subsea pipelines must be heated when transporting heavy or extra heavy oil. All these techniques have added energy costs, and consequently greenhouse gas implications, particularly the direct heating of the heavy oil along the length of the pipeline.

#### **3.3 Quantification of Life cycle Greenhouse gas emissions associated with conventional and unconventional petroleum**

The greenhouse gas emissions associated with both conventional and non-conventional liquid petroleum fuels have been investigated as these industries have grown, and concern

Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions 159

87.5 99.8

Natural gas is the third most abundant fuel in the global energy mix (Table 1), and is projected to overtake coal in projections of energy use to 2035 (IEA, 2011). Power generation is the main sector for gas demand, where gas has been replacing coal (IEA, 2011). In 2007, 33% of the global natural gas production was consumed for electrical generation (EIA, 2010). Figure 5 shows production and consumption of natural gas since 19705. Natural gas production has tripled in the past 35 years and is tightly followed by consumption. The influence of China is seen also in the development of the natural gas market. For example, China is installing liquefied natural gas (LNG) re-gasification terminals and buying shale

Conventional natural gas reservoirs may be found with or without oil. Gas that accumulates with oil is referred to as associated gas, while non-associated gas does not accumulate with oil (Energy Information Agency [EIA], 2010). Conventional gas reservoirs are sandstone or carbonate formations that have both porosity and permeability (interconnected pores so that the gas can flow) (Holditch, 2006). With conventional gas reservoirs, once the well is drilled the gas is easily extracted. Gas production involves drilling to the reservoir and setting up the wellhead and pipelines for gas to flow from the reservoir to market. Processing involves stripping out impurities (e.g. H2S). Pipeline quality natural gas is 95-98% methane (EPA, 2011). The gas is transmitted along high-pressure pipelines with pumping stations

distributed along the length of the pipeline, with final distribution to end-users.

the definition, measurement or conversion of gas supply and demand data." BP, 2011, p. 23.

4 Data from Stratton et al, 2010. The values presented are the baseline scenario; the value for conventional crude is the weighted average of all crude oil fed into U.S. refineries except for Canadian oil sands. 5 "The difference between these world consumption figures and the world production statistics is due to variations in stocks at storage facilities and liquefaction plants, together with unavoidable disparities in

Table 5. Comparative life cycle GHG emissions for gasoline for light duty vehicles and

Unconventional (bitumen) gCO2e/km

Unconventional (bitumen) gCO2e/MJ

(surface mining)

390 (SAGD) 1.23 384 (CSS) 1.21

108.2 (SAGD) 1.24

316.3 354 (surface mining) 1.12

Ratio

Ratio

1.14

Unconventional/ conventional

Unconventional/ conventional

crude gCO2e/km

aviation kerosene for conventional and unconventional oil feedstocks.

crude gCO2e/MJ

Conventional

Conventional

gas resources in North America (IEA, 2011a).

3 Data from (S&T)2 Consultants Inc., 2008.

Light Duty vehicle "well-to-

Aviation Kerosene4 Jet A "Well-to-Wake"

**4. Natural gas** 

wheels"3

over greenhouse gas emissions from these sectors has increased (e.g., (S&T)2 Consultants Inc., 2008; Skone and Gerdes, 2008; Charpentier et al, 2009; Stratton et al, 2010). These are the emissions associated with fuel production, from extraction out of the ground to transportation to the refinery, including flaring of associated natural gas. These emissions include fugitive greenhouse gases from the wells and venting. In the case of the Canadian bitumen this includes all emissions associated with extraction and upgrading, including fugitive methane emissions from open-pit mining operations.

Generally speaking, the ratio of energy used to the energy in the final product (energy produced) is about 6% for conventional crude oil, approximately 20-25% for extra-heavy oil and nearly 30% for bitumen in sandstone (IEA, Energy Technology Systems Analysis Programme [ETSAP], 2010). Comparison of the emissions produced through the production of conventional and unconventional oil, produced from bitumen, are shown in Table 4, below. The values for conventional crude include recovery and transportation emissions, and exclude oil from Angola and Nigeria. The values for the unconventional oil are for Alberta bitumen, extracted either by open pit mining or by in situ processes (SAGD), and then upgraded to synthetic crude oil. The values are taken from the review by Charpentier et al (2009) and Stratton et al (2010).


Table 4. Upstream CO2eq energy by feedstock.

Life cycle analysis can be undertaken for bitumen that ultimately is used in transport vehicles. These GHG life cycle emissions are divided based upon the extraction method; open-pit mining, SAGD, or CCS. For example, the emissions associated with the life cycle of fuel that is ultimately burned in a light duty vehicle (conventional automobile) are referred to as "well-to-wheels"; that is, from the drilled well or mining pit to the fuel tank of the vehicle and subsequent combustion. Similarly, aviation kerosene emissions are referred to as "well-to-wake" (Stratton et al, 2010).

Examples of conventional versus unconventional liquid petroleum life cycle greenhouse gas emissions are presented in Table 5, below. The greenhouse gases considered are combustion and non-combustion related products, CO2, CH4, and N2O, and are reported in total as CO2 equivalents (CO2eq), based on the GWP of each gas (IPCC, 2007). The "well-to-wheels" values are presented as gCO2eq/km for the production and combustion of the final refined product (gasoline) in vehicles ((S&T)2 Consultants Inc., 2008). Values are presented gCO2eq /MJ of energy in the "well-to-wake" aviation fuel (Stratton et al, 2010).


Table 5. Comparative life cycle GHG emissions for gasoline for light duty vehicles and aviation kerosene for conventional and unconventional oil feedstocks.

## **4. Natural gas**

158 Greenhouse Gases – Emission, Measurement and Management

over greenhouse gas emissions from these sectors has increased (e.g., (S&T)2 Consultants Inc., 2008; Skone and Gerdes, 2008; Charpentier et al, 2009; Stratton et al, 2010). These are the emissions associated with fuel production, from extraction out of the ground to transportation to the refinery, including flaring of associated natural gas. These emissions include fugitive greenhouse gases from the wells and venting. In the case of the Canadian bitumen this includes all emissions associated with extraction and upgrading, including

Generally speaking, the ratio of energy used to the energy in the final product (energy produced) is about 6% for conventional crude oil, approximately 20-25% for extra-heavy oil and nearly 30% for bitumen in sandstone (IEA, Energy Technology Systems Analysis Programme [ETSAP], 2010). Comparison of the emissions produced through the production of conventional and unconventional oil, produced from bitumen, are shown in Table 4, below. The values for conventional crude include recovery and transportation emissions, and exclude oil from Angola and Nigeria. The values for the unconventional oil are for Alberta bitumen, extracted either by open pit mining or by in situ processes (SAGD), and then upgraded to synthetic crude oil. The values are taken from the review by Charpentier

fugitive methane emissions from open-pit mining operations.

Fuel Product gCO2eq/MJ

Table 4. Upstream CO2eq energy by feedstock.

"well-to-wake" (Stratton et al, 2010).

Conventional crude 4.5 – 9.6 (Charpentier et al, 2009)

 Mining 9.2 - 26.5 (Charpentier et al, 2009) SAGD 16.2 - 28.7 (Charpentier et al, 2009)

Life cycle analysis can be undertaken for bitumen that ultimately is used in transport vehicles. These GHG life cycle emissions are divided based upon the extraction method; open-pit mining, SAGD, or CCS. For example, the emissions associated with the life cycle of fuel that is ultimately burned in a light duty vehicle (conventional automobile) are referred to as "well-to-wheels"; that is, from the drilled well or mining pit to the fuel tank of the vehicle and subsequent combustion. Similarly, aviation kerosene emissions are referred to as

Examples of conventional versus unconventional liquid petroleum life cycle greenhouse gas emissions are presented in Table 5, below. The greenhouse gases considered are combustion and non-combustion related products, CO2, CH4, and N2O, and are reported in total as CO2 equivalents (CO2eq), based on the GWP of each gas (IPCC, 2007). The "well-to-wheels" values are presented as gCO2eq/km for the production and combustion of the final refined product (gasoline) in vehicles ((S&T)2 Consultants Inc., 2008). Values are presented gCO2eq

/MJ of energy in the "well-to-wake" aviation fuel (Stratton et al, 2010).

5 – 10 (Stratton et al, 2010)

et al (2009) and Stratton et al (2010).

Synthetic crude oil

Natural gas is the third most abundant fuel in the global energy mix (Table 1), and is projected to overtake coal in projections of energy use to 2035 (IEA, 2011). Power generation is the main sector for gas demand, where gas has been replacing coal (IEA, 2011). In 2007, 33% of the global natural gas production was consumed for electrical generation (EIA, 2010). Figure 5 shows production and consumption of natural gas since 19705. Natural gas production has tripled in the past 35 years and is tightly followed by consumption. The influence of China is seen also in the development of the natural gas market. For example, China is installing liquefied natural gas (LNG) re-gasification terminals and buying shale gas resources in North America (IEA, 2011a).

Conventional natural gas reservoirs may be found with or without oil. Gas that accumulates with oil is referred to as associated gas, while non-associated gas does not accumulate with oil (Energy Information Agency [EIA], 2010). Conventional gas reservoirs are sandstone or carbonate formations that have both porosity and permeability (interconnected pores so that the gas can flow) (Holditch, 2006). With conventional gas reservoirs, once the well is drilled the gas is easily extracted. Gas production involves drilling to the reservoir and setting up the wellhead and pipelines for gas to flow from the reservoir to market. Processing involves stripping out impurities (e.g. H2S). Pipeline quality natural gas is 95-98% methane (EPA, 2011). The gas is transmitted along high-pressure pipelines with pumping stations distributed along the length of the pipeline, with final distribution to end-users.

<sup>3</sup> Data from (S&T)2 Consultants Inc., 2008.

<sup>4</sup> Data from Stratton et al, 2010. The values presented are the baseline scenario; the value for conventional crude is the weighted average of all crude oil fed into U.S. refineries except for Canadian oil sands.

<sup>5 &</sup>quot;The difference between these world consumption figures and the world production statistics is due to variations in stocks at storage facilities and liquefaction plants, together with unavoidable disparities in the definition, measurement or conversion of gas supply and demand data." BP, 2011, p. 23.

Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions 161

Unconventional gas reservoirs have low permeability and cannot produce at economic flow rates or economic volumes unless the reservoir is pumped or stimulated by hydraulic fracture treatment, horizontal wellbores, and/or multiple wellbores (Holditch, 2006). Recovery factors for unconventional gas tend to be much lower than for conventional gas, only about 15-30% of gas initially in place (GIIP) versus 80% recovery of GIIP for

Tight gas development started in the 1970's (American Association of Petroleum Geologists [AAPG], 2011). Global tight gas production has risen from 6 billion cubic metres (bcm) in 1993 to 93 bcm in 2009 (Shell, 2011). It is estimated that at least 6000 trillion cubic feet (tcf) are in place in the US alone (NETL, 2011). Tight gas is trapped in impermeable, lowpermeability, or non-porous sedimentary rock (Holditch, 2006; NETL, 2011). The extraction of tight gas requires fracturing and chemical alteration, which makes it costly. To get economic flow rates and/or to recover economic volumes the formation must be stimulated

Methane has historically always been a problem in coalmines, due to the risk of explosion during mining operations (EPA, 2004). In the formation of coal from organic material, methane gas is produced and is adsorbed onto the sides of small pores of the coal (EPA, 2004). Coal is structurally weak and consequently fractures easily so coal beds are typically characterized by networks of connected fractures (EPA, 2004). Water in coal beds contributes to the hydrostatic pressure that keeps methane gas adsorbed to the surface of the

As a natural gas, coalbed methane is difficult to produce, although it is relatively pure and does not contain H2S (Government of Alberta, 2011). Coalbed methane is produced by reducing pressure in the coalbed (National Energy Board [NEB], 2007). When pressure is reduced the methane desorbs, diffuses through the coal, and then flows through the fractures (NETL, 2011). This reduction in pressure may be achieved by drilling and pumping out the groundwater in the coal bed, which brings water to the surface (Berquist et al, 2007; USGS, 2010; NEB, 2007; NETL, 2011). The initial pumping-out of groundwater may need additional flow enhancement and generally is followed by more drilling and hydraulic fracturing to enlarge pre-existing natural fractures in the coal seams (EPA, 2004; NEB, 2007; NETL, 2011). Fracturing fluids containing proppants (usually fine sand) are injected into the coal bed and then pumped back out, leaving the proppants to keep the fractures open, and

Shales are dense, fine-grained rocks, with very limited porosity (Kargbo et al, 2010). Shales hosting natural gas are rich in organic material (Kargbo et al, 2010; Kerr, 2010; Osborn et al, 2011), with natural gas in the pores of these rocks (Kargbo et al, 2010). Shales are so impermeable that they often act as cap seals on reservoirs of conventional oil and gas.

thus increasing the permeability and allowing the methane to flow (EPA, 2004).

conventional deposits (Massachusetts Institute of Technology [MIT], 2010).

using hydraulic fracturing and/or horizontal well bores (AAPG, 2011).

coal (United States Geological Survey, 2000; NETL, 2011).

**4.1.1 Tight gas** 

**4.1.2 Coalbed methane** 

**4.1.3 Shale gas** 

Fig. 5. Global Natural Gas Production and Consumption 1970-2010. Data from BP (2011).

#### **4.1 Unconventional gas**

The principal sources of unconventional gas are tight gas, coal-bed methane, and shale gas (Bourdaire, 2011; Holditch and AdDeenMadani, 2011; IEA, 2011a). Tight gas and coalbed methane (CBM) have been produced economically for four decades now, with shale gas production being relatively recent (Bourdaire, 2011; IEA, 2011a). In the U.S., the production of unconventional gas up to 2008 was distributed approximately as 70% from tight gas, 20% CBM, and 10% from shale gas (Bourdaire, 2011).

Globally, approximately 13% of marketed gas is from unconventional sources (IEA, 2010). Presently, unconventional gas is principally being produced in Canada and the U.S. In 2009, the production of unconventional gas exceeded conventional in the U.S., and it now makes up ~60% of marketed production in that country (IEA, 2011a). The peak of production of conventional gas was shown to have been reached in 2008 (IEA, 2011a, Figure 1.13, p. 36), and consequently, unconventional gas is increasingly meeting global demand. The production of unconventional gas doubled in the last eight years to 350 bcm in 2010 (IEA, 2010).

Deposits of unconventional gas are predicted to increasingly meet the growing demand for natural gas, and there are large quantities of this resource. In China, India, and Australia, unconventional gas development is focused on coal-bed methane, while tight gas is of interest in North Africa and shale gas in Poland (IEA, 2011b). Globally, the share of unconventional gas is predicted to rise from 13% of total natural gas production to 25% in 2035. Total global annual production is predicted to rise from 3.3 trillion cubic metres (tcm) in 2010, to 5.1 tcm in that time period (IEA, 2011a).

Unconventional gas reservoirs have low permeability and cannot produce at economic flow rates or economic volumes unless the reservoir is pumped or stimulated by hydraulic fracture treatment, horizontal wellbores, and/or multiple wellbores (Holditch, 2006). Recovery factors for unconventional gas tend to be much lower than for conventional gas, only about 15-30% of gas initially in place (GIIP) versus 80% recovery of GIIP for conventional deposits (Massachusetts Institute of Technology [MIT], 2010).

## **4.1.1 Tight gas**

160 Greenhouse Gases – Emission, Measurement and Management

Production Consumption

1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

Fig. 5. Global Natural Gas Production and Consumption 1970-2010. Data from BP (2011).

The principal sources of unconventional gas are tight gas, coal-bed methane, and shale gas (Bourdaire, 2011; Holditch and AdDeenMadani, 2011; IEA, 2011a). Tight gas and coalbed methane (CBM) have been produced economically for four decades now, with shale gas production being relatively recent (Bourdaire, 2011; IEA, 2011a). In the U.S., the production of unconventional gas up to 2008 was distributed approximately as 70% from tight gas, 20%

Globally, approximately 13% of marketed gas is from unconventional sources (IEA, 2010). Presently, unconventional gas is principally being produced in Canada and the U.S. In 2009, the production of unconventional gas exceeded conventional in the U.S., and it now makes up ~60% of marketed production in that country (IEA, 2011a). The peak of production of conventional gas was shown to have been reached in 2008 (IEA, 2011a, Figure 1.13, p. 36), and consequently, unconventional gas is increasingly meeting global demand. The production of

Deposits of unconventional gas are predicted to increasingly meet the growing demand for natural gas, and there are large quantities of this resource. In China, India, and Australia, unconventional gas development is focused on coal-bed methane, while tight gas is of interest in North Africa and shale gas in Poland (IEA, 2011b). Globally, the share of unconventional gas is predicted to rise from 13% of total natural gas production to 25% in 2035. Total global annual production is predicted to rise from 3.3 trillion cubic metres (tcm)

unconventional gas doubled in the last eight years to 350 bcm in 2010 (IEA, 2010).

0.0

**4.1 Unconventional gas** 

CBM, and 10% from shale gas (Bourdaire, 2011).

in 2010, to 5.1 tcm in that time period (IEA, 2011a).

50.0

100.0

150.0

200.0

**billion cubic feet**

250.0

300.0

350.0

Tight gas development started in the 1970's (American Association of Petroleum Geologists [AAPG], 2011). Global tight gas production has risen from 6 billion cubic metres (bcm) in 1993 to 93 bcm in 2009 (Shell, 2011). It is estimated that at least 6000 trillion cubic feet (tcf) are in place in the US alone (NETL, 2011). Tight gas is trapped in impermeable, lowpermeability, or non-porous sedimentary rock (Holditch, 2006; NETL, 2011). The extraction of tight gas requires fracturing and chemical alteration, which makes it costly. To get economic flow rates and/or to recover economic volumes the formation must be stimulated using hydraulic fracturing and/or horizontal well bores (AAPG, 2011).

## **4.1.2 Coalbed methane**

Methane has historically always been a problem in coalmines, due to the risk of explosion during mining operations (EPA, 2004). In the formation of coal from organic material, methane gas is produced and is adsorbed onto the sides of small pores of the coal (EPA, 2004). Coal is structurally weak and consequently fractures easily so coal beds are typically characterized by networks of connected fractures (EPA, 2004). Water in coal beds contributes to the hydrostatic pressure that keeps methane gas adsorbed to the surface of the coal (United States Geological Survey, 2000; NETL, 2011).

As a natural gas, coalbed methane is difficult to produce, although it is relatively pure and does not contain H2S (Government of Alberta, 2011). Coalbed methane is produced by reducing pressure in the coalbed (National Energy Board [NEB], 2007). When pressure is reduced the methane desorbs, diffuses through the coal, and then flows through the fractures (NETL, 2011). This reduction in pressure may be achieved by drilling and pumping out the groundwater in the coal bed, which brings water to the surface (Berquist et al, 2007; USGS, 2010; NEB, 2007; NETL, 2011). The initial pumping-out of groundwater may need additional flow enhancement and generally is followed by more drilling and hydraulic fracturing to enlarge pre-existing natural fractures in the coal seams (EPA, 2004; NEB, 2007; NETL, 2011). Fracturing fluids containing proppants (usually fine sand) are injected into the coal bed and then pumped back out, leaving the proppants to keep the fractures open, and thus increasing the permeability and allowing the methane to flow (EPA, 2004).

#### **4.1.3 Shale gas**

Shales are dense, fine-grained rocks, with very limited porosity (Kargbo et al, 2010). Shales hosting natural gas are rich in organic material (Kargbo et al, 2010; Kerr, 2010; Osborn et al, 2011), with natural gas in the pores of these rocks (Kargbo et al, 2010). Shales are so impermeable that they often act as cap seals on reservoirs of conventional oil and gas.

Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions 163

A significant amount of gas is also flared at the wellhead, releasing CO2. In 2010 an estimated 134 bcm of natural gas was flared (IEA, 2011a). The World Bank has initiated the Global Gas Flaring Reduction programme to attempt to reduce this waste from flaring. Given total annual global production of 3193 bcm of natural gas (BP, 2011), losses of 95bcm due to venting and leakages, and flaring of 134 bcm, approximately 7% of annually

For both conventional and unconventional gas production, methane emissions result from drilling, wellhead establishment, normal operations, routine maintenance, and fugitive emissions. A significant amount of raw methane is released to the atmosphere by pressure venting, leakages, or due to accidents (IEA, 2011a). Fugitive methane emissions occur during transportation, storage, distribution, from connections between pipes and vessels, valves, leaks from wellheads, and along transmission lines (Howarth et al, 2011; Wood et al, 2011). Fugitive emissions also include gas that migrates to the surface around the outside of the wellhead casing (IPCC, 2006). The U.S. EPA in 2006 estimated annual releases of 95 bcm of gas due to leaking and venting, with a greater loss due to leakage rather than venting. The gas industry is now using enhanced sensing equipment to locate and seal leaking wells

The amount of fugitive methane produced is much higher for unconventional natural gas than for conventional. Hydraulic fracturing (fracking) fluids used in tight gas and shale gas beds come back up the well as "flow back" fluids, with large quantities of methane in the fluid (Howarth et al, 2011). It is estimated that in conventional well completion, fugitive emissions account for only 0.01% of methane emissions over the lifetime of the well, versus an estimated 0.6-1.3% in tight sandstones, and 1.9% for shale gas (Howarth et al, 2011).

Fugitive methane emissions can continue after well completion from shale gas deposits. Cement casing of wells becomes more challenging with depth and may cause aquifer contamination (Kargbo et al, 2010; Wood et al, 2011). There is documentation of systemic contamination by methane of aquifers overlying the Marcellus and Utica shales where they have been exploited in northeast Pennsylvania and upper New York State (Osborn et al, 2011). Measured methane concentrations increased in drinking water wells when a gas well was within 1 km (Osborn et al, 2011). Isotopic 13C – CH4 data was consistent with the methane being of deep, thermogenic origin, and not biogenic (Osborn et al, 2011). Methane in wells, close to the surface of the earth, can easily migrate and escape to the atmosphere. Wood et al (2011) contains a review of numerous examples of wellbore failure and migration of methane from the fractured rock to the surface, with explosions in some cases. In Quebec, of 31 shale gas exploration sites, 19 leaked methane. All have now been shut and

Since the Industrial Revolution in the mid-18th Century, fossil fuel use has become the dominant energy source for humans on the planet. Fossil fuels now supply 81% of the global energy mix (IEA, 2011a). The primary use of oil is in transportation, and the greatest demand for natural gas is in electrical power generation. Global production of oil has doubled since

1970, and production of natural gas has tripled in the same interval (BP, 2011).

produced natural gas is lost to the atmosphere as greenhouse gases.

a moratorium placed on any shale gas activity (Marsden, 2011).

and pipelines (Kargbo et al, 2010).

**5. Conclusions** 

Therefore, extraction of the gas is very difficult. It is the last unconventional gas to be developed (Bourdaire, 2011).

Shale gas is extracted utilizing directional drilling and hydraulic fracturing techniques, the same as those used for tight-gas and coalbed methane recovery (Osborn et al, 2011). Multiple wellheads are required (Osborn et al, 2011). Multi-stage fracturing techniques are used with horizontally drilled wells (Kargbo et al, 2010). The drilling "fracking" fluid contains gels, acid, biocides, surfactants, and scale inhibitors to prevent precipitation of sulphate and carbonate (Kargbo et al, 2010). With development of the Marcellus Shale in Pennsylvania and New York, the wells are fractured laterally for just under a kilometer (954m) from the wellbore (Kargbo et al, 2010).

With each well, the hydraulic fluid mix is pumped in with millions of gallons of water mixed with chemicals and sand. Sand grains are proppants, keeping the fractures open and artificially creating permeability so the gas can flow (Kerr, 2010). The multistage fracturing techniques use large amounts of water, which must be managed (Kargbo et al, 2010). In the Marcellus Shale, the fracturing of each horizontal well requires 2-10 million gallons of water (7.7-38 ML) (Kargbo et al, 2010), with other estimates of 3-4 million gallons per well (Kerr, 2010). As much as 80% of the fluid may not be recovered (Kargbo et al, 2010). After drilling the wastewater must be treated onsite or trucked to a wastewater treatment facility. Flow from a new well can decline 60-80% in the first year of production (Kerr, 2010) and may have to be repeatedly fractured in order to continue producing.

#### **4.2 Greenhouse gases from conventional and unconventional gas production and transportation**

Greenhouse gases from both conventional and unconventional natural gas include direct and indirect combustion related CO2 emissions and fugitive methane emissions. Direct CO2 emissions are related to end-use combustion (Howarth et al, 2011; Wood et al, 2011). Indirect emissions of CO2 are those produced by extraction, development, and transportation of the natural gas to market (Howarth et al, 2011). This includes combustion of fossil fuels, usually diesel, to drive pumps, drills, compressors, and to transport equipment and resources to and from the well site (Wood et al, 2011).

The energy used as a percentage of the energy produced for conventional gas is about 6% (IEA ETSAP, 2010). When comparing conventional natural gas to unconventional shale gas, the additional direct and indirect "well-to-burner" CO2 emissions are only marginally higher for shale gas (Howarth et al, 2011; IEA, 2011a; Wood et al, 2011). The ratio between energy used to energy produced between conventional natural gas and that from unconventional tight gas, CBM, and shale gas is relatively small (IEA ETSAP, 2010). Indirect emissions from shale gas in the U.S. are estimated to be 1.2 – 1.7 gC/MJ (Marcellus shale; Santoro et al, 2011), compared to 15 gC/MJ for direct end-use combustion (Hayhoe et al, 2002). Wood et al (2011) estimate that additional direct emissions of CO2 from shale gas are only 0.14 to 1.63 tonnes of CO2e/TJ greater than those from combustion of conventional natural gas (57 tonnes of CO2e/TJ). Therefore, for conventional and shale gas, the GHG emissions are dominated by CO2 from direct end-use burner combustion and also by fugitive methane emissions, discussed below (Howarth et al, 2011; Wood et al, 2011).

A significant amount of gas is also flared at the wellhead, releasing CO2. In 2010 an estimated 134 bcm of natural gas was flared (IEA, 2011a). The World Bank has initiated the Global Gas Flaring Reduction programme to attempt to reduce this waste from flaring. Given total annual global production of 3193 bcm of natural gas (BP, 2011), losses of 95bcm due to venting and leakages, and flaring of 134 bcm, approximately 7% of annually produced natural gas is lost to the atmosphere as greenhouse gases.

For both conventional and unconventional gas production, methane emissions result from drilling, wellhead establishment, normal operations, routine maintenance, and fugitive emissions. A significant amount of raw methane is released to the atmosphere by pressure venting, leakages, or due to accidents (IEA, 2011a). Fugitive methane emissions occur during transportation, storage, distribution, from connections between pipes and vessels, valves, leaks from wellheads, and along transmission lines (Howarth et al, 2011; Wood et al, 2011). Fugitive emissions also include gas that migrates to the surface around the outside of the wellhead casing (IPCC, 2006). The U.S. EPA in 2006 estimated annual releases of 95 bcm of gas due to leaking and venting, with a greater loss due to leakage rather than venting. The gas industry is now using enhanced sensing equipment to locate and seal leaking wells and pipelines (Kargbo et al, 2010).

The amount of fugitive methane produced is much higher for unconventional natural gas than for conventional. Hydraulic fracturing (fracking) fluids used in tight gas and shale gas beds come back up the well as "flow back" fluids, with large quantities of methane in the fluid (Howarth et al, 2011). It is estimated that in conventional well completion, fugitive emissions account for only 0.01% of methane emissions over the lifetime of the well, versus an estimated 0.6-1.3% in tight sandstones, and 1.9% for shale gas (Howarth et al, 2011).

Fugitive methane emissions can continue after well completion from shale gas deposits. Cement casing of wells becomes more challenging with depth and may cause aquifer contamination (Kargbo et al, 2010; Wood et al, 2011). There is documentation of systemic contamination by methane of aquifers overlying the Marcellus and Utica shales where they have been exploited in northeast Pennsylvania and upper New York State (Osborn et al, 2011). Measured methane concentrations increased in drinking water wells when a gas well was within 1 km (Osborn et al, 2011). Isotopic 13C – CH4 data was consistent with the methane being of deep, thermogenic origin, and not biogenic (Osborn et al, 2011). Methane in wells, close to the surface of the earth, can easily migrate and escape to the atmosphere. Wood et al (2011) contains a review of numerous examples of wellbore failure and migration of methane from the fractured rock to the surface, with explosions in some cases. In Quebec, of 31 shale gas exploration sites, 19 leaked methane. All have now been shut and a moratorium placed on any shale gas activity (Marsden, 2011).

## **5. Conclusions**

162 Greenhouse Gases – Emission, Measurement and Management

Therefore, extraction of the gas is very difficult. It is the last unconventional gas to be

Shale gas is extracted utilizing directional drilling and hydraulic fracturing techniques, the same as those used for tight-gas and coalbed methane recovery (Osborn et al, 2011). Multiple wellheads are required (Osborn et al, 2011). Multi-stage fracturing techniques are used with horizontally drilled wells (Kargbo et al, 2010). The drilling "fracking" fluid contains gels, acid, biocides, surfactants, and scale inhibitors to prevent precipitation of sulphate and carbonate (Kargbo et al, 2010). With development of the Marcellus Shale in Pennsylvania and New York, the wells are fractured laterally for just under a kilometer

With each well, the hydraulic fluid mix is pumped in with millions of gallons of water mixed with chemicals and sand. Sand grains are proppants, keeping the fractures open and artificially creating permeability so the gas can flow (Kerr, 2010). The multistage fracturing techniques use large amounts of water, which must be managed (Kargbo et al, 2010). In the Marcellus Shale, the fracturing of each horizontal well requires 2-10 million gallons of water (7.7-38 ML) (Kargbo et al, 2010), with other estimates of 3-4 million gallons per well (Kerr, 2010). As much as 80% of the fluid may not be recovered (Kargbo et al, 2010). After drilling the wastewater must be treated onsite or trucked to a wastewater treatment facility. Flow from a new well can decline 60-80% in the first year of production (Kerr, 2010) and may

**4.2 Greenhouse gases from conventional and unconventional gas production and** 

Greenhouse gases from both conventional and unconventional natural gas include direct and indirect combustion related CO2 emissions and fugitive methane emissions. Direct CO2 emissions are related to end-use combustion (Howarth et al, 2011; Wood et al, 2011). Indirect emissions of CO2 are those produced by extraction, development, and transportation of the natural gas to market (Howarth et al, 2011). This includes combustion of fossil fuels, usually diesel, to drive pumps, drills, compressors, and to transport equipment and resources to and

The energy used as a percentage of the energy produced for conventional gas is about 6% (IEA ETSAP, 2010). When comparing conventional natural gas to unconventional shale gas, the additional direct and indirect "well-to-burner" CO2 emissions are only marginally higher for shale gas (Howarth et al, 2011; IEA, 2011a; Wood et al, 2011). The ratio between energy used to energy produced between conventional natural gas and that from unconventional tight gas, CBM, and shale gas is relatively small (IEA ETSAP, 2010). Indirect emissions from shale gas in the U.S. are estimated to be 1.2 – 1.7 gC/MJ (Marcellus shale; Santoro et al, 2011), compared to 15 gC/MJ for direct end-use combustion (Hayhoe et al, 2002). Wood et al (2011) estimate that additional direct emissions of CO2 from shale gas are only 0.14 to 1.63 tonnes of CO2e/TJ greater than those from combustion of conventional natural gas (57 tonnes of CO2e/TJ). Therefore, for conventional and shale gas, the GHG emissions are dominated by CO2 from direct end-use burner combustion and also by

fugitive methane emissions, discussed below (Howarth et al, 2011; Wood et al, 2011).

developed (Bourdaire, 2011).

**transportation**

from the well site (Wood et al, 2011).

(954m) from the wellbore (Kargbo et al, 2010).

have to be repeatedly fractured in order to continue producing.

Since the Industrial Revolution in the mid-18th Century, fossil fuel use has become the dominant energy source for humans on the planet. Fossil fuels now supply 81% of the global energy mix (IEA, 2011a). The primary use of oil is in transportation, and the greatest demand for natural gas is in electrical power generation. Global production of oil has doubled since 1970, and production of natural gas has tripled in the same interval (BP, 2011).

Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions 165

Alberta Chamber of Resources, 2004. Oil Sands Technology Roadmap: Unlocking the

Baynard, C.W., 2011. The landscape infrastructure footprint of oil development:

Bourdair, J-M, 2011. Unconventional Gas. Presented at the 9th International Association for

Brandt, A., and Ferrell, A, 2007. Scraping the bottom of the bararel: greenhouse gas

CERA (Cambridge Energy Research Associates), 2009. Growth in the Canadian Oil Sands:

Campbell, C., and Laherrère, J. H. 1998. The end of cheap oil. *Scientific American* 278:78–83. CAPP, 2009. Crude Oil: Forecast, Markets, and Pipeline Expansions. Canadian Association

CAPP (Canadian Association of Petroleum Producers), 2011. Crude Oil: Forecast, Markets,

Charpentier, A.D., J.A. Bergerson, and H.L. MacLean, 2009. Understanding the Canadian oil

Chow, D.L., T.N. Nasr, R.S. Chow, and R.P. Sawatzky, 2008. Recovery Techniques for

Delmotte, M., Chappellaz, J., Brook, E., Yiou, P., Barnola, J.M., Goujon, C., Raynaud, D., and

Dlugokencky, E., L. Bruhwiler, J. W. C. White, L. K. Emmons, P. C. Novelli, S. A. Montzka,

bed methane development in the Powder River Basin, Wyoming. Environmental

emissions consequences of a transition to low-quality and synthetic petroleum

Finding the new balance. An IHS CERA Special Report, Cambridge, Massachusetts,

of Petroleum Producers, June 2009, 48pp. 2009-2025 Canadian Crude Oil Forecast

and Pipelines, 40pp. http://www.capp.ca/getdoc.aspx?DocId=190838. Accessed

sands industry's greenhouse gas emissions. Environmental Research Letters, vol. 4,

Canada's Heavy Oil and Bitumen Resources. Journal of Canadian Petroleum

Lipenkov, V., 2004. Atmospheric methane during the last four glacial–interglacial cycles: Rapid changes and their link with Antarctic temperature. J. Geophys. Res.

K. A. Masarie, P. M. Lang, A. M. Crotwell, J. B. Miller, and L. V. Gatti, 2009. Observational constraints on recent increases in the atmospheric CH4 burden.

http://www.acr-alberta.com/Portals/0/projects/OSTR\_report.pdf AAPG (American Association of Petroleum Geologists), 2011. Tight Gas Sands.

Venezuela's heavy oil belt. Ecological Indicators, vol. 11, p. 789-810. Bergquist, E., P. Evangelista, T.J. Stohlgren, and N. Alley, 2007. Invasive species and coal

the Study of Peak Oil and Gas Conference, April 27-29, 2011. http://www.aspo9.be/assets/ASPO9\_Wed\_27\_April\_Bourdaire.pdf

BP, 2011. British Petroleum Statistical Review of World Energy June 2011. 49pp.

http://emd.aapg.org/technical\_areas/tightGas.cfm

Monitoring and Assessment, vol. 128, p. 381-394.

resources. Climatic Change, vol. 84, p. 241-263.

http://www.bp.com/statisticalreview

U.S., 12pp.

11pp.

and Market Outlook

September 22, 2011.

109, 12104, 13 pp..

Technology, vol. 47, no. 5, p. 12-17.

Geophysical Research Letters, vol. 36, L18803, 5 pp. Energy Information Agency, EIA, 2010. International Energy Outlook 2010.

**7. References** 

Potential. 92pp.

Oil and natural gas are both present in conventional and non-conventional forms. Conventional oil and natural gas deposits located on continents and nearshore marine shelves - at shallow depths - made them easily, and therefore cheaply, accessible. The late portions of the first decade of the 21st Century saw the peak of production of conventional crude oil (IEA, 2010) and also conventional natural gas (IEA, 2011a). Easily accessed deposits of conventional oil and gas have been largely depleted, and enhanced recovery techniques need to be used on these diminished reserves. Extraction is now focused on deepwater offshore oil and unconventional deposits of oil and gas.

With increased extraction and use of fossil fuels have come amplifications in the atmospheric abundance of carbon dioxide and methane, the two most abundant greenhouse gases. Concentrations of both gases are now at higher levels than at any time in the past 800,000 years (Luthi et al, 2008; Loulergue et al, 2008). Enhanced levels of greenhouse gases in the atmosphere are believed by the majority of scientists to be responsible for warming of the planet, and concomitant climate change. Policy efforts (e.g. the Kyoto Protocol) and price increases in oil have failed to rein-in usage of fossil fuels and the production of combustion related greenhouse gases. The peak of production of conventional oil has not resulted in diminished use of oil; rather it has resulted in increased production of oil from unconventional sources. Similarly, natural gas from unconventional sources is assuming an increasing role in the global gas market (IEA, 2011b). Demand continues to grow, even though prices remain near historical highs for oil.

To meet the demand for these fuels, rates of exploitation of both unconventional oil and natural gas are growing each year. Both unconventional oil and natural gas share common denominators in that they require extra energy and resources, and are more expensive to produce. They both have higher carbon intensity, producing more greenhouse gases per unit of energy delivered as a final product than conventional oil and natural gas products. There is an increased use of unconventional fuels in order to fill the shortfall left by the peak of production of conventional oil and natural gas. Given that each unit of unconventional oil and natural gas has up to 20% more associated greenhouse gas emissions than a conventional equivalent, there is an enhancement of greenhouse gas in the atmosphere as consumption of unconventional fossil fuels increases.

It is difficult to envisage what will stop the juggernaut of fossil fuel consumption and related GHG increases, barring a global economic collapse. Infrastructure exists, and continues to be built, for both established and developing societies dependent upon oil and natural gas. The large potential reserves of unconventional oil and natural gas can fuel industrialized economies well into the future. It is increasingly difficult to be optimistic of any mitigation of atmospheric GHG growth and climate change in light of the ongoing exploitation of unconventional oil and natural gas.

## **6. Acknowledgements**

This chapter is dedicated to my late husband, Jamie Tiller. I thank Andy Jacobson, NOAA Earth System Research Laboratory, for CO2 ice core data references, Robert W. Howarth for shale gas references, Rushdia Mehreen for help with figures, and Myke Wilder for invaluable technical editing.

### **7. References**

164 Greenhouse Gases – Emission, Measurement and Management

Oil and natural gas are both present in conventional and non-conventional forms. Conventional oil and natural gas deposits located on continents and nearshore marine shelves - at shallow depths - made them easily, and therefore cheaply, accessible. The late portions of the first decade of the 21st Century saw the peak of production of conventional crude oil (IEA, 2010) and also conventional natural gas (IEA, 2011a). Easily accessed deposits of conventional oil and gas have been largely depleted, and enhanced recovery techniques need to be used on these diminished reserves. Extraction is now focused on

With increased extraction and use of fossil fuels have come amplifications in the atmospheric abundance of carbon dioxide and methane, the two most abundant greenhouse gases. Concentrations of both gases are now at higher levels than at any time in the past 800,000 years (Luthi et al, 2008; Loulergue et al, 2008). Enhanced levels of greenhouse gases in the atmosphere are believed by the majority of scientists to be responsible for warming of the planet, and concomitant climate change. Policy efforts (e.g. the Kyoto Protocol) and price increases in oil have failed to rein-in usage of fossil fuels and the production of combustion related greenhouse gases. The peak of production of conventional oil has not resulted in diminished use of oil; rather it has resulted in increased production of oil from unconventional sources. Similarly, natural gas from unconventional sources is assuming an increasing role in the global gas market (IEA, 2011b). Demand continues to grow, even

To meet the demand for these fuels, rates of exploitation of both unconventional oil and natural gas are growing each year. Both unconventional oil and natural gas share common denominators in that they require extra energy and resources, and are more expensive to produce. They both have higher carbon intensity, producing more greenhouse gases per unit of energy delivered as a final product than conventional oil and natural gas products. There is an increased use of unconventional fuels in order to fill the shortfall left by the peak of production of conventional oil and natural gas. Given that each unit of unconventional oil and natural gas has up to 20% more associated greenhouse gas emissions than a conventional equivalent, there is an enhancement of greenhouse gas in the atmosphere as

It is difficult to envisage what will stop the juggernaut of fossil fuel consumption and related GHG increases, barring a global economic collapse. Infrastructure exists, and continues to be built, for both established and developing societies dependent upon oil and natural gas. The large potential reserves of unconventional oil and natural gas can fuel industrialized economies well into the future. It is increasingly difficult to be optimistic of any mitigation of atmospheric GHG growth and climate change in light of the ongoing

This chapter is dedicated to my late husband, Jamie Tiller. I thank Andy Jacobson, NOAA Earth System Research Laboratory, for CO2 ice core data references, Robert W. Howarth for shale gas references, Rushdia Mehreen for help with figures, and Myke Wilder for

deepwater offshore oil and unconventional deposits of oil and gas.

though prices remain near historical highs for oil.

consumption of unconventional fossil fuels increases.

exploitation of unconventional oil and natural gas.

**6. Acknowledgements** 

invaluable technical editing.

Alberta Chamber of Resources, 2004. Oil Sands Technology Roadmap: Unlocking the Potential. 92pp.

http://www.acr-alberta.com/Portals/0/projects/OSTR\_report.pdf

AAPG (American Association of Petroleum Geologists), 2011. Tight Gas Sands. http://emd.aapg.org/technical\_areas/tightGas.cfm


Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions 167

IPCC (Intergovernmental Panel on Climate Change) 2006. 2006 IPCC Guidelines for National Greenhouse Gas Inventories, Chapter 4, Fugitive Emissions, 78 pp. IPCC (Intergovernmental Panel on Climate Change) 2007. Climate Change 2007:

Isaacs, E., 2007. The Canadian Oil Sands in the Context of the Global Energy Demand.

Johnson, E., and Miyanishi, K., 2008. Creating new landscapes and ecosystems: The

Kargbo, D. M., R.G. Wilhelm, and D.J. Campbell, 2010. Natural Gas Plays in the Marcellus

Keeling, C. D., S. C. Piper, R. B. Bacastow, M. Wahlen, T. P. Whorf, M. Heimann, and H. A.

Keeling, R. F., S. C. Piper, A. F. Bollenbacher and S. J. Walker, 2011. Atmospheric CO2

Kerr, R.A., 2010. Natural Gas from Shale Bursts Onto the Scene. Science, vol. 328, no. 5986, p.

Lisiecki, L. E., and Raymo, M. E., 2005. A Pliocene-Pleistocene stack of 57 globally

Loulergue, L., A. Schilt, R. Spahni, V. Masson-Delmotte, T. Blunier, B. Lemieux, J.-M.

Luthi, Dieter, Martine Le Floch, Bernhard Bereiter, Thomas Blunier, Jean-Marc Barnola, Urs

Marsden, Wm., 2011. "Quebec outlines shale gas rules". Montreal Gazette, May 6, 2011, p.

Martinez-Palou, R., M. de Lourdes Mosqueira, B. Zapata-Rendon, E. Mar-Juarez, C. Bernal-

Barnola, D. Raynaud, T.F. Stocker, and J. Chappellaz, 2008. Orbital and millennialscale features of atmospheric CH4 over the past 800,000 years. Nature, vol. 453, p.

Siegenthaler, Dominique Raynaud, Jean Jouzel, Hubertus Fischer, Kenji Kawamura & Thomas F. Stocker, 2008. High-resolution carbon dioxide concentration record

Huicochea, J. de la Cruz Clavel-Lopez, and J. Aburto, 2011. Transportation of heavy

distributed benthic d18O records. Paleoceanography 20, PA2007.

650,000–800,000 years before present. Nature, vol. 453, p. 379-382.

http://scrippsco2.ucsd.edu/data/in\_situ\_co2/monthly\_mlo.csv

Change, 52 pp.

October, 200

Technology, vol. 44, p. 5679-5684.

http://scrippsco2.ucsd.edu

120-145.

0244

1624-1626.

383-386.

A11.

Synthesis Report. An Assessment of the Intergovernmental Panel on Climate

Extended Abstract for the 17th Convocation of the International Council of Academies of engineering and Technological Sciences (CAETS), Tokyo, Japan,

Alberta Oil Sands. Annals of the New York Academy of Sciences, vol. 1134, p.

Shale: Challenges and Potential Opportunities. Environmental Science and

Meijer, 2001. Exchanges of atmospheric CO2 and 13CO2 with the terrestrial biosphere and oceans from 1978 to 2000. I. Global aspects, SIO Reference Series, No. 01-06, Scripps Institution of Oceanography, San Diego, 88 pages, 2001.

concentrations (ppm) derived from in situ air measurements at Mauna Loa, Observatory, Hawaii: Latitude 19.5°N Longitude 155.6°W Elevation 3397m. Scripps CO2 Program ( http://scrippsco2.ucsd.edu ) Scripps Institution of Oceanography (SIO), University of California, La Jolla, California USA 92093-

http://205.254.135.24/oiaf/ieo/nat\_gas.html


http://www.iea.org/index\_info.asp?id=1762

IEA ETSAP (Energy Technology Systems Analysis Programme), 2010. Unconventional Oil and Gas Production. IEA ETSAP Technology Brief P02 – May 2010, 8 pp. http://www.iea-etsap.org/web/E-TechDS.asp

http://205.254.135.24/oiaf/ieo/nat\_gas.html

http://www.eia.gov/todayinenergy/detail.cfm?id=110

http://www.epa.gov/gasstar/basic-information/index.html

Fracturing of Coalbed Methane Reservoirs. 32pp.

Government of Alberta, 2011. Alberta Energy: Coalbed Methane. http://www.energy.alberta.ca/NaturalGas/754.asp

s.pdf

Tech.

vol. 46, p. 927-937.

Report to Congress 2008

Agency,Paris, France, 738 pp.

Energy Outlook 2011, 131 pp.

production of 'unconventional' gas in U.S. http://www.iea.org/index\_info.asp?id=1762

http://www.iea-etsap.org/web/E-TechDS.asp

Energy Information Agency, EIA, 2011. The geology of natural gas resources.

EPA (U.S. Environmental Protection Agency), 2011. Natural Gas STAR program, 2011.

EPA (U.S. Environmental Protection Agency), 2004. Characteristics of Coalbed Methane

http://www.epa.gov/ogwdw/uic/pdfs/cbmstudy\_attach\_uic\_ch03\_cbm\_practice

Hayhoe K, Kheshgi HS, Jain AK, Wuebbles DJ (2002) Substitution of natural gas for coal: climatic effects of utility sector emissions. Climatic Change 54:107–139. Holditch, S.A. 2006. *Tight Gas Sands.* SPE Paper 103356. Distinguished Author Series, J Pet

Holditch, S.A., and H. AdDeen Madani, 2011. Global Unconventional Gas—It Is There, But

Howarth, R.W., Renee Santoro, and Anthony Ingraffea, 2011. Methane and the greenhousegas footprint of natural gas from shale formations. Climatic Change, 12pp. Head, I.M., D.M. Jones, and S.R. Larter, 2003. Biological activity in the deep subsurface and

Holowenko, F.M., M.D. MacKinnon, and P.M. Fedorak, 2000. Methanogens and sulfate-

Humphries, M., 2008. North American Oil Sands: History of Development, Prospects for

Hyndman, A.W., and Luhning, 1991. Recovery and upgrading of bitumen and heavy oil in Canada. Journal of Canadian Petroleum Technology, vol. 30, no. 2, p. 61-71. IEA (International Energy Agency), 2010. World Energy Outlook 2010. International Energy

IEA (International Energy Agency), 2011a. Are we entering a Golden Age of Gas? World

IEA (International Energy Agency), 2011b. Global surge of activity follows successful

IEA ETSAP (Energy Technology Systems Analysis Programme), 2010. Unconventional Oil and Gas Production. IEA ETSAP Technology Brief P02 – May 2010, 8 pp.

Hubbert, M. K. 1971. The energy resources of the Earth. *Scientific American* 225:66–70.

reducing bacteria in oil sands fine tailing water. Canadian Journal of Microbiology,

the Future. CRS Report for Congress. Order Code RL34258, 30pp. Marc Humphries

Is It Profitable? http://www.jptonline.org/index.php?id=533

the origin of heavy oil. Nature, vol. 426, p. 344-352.

Hubbert, M. K. 1949. Energy from fossil fuels. *Science* 109:103–109.

Production and Associated Hydraulic Fracturing Practices. Chapter3 In Evaluation of Impacts to Underground Sources June 2004 of Drinking Water by Hydraulic


http://scrippsco2.ucsd.edu/data/in\_situ\_co2/monthly\_mlo.csv


Exploitation of Unconventional Fossil Fuels: Enhanced Greenhouse Gas Emissions 169

Santoro, R., R.W. Howarth, and A. R. Ingraffea, 2011. Indirect Emissions of Carbon Dioxide

http://www.eeb.cornell.edu/howarth/IndirectEmissionsofCarbonDioxidefromMa

Schneider, R.R., J.B. Stelfox, S.Bouton, and S. Wassel, 2003. Managing the Cumulative

 http://www.shell.co.uk/home/content/gbr/aboutshell/media\_centre/annual\_re ports\_and\_publications/swuk/summer\_2011/shale\_tight\_gas.html Siegenthaler, U., Stocker, T., Monnin, E., Luthi, D., Schwander, J., Stauffer, B., Raynaud, D.,

 http://www.netl.doe.gov/energyanalyses/pubs/NETL%20LCA%20Petroleum-Based%20Fuels%20Nov%202008.pdf (accessed September 20, 2011). Soderburgh, B., Robelius, F., and K. Aleklett, 2007. A crash programme scenario for the

Sorrell, S., Speirs, J., Bentley, R., Brandt, A., and Miller, R., 2010. Global Oil depletion: A

Spahni, R., Chappellaz, J., Stocker, T., Loulergue, L., Hausammann, G., Kawamura, K.,

Stearns, M., J.A. Tindall, G. Cronin, M.J. Friedel and E. Bergquist, 2005. Effects of coal-bed

Storey, M., R.A. Duncan, and C.C. Swisher, 2007. Paleocene-Eocene Thermal Maximum and

Stratton, R.W., H.M. Wong, and J.I. Hileman, 2010. PARTNER Project 28 Report: Life cycle

United States Geological Survey (USGS), 2000. Water Produced with Coal-Bed Methane.

http://www.theglobeandmail.com/report-on-business/enbridge-pipeline-plan-

Fluckiger, J., Schwander, J.., Raynaud, D., Masson-Delmotte, V., and Jouzel, J., 2005. Atmospheric methane and nitrous oxide of the Late Pleistocene from Antarctic ice

methane discharge waters on the vegetation an soil econsystem in Powder River

greenhouse gas emissions from alternative jet fuels, 153 pp. Partnership for AiR Transportation Noise and Emissions Reduction Report No. PARTNER-COE-2010- 001http://web.mit.edu/aeroastro/partner/reports/proj28/partner-proj28-2010-

Canadian oil sands industry. Energy Policy, vol. 35, p. 1931-1947.

review of the evidence. Energy Policy, vol. 38, no. 9, p. 5290-5295.

Basin, Wyoming. Water, Air, and Soil Pollution, vol. 168, p. 33–57.

the Opening of the Northeast Atlantic. Science, vol. 316, p. 587-589.

Energy, and Environment Program at Cornell University, 28pp.

Approach. Conservation Ecology, vol. 7, issue 1, article 8, 15pp.

Technology Laboratory: Pittsburgh, Pennsylvania, 2008;

cores. Science, vol. 310, p. 1317–1321.

001.pdf (accessed September 20, 2011).

sparks-opposition/article2175079/

Vanderklippe,N. 2011a. Enbridge pipeline plan sparks opposition.

Vanderklippe, N. 2011b. CN, CP push for a 'pipeline on rails'.

USGS Fact Sheet FS-156-00.

rcellusShaleGasDevelopment\_June302011 .pdf

Shell, 2011. Unlocking natural gas from unconventional sources.

from Marcellus Shale Gas Development. A Technical Report from the Agriculture,

impacts of land-uses in the Western Canadian Sedimentary Basin: A modeling

Barnola, J.-M., Fischer, H., Masson-Delmotte, V., and Jouzel, J., 2005. Stable carbon cycle–climate relationship during the Late Pleistocene. Science 310, 1313–1317. Skone, T and Gerdes, K, *Development of Baseline Data and Analysis of Life Cycle Greenhouse Gas* 

*Emissions of Petroleum-Based Fuels*, DOE/NETL-2009/1346, National Energy and

and extra-heavy crude oil by pipeline: A Review. Journal of Petroleum Science and Engineering, vol. 75, p. 274-282.


http://web.mit.edu/mitei/research/studies/report-natural-gas.pdf


Monnin, E., Indermuhle, A., Dallenback, A., Fluckiger, J., Stauffer, B., Stocker, T., Raynaud,

Montzka, S.A., E.J. Dlugokencky, and J.H. Butler, 2011. Non-CO2 greenhouse gases and

MIT (Massachussets Institute of Technology), 2010. MIT (2010) *The future of natural gas,* an

NETL (National Energy Technology Laboratory), 2011. Future Supply and Emerging

Nisbet, E.G., 2002. Have sudden large releases of methane from geological reservoirs

Osborn, S. G., A. Vengosh, N.R. Warner, and R.B. Jackson, 2011. Methane contamination of

Pepin, L., Raynaud, D., Barnola, J. M. & Loutre, M. F., 2001. Hemispheric roles of climate

Raynaud, D., Barnola, J.-M., Souchex, R., Lorrain, R., Petit, J.-R., Duval, P., and Lipendov, V.,

Ruddiman, W., 2001. Earth's climate: Past and Future. W.H. Freeman, New York, 465 pp. (S&T)2 Consultants Inc., 2008. 2008 GHGENIUS UPDATE, prepared for Natural Resources

and LLN-2D model experiments. J. Geophys. Res. 106, 31885–31892. Petit, J. R., Jouzel, J., Raynaud, d., Barkov, N., Barnola, J.-M., Basile, I., Beders, M.,

Press, F., and Siever, R., 1986. Earth, 4th Edition. W.H. Freeman, New York, 656 pp.

Canada, Office of Energy Efficiency, Ottawa, Ontario, 86pp.

occurred since the last Glacial Maximum, and could such releases occur again? Philosophical Transactions of the Royal Society London A, vol. 360, no. 1793, p.

drinking water accompanying gas-well drilling and hydraulic fracturing. Proceedings of the National Academy of Sciences, vol. 108, no. 20, p. 8172-8176. Patterson, J., and Perl, A., 2007. The End of Cheap Oil: Crossroads for Kyoto. Energy

forcings during glacial–interglacial transitions as deduced from the Vostok record

Chappellaz, J., Davis, M., Delaygue, G., Delmotte, M., Kotlyadov, V., Legrand, M., Lipendov, V., Lorius, C., Pepin, L., Ritz, C., Saltzman, E., and Stievenard, M., 1999. Climate and atmospheric history of the past 420,000 years from the Vostok ice core,

2005. Palaeolclimatology: The record for marine isotopic stage 11. Nature vol. 436,

Engineering, vol. 75, p. 274-282.

termination. Science 291, 112–114.

Resources: Coal Bed Natural Gas.

Sources, Part B, vol. 2, p. 105-111.

Antarctica. Nature 399, 429–436.

581-607.

p. 39–40.

climate change. Nature, vol. 476, p. 43-50.

Initiative ISBN (978-0-9828008-0-5 Copyright MIT 2010. http://web.mit.edu/mitei/research/studies/report-natural-gas.pdf

NEB (National Energy Board), 2007. Coalbed Methane Fact Sheet. http://www.neb.gc.ca/clf-nsi/rnrgynfmtn/nrgyrprt/ntrlgs/ hrsshcnynclbdmthn2007/clbdmthnfctsht-eng.html

> http://www.netl.doe.gov/technologies/oilgas/futuresupply/coalbedng/coalbed\_ng.html

and extra-heavy crude oil by pipeline: A Review. Journal of Petroleum Science and

D., and Barnola, J.-M., 2001. Atmospheric CO2 concentrations over the last glacial

interdisciplinary study the Massachussets Institute of Technology's Energy

Santoro, R., R.W. Howarth, and A. R. Ingraffea, 2011. Indirect Emissions of Carbon Dioxide from Marcellus Shale Gas Development. A Technical Report from the Agriculture, Energy, and Environment Program at Cornell University, 28pp. http://www.eeb.cornell.edu/howarth/IndirectEmissionsofCarbonDioxidefromMa

rcellusShaleGasDevelopment\_June302011 .pdf


**8**

Md Rumi Shammin

*Oberlin College* 

*USA* 

**The Role of US Households** 

**in Global Carbon Emissions** 

United States (US) is currently responsible for about 20% of global carbon emissions. The residential sector is responsible for a little over 20% of this emission. From this perspective, US households account for about 4% of global carbon emissions. Such sector-based approach is commonly used in energy analysis and energy policy. However, this is not necessarily a complete representation of the reality of household carbon emissions. The residential sector includes all energy directly used by homes and related carbon emissions. Two important elements are missing in this approach: energy used for transportation by people living in these homes and the embodied energy in all non-energy goods and services consumed by them. Another approach for reporting household emissions, based on the various end uses of energy in US homes, provides a more detailed understanding of the use of energy for heating, cooling, cooking, appliances, consumer electronics and automobiles. This approach also falls short of identifying emissions beyond the residential sector and personal transportation sub-sector. There is, however, another way of estimating total household carbon emissions. The industrial sector produces products that are transported by the transportation sector and marketed by the commercial sector and eventually consumed by people. Therefore, people in the US consume energy directly in the form of electricity, natural gas, and other fuels for their homes and automobiles. They also consume energy indirectly through the consumptions of various products and services. Combining the emissions related to the direct and indirect consumption of energy, people are accountable for about 71% of US carbon emissions (Shammin & Bullard, 2009) – which is significantly higher than the 20% represented by the residential sector. According to this approach, US households account for about 14% of global carbon dioxide (CO2) emissions – roughly equal to the total emissions of the 27 member states of the European Union. People in the US thus have a significant opportunity to contribute to the reduction of global carbon

This chapter presents a new, more comprehensive, more interesting and above all, more empowering approach to household carbon emissions in the US. It focuses on how US households contribute to greenhouse gas emissions, how they can play an important role in

1 In this chapter *carbon emissions* and *CO2 emissions* are used interchangeably. All data are reported for

**1. Introduction** 

emission1.

CO2.

 http://www.theglobeandmail.com/globe-investor/cn-cp-push-for-a-pipeline-onrails/article1898062/


coop\_shale\_gas\_report\_final.pdf

## **The Role of US Households in Global Carbon Emissions**

Md Rumi Shammin *Oberlin College USA* 

#### **1. Introduction**

170 Greenhouse Gases – Emission, Measurement and Management

http://www.theglobeandmail.com/globe-investor/cn-cp-push-for-a-pipeline-on-

Watkins, E., 2010. Venezuelan oil output set for rise. Oil and Gas Journal, vol. 108, n. 20, p

Wood R, Gilbert P, Sharmina M, Anderson K, Fottitt A, Glynn S, Nicholls F (2011) Shale gas:

Center, University of Manchester, Manchester, England.

http://www.tyndall.ac.uk/sites/default/files/tyndall-

coop\_shale\_gas\_report\_final.pdf

a provisional assessment of climate change and environmental impacts. Tyndall

rails/article1898062/

39-40.

United States (US) is currently responsible for about 20% of global carbon emissions. The residential sector is responsible for a little over 20% of this emission. From this perspective, US households account for about 4% of global carbon emissions. Such sector-based approach is commonly used in energy analysis and energy policy. However, this is not necessarily a complete representation of the reality of household carbon emissions. The residential sector includes all energy directly used by homes and related carbon emissions. Two important elements are missing in this approach: energy used for transportation by people living in these homes and the embodied energy in all non-energy goods and services consumed by them. Another approach for reporting household emissions, based on the various end uses of energy in US homes, provides a more detailed understanding of the use of energy for heating, cooling, cooking, appliances, consumer electronics and automobiles. This approach also falls short of identifying emissions beyond the residential sector and personal transportation sub-sector. There is, however, another way of estimating total household carbon emissions. The industrial sector produces products that are transported by the transportation sector and marketed by the commercial sector and eventually consumed by people. Therefore, people in the US consume energy directly in the form of electricity, natural gas, and other fuels for their homes and automobiles. They also consume energy indirectly through the consumptions of various products and services. Combining the emissions related to the direct and indirect consumption of energy, people are accountable for about 71% of US carbon emissions (Shammin & Bullard, 2009) – which is significantly higher than the 20% represented by the residential sector. According to this approach, US households account for about 14% of global carbon dioxide (CO2) emissions – roughly equal to the total emissions of the 27 member states of the European Union. People in the US thus have a significant opportunity to contribute to the reduction of global carbon emission1.

This chapter presents a new, more comprehensive, more interesting and above all, more empowering approach to household carbon emissions in the US. It focuses on how US households contribute to greenhouse gas emissions, how they can play an important role in

<sup>1</sup> In this chapter *carbon emissions* and *CO2 emissions* are used interchangeably. All data are reported for CO2.

The Role of US Households in Global Carbon Emissions 173

managing carbon emissions for the different sectors. Still, the *sectoral* approach provides a limited understanding of the total carbon emissions and ways of reducing emissions at the household level. It implies that households are responsible for primarily residential emissions and part of the transportation related emissions. Perhaps this is because the *sectoral* approach is not organized around people, their behavior, and their lives; rather it is categorized on the basis of macro-economic activities. Hence, the *sectoral* analysis is limited in its ability to explicitly represent the different ways people interact with the various

21%

It is possible to link people's lives with the economic sectors included in *sectoral* analysis. First, the residential sector emissions are directly attributable to individuals in US households. These are emissions resulting from electricity, natural gas (or propane), fuel oil, wood, and other fuels used as a source of energy by residential consumers. People have a reasonable level of control over their use of these resources – within the constraints of existing infrastructure and resource availability. Second, a large part of emissions from the transportation sector are from personal automobiles. This part of the transportation sector emission is also directly attributable to people. Similar to the first case, people have a lot of control over their transportation emissions: choice of transportation mode, fuel efficiency of automobiles, place of residence relative to workplace and other daily destinations, etc. Together, the two above cases account for about 39% of carbon emissions that can be linked directly to people (based on data for 2009 from EIA, 2011). The remaining 61% is more complicated and requires deeper understanding of the life cycle of products and services. As mentioned earlier, the industrial sector produces products that are transported by the transportation sector and marketed by the commercial sector and eventually consumed by people. Sometimes there are multiple layers in the supply chain of products. Some industrial outputs are transported as parts or input materials for other industries before eventually making it to the marketplace. People are not always direct consumers either. Industries themselves consume various products or services and so do various commercial enterprises and non-governmental organizations. Another big consumer is the government itself – for

28%

Fig. 2. US CO2 emissions by economic sectors (EIA, 2011)

19%

Residential Commercial

Industrial

Transportation

economic sectors.

32%

reducing global carbon emissions, and also how such efforts will potentially make them more resilient in the long run.

## **2. Carbon emissions by US households**

There are multiple ways of estimating household carbon emissions. In fact, currently there is no established system of calculating and reporting total household carbon emissions for US households that includes a comprehensive accounting of the various ways households are directly and indirectly responsible for carbon emissions. US Environmental Protection Agency (EPA) and Energy Information Agency (EIA) both publish yearly reports on US greenhouse gas (GHG) emissions. These reports are organized around the major sectors of the economy (henceforth referred to as the *Sectoral* approach) and provide a macro-level overview of US GHG emissions. Another way household GHG emissions are often reported is based on various energy end uses – such as appliances, HVAC (heating, ventilation and air conditioning) systems, lights, cars, etc. (henceforth referred to as the *End Use* Approach). The sections below investigate the current methods of estimating carbon emissions under both of these approaches. Some boundary conditions need to be established prior to that. About 83% of US GHG emissions are carbon based of which more than 98% is energyrelated (see figure 1). Hence the specific analysis of this paper will focus mainly on energy related carbon emissions. It should, however, be noted that GHG emissions from noncarbon sources also play a significant role in global climate change particularly on a global scale, but they are kept outside of the scope of this analysis.

Fig. 1. 2009 US greenhouse gas emissions by gas in million metric tons (EIA, 2011)

#### **2.1 The sectoral approach**

The *sectoral* approach looks at the four major sectors of the US economy: residential, commercial, industrial and transportation. Emissions from electricity generation are distributed between these sectors, but are also sometimes reported separately. This picture is technically sound and the accounting method is time-tested - resulting in fairly accurate estimation of total carbon emissions for the economy as a whole by adding up the parts (see figure 2). This approach is consistent with government planning, budgeting and other fiscal activities. The *sectoral* approach also helps in the development of appropriate policies for

reducing global carbon emissions, and also how such efforts will potentially make them

There are multiple ways of estimating household carbon emissions. In fact, currently there is no established system of calculating and reporting total household carbon emissions for US households that includes a comprehensive accounting of the various ways households are directly and indirectly responsible for carbon emissions. US Environmental Protection Agency (EPA) and Energy Information Agency (EIA) both publish yearly reports on US greenhouse gas (GHG) emissions. These reports are organized around the major sectors of the economy (henceforth referred to as the *Sectoral* approach) and provide a macro-level overview of US GHG emissions. Another way household GHG emissions are often reported is based on various energy end uses – such as appliances, HVAC (heating, ventilation and air conditioning) systems, lights, cars, etc. (henceforth referred to as the *End Use* Approach). The sections below investigate the current methods of estimating carbon emissions under both of these approaches. Some boundary conditions need to be established prior to that. About 83% of US GHG emissions are carbon based of which more than 98% is energyrelated (see figure 1). Hence the specific analysis of this paper will focus mainly on energy related carbon emissions. It should, however, be noted that GHG emissions from noncarbon sources also play a significant role in global climate change particularly on a global

5,359.6

The *sectoral* approach looks at the four major sectors of the US economy: residential, commercial, industrial and transportation. Emissions from electricity generation are distributed between these sectors, but are also sometimes reported separately. This picture is technically sound and the accounting method is time-tested - resulting in fairly accurate estimation of total carbon emissions for the economy as a whole by adding up the parts (see figure 2). This approach is consistent with government planning, budgeting and other fiscal activities. The *sectoral* approach also helps in the development of appropriate policies for

Fig. 1. 2009 US greenhouse gas emissions by gas in million metric tons (EIA, 2011)

Energy-related carbon dioxide High-GWP gases

Nitrous oxide

Other carbon dioxide

Methane

more resilient in the long run.

**2. Carbon emissions by US households** 

scale, but they are kept outside of the scope of this analysis.

219.6 178.2

87.3

**2.1 The sectoral approach** 

730.9

managing carbon emissions for the different sectors. Still, the *sectoral* approach provides a limited understanding of the total carbon emissions and ways of reducing emissions at the household level. It implies that households are responsible for primarily residential emissions and part of the transportation related emissions. Perhaps this is because the *sectoral* approach is not organized around people, their behavior, and their lives; rather it is categorized on the basis of macro-economic activities. Hence, the *sectoral* analysis is limited in its ability to explicitly represent the different ways people interact with the various economic sectors.

Fig. 2. US CO2 emissions by economic sectors (EIA, 2011)

It is possible to link people's lives with the economic sectors included in *sectoral* analysis. First, the residential sector emissions are directly attributable to individuals in US households. These are emissions resulting from electricity, natural gas (or propane), fuel oil, wood, and other fuels used as a source of energy by residential consumers. People have a reasonable level of control over their use of these resources – within the constraints of existing infrastructure and resource availability. Second, a large part of emissions from the transportation sector are from personal automobiles. This part of the transportation sector emission is also directly attributable to people. Similar to the first case, people have a lot of control over their transportation emissions: choice of transportation mode, fuel efficiency of automobiles, place of residence relative to workplace and other daily destinations, etc. Together, the two above cases account for about 39% of carbon emissions that can be linked directly to people (based on data for 2009 from EIA, 2011). The remaining 61% is more complicated and requires deeper understanding of the life cycle of products and services. As mentioned earlier, the industrial sector produces products that are transported by the transportation sector and marketed by the commercial sector and eventually consumed by people. Sometimes there are multiple layers in the supply chain of products. Some industrial outputs are transported as parts or input materials for other industries before eventually making it to the marketplace. People are not always direct consumers either. Industries themselves consume various products or services and so do various commercial enterprises and non-governmental organizations. Another big consumer is the government itself – for

The Role of US Households in Global Carbon Emissions 175

Life cycle analysis – particularly for all products and services in an economy - appears to be a daunting task. This is where Leontief's work on input-output analysis came in handy. Originally developed for macro-economic analysis, Leontief formulated a mathematical process of inverting the complex matrix of all inter-sector transactions in an economy to derive the total final consumption of individual sectors (Leontief, 1970). This allowed for tracking the flow of money through and between sectors in any given year. Robert Herendeen, Bruce Hannon, Clark Bullard, and others associated with the Energy Resources Group at the University of Illinois carried out the seminal work of using Leontief's method to track the flow of money spent on energy resources (or in some cases the physical flows of energy) within the US economy and then converting the results into the physical quantities of energy used by various industries and enterprises. When this data is combined with national consumer expenditure data, one can actually estimate both the energy intensities of products and services and of households of different income groups (Bullard & Herendeen, 1975; Herendeen & Tanaka, 1976; Herendeen, 1978; Herendeen et al., 1981). Notable followup work that builds on this approach has been done by Manfred Lenzen of University of Sydney, Rutger Hoekstra of Statistics Nederlands, and many others who used this approach to not only estimate energy intensities but also carbon emissions and other environmental impacts. Hoekstra (2010) compiled a database of the development of environmental analysis based on Leontief's input-out method. More recent estimates of energy and carbon intensities are reported in Shammin et al (2010), Shammin & Bullard (2009), and Bin & Dowlatabadi (2005). These papers primarily used the Economic Input Output Life Cycle Analysis (EIOLCA) database developed at Carnegie Melon University2. This method now allows for a much more complete and in-depth understanding of household energy use and carbon emissions that can directly be linked to people's behavior and choices. While details on the methods can be found in the papers cited above, figure 3 provides a generic outline of the process of using economic input-output analysis to estimate carbon intensities for goods and services and combining that with consumer expenditure data to derive total household

An analysis of the US economy on the basis of personal consumption and other expenditures presents a very different perspective than the *sectoral* and *end use* approaches. In this view, based on data from the Bureau of Economic Analysis, personal consumption expenditures in 2003 accounted for about 70% of the gross domestic product (GDP) while the remaining 30% was shared by government expenditure, investment, and net exports3. Here, US households contributed about 4.17 billion metric tons of CO2 emissions through their consumption of various goods and services – about 71% of the national total emissions of 5.86 billion metric tons4. Based on numbers reported in figure 4, a few key indicators can be calculated for 2003: the energy and CO2 intensities of the economy were 10,058 kJ/\$ and

2 Carnegie Mellon University Green Design Institute. Economic Input-Output Life Cycle Assessment (EIO-LCA), Available from: http://www.eiolca.net 3 For consumption based household emissions, all numbers are for the year 2003 for consistency with the results of economic input-output analysis of embodied energy and carbon reported in Shammin et

4 These calculations are based on data from Table 1.5 of the Annual Energy Outlook 2007 published by

carbon emissions.

al (2010) and Shammin & Bullard (2009).

the EIA and results reported in Shammin & Bullard (2009).

**3. Consumption based household CO2 emissions** 

its various organizations including defense (military, air force, navy and various intelligence agencies). Ultimately all these different consumptive activities and related carbon emissions can be linked back to people. They are all intended to provide people with essential and non-essential products (including public infrastructure) and services (including social services and national security). These linkages are not clearly identifiable in the current representation of the *sectoral* analysis of carbon emissions.

## **2.2 The end use approach**

In contrast to the *sectoral* approach, the *end use* approach provides a more detailed overview of micro-level energy use and carbon emissions by US households. Under this approach, the various energy end uses in households are documented and related carbon emissions are quantified. This includes energy used by HVAC systems, cooking, water heating, appliances, lighting, and other devices such as televisions, computers, and other household electronics. EIA routinely conducts the Residential Energy Consumption Survey (RECS) to generate data for *end use* analysis. The results of the *end use* approach is important for understanding the relative magnitudes of energy consumption by different systems within households and identifying opportunities for conservation and efficiency improvements to reduce household GHG emissions. While this approach is crucial for the purposes stated above, it is still limited to direct consumption of energy and related emissions. It provides a more in-depth understanding of residential sector emissions, but does not shed any additional light on the linkages between households and the other economic sectors listed under the *sectoral* approach.

#### **2.3 Towards a more complete accounting of household emissions**

Over the past four decades development of input-out analysis using methods developed by Nobel Laureate Wassily Leontief (Leontief, 1970) has made it possible to use economic input-out analysis to carry out more complete estimates of household energy consumption and carbon emissions. This is based on two related concepts: a) life cycle analysis: a method of estimating the impact of any resources use over its life cycle - from the point of raw material extraction to ultimate disposal of postconsumer wastes; and b) embodied energy (or embodied carbon): energy use or carbon emissions that occur at various stages over the life cycle of products and services that people eventually consume. This is applicable to both direct energy resources (electricity, natural gas, gasoline and other fuel) and non-energy goods and services (food, clothing, entertainment, insurance etc.). The following two examples illustrate this. A life cycle perspective of gasoline demonstrates that it is not just the emissions resulting from burning gasoline in automobiles, but energy is used and carbon emissions occur for finding, drilling, transporting, refining, and marketing gasoline. These add about 25% carbon emissions that are indirectly attributable to gasoline use. Therefore, the embodied CO2 emissions for gasoline is 2.3 kg/liter from direct burning and an additional 0.6 kg/liter from indirect sources – resulting in a total embodied CO2 of 2.9 kg/liter. While this example explains life cycle analysis and embodied emissions for an energy resource, all non-energy goods and services also have similar embodied life cycle emissions. For example, energy related emissions take place throughout the life cycle of the clothes people buy. If all these emissions are added up and normalized for every dollar spent on clothing in the US, the embodied CO2 emissions for clothing amount to 0.43 kg/\$.

its various organizations including defense (military, air force, navy and various intelligence agencies). Ultimately all these different consumptive activities and related carbon emissions can be linked back to people. They are all intended to provide people with essential and non-essential products (including public infrastructure) and services (including social services and national security). These linkages are not clearly identifiable in the current

In contrast to the *sectoral* approach, the *end use* approach provides a more detailed overview of micro-level energy use and carbon emissions by US households. Under this approach, the various energy end uses in households are documented and related carbon emissions are quantified. This includes energy used by HVAC systems, cooking, water heating, appliances, lighting, and other devices such as televisions, computers, and other household electronics. EIA routinely conducts the Residential Energy Consumption Survey (RECS) to generate data for *end use* analysis. The results of the *end use* approach is important for understanding the relative magnitudes of energy consumption by different systems within households and identifying opportunities for conservation and efficiency improvements to reduce household GHG emissions. While this approach is crucial for the purposes stated above, it is still limited to direct consumption of energy and related emissions. It provides a more in-depth understanding of residential sector emissions, but does not shed any additional light on the linkages between households and the other economic sectors listed

Over the past four decades development of input-out analysis using methods developed by Nobel Laureate Wassily Leontief (Leontief, 1970) has made it possible to use economic input-out analysis to carry out more complete estimates of household energy consumption and carbon emissions. This is based on two related concepts: a) life cycle analysis: a method of estimating the impact of any resources use over its life cycle - from the point of raw material extraction to ultimate disposal of postconsumer wastes; and b) embodied energy (or embodied carbon): energy use or carbon emissions that occur at various stages over the life cycle of products and services that people eventually consume. This is applicable to both direct energy resources (electricity, natural gas, gasoline and other fuel) and non-energy goods and services (food, clothing, entertainment, insurance etc.). The following two examples illustrate this. A life cycle perspective of gasoline demonstrates that it is not just the emissions resulting from burning gasoline in automobiles, but energy is used and carbon emissions occur for finding, drilling, transporting, refining, and marketing gasoline. These add about 25% carbon emissions that are indirectly attributable to gasoline use. Therefore, the embodied CO2 emissions for gasoline is 2.3 kg/liter from direct burning and an additional 0.6 kg/liter from indirect sources – resulting in a total embodied CO2 of 2.9 kg/liter. While this example explains life cycle analysis and embodied emissions for an energy resource, all non-energy goods and services also have similar embodied life cycle emissions. For example, energy related emissions take place throughout the life cycle of the clothes people buy. If all these emissions are added up and normalized for every dollar spent on clothing in the US, the embodied CO2 emissions for clothing amount to 0.43 kg/\$.

representation of the *sectoral* analysis of carbon emissions.

**2.3 Towards a more complete accounting of household emissions** 

**2.2 The end use approach** 

under the *sectoral* approach.

Life cycle analysis – particularly for all products and services in an economy - appears to be a daunting task. This is where Leontief's work on input-output analysis came in handy. Originally developed for macro-economic analysis, Leontief formulated a mathematical process of inverting the complex matrix of all inter-sector transactions in an economy to derive the total final consumption of individual sectors (Leontief, 1970). This allowed for tracking the flow of money through and between sectors in any given year. Robert Herendeen, Bruce Hannon, Clark Bullard, and others associated with the Energy Resources Group at the University of Illinois carried out the seminal work of using Leontief's method to track the flow of money spent on energy resources (or in some cases the physical flows of energy) within the US economy and then converting the results into the physical quantities of energy used by various industries and enterprises. When this data is combined with national consumer expenditure data, one can actually estimate both the energy intensities of products and services and of households of different income groups (Bullard & Herendeen, 1975; Herendeen & Tanaka, 1976; Herendeen, 1978; Herendeen et al., 1981). Notable followup work that builds on this approach has been done by Manfred Lenzen of University of Sydney, Rutger Hoekstra of Statistics Nederlands, and many others who used this approach to not only estimate energy intensities but also carbon emissions and other environmental impacts. Hoekstra (2010) compiled a database of the development of environmental analysis based on Leontief's input-out method. More recent estimates of energy and carbon intensities are reported in Shammin et al (2010), Shammin & Bullard (2009), and Bin & Dowlatabadi (2005). These papers primarily used the Economic Input Output Life Cycle Analysis (EIOLCA) database developed at Carnegie Melon University2. This method now allows for a much more complete and in-depth understanding of household energy use and carbon emissions that can directly be linked to people's behavior and choices. While details on the methods can be found in the papers cited above, figure 3 provides a generic outline of the process of using economic input-output analysis to estimate carbon intensities for goods and services and combining that with consumer expenditure data to derive total household carbon emissions.

## **3. Consumption based household CO2 emissions**

An analysis of the US economy on the basis of personal consumption and other expenditures presents a very different perspective than the *sectoral* and *end use* approaches. In this view, based on data from the Bureau of Economic Analysis, personal consumption expenditures in 2003 accounted for about 70% of the gross domestic product (GDP) while the remaining 30% was shared by government expenditure, investment, and net exports3. Here, US households contributed about 4.17 billion metric tons of CO2 emissions through their consumption of various goods and services – about 71% of the national total emissions of 5.86 billion metric tons4. Based on numbers reported in figure 4, a few key indicators can be calculated for 2003: the energy and CO2 intensities of the economy were 10,058 kJ/\$ and

<sup>2</sup> Carnegie Mellon University Green Design Institute. Economic Input-Output Life Cycle Assessment

<sup>(</sup>EIO-LCA), Available from: http://www.eiolca.net 3 For consumption based household emissions, all numbers are for the year 2003 for consistency with the results of economic input-output analysis of embodied energy and carbon reported in Shammin et al (2010) and Shammin & Bullard (2009).

<sup>4</sup> These calculations are based on data from Table 1.5 of the Annual Energy Outlook 2007 published by the EIA and results reported in Shammin & Bullard (2009).

The Role of US Households in Global Carbon Emissions 177

The consumption based perspective on household emissions communicates a very different message to people. It shows that people have the power to directly and indirectly affect a large component of the nations' carbon emissions through behavior and lifestyle changes that would affect their consumption patterns and preferences. People also have the power to affect the remaining 29% of emissions; however, that would require engagement beyond personal choices. Through political activism and voting patterns, people in a democracy have the power to communicate how the government should spend their tax dollars. Regarding domestic investment, increasingly there are options being available to make investment choices on the basis of environmental performance. Finally, export-import policies can be reformed to trade with partners that are more environmentally responsible. None of this is easy; but at least this way of relating people and their behavior with national carbon emissions and mitigation opportunities provides a perspective that is either missing

The most recent results to date for consumption-based carbon emissions for US households are reported in Shammin & Bullard (2009). The methods are based on calculations of energy intensities using input-output analysis described in Shammin et al (2010). These two papers also discuss the assumptions, nuances, and uncertainties associated with this approach. In 2003, the average household in the US spent about \$49,000 of which only 6.5% was spent on direct energy. The total embodied CO2 emission per household was about 37 metric tons of which about 65% was from direct energy. Thus, a small percent of household expenditure is actually responsible for the bulk of its carbon emissions. However, it is also interesting to see that the remainder of household expenditures made on non-energy goods and services were responsible for about 35% of the total embodied carbon emissions by the average household. This is a significant part of household emissions that is associated with life cycle emissions and the linkages between people and the sectors of the economy beyond residential and personal automobiles. The breakdown of total expenditure and total

Shammin & Bullard (2009) reports detailed carbon intensities for all personal consumption categories based on standard classification of the Bureau of Labor Statistics. A list of carbon intensities for major consumption categories is given in table 1. The distribution of carbon emissions between expenditure categories shown in figure 5b and the intensities in table 1 together provide valuable insights into household energy consumption and related

It is important to understand the different implications of the magnitude of CO2 emissions attributable to households (~37 metric tons/household-yr), the percentage share of specific consumption categories (~38% for residential energy), and the above carbon intensities. The total annual carbon emission, which is the grand total derived by summing the products of the carbon intensities and expenditures for individual items as shown earlier in figure 3, is actually the ultimate measure for the carbon footprint of a given household. Energy efficiency and conservation measures are intended to reduce this total emission. It is now a common consensus among most proposals for climate legislation that US needs to reduce

or inadequately addressed in the *sectoral* and *end use* approaches.

**3.1 CO2 emissions by the average US household** 

embodied carbon emissions are shown in figure 5a and 5b.

**3.2 CO2 Intensities of consumption categories** 

conservation opportunities.

0.55 kg/\$ respectively; annual per-capita emission was 20 metric tons/person; and annual per-household emission was 51 metric tons/household.

Fig. 3. Process of calculating carbon intensities using economic input-output analysis and total household carbon emissions by summing the product of carbon intensity and consumer expenditure data for individual sectors across all *n* consumer items. (Shammin, 2009)

Fig. 4. GDP components, energy use and CO2 emissions for the US economy, 2003. (Based on data from EIA and BEA)

0.55 kg/\$ respectively; annual per-capita emission was 20 metric tons/person; and annual

Energy Intensity of goods & services

Annual expenditures on goods & services

Fig. 3. Process of calculating carbon intensities using economic input-output analysis and total household carbon emissions by summing the product of carbon intensity and consumer expenditure data for individual sectors across all *n* consumer items. (Shammin,

Fig. 4. GDP components, energy use and CO2 emissions for the US economy, 2003.

Carbon intensities from the fuel mix

Bureau of Economic Analysis (BEA) transportation and trade margins

*Ci*

Carbon Intensity of Item *i* [Tonnes/\$]

> Expenditure on Item *i* in \$

2003 US Economy 10.3 Trillion \$

Total US Energy Use (103.6 x 1015 kJ at a cost of 708 billion \$ )

5.69 x 109 metric tons of CO2

*Yi*

*i*

*n*

*i <sup>i</sup> YCC* 1

> Total Household Carbon Emissions [Tonnes]

per-household emission was 51 metric tons/household.

500 sector I/O data US Dept. of Commerce

I/O analysis using Leontief's method

Bureau of Labor Statistics (BLS) Consumer Expenditure Survey

> Sample households with yearly data

> > Personal Consumption Expenditure (PCE) 7.2 Trillion \$

Domestic Investment Government Expenditures Net Exports 3.1 Trillion \$

(Based on data from EIA and BEA)

2009)

The consumption based perspective on household emissions communicates a very different message to people. It shows that people have the power to directly and indirectly affect a large component of the nations' carbon emissions through behavior and lifestyle changes that would affect their consumption patterns and preferences. People also have the power to affect the remaining 29% of emissions; however, that would require engagement beyond personal choices. Through political activism and voting patterns, people in a democracy have the power to communicate how the government should spend their tax dollars. Regarding domestic investment, increasingly there are options being available to make investment choices on the basis of environmental performance. Finally, export-import policies can be reformed to trade with partners that are more environmentally responsible. None of this is easy; but at least this way of relating people and their behavior with national carbon emissions and mitigation opportunities provides a perspective that is either missing or inadequately addressed in the *sectoral* and *end use* approaches.

## **3.1 CO2 emissions by the average US household**

The most recent results to date for consumption-based carbon emissions for US households are reported in Shammin & Bullard (2009). The methods are based on calculations of energy intensities using input-output analysis described in Shammin et al (2010). These two papers also discuss the assumptions, nuances, and uncertainties associated with this approach. In 2003, the average household in the US spent about \$49,000 of which only 6.5% was spent on direct energy. The total embodied CO2 emission per household was about 37 metric tons of which about 65% was from direct energy. Thus, a small percent of household expenditure is actually responsible for the bulk of its carbon emissions. However, it is also interesting to see that the remainder of household expenditures made on non-energy goods and services were responsible for about 35% of the total embodied carbon emissions by the average household. This is a significant part of household emissions that is associated with life cycle emissions and the linkages between people and the sectors of the economy beyond residential and personal automobiles. The breakdown of total expenditure and total embodied carbon emissions are shown in figure 5a and 5b.

#### **3.2 CO2 Intensities of consumption categories**

Shammin & Bullard (2009) reports detailed carbon intensities for all personal consumption categories based on standard classification of the Bureau of Labor Statistics. A list of carbon intensities for major consumption categories is given in table 1. The distribution of carbon emissions between expenditure categories shown in figure 5b and the intensities in table 1 together provide valuable insights into household energy consumption and related conservation opportunities.

It is important to understand the different implications of the magnitude of CO2 emissions attributable to households (~37 metric tons/household-yr), the percentage share of specific consumption categories (~38% for residential energy), and the above carbon intensities. The total annual carbon emission, which is the grand total derived by summing the products of the carbon intensities and expenditures for individual items as shown earlier in figure 3, is actually the ultimate measure for the carbon footprint of a given household. Energy efficiency and conservation measures are intended to reduce this total emission. It is now a common consensus among most proposals for climate legislation that US needs to reduce

The Role of US Households in Global Carbon Emissions 179

CO2 emissions by 80% or more. This cannot be achieved without significant reductions at the household level. Share of emissions by specific consumption categories is also important because small percent reductions in large categories can result in a bigger difference than large percent reductions in small categories. Finally, carbon intensities indicate how carbon efficient different consumption categories are and provide an explicit basis for comparisons

Average CO2 Intensity (all categories) 0.80 Average CO2 Intensity of Direct Energy 7.53

Average CO2 Intensity of Indirect Energy 0.32

Table 1. CO2 intensities of major household consumption categories.

Natural gas 6.25 Electricity 8.02 Fuel oil and other fuels 8.07 Gasoline and motor oil 6.87

Housing 0.34 Owned dwellings 0.24 Telephone services 0.17 Water and other public services 0.59 Household operations 0.16 Housekeeping supplies 0.34 Household furnishings and equipment 0.33 Housing structure 0.80 Cars and trucks, new 0.46 Cars and trucks, used 0.50 Other vehicles 0.66 Vehicle finance charges 0.14 Maintenance and repairs 0.29 Vehicle insurance 0.07 Vehicle rental, leases, licenses, other charges 0.19 Public transportation 1.38 Food 0.41 Alcoholic beverages 0.33 Tobacco products and smoking supplies 0.13 Apparel, footwear and related services 0.43 Health care 0.14 Personal care products and services 0.27 Entertainment 0.22 Reading/education 0.21 Cash contributions 0.27 Personal insurance and pensions 0.11 Miscellaneous 0.28

**CO2 Intensity (kg/\$)** 

across categories.

Fig. 5a. Breakdown of annual expenditures for the average US household in 2003

Fig. 5b. Breakdown of annual CO2 emissions for the average US household in 2003

Food/Alc/Tobacco 12.6%

> Apparel 2.5%

> > Housing 31.0%

Food/Alc/Tobacco 6.9%

> Housing 7.6%

Residential Energy 37.9%

Apparel 1.5%

Residential Energy 3.5%

Auto Fuel 3.0%

1.1%

Auto Purch. Maint 13.9%

Fig. 5a. Breakdown of annual expenditures for the average US household in 2003

Fig. 5b. Breakdown of annual CO2 emissions for the average US household in 2003

5.1% Health Care

Asset Gain

Health Care 5.9%

Public Trans 0.9%

Auto Purch. Maint 6.7%

Public Trans 1.6%

Other 19.8% Asset Gain 6.8%

Auto Fuel 26.9%

Other 4.9%

CO2 emissions by 80% or more. This cannot be achieved without significant reductions at the household level. Share of emissions by specific consumption categories is also important because small percent reductions in large categories can result in a bigger difference than large percent reductions in small categories. Finally, carbon intensities indicate how carbon efficient different consumption categories are and provide an explicit basis for comparisons across categories.


Table 1. CO2 intensities of major household consumption categories.

The Role of US Households in Global Carbon Emissions 181

and housing type (about 44% more for single-family homes compared to apartments). These differences, reported in Shammin et al (2010), are estimated for households with the same income having different demographic and lifestyle configurations. While these differences are for total energy consumption, they would yield very similar differences in total

Fig. 7. Price increases for goods and services consumed by households as percent of annual household income due to different prices of carbon emissions. Here, emission data is for

Consumption based approach to household carbon emissions puts households and the people living in them front and center in exploring ways of reducing national carbon emissions. This requires households to develop a comprehensive strategy for carbon emission reductions that involves specific actions to address the various sources of emissions: direct emissions, indirect emissions, and emissions related to government expenditure, investment, and net exports. A pre-requisite for this is the motivation and willingness by members of any given household to undertake the solutions that apply to them. Literature in environmental psychology has several studies that investigate ways of motivating people to change behavior and adopt conservation and efficiency measures to reduce their carbon footprint (Nolan et al, 2008; Parnell & Larsen, 2005). Examples of successful interventions from some recent research include: innovative ways of providing real time feedback on energy and resource use (Petersen et al 2007; Petersen & Frantz, 2009) and offering financial incentives for reducing resource consumption (Suter & Shammin, 2010). Once households are committed to reduce emissions, they have to balance several

carbon and not CO2 and tonne = metric tons. (Shammin & Bullard, 2009)

**4. The role of households in reducing CO2 emissions** 

household carbon emissions as well.

#### **3.3 Predictors of household CO2 emissions**

Energy consumption and related carbon emissions vary across households depending on several key demographic and lifestyle related factors. Household income is the most influential predictor of total household emissions and how those emissions are distributed across various consumption categories. In general, the relationship between income and carbon emissions is non-linear (figure 6) – resulting in regressive impacts on low income households (Shammin & Bullard, 2009).

Fig. 6. Relationship between annual household income and carbon emissions. (Shammin & Bullard, 2009)

There is a large difference in the total annual carbon emissions and the share of indirect carbon emissions between households belonging to the lowest and highest income quintiles. In 2003, the total CO2 emission of the highest quintile (~68 metric tons) was four times higher than that of the lowest quintile (~17 metric tons). At the same time, the share of indirect emission was close to 50% for the highest quintile as opposed to less than 20% for the lowest quintile. The latter has important implications: if price of direct energy resources go up as a result of climate change legislations, the effect on low income households will be disproportionately higher than high income households. Figure 6 shows that if a new capand-trade or carbon tax policy puts a price on carbon emissions at \$100/metric ton, the impact on low income households can be more than 4% of their income as opposed to less than 2% for high income households (Shammin & Bullard, 2009).

Another important predictor of household emissions is the location of residence. Households in rural areas consumed about 17% higher total energy compared to households of the same income level residing in urban locations. Bigger homes, longer commutes, greater use of outdoor power equipment, etc. are typically responsible for this difference. Other predictors that affect total household energy consumption include: family size (about 28% more for a family of 4 compared to single-occupant households), number of cars (about 27% more for a household with two cars compared to households with no cars),

Energy consumption and related carbon emissions vary across households depending on several key demographic and lifestyle related factors. Household income is the most influential predictor of total household emissions and how those emissions are distributed across various consumption categories. In general, the relationship between income and carbon emissions is non-linear (figure 6) – resulting in regressive impacts on low income

Fig. 6. Relationship between annual household income and carbon emissions. (Shammin &

There is a large difference in the total annual carbon emissions and the share of indirect carbon emissions between households belonging to the lowest and highest income quintiles. In 2003, the total CO2 emission of the highest quintile (~68 metric tons) was four times higher than that of the lowest quintile (~17 metric tons). At the same time, the share of indirect emission was close to 50% for the highest quintile as opposed to less than 20% for the lowest quintile. The latter has important implications: if price of direct energy resources go up as a result of climate change legislations, the effect on low income households will be disproportionately higher than high income households. Figure 6 shows that if a new capand-trade or carbon tax policy puts a price on carbon emissions at \$100/metric ton, the impact on low income households can be more than 4% of their income as opposed to less

Another important predictor of household emissions is the location of residence. Households in rural areas consumed about 17% higher total energy compared to households of the same income level residing in urban locations. Bigger homes, longer commutes, greater use of outdoor power equipment, etc. are typically responsible for this difference. Other predictors that affect total household energy consumption include: family size (about 28% more for a family of 4 compared to single-occupant households), number of cars (about 27% more for a household with two cars compared to households with no cars),

than 2% for high income households (Shammin & Bullard, 2009).

**3.3 Predictors of household CO2 emissions** 

households (Shammin & Bullard, 2009).

Bullard, 2009)

and housing type (about 44% more for single-family homes compared to apartments). These differences, reported in Shammin et al (2010), are estimated for households with the same income having different demographic and lifestyle configurations. While these differences are for total energy consumption, they would yield very similar differences in total household carbon emissions as well.

Fig. 7. Price increases for goods and services consumed by households as percent of annual household income due to different prices of carbon emissions. Here, emission data is for carbon and not CO2 and tonne = metric tons. (Shammin & Bullard, 2009)

## **4. The role of households in reducing CO2 emissions**

Consumption based approach to household carbon emissions puts households and the people living in them front and center in exploring ways of reducing national carbon emissions. This requires households to develop a comprehensive strategy for carbon emission reductions that involves specific actions to address the various sources of emissions: direct emissions, indirect emissions, and emissions related to government expenditure, investment, and net exports. A pre-requisite for this is the motivation and willingness by members of any given household to undertake the solutions that apply to them. Literature in environmental psychology has several studies that investigate ways of motivating people to change behavior and adopt conservation and efficiency measures to reduce their carbon footprint (Nolan et al, 2008; Parnell & Larsen, 2005). Examples of successful interventions from some recent research include: innovative ways of providing real time feedback on energy and resource use (Petersen et al 2007; Petersen & Frantz, 2009) and offering financial incentives for reducing resource consumption (Suter & Shammin, 2010). Once households are committed to reduce emissions, they have to balance several

The Role of US Households in Global Carbon Emissions 183

reasonably achievable household emissions reduction in the US can be approximately 20%

In addition to the direct emissions, consumption based approach also allows for households to understand, estimate, monitor and reduce indirect carbon emissions. US society has been on a treadmill of consumption for several decades where more consumption is considered a desirable goal. The core message in this approach is that consumption of non-energy goods and services has associated life cycle carbon emissions and thus reducing consumption would help reduce carbon emissions. As shown in table 1, there is very little difference in the carbon intensity of the various consumption categories responsible for indirect carbon emissions. Any reduction in consumption, irrespective of which items are avoided, would yield similar reductions in a household's carbon footprint. Notable exceptions are water and public transportation. Another related issue is rebound effects. If money saved from the reduction of direct energy is re-spent on other goods and services, part of the conservation savings would be offset. For example, if a household saves \$1,500 by conserving direct energy consumption, they would reduce their CO2 emissions by about 11 metric tons. If that money is re-spent on other goods and services, that would generate about 0.5 metric tons of additional CO2 – resulting in a net savings of 10.5 metric tons. For a single household this may appear to be a small effect, but added across the economy this addition amounts to more than 50 million metric tons. This effect can become much larger if this money is re-spent on more carbon

Influencing the components of GDP beyond personal consumption (see figure 4) requires a different approach – since these involve decision making entities that are exogenous to individual households. A democratic society has avenues for people to influence decisions made by the government about how public tax dollars should be spent – through voting patterns, writing letters to representatives, and other types of civic engagement and political activism. If government expenditure on building roads is shifted towards the development of high speed rail or government subsidy to fossil fuels is shifted towards new incentives for renewable energy projects, significant reductions in carbon emissions can be achieved in the government expenditure component. It is also important to note that reductions of direct and indirect carbon emissions by households would not change the carbon intensity of the underlying infrastructure such as the source mix of power generation or transportation driven by internal combustion engines. Reduction in household electricity use would only go so far if more than 80% of the electricity is generated from coal (which is the case in many US states such as Ohio). Through activism and engagement, people have the opportunity to influence a shift from carbon intensive fossil fuel based sources to carbon-neutral or carbonfree renewable sources. This will have a large impact on economy-wide reductions in carbon

In terms of the investment component of GDP, many investment portfolios now make information on environmental performance or carbon offsets available to investors. If more and more people invest in these green stocks, the carbon footprint of investment can go down. Perhaps the most complicated component of GDP for people to influence is net exports – as this involves carbon emissions associated with industries and commercial ventures in other countries. Reforms in trade policies can allow more partnerships with countries, industries and multi-national companies that promote climate friendly operations. If policy is ultimately reflective of the will of the people, then households can

within 10 years if the most effective non-regulatory interventions are used.

intensive choices such as flying to far-away places for family vacations.

emissions.

different approaches to address their total emissions. Throughout this process, households need to monitor their carbon footprint and track their progress in order to achieve most effective results. People are used to budget their income and expenses on a regular basis. Current challenges of climate change require people to go beyond financial budgeting and begin to develop methods to monitor their energy use and carbon emissions. The paradigm shift and associated challenges required for this to become mainstream involve discussions that are beyond the scope of this chapter.

The *end use* approach mentioned earlier provides the basis for reductions of direct emissions for households within the scope of the residential sector and personal transportation subsector. These mostly include direct conservation and efficiency measures that households can take.

*Conservation measures:* these involve reducing carbon intensive behaviors or replacing carbon intensive behaviors with emission-free options. Examples include:


*Efficiency measures:* these involve replacing less efficient equipment with more efficient ones to achieve the same task. Examples include:


The interplay between conservation and efficiency measures is also important to understand. Here, the ultimate goal is to reduce carbon emissions. However, overemphasis on efficiency measures may lead to *Jevon's Paradox* (people replacing inefficient cars with efficient ones and then driving more miles than before to offset or overshoot energy/carbon savings). On the contrary, when conservation and efficiency measures are coupled, households will be able to maximize their emission reductions. Dietz et al (2009) shows that

different approaches to address their total emissions. Throughout this process, households need to monitor their carbon footprint and track their progress in order to achieve most effective results. People are used to budget their income and expenses on a regular basis. Current challenges of climate change require people to go beyond financial budgeting and begin to develop methods to monitor their energy use and carbon emissions. The paradigm shift and associated challenges required for this to become mainstream involve discussions

The *end use* approach mentioned earlier provides the basis for reductions of direct emissions for households within the scope of the residential sector and personal transportation subsector. These mostly include direct conservation and efficiency measures that households

*Conservation measures:* these involve reducing carbon intensive behaviors or replacing carbon

d. making weather appropriate clothing choices indoors and airflow/shade management

*Efficiency measures:* these involve replacing less efficient equipment with more efficient ones

d. replacing incandescent light bulbs with compact fluorescent or light emitting diode

e. taking inventory of household electronics, eliminate unnecessary ones, and using

The interplay between conservation and efficiency measures is also important to understand. Here, the ultimate goal is to reduce carbon emissions. However, overemphasis on efficiency measures may lead to *Jevon's Paradox* (people replacing inefficient cars with efficient ones and then driving more miles than before to offset or overshoot energy/carbon savings). On the contrary, when conservation and efficiency measures are coupled, households will be able to maximize their emission reductions. Dietz et al (2009) shows that

h. regular maintenance of home appliances, HVAC equipment, and automobiles i. installing renewable energy systems in homes such as solar, wind, geothermal, etc. j. building homes that are designed to maximize the use of passive solar energy

intensive behaviors with emission-free options. Examples include:

b. reducing number of daily trips using motorized transportation

e. lowering thermostat setting in winter and raising it in summer

power strips and on/off switches to avoid phantom loads f. using public transportation instead of personal automobile g. replacing fuel-inefficient vehicles with more fuel-efficient vehicles

a. walking/biking instead of using automobiles

c. using clotheslines instead of dryers

instead of using air conditioning

f. lowering water heater temperature g. choosing to live in a smaller home h. choosing to drive a smaller automobile

to achieve the same task. Examples include:

c. using more efficient water heaters

light bulbs

a. upgrading inefficient HVAC systems and appliances b. improving insulation and reducing leakage in homes

that are beyond the scope of this chapter.

can take.

reasonably achievable household emissions reduction in the US can be approximately 20% within 10 years if the most effective non-regulatory interventions are used.

In addition to the direct emissions, consumption based approach also allows for households to understand, estimate, monitor and reduce indirect carbon emissions. US society has been on a treadmill of consumption for several decades where more consumption is considered a desirable goal. The core message in this approach is that consumption of non-energy goods and services has associated life cycle carbon emissions and thus reducing consumption would help reduce carbon emissions. As shown in table 1, there is very little difference in the carbon intensity of the various consumption categories responsible for indirect carbon emissions. Any reduction in consumption, irrespective of which items are avoided, would yield similar reductions in a household's carbon footprint. Notable exceptions are water and public transportation. Another related issue is rebound effects. If money saved from the reduction of direct energy is re-spent on other goods and services, part of the conservation savings would be offset. For example, if a household saves \$1,500 by conserving direct energy consumption, they would reduce their CO2 emissions by about 11 metric tons. If that money is re-spent on other goods and services, that would generate about 0.5 metric tons of additional CO2 – resulting in a net savings of 10.5 metric tons. For a single household this may appear to be a small effect, but added across the economy this addition amounts to more than 50 million metric tons. This effect can become much larger if this money is re-spent on more carbon intensive choices such as flying to far-away places for family vacations.

Influencing the components of GDP beyond personal consumption (see figure 4) requires a different approach – since these involve decision making entities that are exogenous to individual households. A democratic society has avenues for people to influence decisions made by the government about how public tax dollars should be spent – through voting patterns, writing letters to representatives, and other types of civic engagement and political activism. If government expenditure on building roads is shifted towards the development of high speed rail or government subsidy to fossil fuels is shifted towards new incentives for renewable energy projects, significant reductions in carbon emissions can be achieved in the government expenditure component. It is also important to note that reductions of direct and indirect carbon emissions by households would not change the carbon intensity of the underlying infrastructure such as the source mix of power generation or transportation driven by internal combustion engines. Reduction in household electricity use would only go so far if more than 80% of the electricity is generated from coal (which is the case in many US states such as Ohio). Through activism and engagement, people have the opportunity to influence a shift from carbon intensive fossil fuel based sources to carbon-neutral or carbonfree renewable sources. This will have a large impact on economy-wide reductions in carbon emissions.

In terms of the investment component of GDP, many investment portfolios now make information on environmental performance or carbon offsets available to investors. If more and more people invest in these green stocks, the carbon footprint of investment can go down. Perhaps the most complicated component of GDP for people to influence is net exports – as this involves carbon emissions associated with industries and commercial ventures in other countries. Reforms in trade policies can allow more partnerships with countries, industries and multi-national companies that promote climate friendly operations. If policy is ultimately reflective of the will of the people, then households can

The Role of US Households in Global Carbon Emissions 185

arguments presented in this paper. Special thanks go to Robert A. Herendeen at the University of Vermont and Clark W. Bullard at the University of Illinois for their

Bin, S. & Dowlatabadi, H. (2005). Consumer lifestyle approach to U.S. energy use and the

Bullard, C.W. & Herendeen, R.A. (1975). The energy cost of goods and services. *Energy* 

Dietz, T., Gardner, G., Gilligan,, J., Stern, P. & Vandenbergh, M. (2009). Household actions

EIA (2011). Annual Energy Outlook 2011. *US Energy Information Agency*. Washington D.C. Herendeen, R.A. (1978). Total energy cost of household consumption in Norway, 1973.

Hoekstra, R. (2010). Towards a complete database of peer reviewed articles on

Nolan, J., Schultz, P., Cialdini, R., Goldstein, N. & Griskevicius, V. (2008). Normative social

Parnell, R. & Larsen, O. P. (2005). Informing the Development of Domestic Energy Efficiency

Petersen, J.E. & Frantz, C. (2009). Employing multiple modes and scales of real-time

Petersen, J. E., Shunturov, V., Janda, K., Platt, G. & Weinberger, K. (2007). Dormitory

Shammin, M., 2009. Embodied Carbon Emissions by US Households: Missing emissions and

Shammin, M., Herendeen, R. A., Hanson, M. & Wilson, E. (2010). A Multivariate Analysis of

environmentally extended input-out analysis. Paper presented at the 18th

influence is underdetected. *Personality and Social Psychology*, Bulletin 34, pp. 913 –

Initiatives: An Everyday Householder-Centered Framework. *Environment and* 

feedback to engage, educate, motivate and empower electricity and water conservation. Paper presented at the *Behavior, Energy and Climate Change (BECC)*

residents reduce electricity consumption when exposed to real-time visual feedback and incentives. *International Journal of Sustainability in Higher Education,*

equity impacts of calculators used in carbon offsets. Paper presented at the 5th Bi‐ Annual Meeting of the US Society for Ecological Economics: Science and Policy for

the Energy Intensity of Sprawl versus Compact Living in the US for 2003. *Ecological* 

Herendeen, R.A. & Tanaka, J. (1976). Energy cost of living. *Energy*, Vol. 1, pp. 163–178. Herendeen, R.A., Ford, C., & Hannon, B. (1981). Energy cost of living, 1972–73. *Energy*, Vol.

*International Input-output Conference*, June 20 – 25, Sydney, Australia Leontief, W. (1970). Environmental Repercussions and the Economic Structure: An Input-Output Approach. *Review of Economic Statistics*, Vol. 52, pp. 262-277

can provide a behavioral wedge to rapidly reduce US carbon emissions. *Proceedings* 

mentorship and collaboration on the research that has made this chapter possible.

related CO2 emissions. *Energy Policy,* Vol. 33, pp. 197–208

*of the National Academy of Sciences*, Vol. 44, pp. 18452–18456

**7. References** 

*Policy,* Vol. 3, pp. 268–278

*Energy*, Vol. 4, pp. 615–630

*Behavior,* Vol. 37, No. 6, pp. 787-807

Vol. 8, No. 1, pp. 16-33

meeting, November 15-18, Washington D.C.

*Economics*, Vol. 69, Issue 12, pp.2363–2373

a Sustainable Future, May 31 – June 3, Washington DC.

6, pp. 1433–1450

923.

play a role in paving the way for such transitions from carbon intensive to a low carbon or carbon-free economy.

Finally, reducing carbon emissions is not necessarily about compromises and sacrifices. There are multiple benefits of low carbon lifestyle in a low carbon economy for households. First, humanity is currently threatened by the grim prospect of catastrophic consequences unless human-induced climate change is slowed, halted or reversed. While there are theories about winners and losers in a post climate change world, in reality everyone is at risk as the global economy is now more interconnected than ever before. We have already seen how crisis in East Asian markets had ripple effects throughout the world and how economic downturn in the US is affecting other countries. Locally, households, communities and regions with low carbon footprint will be more resilient against increased prices of carbon intensive energy resources and consequent increases in the price of goods and services. Thus, a more comprehensive and aggressive strategy for reducing carbon emissions by households, particularly in a carbon intensive nation such as the US, makes sense on many levels: for the sustainability of human race, for a healthy environment for future generations, for economic stability, for social security, and for the development of more engaged and resilient communities.

## **5. Conclusion**

The daunting task of combating climate change is a defining challenge of the present generation. Reducing global carbon emissions is the most important aspect of that challenge. The US is a major player in global climate mitigation initiatives – since it is responsible for more than 20% of global carbon emissions. While the residential sector in the US accounts for about 20% of US emissions, this chapter demonstrates that households can directly and indirectly play a very important role in reducing all of the nations' carbon emissions. They have direct control over about 46% of embodied carbon emissions in the US by managing their consumption of energy resources and indirect control over another 25% of emissions by managing their consumption patterns. They can also play a role in influencing the remaining 29% by promoting and/or supporting initiatives to reduce the carbon footprint of energy systems, government expenditures, investment portfolios, and even businesses and industries beyond US borders. They can do this by becoming more engaged citizens and exercising their democratic privileges. The *sectoral* and *end use* approaches used to represent household emissions in the US are important, but limited in terms of helping people fully understand how their lives are connected to all sectors of the economy. The consumption based approach presented in this chapter constitute a more comprehensive accounting of household emissions as it includes embodied carbon over the life cycle of products and services that people consume to support their lifestyle. Most importantly, this approach offers people much more direct ways of relating personal choices with large scale reductions of national carbon emissions. This is a perspective that has the potential to empower people to become proactive agents of change and provide more explicit tools to make a difference in the battle against climate change.

## **6. Acknowledgment**

I would like to acknowledge my co-authors of the two papers (Shammin & Bullard, 2009; Shammin et al, 2010) that formulate the conceptual and analytical basis for the main arguments presented in this paper. Special thanks go to Robert A. Herendeen at the University of Vermont and Clark W. Bullard at the University of Illinois for their mentorship and collaboration on the research that has made this chapter possible.

## **7. References**

184 Greenhouse Gases – Emission, Measurement and Management

play a role in paving the way for such transitions from carbon intensive to a low carbon or

Finally, reducing carbon emissions is not necessarily about compromises and sacrifices. There are multiple benefits of low carbon lifestyle in a low carbon economy for households. First, humanity is currently threatened by the grim prospect of catastrophic consequences unless human-induced climate change is slowed, halted or reversed. While there are theories about winners and losers in a post climate change world, in reality everyone is at risk as the global economy is now more interconnected than ever before. We have already seen how crisis in East Asian markets had ripple effects throughout the world and how economic downturn in the US is affecting other countries. Locally, households, communities and regions with low carbon footprint will be more resilient against increased prices of carbon intensive energy resources and consequent increases in the price of goods and services. Thus, a more comprehensive and aggressive strategy for reducing carbon emissions by households, particularly in a carbon intensive nation such as the US, makes sense on many levels: for the sustainability of human race, for a healthy environment for future generations, for economic stability, for social security, and for the development of

The daunting task of combating climate change is a defining challenge of the present generation. Reducing global carbon emissions is the most important aspect of that challenge. The US is a major player in global climate mitigation initiatives – since it is responsible for more than 20% of global carbon emissions. While the residential sector in the US accounts for about 20% of US emissions, this chapter demonstrates that households can directly and indirectly play a very important role in reducing all of the nations' carbon emissions. They have direct control over about 46% of embodied carbon emissions in the US by managing their consumption of energy resources and indirect control over another 25% of emissions by managing their consumption patterns. They can also play a role in influencing the remaining 29% by promoting and/or supporting initiatives to reduce the carbon footprint of energy systems, government expenditures, investment portfolios, and even businesses and industries beyond US borders. They can do this by becoming more engaged citizens and exercising their democratic privileges. The *sectoral* and *end use* approaches used to represent household emissions in the US are important, but limited in terms of helping people fully understand how their lives are connected to all sectors of the economy. The consumption based approach presented in this chapter constitute a more comprehensive accounting of household emissions as it includes embodied carbon over the life cycle of products and services that people consume to support their lifestyle. Most importantly, this approach offers people much more direct ways of relating personal choices with large scale reductions of national carbon emissions. This is a perspective that has the potential to empower people to become proactive agents of change and provide more explicit tools to make a difference

I would like to acknowledge my co-authors of the two papers (Shammin & Bullard, 2009; Shammin et al, 2010) that formulate the conceptual and analytical basis for the main

carbon-free economy.

**5. Conclusion** 

more engaged and resilient communities.

in the battle against climate change.

**6. Acknowledgment** 


**9**

*Ukraine* 

**The Uncertainty Estimation and Use of** 

*1State Enterprise "All-Ukrainian State Scientific and Production Centre for Standardization, Metrology, Certification and Protection of Consumer"* 

*2State Enterprise "Ukrainian Scientific-Research and Educational Centre of Standardization, Certification and Problems of Quality" (SE "UkrSREC")* 

A global policy for environmental protection is needed and is under discussion. One of the effects of modern global policy for environmental protection includes the multitude of measurements that are part of the process of environment protection, including the estimation of greenhouse gas (GHG) emission. Global climate studies bring together an enormous range of sciences and for a sound model to be developed it is necessary that the data from all these areas be comparable. The only way for this to be assured is for measurements in all areas of science to be made in terms of a well-defined system of units,

Human activities to have major impacts on the global climate change which caused by an increase of GHG in the atmosphere. In general, there is now a demand for people to have confidence in the credibility of the results of measurements because in so many ways decisions based on the data that come from measurements are increasingly seen to have a direct influence on the economy, human health and safety, and welfare. The United Nations (UN) and its member states adopted the UN Framework Convention on Climate Change (UNFCCC). Parties of the UNFCCC must estimate GHG anthropogenic emissions and to

The governing bodies of the World Meteorological Organization (WMO) and of the UN Environment Programme (UNEP) created a body, the Intergovernmental Panel on Climate Change (IPCC), to marshal and assess scientific information on the subject. For monitoring of global climate change and providing reliable data for climate modelling, a Global Atmospheric Watch (GAW) programme has started by the WMO. The UNFCCC is also starting probably the largest environmental monitoring programme in the world. Parties of UNFCCC can estimate GHG emissions in using two general approaches: direct

measurement or proxy data (Velychko O. & Gordiyenko T., 2007a, 2011).

**1. Introduction** 

namely the International System of Units (SI).

develop annual national GHG inventories.

Oleh Velychko1 and Tetyana Gordiyenko2

*(SE "Ukrmetrteststandard")* 

**Measurement Units in National Inventories of Anthropogenic Emission of Greenhouse Gas** 


*Ukraine* 

## **The Uncertainty Estimation and Use of Measurement Units in National Inventories of Anthropogenic Emission of Greenhouse Gas**

Oleh Velychko1 and Tetyana Gordiyenko2

*1State Enterprise "All-Ukrainian State Scientific and Production Centre for Standardization, Metrology, Certification and Protection of Consumer" (SE "Ukrmetrteststandard") 2State Enterprise "Ukrainian Scientific-Research and Educational Centre of Standardization, Certification and Problems of Quality" (SE "UkrSREC")* 

## **1. Introduction**

186 Greenhouse Gases – Emission, Measurement and Management

Shammin, M. & Bullard, C. (2009). Impact of Cap-and-trade Policies for Reducing

Suter, J and Shammin, M. 2010. Estimating Payback to Residential Energy Efficiency

Annual Meeting, Denver, Colorado, July 25-27, 2010.

2432–2438.

Greenhouse Gas Emissions on U.S. Households. *Ecological Economics*, Vol. 68, pp.

Measures: A Field Experiment. Selected paper prepared for presentation at the Agricultural & Applied Economics Association 2010 AAEA, CAES, & WAEA Joint

> A global policy for environmental protection is needed and is under discussion. One of the effects of modern global policy for environmental protection includes the multitude of measurements that are part of the process of environment protection, including the estimation of greenhouse gas (GHG) emission. Global climate studies bring together an enormous range of sciences and for a sound model to be developed it is necessary that the data from all these areas be comparable. The only way for this to be assured is for measurements in all areas of science to be made in terms of a well-defined system of units, namely the International System of Units (SI).

> Human activities to have major impacts on the global climate change which caused by an increase of GHG in the atmosphere. In general, there is now a demand for people to have confidence in the credibility of the results of measurements because in so many ways decisions based on the data that come from measurements are increasingly seen to have a direct influence on the economy, human health and safety, and welfare. The United Nations (UN) and its member states adopted the UN Framework Convention on Climate Change (UNFCCC). Parties of the UNFCCC must estimate GHG anthropogenic emissions and to develop annual national GHG inventories.

> The governing bodies of the World Meteorological Organization (WMO) and of the UN Environment Programme (UNEP) created a body, the Intergovernmental Panel on Climate Change (IPCC), to marshal and assess scientific information on the subject. For monitoring of global climate change and providing reliable data for climate modelling, a Global Atmospheric Watch (GAW) programme has started by the WMO. The UNFCCC is also starting probably the largest environmental monitoring programme in the world. Parties of UNFCCC can estimate GHG emissions in using two general approaches: direct measurement or proxy data (Velychko O. & Gordiyenko T., 2007a, 2011).

The Uncertainty Estimation and Use of Measurement Units

national inventories.

Fig. 1. The TACCC criteria

terms are provided in Table 1.

(ISO 5725-1, ISO 3534-1).

closeness of agreement between

closeness of agreement between a measured quantity value and a true quantity value of a measurand (VIM

closeness of agreement between a test result and the accepted reference value

indications or measured quantity values obtained by replicate measurements on the same or similar objects under specified conditions (VIM 2007); the closeness of agreement between independent test results obtained under stipulated conditions (ISO 5725-1, ISO

A*ccuracy*:

2007);

*Precision*:

3534-1)

in National Inventories of Anthropogenic Emission of Greenhouse Gas 189

*transparency* (disclosing sufficient and appropriate GHG-related information to allow intended users to make decisions with reasonable confidence); *accuracy* (reducing bias and uncertainties as far as is practical); *completeness* (including all relevant GHG emissions and removals); *consistency* (enabling meaningful comparisons in GHG-related information) and *comparability* (estimates of emissions and removals reported by countries in inventories should be comparable among countries) (ISO 14064-1…3, IPCC 2006). For this purpose, countries should use agreed methodologies and formats for estimating and reporting

*Accuracy* 

Completeness Comparability

Definitions associated with conducting an uncertainty analysis include *accuracy*, *precision*, *uncertainty*, and *error* are described in GPG 2000, IPCC 2006, ISO 14064-1…3, VIM 2007, ISO 5725-1, ISO 3534-1 (Velychko O. & Gordiyenko T., 2005, 2007a, 2007b). Comparison of same metrological (and some statistical) and environmental guides and international standards

*Accuracy*:

*Precision*:

Transparency

a general term which describes the degree to which an estimate of a quantity is unaffected by bias due to systematic error (GPG 2000); a relative measure of the exactness of an emission or removal estimate (IPCC 2006); reducing bias and uncertainties as far as is

the inverse of uncertainty in the sense that the more precise something is, the less

closeness of agreement between independent results of measurements obtained under stipulated conditions (IPCC 2006).

practical (ISO 14064-1…3).

uncertain it is (GPG 2000);

Consistency TACCC

**Metrological terms Environmental terms** 

Today's global economy depends on reliable measurements and tests, which are trusted and accepted internationally. Metrology is the scientific study of measurement. Measurements have always been essential in supporting international trade and regulation. Metrology delivers the basis for the comparability of test results, e. g. by defining the units of the measurement and by providing traceability and associated uncertainty of the measurement results. Measurement results may be used provided that the corresponding characteristics of measurement uncertainty are known.

The tasks of the Joint Committee for Guides in Metrology (JCGM) are to maintain and promote the use of the Guide to the Expression of Uncertainty in Measurement (known as the GUM) and the International Vocabulary of Metrology (known as the VIM). The JCGM has taken over responsibility for these two documents, who originally published them under the auspices of the International Bureau of Weights and Measures (BIPM), the International Organization of Legal Metrology (OIML), the International Organization for Standardization (ISO), the International Electrotechnical Commission (IEC), the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC), the International Union of Pure and Applied Chemistry (IUPAC), the International Union of Pure and Applied Physics (IUPAP), the International Laboratory Accreditation Cooperation (ILAC).

The General Conference on Weights and Measures (CGPM) adopted SI, for the recommended practical system of units of measurement. The nearly universal use of the SI has brought coherence to all scientific and technological measurements, a worldwide consensus on the evaluation and expression of uncertainty in measurement would permit the significance of a vast spectrum of measurement results in science, engineering, commerce, industry, and regulation to be readily understood and properly interpreted. Many of the quantities, their recommended names and symbols, and the equations relating them, are listed in the international standards ISO/IEC 80000, in which it is proposed that the quantities and equations used with the SI. The IUPAP recognizes the SI for expressing the quantitative results of measurements in physics. The IUPAC serves to advance the worldwide aspects of the chemical sciences and to contribute to the application of chemistry in the service of Mankind (Velychko O. & Gordiyenko T., 2007a, 2010).

Some metrological terms are used in special guides of the IPCC, which to use for preparation of national inventories of GHG. Therefore it is important to compare uncertainty estimation with international ecological and metrological guides, and to consider peculiarities of their using also. It is also important to consider peculiarities of SI units used in those ecological guides.

## **2. The use metrological terms in international environmental guides**

All branches of science and technology need to choose their vocabulary with care. Each term must have the same meaning for all of its users. In order to try and resolve this problem in field of metrology at an international level, eight international organizations developed VIM. The IPCC and the UNFCCC resolved this problem in environmental field.

Throughout the review process, it is asked to assess the quality of the each Party's UNFCCC national inventory submission, with quality being determined by criteria (TACCC – Fig. 1): *transparency* (disclosing sufficient and appropriate GHG-related information to allow intended users to make decisions with reasonable confidence); *accuracy* (reducing bias and uncertainties as far as is practical); *completeness* (including all relevant GHG emissions and removals); *consistency* (enabling meaningful comparisons in GHG-related information) and *comparability* (estimates of emissions and removals reported by countries in inventories should be comparable among countries) (ISO 14064-1…3, IPCC 2006). For this purpose, countries should use agreed methodologies and formats for estimating and reporting national inventories.

#### Fig. 1. The TACCC criteria

188 Greenhouse Gases – Emission, Measurement and Management

Today's global economy depends on reliable measurements and tests, which are trusted and accepted internationally. Metrology is the scientific study of measurement. Measurements have always been essential in supporting international trade and regulation. Metrology delivers the basis for the comparability of test results, e. g. by defining the units of the measurement and by providing traceability and associated uncertainty of the measurement results. Measurement results may be used provided that the corresponding characteristics of

The tasks of the Joint Committee for Guides in Metrology (JCGM) are to maintain and promote the use of the Guide to the Expression of Uncertainty in Measurement (known as the GUM) and the International Vocabulary of Metrology (known as the VIM). The JCGM has taken over responsibility for these two documents, who originally published them under the auspices of the International Bureau of Weights and Measures (BIPM), the International Organization of Legal Metrology (OIML), the International Organization for Standardization (ISO), the International Electrotechnical Commission (IEC), the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC), the International Union of Pure and Applied Chemistry (IUPAC), the International Union of Pure and Applied Physics (IUPAP), the International Laboratory Accreditation Cooperation

The General Conference on Weights and Measures (CGPM) adopted SI, for the recommended practical system of units of measurement. The nearly universal use of the SI has brought coherence to all scientific and technological measurements, a worldwide consensus on the evaluation and expression of uncertainty in measurement would permit the significance of a vast spectrum of measurement results in science, engineering, commerce, industry, and regulation to be readily understood and properly interpreted. Many of the quantities, their recommended names and symbols, and the equations relating them, are listed in the international standards ISO/IEC 80000, in which it is proposed that the quantities and equations used with the SI. The IUPAP recognizes the SI for expressing the quantitative results of measurements in physics. The IUPAC serves to advance the worldwide aspects of the chemical sciences and to contribute to the application of chemistry

Some metrological terms are used in special guides of the IPCC, which to use for preparation of national inventories of GHG. Therefore it is important to compare uncertainty estimation with international ecological and metrological guides, and to consider peculiarities of their using also. It is also important to consider peculiarities of SI

All branches of science and technology need to choose their vocabulary with care. Each term must have the same meaning for all of its users. In order to try and resolve this problem in field of metrology at an international level, eight international organizations developed

Throughout the review process, it is asked to assess the quality of the each Party's UNFCCC national inventory submission, with quality being determined by criteria (TACCC – Fig. 1):

**2. The use metrological terms in international environmental guides** 

VIM. The IPCC and the UNFCCC resolved this problem in environmental field.

in the service of Mankind (Velychko O. & Gordiyenko T., 2007a, 2010).

measurement uncertainty are known.

units used in those ecological guides.

(ILAC).

Definitions associated with conducting an uncertainty analysis include *accuracy*, *precision*, *uncertainty*, and *error* are described in GPG 2000, IPCC 2006, ISO 14064-1…3, VIM 2007, ISO 5725-1, ISO 3534-1 (Velychko O. & Gordiyenko T., 2005, 2007a, 2007b). Comparison of same metrological (and some statistical) and environmental guides and international standards terms are provided in Table 1.


The Uncertainty Estimation and Use of Measurement Units

interval containing the set of true quantity values of a measurand with a stated probability, based on the information

means of the mutual dependence of two

the relationship between two or several random variables within a distribution of two or more random variables (ISO

dependence of two variables, equal to the ratio of their covariances to the positive square root of the product of their

PDF

Fig. 2. Relationship between accuracy and precision

random variables (GUM 1993).

measure of the relative mutual

*Coverage interval*:

available (VIM 2007).

*Correlation coefficient*:

variances (GUM 1993).

*Covariance*:

*Correlation*:

3534-1).

**Metrological terms Environmental terms** 

in National Inventories of Anthropogenic Emission of Greenhouse Gas 191

*Confidence interval*:

(IPCC 2006).

two variables (GPG 2000).

(GPG 2000, IPCC 2006).

*Correlation coefficient*:

(GPG 2000, IPCC 2006).

Value RV

*Precision*

The *accuracy* and *precision* of individual measurements will depend upon the equipment and protocols used to make the measurements. A measurement system (equipment) is designated valid if it is both *accurate* and *precise*. The relationship between accuracy and precision is shown in Fig. 2 (PDF is probability density function; RV is reference value).

The concept "*measurement accuracy*" is not a quantity and is not given a numerical quantity value (VIM 2007). The term *accuracy*, when applied to a set of test results, involves a

*Covariance*:

*Correlation*:

Table 1. Metrological and environmental guides and international standard terms

*Accuracy* 

the range in which it is believed that the true

the true value of the quantity for which the interval is to be estimated is a fixed but unknown constant, such as the annual total emissions in a given year for a given country

a measure of the mutual dependence between

mutual dependence between two quantities

a number laying between –1 and +1 which measures the mutual dependence between two variables which are observed together

value of a quantity lies (GPG 2000);


*Uncertainty*:

quantity (GPG 2000);

amount (ISO 14064-1…3).

*Error*:

(GPG 2000).

2006).

*Random error*:

IPCC 2006).

*Systematic error*:

an uncertainty is a parameter, associated with the result of measurement that characterises the dispersion of the values that could be reasonably attributed to the measured

lack of knowledge of the true value of a variable that can be described as a probability density function characterizing the range and likelihood of possible values (IPCC 2006); parameter associated with the result of quantification which characterizes the dispersion of the values that could be reasonably attributed to the quantified

a general term referring to the difference between an observed (measured) value of a

the difference between the true, but usually

unknown, value of a quantity being estimated, and the mean observed value as would be estimated by the sample mean of an infinite set of observations (GPG 2000, IPCC

the random error of an individual

measurement is the difference between an individual measurement and the above limiting value of the sample mean (GPG 2000,

quantity and its "true" (but usually unknown) value and does not carry the pejorative sense of a mistake or blunder

**Metrological terms Environmental terms** 

*Uncertainty*:

5725-1, ISO 3534-1).

*Error*:

3534-1).

2007);

ISO 3534-1).

*Random error*:

3534-1).

*Systematic error*:

non-negative parameter characterizing the dispersion of the quantity values being attributed to a measurand, based on

an estimate attached to a test result which characterizes the range of values within which the true value is asserted to lie (ISO

the information used (VIM 2007);

measured quantity value minus a reference quantity value (VIM 2007); the test result minus the accepted reference value (of the characteristic), which is the sum of random errors and systematic errors (ISO 5725-1, ISO

component of measurement error that in replicate measurements remains constant or varies in a predictable manner (VIM

a component of the error which, in the course of a number of test results for the same characteristic, remains constant or varies in a predictable way (ISO 5725-1,

component of measurement error that in replicate measurements varies in an unpredictable manner (VIM 2007); a component of the error which, in the course of a number of test results for the

same characteristic, varies in an unpredictable way (ISO 5725-1, ISO


Table 1. Metrological and environmental guides and international standard terms

Fig. 2. Relationship between accuracy and precision

The *accuracy* and *precision* of individual measurements will depend upon the equipment and protocols used to make the measurements. A measurement system (equipment) is designated valid if it is both *accurate* and *precise*. The relationship between accuracy and precision is shown in Fig. 2 (PDF is probability density function; RV is reference value).

The concept "*measurement accuracy*" is not a quantity and is not given a numerical quantity value (VIM 2007). The term *accuracy*, when applied to a set of test results, involves a

The Uncertainty Estimation and Use of Measurement Units

of the dispersion (ISO 14064-1…3).

is not possible to correct (ISO 3534-1).

Uncorrected arithmetic mean of observations

Correction for all recognized systematic effects

Measurand value due to incomplete definition

Fig. 4. Relationship between error and uncertainty

Uncorrected observations

*Result of measurement*

Remaining error (unknowable)

*Final result of measurement*

(GUM 1993, VIM 2007).

Measurand value

(unknowable)

in National Inventories of Anthropogenic Emission of Greenhouse Gas 193

series of measurements and can be characterized by *standard deviations*. Estimates of other

Uncertainty depends on the analyst's state of knowledge, which in turn depends on the quality and quantity of applicable data as well as knowledge of underlying processes and inference methods (IPCC 2006). Uncertainty information typically specifies quantitative estimates of the likely dispersion of values and a qualitative description of the likely causes

Uncertainty should be distinguished from an estimate attached to a test result which characterizes the range of values within which the expectation is asserted to lie. This latter estimate is a measure of precision rather than of accuracy and should be used only when the true value is not defined. When the expectation is used instead of the true value the

*Systematic measurement error*, and its causes, can be known or unknown. A correction can be applied to compensate for a known systematic measurement error. *Random measurement errors* of a set of replicate measurements form a distribution that can be summarized by its

*Error* of result is the test result minus the accepted RV (of the characteristic). Error is the sum of systematic and random errors. *Systematic error* may be known or unknown; *random error* it

*Quantity Value Variance*

To express of the measurement result is used the expanded measurement uncertainty with a specified coverage interval which does not need to be centred on the chosen measured quantity value. This interval should not be termed "confidence interval" to avoid confusion with the statistical concept and can be derived from an expanded measurement uncertainty

(single observation) (arithmetic mean)

(does not include variance due to incomplete measurand definition)

expectation, which is generally assumed to be zero, and its variance (VIM 2007).

The relationship between error and uncertainty is shown in Fig. 4 (GUM 1993).

components can only be based on experience or other information (ISO 3534-1).

expression "random component of uncertainty" should be used.

combination of random components and a common systematic error or bias component (ISO 5725-1, ISO 3534-1). Estimates should be accurate in the sense that they are systematically neither over nor under true emissions or removals, so far as can be judged, and that uncertainties are reduced so far as is practicable (IPCC 2006).

*Accepted reference value* is a value that serves as an agreed-upon reference for comparison, and which is derived as: a theoretical or established value, based on scientific principles; an assigned or certified value, based on experimental work of some national or international organization; a consensus or certified value, based on collaborative experimental work under the auspices of a scientific or engineering group; when the first three are not available, the expectation of the (measurable) quantity, i.e. the mean of a specified population of measurements.

*Measurement precision* is usually expressed numerically by measures of imprecision, such as standard deviation, variance, or coefficient of variation under the specified conditions of measurement (VIM 2007). *Precision* is the inverse of uncertainty in the sense that the more precise something is, the less uncertain it is (IPCC 2006).

Precision depends only on the distribution of random errors and does not relate to the true value or the specified value. The measure of precision usually is expressed in terms of imprecision and computed as a standard deviation of the test results. Less precision is reflected by a larger standard deviation. "Independent results" means results obtained in a manner not influenced by any previous result on the same or similar test object. Quantitative measures of precision depend critically on the stipulated conditions (ISO 3534-1). The relationship between small or large accuracy and precision is shown in Fig. 3.

Fig. 3. Small or large accuracy and precision (1-small accuracy and large precision; 2-large accuracy and large precision; 3-large accuracy and small precision; 4-small accuracy and small precision)

The parameter (for uncertainty) may be, for example, a standard deviation called standard measurement uncertainty (or a specified multiple of it), or the half-width of an interval, having a stated coverage probability. In general, for a given set of information, it is understood that the measurement uncertainty is associated with a stated quantity value attributed to the measurand (VIM 2007).

*Uncertainty* of measurement comprises, in general, many components. Some of these components may be estimated on the basis of the statistical distribution of the results of a

combination of random components and a common systematic error or bias component (ISO 5725-1, ISO 3534-1). Estimates should be accurate in the sense that they are systematically neither over nor under true emissions or removals, so far as can be judged,

*Accepted reference value* is a value that serves as an agreed-upon reference for comparison, and which is derived as: a theoretical or established value, based on scientific principles; an assigned or certified value, based on experimental work of some national or international organization; a consensus or certified value, based on collaborative experimental work under the auspices of a scientific or engineering group; when the first three are not available, the expectation of the (measurable) quantity, i.e. the mean of a specified

*Measurement precision* is usually expressed numerically by measures of imprecision, such as standard deviation, variance, or coefficient of variation under the specified conditions of measurement (VIM 2007). *Precision* is the inverse of uncertainty in the sense that the more

Precision depends only on the distribution of random errors and does not relate to the true value or the specified value. The measure of precision usually is expressed in terms of imprecision and computed as a standard deviation of the test results. Less precision is reflected by a larger standard deviation. "Independent results" means results obtained in a manner not influenced by any previous result on the same or similar test object. Quantitative measures of precision depend critically on the stipulated conditions (ISO 3534-1).

RV Value

3 4

The relationship between small or large accuracy and precision is shown in Fig. 3.

Fig. 3. Small or large accuracy and precision (1-small accuracy and large precision; 2-large accuracy and large precision; 3-large accuracy and small precision; 4-small accuracy and

The parameter (for uncertainty) may be, for example, a standard deviation called standard measurement uncertainty (or a specified multiple of it), or the half-width of an interval, having a stated coverage probability. In general, for a given set of information, it is understood that the measurement uncertainty is associated with a stated quantity value

*Uncertainty* of measurement comprises, in general, many components. Some of these components may be estimated on the basis of the statistical distribution of the results of a

and that uncertainties are reduced so far as is practicable (IPCC 2006).

precise something is, the less uncertain it is (IPCC 2006).

1

attributed to the measurand (VIM 2007).

2

population of measurements.

PDF

small precision)

series of measurements and can be characterized by *standard deviations*. Estimates of other components can only be based on experience or other information (ISO 3534-1).

Uncertainty depends on the analyst's state of knowledge, which in turn depends on the quality and quantity of applicable data as well as knowledge of underlying processes and inference methods (IPCC 2006). Uncertainty information typically specifies quantitative estimates of the likely dispersion of values and a qualitative description of the likely causes of the dispersion (ISO 14064-1…3).

Uncertainty should be distinguished from an estimate attached to a test result which characterizes the range of values within which the expectation is asserted to lie. This latter estimate is a measure of precision rather than of accuracy and should be used only when the true value is not defined. When the expectation is used instead of the true value the expression "random component of uncertainty" should be used.

*Systematic measurement error*, and its causes, can be known or unknown. A correction can be applied to compensate for a known systematic measurement error. *Random measurement errors* of a set of replicate measurements form a distribution that can be summarized by its expectation, which is generally assumed to be zero, and its variance (VIM 2007).

*Error* of result is the test result minus the accepted RV (of the characteristic). Error is the sum of systematic and random errors. *Systematic error* may be known or unknown; *random error* it is not possible to correct (ISO 3534-1).

The relationship between error and uncertainty is shown in Fig. 4 (GUM 1993).

#### Fig. 4. Relationship between error and uncertainty

To express of the measurement result is used the expanded measurement uncertainty with a specified coverage interval which does not need to be centred on the chosen measured quantity value. This interval should not be termed "confidence interval" to avoid confusion with the statistical concept and can be derived from an expanded measurement uncertainty (GUM 1993, VIM 2007).

The Uncertainty Estimation and Use of Measurement Units

importance of these factors.

statistical uncertainty associated with the resulting emission estimates.

in National Inventories of Anthropogenic Emission of Greenhouse Gas 195

GHG emissions can be measured either directly or indirectly. The indirect approach usually involves the use of an estimation model (e.g., AD and an EF), while the direct approach requires that emissions to the atmosphere be measured directly by some form of instrumentation (e.g., continuous emissions monitor). As the data used in the direct or indirect measurement of GHG emissions are subject to random variation there is always

The uncertainty in this relationship must be considered as well as the accuracy and precision in measurements in the proxy data itself. An uncertainty is a parameter, associated with the result of measurement that characterizes the dispersion of the values that could be reasonably attributed to the measured quantity (GPG 2000). An *uncertainty analysis* of a model aims to provide quantitative measures of the uncertainty of output values caused by uncertainties in the model itself and in its input values, and to examine the relatively

The IPCC guides (GPG 2000, IPCC 2006) use two main statistical concepts: the PDF and *confidence limits*. On Fig. 5 show PDF and cumulative distribution function (CDF) graphs. The PDF describes the range and relative likelihood of possible values; confidence limits give the range within which the underlying value of an uncertain quantity is thought to lie (confidence interval). The IPCC Guides suggest the use of a 95 % confidence interval, which

PDF is a mathematical function which characterizes the probability behaviour of population. It is a function *f*(*x*) which specifies the relative likelihood of a continuous random variable *X* taking a value near *x*, and is defined as the probability that *X* takes a value between *x* and *x+dx*, divided by *dx*, where *dx* is an infinitesimally small number. Most PDFs require one or

The probability that a continuous random variable *X* lies in between the values *a* and *b* is

EF

EF

is the interval that has a 95 % probability of containing the unknown true value.

0.00 0.05 0.10 0.15 0.20 0.25 0.30

0.00

given by the interval of the PDF, *f*(*x*), over the range between *a* and *b*:

0.50

1.00

PDF

CDF

Fig. 5. PDF (a) and CDF (b) graphs

more parameters to specify them fully.

The level of belief is expressed by the probability, whose value is related to the size of the interval. It is one of the ways in which uncertainty can be expressed. In practice a confidence interval is defined by a probability value, say 95%, and confidence limits on either side of the mean value *x*. In this case the confidence limits would be calculated from the PDF such that there was a 95% chance of the true value of the quantity being estimated by *x* lying between those limits. Commonly limits are the 2.5 percentile and 97.5 percentile respectively (GPG 2000).

The *confidence interval* is a range that encloses the true value of this unknown fixed quantity with a specified confidence (probability). Typically, a 95 % confidence interval is used in greenhouse gas inventories. From a traditional statistical perspective, the 95 % confidence interval has a 95 % probability of enclosing the true but unknown value of the quantity. An alternative interpretation is that the confidence interval is a range that may safely be declared to be consistent with observed data or information. The 95 %confidence interval is enclosed by the 2.5th and 97.5th percentiles of the PDF (IPCC 2006).

Dependencies among input sources will matter only if the dependencies exist between two sources to which the uncertainty in the GHG national inventory is sensitive and if the dependencies are sufficiently strong. For the quantities evaluation of dependence of two or more input sources used the correlation coefficient.

A value of +1 of *correlation coefficient* means that the variables have a perfect linear relationship; a value of –1 of correlation coefficient means that there is a perfect inverse linear relation; and a value of 0 of correlation coefficient means that there is no straight line relation. It is defined as the covariance of the two variables divided by the product of their *standard deviations* (σ). The population standard deviation is the positive square root of the variance*.* It is estimated by the sample standard deviation that is the positive square root of the sample variance (GPG 2000, IPCC 2006).

For the preparation of GHG national inventories used the activity data (AD) and emission factor (EF). *Activity data* is data on the magnitude of a human activity resulting in emissions or removals taking place during a given period of time. Data on energy use, metal production, land areas, management systems, lime and fertilizer use and waste arisings are examples of AD. *Emission factor* is a coefficient that quantifies the emissions or removals of a gas per unit activity. EF is often based on a sample of measurement data, averaged to develop a representative rate of emission for a given activity level under a given set of operating conditions (IPCC 2006).

## **3. Uncertainty estimation in international environmental and metrological guides**

Parties of UNFCCC can estimate GHG emissions in using two general approaches: direct measurement or proxy data. The concept of uncertainty in direct measurements is more consistent with a statistical concept of uncertainty. The statistical issues include precision and calibration of measurement equipment, fraction of population captured, frequency of sampling, etc. In contrast, proxy data is more typically in the form of AD and EF. The proxy data approach requires assumptions as to the relationship between some activity and actual emissions (IPCC 2006).

The level of belief is expressed by the probability, whose value is related to the size of the interval. It is one of the ways in which uncertainty can be expressed. In practice a confidence interval is defined by a probability value, say 95%, and confidence limits on either side of the mean value *x*. In this case the confidence limits would be calculated from the PDF such that there was a 95% chance of the true value of the quantity being estimated by *x* lying between those limits. Commonly limits are the 2.5 percentile and 97.5 percentile respectively

The *confidence interval* is a range that encloses the true value of this unknown fixed quantity with a specified confidence (probability). Typically, a 95 % confidence interval is used in greenhouse gas inventories. From a traditional statistical perspective, the 95 % confidence interval has a 95 % probability of enclosing the true but unknown value of the quantity. An alternative interpretation is that the confidence interval is a range that may safely be declared to be consistent with observed data or information. The 95 %confidence interval is

Dependencies among input sources will matter only if the dependencies exist between two sources to which the uncertainty in the GHG national inventory is sensitive and if the dependencies are sufficiently strong. For the quantities evaluation of dependence of two or

A value of +1 of *correlation coefficient* means that the variables have a perfect linear relationship; a value of –1 of correlation coefficient means that there is a perfect inverse linear relation; and a value of 0 of correlation coefficient means that there is no straight line relation. It is defined as the covariance of the two variables divided by the product of their *standard deviations* (σ). The population standard deviation is the positive square root of the variance*.* It is estimated by the sample standard deviation that is the positive square root of

For the preparation of GHG national inventories used the activity data (AD) and emission factor (EF). *Activity data* is data on the magnitude of a human activity resulting in emissions or removals taking place during a given period of time. Data on energy use, metal production, land areas, management systems, lime and fertilizer use and waste arisings are examples of AD. *Emission factor* is a coefficient that quantifies the emissions or removals of a gas per unit activity. EF is often based on a sample of measurement data, averaged to develop a representative rate of emission for a given activity level under a given set of

**3. Uncertainty estimation in international environmental and metrological** 

Parties of UNFCCC can estimate GHG emissions in using two general approaches: direct measurement or proxy data. The concept of uncertainty in direct measurements is more consistent with a statistical concept of uncertainty. The statistical issues include precision and calibration of measurement equipment, fraction of population captured, frequency of sampling, etc. In contrast, proxy data is more typically in the form of AD and EF. The proxy data approach requires assumptions as to the relationship between some activity and actual

enclosed by the 2.5th and 97.5th percentiles of the PDF (IPCC 2006).

more input sources used the correlation coefficient.

the sample variance (GPG 2000, IPCC 2006).

operating conditions (IPCC 2006).

**guides** 

emissions (IPCC 2006).

(GPG 2000).

GHG emissions can be measured either directly or indirectly. The indirect approach usually involves the use of an estimation model (e.g., AD and an EF), while the direct approach requires that emissions to the atmosphere be measured directly by some form of instrumentation (e.g., continuous emissions monitor). As the data used in the direct or indirect measurement of GHG emissions are subject to random variation there is always statistical uncertainty associated with the resulting emission estimates.

The uncertainty in this relationship must be considered as well as the accuracy and precision in measurements in the proxy data itself. An uncertainty is a parameter, associated with the result of measurement that characterizes the dispersion of the values that could be reasonably attributed to the measured quantity (GPG 2000). An *uncertainty analysis* of a model aims to provide quantitative measures of the uncertainty of output values caused by uncertainties in the model itself and in its input values, and to examine the relatively importance of these factors.

The IPCC guides (GPG 2000, IPCC 2006) use two main statistical concepts: the PDF and *confidence limits*. On Fig. 5 show PDF and cumulative distribution function (CDF) graphs. The PDF describes the range and relative likelihood of possible values; confidence limits give the range within which the underlying value of an uncertain quantity is thought to lie (confidence interval). The IPCC Guides suggest the use of a 95 % confidence interval, which is the interval that has a 95 % probability of containing the unknown true value.

#### Fig. 5. PDF (a) and CDF (b) graphs

PDF is a mathematical function which characterizes the probability behaviour of population. It is a function *f*(*x*) which specifies the relative likelihood of a continuous random variable *X* taking a value near *x*, and is defined as the probability that *X* takes a value between *x* and *x+dx*, divided by *dx*, where *dx* is an infinitesimally small number. Most PDFs require one or more parameters to specify them fully.

The probability that a continuous random variable *X* lies in between the values *a* and *b* is given by the interval of the PDF, *f*(*x*), over the range between *a* and *b*:

The Uncertainty Estimation and Use of Measurement Units

*n <sup>s</sup> <sup>u</sup> <sup>A</sup>* ,

Type A of GUM 1993 regulated using assessment of uncertainty (components evaluated by statistical methods to a series of repeated determinations) and use

Type B of GUM 1993 regulated using assessment of uncertainty (components evaluated by other means) and use

*B*

*U* is the expanded uncertainty given in

*k* is the coverage factor (typically *k* = 2).

In GUM 1993 the overall uncertainty arising from the combination of type A and type B uncertainties calculated used

<sup>2</sup> or

*ui* is the components of uncertainty; *U <sup>p</sup>* is the expanded uncertainty.

2 2 ˆ ˆˆ *c AB u uu* ,

 

*i <sup>c</sup> <sup>i</sup> uu* 1

*uc* is the total uncertainty;

*m*

*u*

*U*

*<sup>k</sup>* ,

*s* is the standard deviation; *n* is the number of measurements.

equation:

where:

equation:

where:

Certificate;

equation:

where:

and*U ku p c*

**Metrological guides Environmental guides** 

in National Inventories of Anthropogenic Emission of Greenhouse Gas 197

equation:

where:

equation:

where:

where:

*U*

sum of the quantities\*;

them, respectively.

Rule А of GPG 2000 regulated using assessment of uncertainty and use

*total xxx*

Rule B of GPG 2000 regulated using assessment of uncertainty (if impossible use statistical processing) and use

> 2 <sup>1</sup> ... *total UUUU <sup>n</sup>* ,

*Utotal* is percentage of uncertainty in the

*Ui* are percentage of uncertainties associated with each of the quantities.

In IPCC 2006 the overall uncertainty arising from the combination of EF and AD uncertainty calculated used equation:

*UE* is percentage of uncertainties

*U <sup>A</sup>* is percentage of uncertainties

associated with the AD, so long as *UE* ,

associated with the EF;

*<sup>U</sup> <sup>A</sup>* < 60 %\*\*.

\* half the 95 % confidence interval divided by the total and expressed as a percentage; \*\* the 60 % limit is imposed because the rule suggested for *UT* requires σ to be less than about 30 % of the central estimate, and we are interpreting the quoted range as ± 2σ.

Table 2. Uncertainty estimation in metrological and environmental guides

product of the quantities\*;

2

*n*

2 2

2

)( <sup>22</sup> *<sup>T</sup> UUU AE* ,

*xUxUxU*

11 ,

22

 ... )(...)()(

21

*Utotal* is percentage of uncertainty in the

*<sup>i</sup> x* and *Ui* are the uncertain quantities and percentage of uncertainties associated with

*nn*

2 2

$$\Pr(a \le x < b) = \bigcap\_{b}^{a} f(\mathbf{x}) d\mathbf{x} \text{ .}$$

The PDF is the derivative (when it exists) of the distribution function (*F*(*x*) for a random variable *X* specifies the probability Pr(*X* ≤ *x*) that *X* is less than or equal to *x*):

$$f(\mathbf{x}) = \frac{dF(\mathbf{x})}{d\mathbf{x}}\,\,\,\,\,\,$$

In practical situations, the PDF used is chosen from a relatively small number of standard PDFs and the main statistical task is to estimate its parameters. Thus, for inventory applications, a knowledge of which PDF has been used is a necessary item in the documentation of an uncertainty assessment (GPG 2000).

Uncertainty information on the EF, AD and other parameters used for the uncertainty analysis must be collected to create PDF (for the Monte Carlo method) or mean and standard deviation of the data (for the error propagation method). As this uncertainty data is collected, the correlations between parameters should also be considered.

The measurement error is one of the first types of uncertainty of GHG emission inventories, which may be: results from errors in measuring, recording and transmitting information; finite instrument resolution; inexact values of measurement standards and reference materials; inexact values of constants and other parameters obtained from external sources and used in the data-reduction algorithm; approximations and assumptions incorporated in the measurement method and estimation procedure; and/or variations in repeated observations of the emission or uptake or associated quantity under apparently identical conditions. Measurement error can be reduced using more precise measurement methods, avoiding simplifying assumptions and ensuring, that measurement technologies are appropriately used and calibrated.

Uncertainties also may be a result of: measurements were attempted but no value was available (missing data); and measurement data are not available either because the process is not yet recognized or a measurement method does not yet exist (lack of completeness). Where a PDF can be identified, sources of uncertainty can be addressed by statistical means (*type A of uncertainty* for GUM 1993). There can be structural uncertainties that are not easily incorporated into a quantitative uncertainty analysis in the form of a PDF. These types of situations are typically outside the scope of statistics (*type B of uncertainty* for GUM 1993).

Comparison of uncertainty estimation in metrological and environmental guides are provided in Table 2 (Velychko O. & Gordiyenko T., 2005; Gordiyenko T. & Velychko O., 2006; Velychko O. & Gordiyenko T., 2007a; Velichko О. N. & Gordienko T. B., 2007c, 2009; Velychko O. M. & Gordiyenko T. B., 2008; Velychko O. & Gordiyenko T., 2009).

The pragmatic approach for producing quantitative uncertainty estimation is using the best available estimates, which are often a combination of measured data, published information, model outputs, and expert judgement. Although uncertainties determined from measured data are often perceived to be mote rigorous than uncertainty estimates based on models, and similarly.

*a b a x b f x dx* .

Pr( ) ( )

variable *X* specifies the probability Pr(*X* ≤ *x*) that *X* is less than or equal to *x*):

is collected, the correlations between parameters should also be considered.

documentation of an uncertainty assessment (GPG 2000).

appropriately used and calibrated.

and similarly.

The PDF is the derivative (when it exists) of the distribution function (*F*(*x*) for a random

( ) ( ) *dF x f x dx* .

In practical situations, the PDF used is chosen from a relatively small number of standard PDFs and the main statistical task is to estimate its parameters. Thus, for inventory applications, a knowledge of which PDF has been used is a necessary item in the

Uncertainty information on the EF, AD and other parameters used for the uncertainty analysis must be collected to create PDF (for the Monte Carlo method) or mean and standard deviation of the data (for the error propagation method). As this uncertainty data

The measurement error is one of the first types of uncertainty of GHG emission inventories, which may be: results from errors in measuring, recording and transmitting information; finite instrument resolution; inexact values of measurement standards and reference materials; inexact values of constants and other parameters obtained from external sources and used in the data-reduction algorithm; approximations and assumptions incorporated in the measurement method and estimation procedure; and/or variations in repeated observations of the emission or uptake or associated quantity under apparently identical conditions. Measurement error can be reduced using more precise measurement methods, avoiding simplifying assumptions and ensuring, that measurement technologies are

Uncertainties also may be a result of: measurements were attempted but no value was available (missing data); and measurement data are not available either because the process is not yet recognized or a measurement method does not yet exist (lack of completeness). Where a PDF can be identified, sources of uncertainty can be addressed by statistical means (*type A of uncertainty* for GUM 1993). There can be structural uncertainties that are not easily incorporated into a quantitative uncertainty analysis in the form of a PDF. These types of situations are typically outside the scope of statistics (*type B of uncertainty* for GUM 1993).

Comparison of uncertainty estimation in metrological and environmental guides are provided in Table 2 (Velychko O. & Gordiyenko T., 2005; Gordiyenko T. & Velychko O., 2006; Velychko O. & Gordiyenko T., 2007a; Velichko О. N. & Gordienko T. B., 2007c, 2009;

The pragmatic approach for producing quantitative uncertainty estimation is using the best available estimates, which are often a combination of measured data, published information, model outputs, and expert judgement. Although uncertainties determined from measured data are often perceived to be mote rigorous than uncertainty estimates based on models,

Velychko O. M. & Gordiyenko T. B., 2008; Velychko O. & Gordiyenko T., 2009).


\*\* the 60 % limit is imposed because the rule suggested for *UT* requires σ to be less than about 30 % of the central estimate, and we are interpreting the quoted range as ± 2σ.

Table 2. Uncertainty estimation in metrological and environmental guides

The Uncertainty Estimation and Use of Measurement Units

*m* is mode (most likely position), subject to *amb* .

by:

distribution is given by:

(GPG 2000).

in National Inventories of Anthropogenic Emission of Greenhouse Gas 199

The *normal* (*or Gaussian*) *distribution* is most appropriate when the range of uncertainty is small, and symmetric relative to the mean. This distribution arises in situations where many individual inputs contribute to an overall uncertainty, and in which none of the individual uncertainties dominates the total uncertainty. The PDF of the normal distribution is given

> <sup>2</sup> ( ) <sup>1</sup> ( ) , for . <sup>2</sup>

The *lognormal distribution* may be appropriate when uncertainties are large for a nonnegative variable and known to be positively skewed. If many uncertain variables are multiplied, the product asymptotically approaches lognormality. The PDF of the lognormal

<sup>1</sup> ( ) , for 0 . <sup>2</sup>

The parameters required to specify the function are: *l* the mean of the natural log transform of the data; and <sup>2</sup> *l* the variance of the natural log transform of the data. The data and information that the inventory compiler can use to determine the input parameters are:

and

*Fractile distribution* is a type of empirical distribution in which judgements are made regarding the relative likelihood of different ranges of values for a variable (GPG 2000).

The rules for *uncertainties propagation* specify how to algebraically combine the quantitative measures of uncertainty associated with the input values to the mathematical formulae used in GHG national inventory compilation, so as to obtain corresponding measures of uncertainty for the output values. The *Monte Carlo analysis* is suitable for detailed category-by-category assessment of uncertainty, particularly where uncertainties are large, distribution is non-normal (non-Gaussian), the algorithms are complex functions and/or there are correlations between some of the activity sets, EF, or both

Simplified estimation of expanded measurement uncertainty for Type A is shown on Fig. 7.

1 2 ( , ,..., ) *Y fXX X <sup>m</sup>* ,

*fx e x*

*x*

2 2 (ln )

> 2 <sup>2</sup> ln 1 *<sup>l</sup>*

.

*l l*

*x fx e x* 

*l*

*l* is mean; <sup>2</sup> variance; and the relationships:

ln *<sup>l</sup>*

Measurement uncertainty of the values equation is used:

*x*

σ

 

Probability distribution is a function giving the probability that a random variable takes any given value or belongs to a given set of values. The probability on the whole set of values of the random variable equals 1. Many commonly used PDF distributions of practical important are: uniform; triangular, normal; lognormal; and fractile (Fig. 6).

Fig. 6. Some commonly used PDF models

*Uniform distribution* describes an equal likelihood of obtaining any value within a range. Sometimes the uniform distribution is useful for representing physically-bounded quantities. The PDF of the uniform distribution is given by:

$$f(\mathbf{x}) = \begin{cases} 1 / \, (b - a) \text{, for } a \le \mathbf{x} \le b \, \mu \\ 0 \quad \text{elsewhere,} \end{cases}$$

where:

 *a b* 2 is a mean and; *a b* 12 is the variance.

The *triangular distribution* is appropriate where upper and lower limits and a preferred value are provided by experts but there is no other information about the PDF. The triangular distribution can be asymmetrical. The PDF of the triangular distribution is given by:

$$f(\mathbf{x}) = \begin{cases} 2(\mathbf{x} - a) / \left\{ (b - a)(m - a) \right\}, \text{ when } a \le \mathbf{x} \le m \text{ and } a < m \le b, \\ 2(b - \mathbf{x}) / \left\{ (b - a)(b - m) \right\}, \text{ when } m \le \mathbf{x} \le b \text{ and } a \le m < b, \\ 0 & \text{elsewhere,} \end{cases}$$

where:

*a b*, are minimum and maximum value respectively;

*m* is mode (most likely position), subject to *amb* .

198 Greenhouse Gases – Emission, Measurement and Management

Probability distribution is a function giving the probability that a random variable takes any given value or belongs to a given set of values. The probability on the whole set of values of the random variable equals 1. Many commonly used PDF distributions of practical

Value of variable Value of variable

d) lognormal

Value of variable Value of variable

important are: uniform; triangular, normal; lognormal; and fractile (Fig. 6).

PDF a) uniform PDF b) triangular

c) normal

PDF PDF

quantities. The PDF of the uniform distribution is given by:

e) fractile

Fig. 6. Some commonly used PDF models

PDF

where:

where:

*a b* 2 is a mean and;

*a b* 12 is the variance.

Value of variable

*Uniform distribution* describes an equal likelihood of obtaining any value within a range. Sometimes the uniform distribution is useful for representing physically-bounded

1 /( ), for , ( ) 0 elsewhere,

The *triangular distribution* is appropriate where upper and lower limits and a preferred value are provided by experts but there is no other information about the PDF. The triangular

( ) 2( ) / ( )( ) , when and ,

*f x b x babm m x b a m b* 

2( ) / ( )( ) , when and ,

*x a bama a x m a m b*

distribution can be asymmetrical. The PDF of the triangular distribution is given by:

 

0 elsewhere,

 

*a b*, are minimum and maximum value respectively;

*f x*

*ba axb*

The *normal* (*or Gaussian*) *distribution* is most appropriate when the range of uncertainty is small, and symmetric relative to the mean. This distribution arises in situations where many individual inputs contribute to an overall uncertainty, and in which none of the individual uncertainties dominates the total uncertainty. The PDF of the normal distribution is given by:

$$f(\mathbf{x}) = \frac{1}{\sigma\sqrt{2\pi}}e^{-\frac{(\mathbf{x}-\mu)^2}{2\sigma^2}}, \text{ for } -\infty \le \mathbf{x} \le \infty.$$

The *lognormal distribution* may be appropriate when uncertainties are large for a nonnegative variable and known to be positively skewed. If many uncertain variables are multiplied, the product asymptotically approaches lognormality. The PDF of the lognormal distribution is given by:

$$f(\mathbf{x}) = \frac{1}{\sigma\_l \mathbf{x} \sqrt{2\pi}} e^{-\frac{\left(\ln \|\mathbf{x} - \boldsymbol{\mu}\_l\|\right)^2}{2\sigma\_l^2}}, \text{ for } 0 \le \mathbf{x} \le \infty.$$

The parameters required to specify the function are: *l* the mean of the natural log transform of the data; and <sup>2</sup> *l* the variance of the natural log transform of the data. The data and information that the inventory compiler can use to determine the input parameters are: *l* is mean; <sup>2</sup> variance; and the relationships:

$$\mu\_I = \ln \frac{\mu^2}{\sqrt{\left(\sigma^2 + \mu^2\right)}} \qquad \text{and} \quad \sigma\_I = \sqrt{\ln \left(\frac{\sigma^2}{\mu^2} + 1\right)} \quad . \dots$$

*Fractile distribution* is a type of empirical distribution in which judgements are made regarding the relative likelihood of different ranges of values for a variable (GPG 2000).

The rules for *uncertainties propagation* specify how to algebraically combine the quantitative measures of uncertainty associated with the input values to the mathematical formulae used in GHG national inventory compilation, so as to obtain corresponding measures of uncertainty for the output values. The *Monte Carlo analysis* is suitable for detailed category-by-category assessment of uncertainty, particularly where uncertainties are large, distribution is non-normal (non-Gaussian), the algorithms are complex functions and/or there are correlations between some of the activity sets, EF, or both (GPG 2000).

Simplified estimation of expanded measurement uncertainty for Type A is shown on Fig. 7.

Measurement uncertainty of the values equation is used:

$$Y = f(X\_1, X\_2, \dots, X\_m)\_{\prime\prime}$$

The Uncertainty Estimation and Use of Measurement Units

by means of the equation from Table 2 (GUM 1993).

within the identified uncertainty range.

Table 3. *t*-factors for the 95% confidence level

estimations of emission foul and GHG.

**4. Uncertainty estimation with correlation values** 

measurement (the evaluation of type B covariance).

be done by using the Table 3.

in National Inventories of Anthropogenic Emission of Greenhouse Gas 201

The overall standard uncertainty *uc* (Table 2) determined by a combination of uncertainties components; estimation of the expanded measurement uncertainty *U <sup>p</sup>* may be calculated

Statistical uncertainty in the context of GHG inventories is usually presented by giving an uncertainty range expressed in a percentage of the expected mean value of the emission. This range can be determined by calculating the "confidence limits", within which the underlying value of an uncertain quantity is thought to lie for a specified probability. The "confidence level" determines the probability, that the true value of emission is situated

Determining the *t*-factor *t* (standard error that is to be estimated follows a *t*-distribution) can

**Number of measurements (***n***)** *t***-factor (***t***) for confidence level 95 %**  3 4,30 5 2,78 8 2,37 10 2,26 50 2,01 100 1,98 1,96

Necessity in the analysis of covariance and autocorrelation for the uncertainty estimates of emission foul and greenhouse gases exists. It is importantly to investigate the correlations of the estimated values relevant to emissions as in the context one estimation and the various

If correlation exists between input values then this correlation is essential and can not be ignored. Covariance of input values may be estimated experimentally, on condition of possible the change of input correlated values (the estimation of type A covariance), or with the using of necessary data on the correlation changeability of values which relate to this

In practical cases input values often appear to be correlated, as far as the evaluation of their values is used by the same standards, measuring instruments, standard data and even method of measurement which possess peculiar uncertainty. If direct measurement results

Covariance with the estimations of two input values may be equal to zero or selected as negligible, if: these values are not correlated (values measurement are run repeatedly in various independent experiments, or they present the various estimations of values which made independently); any values may be is accepted as an constant; there is negligible data

Comparison of the uncertainties estimations of input values with correlation in metrological and environmental guides are driven in Table 4 (Velychko O. & Gordiyenko T., 2007b).

are not correlated then calculated value of covariance is expected to be close to zero.

for calculation of covariance related with the estimations of these values.

Fig. 7. Simplified estimation of expanded measurement uncertainty

where:

*X1, …, Xm* are entrance value (direct measured value or other value which have an influence on measurement results);

*m* is quantity values;

*f* is functional dependence type.

Estimation of the standard measurement uncertainty *иА*(*хi*) of measurement *i*-th input values without correlation input values may be calculated by means of the following equation:

$$
\mu\_{\mathcal{A}}(\mathbf{x}\_{i}) = \sqrt{\frac{1}{n\_{i}(n\_{i}-1)} \sum\_{q=1}^{n\_{i}} \left(\mathbf{x}\_{iq} - \overline{\mathbf{x}\_{i}}\right)^{2}},
$$

where:

*iq x* is the measurement results of *i*-th input values;

$$\overline{\mathbf{x}\_i} = \frac{1}{n} \sum\_{q=1}^{n\_i} \mathbf{x}\_{iq} \quad \text{is the arithmetic median of measurement results of } i\text{-th input values.}$$

Fig. 7. Simplified estimation of expanded measurement uncertainty

( )

*iq x* is the measurement results of *i*-th input values;

*X1, …, Xm* are entrance value (direct measured value or other value which have an influence

Estimation of the standard measurement uncertainty *иА*(*хi*) of measurement *i*-th input values without correlation input values may be calculated by means of the following equation:

1

( 1)

is the arithmetic median of measurement results of *i*-th input values.

*A i iq i <sup>q</sup> i i u x x x n n* 

<sup>2</sup>

,

1

*i n*

where:

where:

1

1 *<sup>i</sup> n <sup>i</sup> iq <sup>q</sup> x x n*

on measurement results); *m* is quantity values;

*f* is functional dependence type.

The overall standard uncertainty *uc* (Table 2) determined by a combination of uncertainties components; estimation of the expanded measurement uncertainty *U <sup>p</sup>* may be calculated by means of the equation from Table 2 (GUM 1993).

Statistical uncertainty in the context of GHG inventories is usually presented by giving an uncertainty range expressed in a percentage of the expected mean value of the emission. This range can be determined by calculating the "confidence limits", within which the underlying value of an uncertain quantity is thought to lie for a specified probability. The "confidence level" determines the probability, that the true value of emission is situated within the identified uncertainty range.

Determining the *t*-factor *t* (standard error that is to be estimated follows a *t*-distribution) can be done by using the Table 3.


Table 3. *t*-factors for the 95% confidence level

## **4. Uncertainty estimation with correlation values**

Necessity in the analysis of covariance and autocorrelation for the uncertainty estimates of emission foul and greenhouse gases exists. It is importantly to investigate the correlations of the estimated values relevant to emissions as in the context one estimation and the various estimations of emission foul and GHG.

If correlation exists between input values then this correlation is essential and can not be ignored. Covariance of input values may be estimated experimentally, on condition of possible the change of input correlated values (the estimation of type A covariance), or with the using of necessary data on the correlation changeability of values which relate to this measurement (the evaluation of type B covariance).

In practical cases input values often appear to be correlated, as far as the evaluation of their values is used by the same standards, measuring instruments, standard data and even method of measurement which possess peculiar uncertainty. If direct measurement results are not correlated then calculated value of covariance is expected to be close to zero.

Covariance with the estimations of two input values may be equal to zero or selected as negligible, if: these values are not correlated (values measurement are run repeatedly in various independent experiments, or they present the various estimations of values which made independently); any values may be is accepted as an constant; there is negligible data for calculation of covariance related with the estimations of these values.

Comparison of the uncertainties estimations of input values with correlation in metrological and environmental guides are driven in Table 4 (Velychko O. & Gordiyenko T., 2007b).

The Uncertainty Estimation and Use of Measurement Units

imply an expected change in the other.

*Correlation and covariance for entrance value* 

the following equation:

the following equation:

1 (, ) ( ) *L i k il kl l l ux x c c u Q* 

equations (GUM 1993):

*<sup>i</sup> c* are sensitivity coefficients;

*y* is estimation of measurable value *Y*; <sup>1</sup> *x x* ,..., *<sup>m</sup>* are input values ( *i m* 1, ).

where:

where:

are equal 1.

For correlation and covariance with type *А* using

1 (, ) 1 ( 1) ( )( ) *n i k ij i kj k j ux x n n x xx x* .

2

*il c* , *kl c* are sensitivity coefficients respectively; ( ) *u Ql* is standard uncertainty of variables *Ql* .

Uncertainties contribution ( ) *u y <sup>i</sup>* every input

values *Xi* to uncertainty *u y*( ) using the following

() ( ) *u y c ux i ii* ; 1 2 *<sup>i</sup> <sup>i</sup>* , ,..., *<sup>m</sup> yx YXix x x <sup>c</sup>*

For direct measurement all sensitivity coefficients

*Uncertainties contribution every input values and sensitivity coefficient* 

,

For correlation and covariance with type *B* using

For correlation coefficient (,) *<sup>i</sup> <sup>j</sup> rx x* = 1 uncertainties contribution is:

values in metrological and environmental guides are driven in Table 5.

**Metrological guides Environmental guides** 

in National Inventories of Anthropogenic Emission of Greenhouse Gas 203

, () () () ( ) ( ) *u y u y u y cux cux <sup>i</sup> <sup>j</sup> <sup>i</sup> <sup>j</sup> i i j j* ; if the estimates *<sup>i</sup> x* and *<sup>j</sup> x* are independent, (,) *<sup>i</sup> <sup>j</sup> rx x* = 0, and a change in one does not

Uncertainties aggregation arises of two various processes: aggregation of emissions one gas which complies with the law of propagation of uncertainty; aggregation of emissions bound with several gases. Into second case emission must be result in common scale, and been used for this process consists in the application of Global Warming Potentials (GWP).

Comparison of the emission sources with allowance for correlation and covariance of input

The sample covariance of paired sample

calculated using the following equation

*xi, yi, i = 1,…,n* are items in the sample;

using the following equation (GPG

*ET* is the aggregated emissions; *a* is input quantity (or parameter). Dispersion of tendency for emission two different time *E t*( ) and *Et t* ( ) with *t* using the following equation:

2 2 ( ) 2 (1 ( ))

 *E rt <sup>E</sup>* . where: *r*(∆*t*) is correlation coefficient

calculated

of random variables *X* and *Y* is

(GPG 2000, IPCC 2006): <sup>2</sup> <sup>1</sup> ( )( ) *n xy i i i s x xy y*

,

*x* and *y* are sample means

Sensitivity coefficient

*n*

respectively.

 *E a <sup>T</sup>* , where:

where:

2000): 

Table 5. Comparison of uncertainty contributions in metrological and environmental guides

Some variables which are necessary aggregation, do not are Gauss, large dispersion and correlated with other variables are have. In this case the application of Monte-Carlo method


Table 4. Uncertainty estimation with correlation in metrological and environmental guides

*U*

where:

Using estimation of uncertainty Rule А and use equation (GPG 2000, IPCC 2006):

<sup>11</sup> ,

*Utotal* is percentage of uncertainty in the

*<sup>i</sup> <sup>x</sup>* , *Ui* are uncertain quantities and percentage of uncertainties associated

Not numerical estimated in GPG 2000

A value of correlation coefficient of +1 means that the variables have a perfect direct straight line relation; a value of –1 means that there is a perfect inverse straight line relation; and a value of 0 means that there is no straight line

It is defined as the covariance of the two variables divided by the product of their

relation (GPG 2000).

standard deviations.

Table 4. Uncertainty estimation with correlation in metrological and environmental guides

*xUxUxU*

22

 ... )(...)()(

21

*total xxx*

sum of the quantities;

with them, respectively.

and IPCC 2006.

2

*n*

*nn*

2 2

**Metrological guides Environmental guides** 

*Entrance value is non correlated*  Using estimation of uncertainty in

where:

or

where:

accordance with the law of propagation of uncertainty and use equation (GUM 1993):

2 22 22 2 2 11 2 ( ) ( ) ( ) ... ( ), *u y cu x cu x cu x c m <sup>m</sup>*

( ) *u xi* , ( ) *u y <sup>i</sup>* are standard uncertainty input ( *i m* 1, ) and output value respectively; *y*, *<sup>i</sup> x* are estimations of measurable value *Y*

Using estimation of uncertainty in accordance with the law of propagation of uncertainty

1 1 1 ( ) ()2 (, ) *m m <sup>m</sup> c ii <sup>i</sup> <sup>j</sup> i i <sup>j</sup> <sup>i</sup> u y f x u x ux x* 

<sup>1</sup> <sup>2</sup> 1 11 () () 2 ( , ) *m mm <sup>c</sup> <sup>i</sup> ik i <sup>j</sup> i i <sup>j</sup> <sup>i</sup> u y u y cc u x x* 

(,) *ux x <sup>i</sup> <sup>j</sup>* is estimation covariance with two

*y* is estimation of measurable value *Y* ; *<sup>i</sup> x* , *<sup>j</sup> x* are estimation input values *Xi* and

The degree of correlation between *<sup>i</sup> x* and *<sup>j</sup> x* is characterized by the estimated correlation

(,) (,) ( , ) [ ( ) ( )] *i j ji ux x ux ux ij i j rx x rx x* ,

where: *rx x r x x* (,) , *<sup>i</sup> j j <sup>i</sup>* ; –1 (,) *<sup>i</sup> <sup>j</sup> rx x* 1.

<sup>1</sup> <sup>2</sup>

and input value *Xi* respectively.

and use equation (GUM 1993):

input estimations *<sup>i</sup> x* and *<sup>j</sup> x* ;

*Xj* respectively.

coefficient (GUM 1993):

*Entrance value is correlated* 

 <sup>2</sup> 2 2 1 1 ( ) , ( ) () *m m c i i i i i u y f x ux uy* or For correlation coefficient (,) *<sup>i</sup> <sup>j</sup> rx x* = 1 uncertainties contribution is:

$$\mu\_{i,j}(y) = \left| \mu\_i(y) \pm \mu\_j(y) \right| = \left| c\_i \mu(\boldsymbol{\chi}\_i) \pm c\_j \mu(\boldsymbol{\chi}\_j) \right| : \boldsymbol{\chi}$$

if the estimates *<sup>i</sup> x* and *<sup>j</sup> x* are independent, (,) *<sup>i</sup> <sup>j</sup> rx x* = 0, and a change in one does not imply an expected change in the other.

Uncertainties aggregation arises of two various processes: aggregation of emissions one gas which complies with the law of propagation of uncertainty; aggregation of emissions bound with several gases. Into second case emission must be result in common scale, and been used for this process consists in the application of Global Warming Potentials (GWP).

Comparison of the emission sources with allowance for correlation and covariance of input values in metrological and environmental guides are driven in Table 5.


Table 5. Comparison of uncertainty contributions in metrological and environmental guides

Some variables which are necessary aggregation, do not are Gauss, large dispersion and correlated with other variables are have. In this case the application of Monte-Carlo method

The Uncertainty Estimation and Use of Measurement Units

where: *сА* is Type A sensitivity coefficient.

where: *сВ* is Type B sensitivity coefficient.

tendency of overall GHG emission use equation (GPG 2000):

**5. Greenhouse Gas Protocol Uncertainty Tool** 

(estimation) of uncertainties in GHG emission inventory.

equation:

use equation:

multiply by 2 .

requirements of GUM 1993.

Gordiyenko T., 2005, 2007a).

in National Inventories of Anthropogenic Emission of Greenhouse Gas 205

For uncertainty of tendency of the GHG emission with EF uncertainty *UEFt* (percentage) use

*U cU EF A EF <sup>t</sup>* ,

If between EF have not correlation necessary use Type B sensitivity and to multiply by 2 . For uncertainty of tendency of the GHG emission with AD uncertainty *UADt* (percentage)

2 *U cU AD B AD <sup>t</sup>* ,

If between AD have correlation necessary use Type A sensitivity and not necessary to

For estimation of uncertainty contribution *Utdi* (percentage) which to make one's on tendency of overall GHG emission for each emission categories for Rule B use equation:

2 2 *U UU tdi EF AD t t* .

For estimation of overall uncertainty contribution *Utd* (percentage) which to make one's on

On Fig. 9 gives developed an uncertainty estimation algorithm with tendency, correlation and covariance according to GPG 2000 and ІРСС 2006 which taking into consideration main

The IPCC 2006 for assessment GHG emissions used statistical AD of fuel combustion activities for different sectors and sources category and take account of direct and indirect GHG: CO2, CH4, N2O, CO, NOх, NMVOCs. Important element used IPCC 2006 is determination and/or selection EF which is take from IPCC 2006 ("default") or calculated as local for country, sectors, sources category or process. Accounts data submit in Common Reporting Format (CRF) which is standard data tables. IPCC 2006 contain chapter of key conceptions uncertainties, describe being types uncertainties, methods assessment

The GHG Protocol Uncertainty Tool is based on the IPCC 2006 and should be considered as an addition to the calculation tools provided by the GHG Protocol Initiative. The GHG Protocol is to describe the functionality of the tool and to give user a better understanding of how to prepare, interpret, and utilize inventory uncertainty estimation (Velychko O. &

2 *td tdi i U U* .

for uncertainties aggregation is presented the most preferable. The Monte Carlo analysis can be performed at the source category level, for aggregations of source categories, or for the inventory as a whole. It analysis can deal with PDF of any physically possible shape and width, can handle varying degrees of correlation (both in time and between source categories) and can deal with more complex models.

Algorithm of uncertainty estimation with correlation according to GUM 1993 is present on Fig. 8.

Fig. 8. Uncertainty estimation algorithm with correlation according to GUM 1993

Uncertainty results in compliance with GUM 1993 can be calculated with use a few well known commercially available software. Special software can be a useful tool for uncertainty estimation (Velychko O., 2008).

Overall uncertainty in total national GHG emissions in the current year, calculated using Rule A (corresponding Type A in accordance with GUM) and use equation

$$U\_T = E\_i \cdot U\_{\sum\_{B} B} / \sum E\_{T'} \cdot \frac{1}{2}$$

where:

*<sup>B</sup> <sup>U</sup>* is estimation of overall uncertainty Rule B;

*Еі* is GHG emissions from certain source category in СО2-equivalent, Gg;

*ЕТ* is total GHG emissions from all source categories in СО2-equivalent, Gg.

If overall uncertainty is correlated across years using estimation of overall uncertainty Rule B with uncertainty of EF only (assume AD to be equal 0 %).

For uncertainty of tendency of the GHG emission with EF uncertainty *UEFt* (percentage) use equation:

$$U\_{EFt} = \mathbf{c}\_A \cdot U\_{EF't'}$$

where: *сА* is Type A sensitivity coefficient.

204 Greenhouse Gases – Emission, Measurement and Management

for uncertainties aggregation is presented the most preferable. The Monte Carlo analysis can be performed at the source category level, for aggregations of source categories, or for the inventory as a whole. It analysis can deal with PDF of any physically possible shape and width, can handle varying degrees of correlation (both in time and between source

Algorithm of uncertainty estimation with correlation according to GUM 1993 is present on

Fig. 8. Uncertainty estimation algorithm with correlation according to GUM 1993

Rule A (corresponding Type A in accordance with GUM) and use equation

*Еі* is GHG emissions from certain source category in СО2-equivalent, Gg; *ЕТ* is total GHG emissions from all source categories in СО2-equivalent, Gg.

Uncertainty results in compliance with GUM 1993 can be calculated with use a few well known commercially available software. Special software can be a useful tool for

Overall uncertainty in total national GHG emissions in the current year, calculated using

*iT <sup>B</sup> EUEU <sup>T</sup>* / ,

If overall uncertainty is correlated across years using estimation of overall uncertainty Rule

categories) and can deal with more complex models.

uncertainty estimation (Velychko O., 2008).

*<sup>B</sup> <sup>U</sup>* is estimation of overall uncertainty Rule B;

B with uncertainty of EF only (assume AD to be equal 0 %).

Fig. 8.

where:

If between EF have not correlation necessary use Type B sensitivity and to multiply by 2 .

For uncertainty of tendency of the GHG emission with AD uncertainty *UADt* (percentage) use equation:

$$U\_{ADr} = \sqrt{2}c\_B \cdot U\_{AD'}$$

where: *сВ* is Type B sensitivity coefficient.

If between AD have correlation necessary use Type A sensitivity and not necessary to multiply by 2 .

For estimation of uncertainty contribution *Utdi* (percentage) which to make one's on tendency of overall GHG emission for each emission categories for Rule B use equation:

$$U\_{tdi} = \sqrt{U\_{EF\_l}^2 + U\_{ADr}^2} \cdot \frac{1}{2}$$

For estimation of overall uncertainty contribution *Utd* (percentage) which to make one's on tendency of overall GHG emission use equation (GPG 2000):

$$U\_{td} = \sqrt{\sum\_{i} U\_{\phantom{i}}^{2}} \,\, \_{tdi}$$

On Fig. 9 gives developed an uncertainty estimation algorithm with tendency, correlation and covariance according to GPG 2000 and ІРСС 2006 which taking into consideration main requirements of GUM 1993.

## **5. Greenhouse Gas Protocol Uncertainty Tool**

The IPCC 2006 for assessment GHG emissions used statistical AD of fuel combustion activities for different sectors and sources category and take account of direct and indirect GHG: CO2, CH4, N2O, CO, NOх, NMVOCs. Important element used IPCC 2006 is determination and/or selection EF which is take from IPCC 2006 ("default") or calculated as local for country, sectors, sources category or process. Accounts data submit in Common Reporting Format (CRF) which is standard data tables. IPCC 2006 contain chapter of key conceptions uncertainties, describe being types uncertainties, methods assessment (estimation) of uncertainties in GHG emission inventory.

The GHG Protocol Uncertainty Tool is based on the IPCC 2006 and should be considered as an addition to the calculation tools provided by the GHG Protocol Initiative. The GHG Protocol is to describe the functionality of the tool and to give user a better understanding of how to prepare, interpret, and utilize inventory uncertainty estimation (Velychko O. & Gordiyenko T., 2005, 2007a).

The Uncertainty Estimation and Use of Measurement Units

uncertainty for one year and the uncertainty in the trend.

and to calculate the corresponding values, e.g., emissions).

basic uncertainty estimation for GHG inventory data.

estimates are associated with uncertainty estimation.

*statistical* uncertainties (GHG Protocol).

AD parameter.

in National Inventories of Anthropogenic Emission of Greenhouse Gas 207

GPG 2000 describes two tiers of uncertainty estimation for GHG emission inventories for provided for combining source category uncertainties into uncertainty estimation for total national emissions and for emission trends: *Tier 1* – estimation of uncertainties by source category using the error propagation equation via Rules A and B and simple combination of uncertainties by source category to estimate overall uncertainty for one year and the uncertainty in the trend; *Tier 2* – estimation of uncertainties by source category using the Monte Carlo analysis, followed by the use of the Monte Carlo techniques to estimate overall

IPCC 2006 describes two tiers: *Tier 1* – for combining uncertainties in inventory data is to use the error propagation method. This method has limitations, in that it assumes normality in the input PDF (it cannot easily deal with correlations between datasets or across time and dependency between source categories that may occur because the same AD or EF may be used for multiple estimates); *Tier 2* – to use the Monte Carlo analysis which avoids all the limitations of the error propagation method (the principle of the Monte Carlo analysis is to select random values of each parameter, e.g., EF and AD, from within their individual PDF,

The GHG Protocol Uncertainty Tool is based on the IPCC 2006 and should be considered as an addition to the calculation tools provided by the GHG Protocol Initiative. The GHG Protocol Initiative has developed this guidance along with a calculation tool based on Excel spreadsheets. This calculation tool automates the aggregation steps involved in developing

The GHG Protocol display uncertainties associated with GHG inventories: scientific uncertainty and uncertainty estimation last can be further classified into two types: model uncertainty and parameter uncertainty. *Scientific uncertainty* arises when the science of the actual emission and/or removal process is not sufficiently understood. *Uncertainty estimation* arises any time GHG emissions are quantified. Therefore all emission or removal

*Model uncertainty* refers to the uncertainty associated with the mathematical equations (i.e., models) used to characterize the relationships between various parameters and emission processes. Emission estimation models that consist of only AD times an EF only involve parameter uncertainties, assuming that emissions are perfectly linearly correlated with the

*Parameter uncertainty* refers to the uncertainty associated with quantifying the parameters used as inputs (e.g., AD and EF) into estimation models. This uncertainty can be evaluated through statistical analysis, measurement equipment precision determinations, and expert judgment. Emission estimated from direct emissions monitoring will generally involve only parameter uncertainty (e.g., equipment measurement error). The type of uncertainty most amenable to assessment of inventory is the uncertainties associated with parameters (e.g. AD, EF, and other parameters) used as inputs in an emission estimation model. GHG Protocol identified two types of parameter uncertainties in this context: *systematic* and

*Systematic uncertainty* occurs if data are systematically biased (the average of the measured or estimated value is always less or greater than the true value). Biases can arise, because EF

Fig. 9. Developed uncertainty estimation algorithm according to GPG 2000 and ІРСС 2006

All IPCC guides use elements and reference to GUM 1993. GPG 2000 is the response to the request from the UNFCCC for the IPCC to complete its work on uncertainty and prepare a report on good practice in inventory management. Most of the statistical definitions given GPG 2000 lie within the context of "classical" frequency-based statistical inference, although it is acknowledged that this is not the only theory of statistical inference.

Fig. 9. Developed uncertainty estimation algorithm according to GPG 2000 and ІРСС 2006

it is acknowledged that this is not the only theory of statistical inference.

All IPCC guides use elements and reference to GUM 1993. GPG 2000 is the response to the request from the UNFCCC for the IPCC to complete its work on uncertainty and prepare a report on good practice in inventory management. Most of the statistical definitions given GPG 2000 lie within the context of "classical" frequency-based statistical inference, although GPG 2000 describes two tiers of uncertainty estimation for GHG emission inventories for provided for combining source category uncertainties into uncertainty estimation for total national emissions and for emission trends: *Tier 1* – estimation of uncertainties by source category using the error propagation equation via Rules A and B and simple combination of uncertainties by source category to estimate overall uncertainty for one year and the uncertainty in the trend; *Tier 2* – estimation of uncertainties by source category using the Monte Carlo analysis, followed by the use of the Monte Carlo techniques to estimate overall uncertainty for one year and the uncertainty in the trend.

IPCC 2006 describes two tiers: *Tier 1* – for combining uncertainties in inventory data is to use the error propagation method. This method has limitations, in that it assumes normality in the input PDF (it cannot easily deal with correlations between datasets or across time and dependency between source categories that may occur because the same AD or EF may be used for multiple estimates); *Tier 2* – to use the Monte Carlo analysis which avoids all the limitations of the error propagation method (the principle of the Monte Carlo analysis is to select random values of each parameter, e.g., EF and AD, from within their individual PDF, and to calculate the corresponding values, e.g., emissions).

The GHG Protocol Uncertainty Tool is based on the IPCC 2006 and should be considered as an addition to the calculation tools provided by the GHG Protocol Initiative. The GHG Protocol Initiative has developed this guidance along with a calculation tool based on Excel spreadsheets. This calculation tool automates the aggregation steps involved in developing basic uncertainty estimation for GHG inventory data.

The GHG Protocol display uncertainties associated with GHG inventories: scientific uncertainty and uncertainty estimation last can be further classified into two types: model uncertainty and parameter uncertainty. *Scientific uncertainty* arises when the science of the actual emission and/or removal process is not sufficiently understood. *Uncertainty estimation* arises any time GHG emissions are quantified. Therefore all emission or removal estimates are associated with uncertainty estimation.

*Model uncertainty* refers to the uncertainty associated with the mathematical equations (i.e., models) used to characterize the relationships between various parameters and emission processes. Emission estimation models that consist of only AD times an EF only involve parameter uncertainties, assuming that emissions are perfectly linearly correlated with the AD parameter.

*Parameter uncertainty* refers to the uncertainty associated with quantifying the parameters used as inputs (e.g., AD and EF) into estimation models. This uncertainty can be evaluated through statistical analysis, measurement equipment precision determinations, and expert judgment. Emission estimated from direct emissions monitoring will generally involve only parameter uncertainty (e.g., equipment measurement error). The type of uncertainty most amenable to assessment of inventory is the uncertainties associated with parameters (e.g. AD, EF, and other parameters) used as inputs in an emission estimation model. GHG Protocol identified two types of parameter uncertainties in this context: *systematic* and *statistical* uncertainties (GHG Protocol).

*Systematic uncertainty* occurs if data are systematically biased (the average of the measured or estimated value is always less or greater than the true value). Biases can arise, because EF

The Uncertainty Estimation and Use of Measurement Units

Table 6. The GHG Protocol Uncertainty Tool

in National Inventories of Anthropogenic Emission of Greenhouse Gas 209

are constructed from non-representative samples, all relevant source activities or categories have not been identified, or incorrect or incomplete estimation methods or faulty measurement equipment have been used.

*Statistical uncertainty* results from natural variations (e.g. random human errors in the measurement process and fluctuations in measurement equipment). This uncertainty can be detected through repeated experiments or sampling of data. Complete and robust sample data will not always be available to assess the statistical uncertainty in every parameter. For most parameters only a single data point may be available. Random uncertainty can be detected through repeated experiments or sampling of data.

The different uncertainties associated with GHG inventories according GHG Protocol shown on Fig. 10.

Fig. 10. Uncertainties associated with GHG inventories

The GHG Protocol is designed to aggregate statistical uncertainty assuming a *normal distribution* of the relevant variables and uses the first order error propagation method (Gaussian method), which corresponds to Tier 1 of the GPG 2000. This method should only be applied if the following assumptions are fulfilled: the errors in each parameter must be normally distributed (i.e. Gaussian); there must be no biases in the estimator function (i.e. that the estimated value is the mean value); the estimated parameters must be uncorrelated (i.e. all parameters are fully independent); individual uncertainties in each parameter must be less than 60 % of the mean. This procedure is repeated many times, using a computer, and the results of each calculation run build up the overall emission PDF.

A second approach is to use a technique based on a Monte Carlo simulation that allows uncertainties with any probability distribution, range, and correlation structure to be combined, provided they have been suitably quantified. This method, which corresponds to Tier 2 of the GPG 2000, can be used to estimate the uncertainty of single sources as well as to aggregate uncertainties for a site.

Calculation and ranking of uncertainties of indirectly measured emissions are shown in Table 6.


Table 6. The GHG Protocol Uncertainty Tool

are constructed from non-representative samples, all relevant source activities or categories have not been identified, or incorrect or incomplete estimation methods or faulty

*Statistical uncertainty* results from natural variations (e.g. random human errors in the measurement process and fluctuations in measurement equipment). This uncertainty can be detected through repeated experiments or sampling of data. Complete and robust sample data will not always be available to assess the statistical uncertainty in every parameter. For most parameters only a single data point may be available. Random uncertainty can be

The different uncertainties associated with GHG inventories according GHG Protocol

The GHG Protocol is designed to aggregate statistical uncertainty assuming a *normal distribution* of the relevant variables and uses the first order error propagation method (Gaussian method), which corresponds to Tier 1 of the GPG 2000. This method should only be applied if the following assumptions are fulfilled: the errors in each parameter must be normally distributed (i.e. Gaussian); there must be no biases in the estimator function (i.e. that the estimated value is the mean value); the estimated parameters must be uncorrelated (i.e. all parameters are fully independent); individual uncertainties in each parameter must be less than 60 % of the mean. This procedure is repeated many times, using a computer,

A second approach is to use a technique based on a Monte Carlo simulation that allows uncertainties with any probability distribution, range, and correlation structure to be combined, provided they have been suitably quantified. This method, which corresponds to Tier 2 of the GPG 2000, can be used to estimate the uncertainty of single sources as well as to

Calculation and ranking of uncertainties of indirectly measured emissions are shown in

and the results of each calculation run build up the overall emission PDF.

measurement equipment have been used.

shown on Fig. 10.

detected through repeated experiments or sampling of data.

Fig. 10. Uncertainties associated with GHG inventories

aggregate uncertainties for a site.

Table 6.

The Uncertainty Estimation and Use of Measurement Units

O. & Gordiyenko T., 2005; Velychko O. & Gordiyenko T., 2007a).

Units for weight Gram (g): Mg, Gg, Tg, Pg

measurements.

Units for substances\*

in National Inventories of Anthropogenic Emission of Greenhouse Gas 211

everyday use, in particular the traditional units of time and of angle, together with a few

The series of international standards on air quality includes the standardization of methods for the measurement of gases, vapours and particles (for example, ISO 4226). In order to enable results to be compared either between countries, it is essential to use agreed of measurement units to report the results and other relevant information. It is also desirable to keep the number of measurement units to a minimum. Those international standards lay down the units and symbols to be used when reporting results of air quality

Used in environmental guides and international standards (GPG 2000, IPCC 2006, ISO 4226) SI units, and also non-SI units and their conversion factors are shown in Table 7 (Velychko

Pound (lb) 1 lb = 454 g

Short ton (sh t) 1 sh t = 0.9072 t

– –

Milligram per litre (mg/l)

1 g = 0.002205 lb 1 mg = 10-3 g 1 μg = 10-6 g 1 ng = 10-9 g 1 pg = 10-12 g 1 Mg = 106 g 1 Gg = 109 g 1 Tg = 1012 g 1 Pg = 1015 g

1 t = 103 kg = 1 Mg 1 kg = 2.2046 lb 1 kt = 103 t = 1 Gg 1 Mt = 106 t = 1 Tg 1 Gt = 109 t = 1 Pg

1 t = 1.1023 sh t

1 mg/l = 103 mg/m3

**Units for quantity SI units Non-SI units Conversion factor** 

Kilogram (kg) Tonne (t): kt, Mt,

Units for length Metre (m): μm – 1 μm = 10-6 m

Percent (% by mass) – –

Percent (% by volume)

Milligram per cubic metre (mg/m3) Microgram per cubic metre (μg/m3) Nanogram per cubic metre (ng/m3) Picogram per cubic metre (pg/m3)

Gt

other units, which have assumed increasing technical importance (Tailor B. N., 1995).

Calculating aggregation of uncertainties uses equation:

$$\pm U = \pm \sqrt{\sum\_{i=1}^{n} \left(H\_i \cdot I\_i\right)^2} \left/M, \right.$$

where:

*Hi* is CO2 emissions from *i*-th source, tones;

*Ii* is percentage of uncertainty of calculated emissions from *i*-th source;

*М* is total CO2 emissions, tones.

According GUM 1993 the symbol "±" should be avoided whenever possible because it has traditionally been used to indicate an interval corresponding with expanded uncertainty.

In GHG Protocol *measurement uncertainty* is usually presented as an uncertainty range, i.e. an interval expressed in ± percent of the mean value reported (e.g. 100 t ± 5 %). The likely causes of *uncertainty with direct measurement* are generally related to the measurement techniques used. Methods with a high degree of variability will typically lead to a high degree of statistical uncertainty in the final estimates. In the case of *indirect measurement* the uncertainties are related to the AD, and the EF.

The aggregation of uncertainties using this approach is facilitated by the GHG Protocol, which provides automated worksheets for directly and indirectly measured emissions.

For user that characterizes uncertainty numerically, a sum of squares approach may be used to calculate the confidence interval for the product of two or more factors. This approach is only valid if the uncertainties follow a normal distribution and if the individual uncertainties are less than 60%. The relative confidence interval (the ± percent) of the product is the square root of the sum of the squares of the relative (percent) confidence intervals of each factor.

### **6. The use of measurement units in environmental guides**

SI units are recommended for use throughout science, technology and commerce. Each physical quantity has only one SI unit, even if this unit can be expressed in different forms. In some case the same SI unit can be used to express the values of several different quantities. The SI adopted series of prefixes for use in forming the decimal multiples and submultiples of SI units.

The International Committee of Weights and Measures (CIPM), recognizing that users would wish to employ the SI with units which are not part of it but are important and widely used, listed three categories of non-SI units: units to be maintained; to be tolerated temporarily; and to be avoided.

In reviewing this categorization the CIPM agreed a new classification of non-SI units: units accepted for use with the SI (for environmental guides – tonne or "metric ton"); units accepted for use with the SI whose values are obtained experimentally (-); and other units currently accepted for use with the SI to satisfy the needs of special interests (hectare). Non-SI unit tonne is accepted for use with the SI. It includes units, which are in continuous

1

According GUM 1993 the symbol "±" should be avoided whenever possible because it has traditionally been used to indicate an interval corresponding with expanded uncertainty.

In GHG Protocol *measurement uncertainty* is usually presented as an uncertainty range, i.e. an interval expressed in ± percent of the mean value reported (e.g. 100 t ± 5 %). The likely causes of *uncertainty with direct measurement* are generally related to the measurement techniques used. Methods with a high degree of variability will typically lead to a high degree of statistical uncertainty in the final estimates. In the case of *indirect measurement* the

The aggregation of uncertainties using this approach is facilitated by the GHG Protocol, which provides automated worksheets for directly and indirectly measured emissions.

For user that characterizes uncertainty numerically, a sum of squares approach may be used to calculate the confidence interval for the product of two or more factors. This approach is only valid if the uncertainties follow a normal distribution and if the individual uncertainties are less than 60%. The relative confidence interval (the ± percent) of the product is the square root of the sum of the squares of the relative (percent) confidence

SI units are recommended for use throughout science, technology and commerce. Each physical quantity has only one SI unit, even if this unit can be expressed in different forms. In some case the same SI unit can be used to express the values of several different quantities. The SI adopted series of prefixes for use in forming the decimal multiples and

The International Committee of Weights and Measures (CIPM), recognizing that users would wish to employ the SI with units which are not part of it but are important and widely used, listed three categories of non-SI units: units to be maintained; to be tolerated

In reviewing this categorization the CIPM agreed a new classification of non-SI units: units accepted for use with the SI (for environmental guides – tonne or "metric ton"); units accepted for use with the SI whose values are obtained experimentally (-); and other units currently accepted for use with the SI to satisfy the needs of special interests (hectare). Non-SI unit tonne is accepted for use with the SI. It includes units, which are in continuous

*i U HI M* 

*Ii* is percentage of uncertainty of calculated emissions from *i*-th source;

**6. The use of measurement units in environmental guides** 

*n*

2

(),

*i i*

Calculating aggregation of uncertainties uses equation:

*Hi* is CO2 emissions from *i*-th source, tones;

uncertainties are related to the AD, and the EF.

*М* is total CO2 emissions, tones.

intervals of each factor.

submultiples of SI units.

temporarily; and to be avoided.

where:

everyday use, in particular the traditional units of time and of angle, together with a few other units, which have assumed increasing technical importance (Tailor B. N., 1995).

The series of international standards on air quality includes the standardization of methods for the measurement of gases, vapours and particles (for example, ISO 4226). In order to enable results to be compared either between countries, it is essential to use agreed of measurement units to report the results and other relevant information. It is also desirable to keep the number of measurement units to a minimum. Those international standards lay down the units and symbols to be used when reporting results of air quality measurements.

Used in environmental guides and international standards (GPG 2000, IPCC 2006, ISO 4226) SI units, and also non-SI units and their conversion factors are shown in Table 7 (Velychko O. & Gordiyenko T., 2005; Velychko O. & Gordiyenko T., 2007a).


The Uncertainty Estimation and Use of Measurement Units

prepared with preferably employment of SI units.

statistical terms. *ISO*, 1993.

and removals. *ISO*, 2006.

Inventories (GPG 2000). *IPCC*, Switzerland, 2000.

Part 1: General principles and definitions. *ISO*, 1994.

emission reductions or removal enhancements. *ISO*, 2006.

and verification of greenhouse gas assertions. *ISO*, 2006.

*Measurement* (AMUEM 2006), Sardagna-Trento, Italy, 5 р.

Sustainable Development, *World Resources Institute*, 2003.

The International Systems of Units (SI): 8th Edition. *BIPM*, 2006, 180 p.

uncertainty in measurement (GUM 1993). *ISO*, 2008.

and associated terms (VIM 2007). *ISO*, 2007.

Publication 811, 1995 Edition, 84 p.

guides is recommended.

**8. References** 

2006.

in National Inventories of Anthropogenic Emission of Greenhouse Gas 213

analysis of uncertainty. Also need to use approaches international metrological guide GUM 1993 for uncertainty assessment for GHG inventory with correlation or covariance of input values in the development of new and reconsideration of old international environmental

SI units as well as traditional non-SI units are used in environmental guides. It considerably complicates the preparation of national experts of ecological information and its comparative analysis. Environmental guides that are used on international level must be

2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006). *IGES,* Japan,

IPCC Good Practice Guidance and Uncertainty Management in National Greenhouse Gas

ISO3534-1:1993. Statistics. – Vocabulary and symbols. – Part 1: Probability and general

ISO 5725-1:1994. Accuracy (trueness and precision) of measurement methods and results. –

ISO 14064-1:2006. Greenhouse gas emissions. Part 1: Specification with guidance at the

ISO 14064-2:2006. Greenhouse gas emissions. Part 2: Specification with guidance at the

ISO 14064-3:2006. Greenhouse gases. Part 3: Specification with guidance for the validation

ISO/IEC Guide 98-3:2008 Uncertainty of measurement. – Part 3: Guide to the expression of

ISO/IEC Guide 99:2007 International vocabulary of metrology. – Basic and general concepts

Gordiyenko T. & Velychko O. (2006) Peculiarities of Using Uncertainty in Environmental

Greenhouse Gas Protocol Guidance on uncertainty assessment in GHG inventories and

Tailor B. N. (1995) Guide for the Use of the International System of Units (SI), *NIST* Special

Velychko O. & Gordiyenko T. (2005) Peculiarities of using metrological terms and SI units in

Guides, *International Workshop on Advanced Methods for Uncertainty Estimation in* 

calculating statistical parameter uncertainty (GHG Protocol). Business Council for

environmental guides. *Joint International IMEKO TC1+TC7 Symposium "Metrology and Measurement Education in the Internet Era"*. Ilmenau, Germany, pp. 124–127.

organization level for quantification and reporting of greenhouse gas emissions

project level for quantification, monitoring and reporting of greenhouse gas

ISO 4226:1993. Air quality. – General aspects. – Units of measurement. *ISO*, 1993.


Table 7. Used units for quantities and their conversion factors in environmental guides and international standards

## **7. Conclusion**

Metrological and environmental international guides and standards comparison has shown peculiarities of used terms and uncertainty analysis with correlation and covariance of input values. Metrological and environmental guides use two types of uncertainty, and also used the Monte Carlo analysis of uncertainty. Environmental guides use more simplified approaches of the uncertainty analysis and also specifically use of data, which based on many models. Specific approaches to the uncertainty estimation in trend of GHG emissions imply realization of long-term observation and also environmental guide's consideration correlation of data across years.

It is necessary to apply international metrological guide to the expression of uncertainty in measurement – GUM 1993 and international vocabulary of metrology – VIM 2007, which are developed by eight international organizations in the field metrology, standardization, physicists and chemistry, during the preparation of environmental guides concerning the analysis of uncertainty. Also need to use approaches international metrological guide GUM 1993 for uncertainty assessment for GHG inventory with correlation or covariance of input values in the development of new and reconsideration of old international environmental guides is recommended.

SI units as well as traditional non-SI units are used in environmental guides. It considerably complicates the preparation of national experts of ecological information and its comparative analysis. Environmental guides that are used on international level must be prepared with preferably employment of SI units.

## **8. References**

212 Greenhouse Gases – Emission, Measurement and Management

British thermal unit (Btu)

Tonne of oil equivalent (toe)

Kilowatt-hour

Horse power

Year (yr) –

(kWh)

(HP)

(gal)

(°C)

(atm)

\* if concentrations are expressed in terms of mass per unit volume, temperature and

Table 7. Used units for quantities and their conversion factors in environmental guides and

Metrological and environmental international guides and standards comparison has shown peculiarities of used terms and uncertainty analysis with correlation and covariance of input values. Metrological and environmental guides use two types of uncertainty, and also used the Monte Carlo analysis of uncertainty. Environmental guides use more simplified approaches of the uncertainty analysis and also specifically use of data, which based on many models. Specific approaches to the uncertainty estimation in trend of GHG emissions imply realization of long-term observation and also environmental guide's consideration

It is necessary to apply international metrological guide to the expression of uncertainty in measurement – GUM 1993 and international vocabulary of metrology – VIM 2007, which are developed by eight international organizations in the field metrology, standardization, physicists and chemistry, during the preparation of environmental guides concerning the

1 Btu = 1055.056 J 1 GJ = 109 J 1 Tj = 1012 J

1 toe = 1·1010 calIT =

1 ktoe = 41.868 TJ

1 kWh = 3,6·106 J 1 TJ = 2.78·105 kWh

1 kW = 103 W 1 MW = 106 W 1 GW = 109 W 1 HP = 735.499 W

1 dm3 = 10-3 m3 1 gal dry(US) = 4.405 dm3

Hour (h) 1 h = 60 min = 3600 s Day (d) 1 d = 24 h = 1440 min

1 °C = 1 K

1 kPa = 103 Pa 1 atm = 101.325 kPa

41.868 GJ

**Units for quantity SI units Non-SI units Conversion factor**  Units for energy Joule (J): GJ, TJ CalorieIT (calIT) 1 calIT = 4.1868 J

Units for square Square metre (m2) Hectare (ha) 1 ha = 104 m2

Units for time Second (s) Minute (min) 1 min = 60 s

Kelvin (K) Degree Celsius

Units for volume Cubic metre (m3) Dry gallon US

Units for pressure Pascal (Pa): kPa Atmosphere

pressure (as well as humidity) are required.

Units for power Watt (W): kW, MW, GW

Units for temperature

international standards

correlation of data across years.

**7. Conclusion** 


**10**

*Brazil* 

Marcelo Sthel et al.\*

**Detection of Greenhouse Gases Using**

*State University of the Northern Fluminense Darcy Ribeiro (UENF)* 

Global warming is one of the main environmental problems of the XXI Century. It is generated by the intensive use of fossil fuels that leads to the so-called greenhouse effect (Meinshausen et al.,2009; Allen et al., 2009; Sthel et al., 2010; Hansen et al., 2008; Hansen & Makiko, 2004; Rosenzweig et al, 2008). It causes climate changes (Steffen et al., 2004; Solomon et al.,2009; Kevin et al., 2006; Kurz et al.,2008; Greene et al., 2009; Nathan P et al., 2008; Siddall et al., 2009; Sander et al, 2006; Emanuel, 2005; Peza & Simmonds, 2005; Janssen, 1998; Mann et al., 1998) and produces significant damages to the human society and biodiversity, such as the melting of the poles with the consequent increasing of oceans level, the intensity increasing of hurricanes, extreme events, changes in the rainfall patterns (floods, desertification), oceans acidification and biodiversity decreasing. Recently, the Intergovernmental Panel on Climate Change (IPCC) report (IPCC, 2007), published in February 2007, indicated based on meteorological studies, that the earth's global average temperature rose about 0.8oC in the last 150 years, mainly due to human activities. The global atmospheric concentrations of carbon dioxide(CO2), methane (CH4) and nitrous oxide (N2O)have risen considerably as a result of human activities since 1750. Increases in global carbon dioxide concentration are primarily due to the use of fossil fuels and changes in the use of the soil through practices such as deforestation, biomass burning and biomass decomposition. Practices such as logging, peat decomposition and burning and other techniques employed in the modern agriculture are resulting in an increased concentration of methane and nitrous oxide. While the global atmospheric concentration of carbon dioxide has increased from a pre-industrial value of about 280 ppmv to 379 ppmv in 2005, the global atmospheric concentration of methane (CH4) has increased from a pre-industrial value of about 715 ppbv to 1774 ppbv in 2005 and the global atmospheric concentration of nitrous

oxide has risen from pre-industrial value of about 270 ppbv to 319 ppbv in 2005.

Marcelo Gomes1, Guilherme Lima1, Mila Vieira1, Juliana Rocha1, Delson Schramm1,

<sup>1</sup>*State University of the Northern Fluminense Darcy Ribeiro (UENF), Brazil*

Maria Priscila Castro1, Andras Miklos2, Helion Vargas1

2*Fraunhofer Institute for Building Physics, Germany*

Therefore, it is necessary to use suitable analytical techniques to identify the atmospheric components and to determinate their trace concentrations. A trace sensor of atmospheric pollutants must meet a set of fundamental requisites. High selectivity is necessary to

**1. Introduction** 

\*

**the Photoacoustic Spectroscopy**


## **Detection of Greenhouse Gases Using the Photoacoustic Spectroscopy**

Marcelo Sthel et al.\*

*State University of the Northern Fluminense Darcy Ribeiro (UENF) Brazil* 

## **1. Introduction**

214 Greenhouse Gases – Emission, Measurement and Management

Velychko O. & Gordiyenko T. (2007a) The use of metrological terms and SI units in

Velychko O. & Gordiyenko T. (2007b) Estimation of uncertainty with correlation values in

Velichko О. N. & Gordienko T. B. (2007c) Comparisons of uncertainty assessment in

Velychko O. M. & Gordiyenko T. B. (2008) Uncertainty estimation at determination of pollutant emission, *7th Intern. Symp. "Metrology 2008"*. Havana, Cuba, 12 p. Velychko O. (2008) Using of Validated Software for Uncertainty Analyses Tools in Accredited Laboratories, *Key Engineering Materials*, Vols. 381–382, pp. 599–602. Velychko O. & Gordiyenko T. (2009) The use of guide to the expression of uncertainty in

Velichko O. N. & Gordienko T. B. (2009) Methodologies of calculating of pollutants emission

Velychko O. & Gordiyenko T. (2010) The implementation of general guides and standards

Velychko O. & Gordiyenko T. (2011) Greenhouse Gases Global Monitoring Systems:

(IMEKO-MI2011), Cavtat, Dubrovnik Riviera, Croatia, p. 125–130.

(February 2007), pp. 202–212.

Vol.50, No 5 (2007), pp. 490–495.

No2 (February 2009), pp. 193–199.

238., Numb. 1, 012044, 6 p.

Romania, pp. 8–12.

517.

environmental guides and international standards. *Measurement*. Vol.40, Issue2

international metrological and environmental guides, *1st IMEKO TC19 Intern. Symp. on Measurements and Instrumentation for Environmental Monitoring*. Proceeding. Iasi,

international metrological and environmental guides, *Measurement Techniques*,

measurement for uncertainty management in National Greenhouse Gas Inventories, *International Journal of Greenhouse Gas Control*, Vol.3, Issue4, pp. 514-

in atmosphere and their uncertainty assessment, *Measurement Techniques*. Vol.52,

on regional level in the field of metrology. Journal of Physics: Conf. Series, Vol.

Ecological and Metrological Aspects. Proceedings Joint IMEKO TC 11–TC 19–TC 20 Intern. Symp. "*Metrological Infrastructure, Environmental and Energy Measurement*" Global warming is one of the main environmental problems of the XXI Century. It is generated by the intensive use of fossil fuels that leads to the so-called greenhouse effect (Meinshausen et al.,2009; Allen et al., 2009; Sthel et al., 2010; Hansen et al., 2008; Hansen & Makiko, 2004; Rosenzweig et al, 2008). It causes climate changes (Steffen et al., 2004; Solomon et al.,2009; Kevin et al., 2006; Kurz et al.,2008; Greene et al., 2009; Nathan P et al., 2008; Siddall et al., 2009; Sander et al, 2006; Emanuel, 2005; Peza & Simmonds, 2005; Janssen, 1998; Mann et al., 1998) and produces significant damages to the human society and biodiversity, such as the melting of the poles with the consequent increasing of oceans level, the intensity increasing of hurricanes, extreme events, changes in the rainfall patterns (floods, desertification), oceans acidification and biodiversity decreasing. Recently, the Intergovernmental Panel on Climate Change (IPCC) report (IPCC, 2007), published in February 2007, indicated based on meteorological studies, that the earth's global average temperature rose about 0.8oC in the last 150 years, mainly due to human activities. The global atmospheric concentrations of carbon dioxide(CO2), methane (CH4) and nitrous oxide (N2O)have risen considerably as a result of human activities since 1750. Increases in global carbon dioxide concentration are primarily due to the use of fossil fuels and changes in the use of the soil through practices such as deforestation, biomass burning and biomass decomposition. Practices such as logging, peat decomposition and burning and other techniques employed in the modern agriculture are resulting in an increased concentration of methane and nitrous oxide. While the global atmospheric concentration of carbon dioxide has increased from a pre-industrial value of about 280 ppmv to 379 ppmv in 2005, the global atmospheric concentration of methane (CH4) has increased from a pre-industrial value of about 715 ppbv to 1774 ppbv in 2005 and the global atmospheric concentration of nitrous oxide has risen from pre-industrial value of about 270 ppbv to 319 ppbv in 2005.

Therefore, it is necessary to use suitable analytical techniques to identify the atmospheric components and to determinate their trace concentrations. A trace sensor of atmospheric pollutants must meet a set of fundamental requisites. High selectivity is necessary to

<sup>\*</sup> Marcelo Gomes1, Guilherme Lima1, Mila Vieira1, Juliana Rocha1, Delson Schramm1,

Maria Priscila Castro1, Andras Miklos2, Helion Vargas1

<sup>1</sup>*State University of the Northern Fluminense Darcy Ribeiro (UENF), Brazil*

<sup>2</sup>*Fraunhofer Institute for Building Physics, Germany*

Detection of Greenhouse Gases Using the Photoacoustic Spectroscopy 217

Lasers, sensitive microphones and lock-in amplifiers permitted this technique to have a great technical development. In 1968, L. B. Kerr and J. G. Atwood conducted the first photoacoustic experiments using Laser as the radiation source (Kerr & Atwood, 1968). Using a continuous wave CO2 laser as excitation source, they achieved a minimum absorption coefficient of 1.2 x 10-7 cm-1 of CO2 diluted in nitrogen. In 1971, L. B. Kreuzer, using a HeNe laser operating on 3.39 μm, attained a detection limit of 10ppbv of methane diluted in nitrogen (Kreuzer, 1971). Kreuzer, N. D. Kenyon and C. N. K. Patel used CO2 and CO lasers as radiation sources to perform photoacoustic measurements of many trace gases (Kreuzer et al., 1972). The use of a photoacoustic cell operating at in resonant mode, associated with the modulation of the excitation beam in the acoustic resonance frequencies of the cell, was introduced in 1973 by CF Dewey Jr., RD Kahn and CE Hackett (Dewey et al., 1973). The success achieved by these pioneering studies provided a great interest in the application of photoacoustic spectroscopy in trace gases analysis in many

Advances in laser technology allowed the development of new infrared radiation sources continuously operating in a wide range of wavelengths, especially in the mid-infrared region. Among these new Lasers, the most promising for gas sensing are the optical parametric oscillator (OPO) and the quantum cascade laser (QCL). The combination of these new radiation sources with photoacoustic cell optimized settings (differential photoacoustic cell, intra-cavity arrangements or multi-pass) enabled major advances in trace gases

The photoacoustic signal generated and detection in gases been studied mainly by Kreuzer (Kreuzer, 1971) and revised and expanded by several authors( Harren & Reuss, 1997; Miklos & Hess, 2001; Hess, 1992; West, 1983; Meyer & Sigrist, 1990). Molecular absorption of photons results in the excitation of molecular energy levels (rotational, vibrational and electronic) degrees freedom. The excited state loses its energy by radiative processes, such as spontaneous or stimulated emission, and/or by collisional relaxation, in which the state

In the case of vibrational excitation, radiative emission and chemical reactions do not play an important role, because the radiative lifetimes of vibrational levels are long compared with the time needed for collisional deactivation at pressures used in photoacoustic. Furthermore, the photon energy is too small to induce chemical reactions. For 1 atm pressure, the vibrational-translational non-radiative decay time is typically around 10-6-10-9s, whereas the radiative lifetime is between 10-1 and 10-3 s (Hess, 1983). Thus, in practice, the absorbed energy is completely released as heat, appearing as translational (kinetic) energy

By modulating the intensity or the wavelength of the incident radiation, the sample local heating and expansion become periodic. If radiation intensity is modulated (without optical saturation), the heat density in the sample (H) is directly proportional to the volumetric density of molecules (N), to the absorption cross section of the absorbing molecule (σ) and to the incident laser radiation intensity (I0). Therefore, the gas heat production is given by:

scientific areas, especially in pollutant gas studies.

photoacoustic detection.

of the gas molecules.

**2.1 Photoacoustic effect in gases** 

energy is transformed into translational one.

distinguish the gas species present in a multicomponent gas mixture, such as air, and high sensitivity is essential to detect very low concentrations of substances. A large dynamic range is important to monitor the gas components at high and low concentrations using a unique instrument. In addition, a good time resolution ensures the possibility of on-line analyses controlled by a computer. Photoacoustic spectroscopy meets these requirements that enable this technique to offer important advantages in pollutant gas monitoring.

In conventional spectroscopy, the absorption of radiation is measured from the power transmitted through the sample. On the contrary, in photoacoustic spectroscopy, the absorbed radiation power is determined directly via its heat and hence the sound produced in the sample. This methodology is based on the so called photoacoustic effect which consists on the generation and detection of pressure waves (sound) inside a resonant cell, where the gas samples are placed. These samples are exposed to the incidence of modulated radiation, absorbing it at determined wavelengths. The resonant absorption of radiation generates a modulated heating in the sample and, therefore, a sound signal is produced (photoacoustic effect) and detected by highly sensitive microphones coupled, inside the cell. These microphones convert the sound signal into an electric signal, which is filtered and detected by a lock-in amplifier. Photoacoustic spectroscopy is widely used for the detection of several gases in the concentration range of ppbv and sub-ppbv( Mothé et al., 2010; Berrou et al., 2010; Sthel et al.,2011; Harren et al.,2008; Thomas, 2006; Rocha et al., 2010; Elia et al., 2009; Sorvajärvi et al., 2009; Sigrist et al., 2008; Angelmahr et al.,2006; Filho et al., 2006; Schramm et al., 2003; Harren et al., 2000; Miklos et al., 2001; Gondal, 1997; Repond & Sigrist, 1996; Sigrist, 1994a, 1994b). There are some types of trace gas detection systems based on continuous wave (CW) CO2 laser, optical parametric oscillator (OPO) in combination with photoacoustic spectroscopy and quantum cascade laser (QCL). These experimental arrangements are efficient in the detection of greenhouse gases and their precursors.

## **2. Photoacoustic spectroscopy**

Currently, photoacoustic spectroscopy has been consolidated as an effective option for trace gases analysis for high sensitivity trace gases analysis (detection limits in the range of subppbv and ppbv), good selectivity, possibility of in situ measurements and continuous flow systems associated with the possibility of non-destructive analysis are attributes that allow photoacoustic to be a powerful analytical tool for gases monitoring. The photoacoustic effect consists on the generation of sound waves from the absorption of a pulsed modulated radiation. It was discovered in 1880 by Alexander Graham Bell (Bell, 1880). This discovery raised the interests of other researchers, such as John Tyndall (Tyndall, 1881), Wilhelm Röntgen (Röntgen, 1881) and Lord Rayleigh (Rayleigh,1881). However, the lack of equipment (radiation sources, microphones, amplifiers, etc.) prevented the immediate development of this new research field, and soon the photoacoustic effect has become a mere scientific curiosity, remained virtually forgotten for over half a century.

In the late 1930s, Viengerov (Viengerov et al., 1938) introduced an infrared absorption photoacoustic system to analyze gases infrared absorption. Then, Luft (Luft,1943) improved the sensitivity of this technique, allowing absorption measurements of gaseous species in the concentration range of ppmV (10-6). Since the 1960s, the development of

distinguish the gas species present in a multicomponent gas mixture, such as air, and high sensitivity is essential to detect very low concentrations of substances. A large dynamic range is important to monitor the gas components at high and low concentrations using a unique instrument. In addition, a good time resolution ensures the possibility of on-line analyses controlled by a computer. Photoacoustic spectroscopy meets these requirements

In conventional spectroscopy, the absorption of radiation is measured from the power transmitted through the sample. On the contrary, in photoacoustic spectroscopy, the absorbed radiation power is determined directly via its heat and hence the sound produced in the sample. This methodology is based on the so called photoacoustic effect which consists on the generation and detection of pressure waves (sound) inside a resonant cell, where the gas samples are placed. These samples are exposed to the incidence of modulated radiation, absorbing it at determined wavelengths. The resonant absorption of radiation generates a modulated heating in the sample and, therefore, a sound signal is produced (photoacoustic effect) and detected by highly sensitive microphones coupled, inside the cell. These microphones convert the sound signal into an electric signal, which is filtered and detected by a lock-in amplifier. Photoacoustic spectroscopy is widely used for the detection of several gases in the concentration range of ppbv and sub-ppbv( Mothé et al., 2010; Berrou et al., 2010; Sthel et al.,2011; Harren et al.,2008; Thomas, 2006; Rocha et al., 2010; Elia et al., 2009; Sorvajärvi et al., 2009; Sigrist et al., 2008; Angelmahr et al.,2006; Filho et al., 2006; Schramm et al., 2003; Harren et al., 2000; Miklos et al., 2001; Gondal, 1997; Repond & Sigrist, 1996; Sigrist, 1994a, 1994b). There are some types of trace gas detection systems based on continuous wave (CW) CO2 laser, optical parametric oscillator (OPO) in combination with photoacoustic spectroscopy and quantum cascade laser (QCL). These experimental arrangements are efficient in the

Currently, photoacoustic spectroscopy has been consolidated as an effective option for trace gases analysis for high sensitivity trace gases analysis (detection limits in the range of subppbv and ppbv), good selectivity, possibility of in situ measurements and continuous flow systems associated with the possibility of non-destructive analysis are attributes that allow photoacoustic to be a powerful analytical tool for gases monitoring. The photoacoustic effect consists on the generation of sound waves from the absorption of a pulsed modulated radiation. It was discovered in 1880 by Alexander Graham Bell (Bell, 1880). This discovery raised the interests of other researchers, such as John Tyndall (Tyndall, 1881), Wilhelm Röntgen (Röntgen, 1881) and Lord Rayleigh (Rayleigh,1881). However, the lack of equipment (radiation sources, microphones, amplifiers, etc.) prevented the immediate development of this new research field, and soon the photoacoustic effect has become a

In the late 1930s, Viengerov (Viengerov et al., 1938) introduced an infrared absorption photoacoustic system to analyze gases infrared absorption. Then, Luft (Luft,1943) improved the sensitivity of this technique, allowing absorption measurements of gaseous species in the concentration range of ppmV (10-6). Since the 1960s, the development of

mere scientific curiosity, remained virtually forgotten for over half a century.

that enable this technique to offer important advantages in pollutant gas monitoring.

detection of greenhouse gases and their precursors.

**2. Photoacoustic spectroscopy** 

Lasers, sensitive microphones and lock-in amplifiers permitted this technique to have a great technical development. In 1968, L. B. Kerr and J. G. Atwood conducted the first photoacoustic experiments using Laser as the radiation source (Kerr & Atwood, 1968). Using a continuous wave CO2 laser as excitation source, they achieved a minimum absorption coefficient of 1.2 x 10-7 cm-1 of CO2 diluted in nitrogen. In 1971, L. B. Kreuzer, using a HeNe laser operating on 3.39 μm, attained a detection limit of 10ppbv of methane diluted in nitrogen (Kreuzer, 1971). Kreuzer, N. D. Kenyon and C. N. K. Patel used CO2 and CO lasers as radiation sources to perform photoacoustic measurements of many trace gases (Kreuzer et al., 1972). The use of a photoacoustic cell operating at in resonant mode, associated with the modulation of the excitation beam in the acoustic resonance frequencies of the cell, was introduced in 1973 by CF Dewey Jr., RD Kahn and CE Hackett (Dewey et al., 1973). The success achieved by these pioneering studies provided a great interest in the application of photoacoustic spectroscopy in trace gases analysis in many scientific areas, especially in pollutant gas studies.

Advances in laser technology allowed the development of new infrared radiation sources continuously operating in a wide range of wavelengths, especially in the mid-infrared region. Among these new Lasers, the most promising for gas sensing are the optical parametric oscillator (OPO) and the quantum cascade laser (QCL). The combination of these new radiation sources with photoacoustic cell optimized settings (differential photoacoustic cell, intra-cavity arrangements or multi-pass) enabled major advances in trace gases photoacoustic detection.

#### **2.1 Photoacoustic effect in gases**

The photoacoustic signal generated and detection in gases been studied mainly by Kreuzer (Kreuzer, 1971) and revised and expanded by several authors( Harren & Reuss, 1997; Miklos & Hess, 2001; Hess, 1992; West, 1983; Meyer & Sigrist, 1990). Molecular absorption of photons results in the excitation of molecular energy levels (rotational, vibrational and electronic) degrees freedom. The excited state loses its energy by radiative processes, such as spontaneous or stimulated emission, and/or by collisional relaxation, in which the state energy is transformed into translational one.

In the case of vibrational excitation, radiative emission and chemical reactions do not play an important role, because the radiative lifetimes of vibrational levels are long compared with the time needed for collisional deactivation at pressures used in photoacoustic. Furthermore, the photon energy is too small to induce chemical reactions. For 1 atm pressure, the vibrational-translational non-radiative decay time is typically around 10-6-10-9s, whereas the radiative lifetime is between 10-1 and 10-3 s (Hess, 1983). Thus, in practice, the absorbed energy is completely released as heat, appearing as translational (kinetic) energy of the gas molecules.

By modulating the intensity or the wavelength of the incident radiation, the sample local heating and expansion become periodic. If radiation intensity is modulated (without optical saturation), the heat density in the sample (H) is directly proportional to the volumetric density of molecules (N), to the absorption cross section of the absorbing molecule (σ) and to the incident laser radiation intensity (I0). Therefore, the gas heat production is given by:

$$H(\mathbf{r}, t) = N \sigma I\_0 \mathbf{e}^{i \alpha t} \tag{1}$$

Detection of Greenhouse Gases Using the Photoacoustic Spectroscopy 219

In order to determine the concentration of various gaseous species in a multicomponent sample, the photoacoustic signal can be obtained for different wavelengths, corresponding to the absorption of each analyzed component. In this case, the signal can be obtained by the

() ()

at λi. This equation solution for each gas species concentration is given by:

*c*

photoacoustic spectrum of multicomponent samples (Moeckli, 1998).

**2.2 Lasers as radiation sources** 

Parametric Oscillator lasers (OPO).

resolution.

*S CP N c*

1

1

*S*

*i i*

 

(5)

(6)

*n*

*i i j ij j*

with i = 1, 2, ..., m and j = 1, 2, ..., n e m ≥ n.Here, Pi =P(λi) represents the laser power at wavelength λi and cj is the concentration of the component j with absorption cross section σij

> 1 <sup>1</sup> ( ) *<sup>m</sup> <sup>i</sup>*

Where (σij)-1 is the inverse matrix of the matrix σij. The success of this method in multicomponent mixture analysis strongly depends on the nature of the matrix σij (Sigrist, 1994a; Meyer et al., 1988). The trivial case is represented by a diagonal matrix (σij), where a set of wavelengths λi can be selected and the absorption by a single gas component occurs in each λi. However, this ideal case without interference between the absorptions due to different gas species in a multicomponent mixture hardly occurs. Instead, considerable effort is needed to discriminate the various components, as well as to identify any unexpected component. Iterative algorithms using least squares regression and based on prior knowledge of the absorption cross sections (σij) have been developed to fit the

The selectivity required for the identification of the components of a gaseous mixture is achieved in spectroscopy when lasers are employed. Due to its high radiance, narrow linewidth and wide spectral range of emission, lasers are indeed suitable for trace gas monitoring. They allow a selective detection of different gases in trace level concentration (lower limit: parts per trillion by volume) even when there is overlapping of spectrum lines from different molecules. It is known that for molecules, absorption spectra in the near infrared (NIR) and the infrared (IR) ranges work as fingerprints, that is, they are unique for each molecular species. Currently, there are commercially available lasers that emit in these regions, for example, semiconductor lasers (DFB diode lasers) and gas lasers (CO and CO2), respectively. The great advantage of using these light sources is the resulting high spectral

Promising new sources of light are the quantum cascade lasers (QCL). These lasers emit in the mid-IR range, where the molecules have higher absorption coefficients, and operate at room temperature. Since the photoacoustic signal intensity is also a function of the molecular absorption coefficients, the low power these lasers emit is hence compensated. Another important feature and advantage of these lasers is the high spectral resolution. Furthermore, the emission wavelength can be selected to match two, or more, absorption lines of a given molecule. Third generations of extremely promising lasers are the Optical

*CN P* 

*j ij*

relation:

The sample pressure oscillations (p) are related to the heat production by the following equation:

$$\nabla^2 p - \frac{1}{\nu^2} \frac{\partial^2 p}{\partial t^2} = -\frac{\mathcal{V} - 1}{\nu^2} \frac{\partial H}{\partial t} \tag{2}$$

where ν, γ, and H are the sound velocity, the adiabatic co- efficient of the gas, and the heat density deposited in the gas by light absorption, respectively. All dissipative terms of heat diffusion to the cell walls and viscosity were despised.

For many applications, this simplified approach is sufficient, although in some cases, more stringent treatment that takes into account relaxation effects is necessary. Since the amplitude of the acoustic waves produced depends on both the nature of the absorbing gases and their concentrations, the photoacoustic detection allows qualitative and quantitative analysis of gas mixtures.

The acoustic wave is generated in places where light is absorbed by the monitored species. The photoacoustic signal S(λ) produced by a single absorbing gaseous species diluted in a non absorbing gas can be expressed as:

$$S(\mathcal{A}) = \mathbb{C}P(\mathcal{A})N\_{\text{byl}}\mathbb{C}\sigma(\mathcal{A}) = \mathbb{C}P(\mathcal{A})\alpha(\mathcal{A})\mathfrak{c} \tag{3}$$

Where C is the cell constant, which depends on the cell geometry, the microphone responsivity and on the nature of the acoustic mode, P(λ) is the laser power at the wavelength λ, Ntot is the total density of molecules, considering a pressure of 1013hPa and a temperature of 20°C, Ntot ≈ 2.5 x 1019 molecules/cm3, c and σ (λ) are, respectively, the concentration (mole fraction) and the absorption cross section of the absorbing molecule and α(λ)=Ntot.cσ(λ) is the so called gas absorption coefficient (cm-1).

Using equation 3, it is possible to determine the minimum detectable gas concentration in the photoacoustic spectrometer, through:

$$\mathcal{L}\_{\text{min}} = \frac{\mathcal{S}\_{\text{min}}}{N\_{\text{tot}} \text{CP} \sigma} = \frac{\mathcal{S}\_{\text{min}}}{\text{CPa}} \tag{4}$$

where Smin (λ) is the minimum detectable signal, which is usually measured by passing a flow of non-absorbing inert gas (usually nitrogen or synthetic air) in the photoacoustic cell. The minimum signal is produced by the various sources of noise, which are always present in photoacoustic detection, determining its limitations. The most common sources of noise in photoacoustic systems include acoustic signals caused by the windows heating, the absorption and scattering of radiation on the cell resonator walls, by molecules adsorbed on them, the noise caused by the gas flow and electronic noise.

We can notice that the signal is obtained at a given wavelength, which is specific for the molecule we wish to detect. Moreover, the obtained signal is directly proportional to the concentration of the absorbing gas. Thus, it is possible to obtain the concentration as a function of the generated signal.

<sup>0</sup> (,) e*i t H t NI*

 **r** 

The sample pressure oscillations (p) are related to the heat production by the following

22 2 1 1 *p H <sup>p</sup> <sup>t</sup> <sup>t</sup>* 

where ν, γ, and H are the sound velocity, the adiabatic co- efficient of the gas, and the heat density deposited in the gas by light absorption, respectively. All dissipative terms of heat

For many applications, this simplified approach is sufficient, although in some cases, more stringent treatment that takes into account relaxation effects is necessary. Since the amplitude of the acoustic waves produced depends on both the nature of the absorbing gases and their concentrations, the photoacoustic detection allows qualitative and

The acoustic wave is generated in places where light is absorbed by the monitored species. The photoacoustic signal S(λ) produced by a single absorbing gaseous species diluted in a

() () () ()() *S CP N c CP c*

Where C is the cell constant, which depends on the cell geometry, the microphone responsivity and on the nature of the acoustic mode, P(λ) is the laser power at the wavelength λ, Ntot is the total density of molecules, considering a pressure of 1013hPa and a temperature of 20°C, Ntot ≈ 2.5 x 1019 molecules/cm3, c and σ (λ) are, respectively, the concentration (mole fraction) and the absorption cross section of the absorbing molecule and

Using equation 3, it is possible to determine the minimum detectable gas concentration in

*tot*

where Smin (λ) is the minimum detectable signal, which is usually measured by passing a flow of non-absorbing inert gas (usually nitrogen or synthetic air) in the photoacoustic cell. The minimum signal is produced by the various sources of noise, which are always present in photoacoustic detection, determining its limitations. The most common sources of noise in photoacoustic systems include acoustic signals caused by the windows heating, the absorption and scattering of radiation on the cell resonator walls, by molecules adsorbed on

We can notice that the signal is obtained at a given wavelength, which is specific for the molecule we wish to detect. Moreover, the obtained signal is directly proportional to the concentration of the absorbing gas. Thus, it is possible to obtain the concentration as a

min min

*S S*

*N CP CP* 

 

 

min

*c*

 

(2)

*tot* (3)

(4)

2

2

diffusion to the cell walls and viscosity were despised.

α(λ)=Ntot.cσ(λ) is the so called gas absorption coefficient (cm-1).

them, the noise caused by the gas flow and electronic noise.

quantitative analysis of gas mixtures.

non absorbing gas can be expressed as:

the photoacoustic spectrometer, through:

function of the generated signal.

equation:

(1)

In order to determine the concentration of various gaseous species in a multicomponent sample, the photoacoustic signal can be obtained for different wavelengths, corresponding to the absorption of each analyzed component. In this case, the signal can be obtained by the relation:

$$S(\mathcal{A}\_i) = \text{CP}(\mathcal{A}\_i) N \sum\_{j=1}^n c\_j \sigma\_{ij} \tag{5}$$

with i = 1, 2, ..., m and j = 1, 2, ..., n e m ≥ n.Here, Pi =P(λi) represents the laser power at wavelength λi and cj is the concentration of the component j with absorption cross section σij at λi. This equation solution for each gas species concentration is given by:

$$\sigma\_j = \frac{1}{\text{CN}} \sum\_{i=1}^{m} (\sigma\_{ij})^{-1} \left(\frac{S\_i}{P\_i}\right) \tag{6}$$

Where (σij)-1 is the inverse matrix of the matrix σij. The success of this method in multicomponent mixture analysis strongly depends on the nature of the matrix σij (Sigrist, 1994a; Meyer et al., 1988). The trivial case is represented by a diagonal matrix (σij), where a set of wavelengths λi can be selected and the absorption by a single gas component occurs in each λi. However, this ideal case without interference between the absorptions due to different gas species in a multicomponent mixture hardly occurs. Instead, considerable effort is needed to discriminate the various components, as well as to identify any unexpected component. Iterative algorithms using least squares regression and based on prior knowledge of the absorption cross sections (σij) have been developed to fit the photoacoustic spectrum of multicomponent samples (Moeckli, 1998).

#### **2.2 Lasers as radiation sources**

The selectivity required for the identification of the components of a gaseous mixture is achieved in spectroscopy when lasers are employed. Due to its high radiance, narrow linewidth and wide spectral range of emission, lasers are indeed suitable for trace gas monitoring. They allow a selective detection of different gases in trace level concentration (lower limit: parts per trillion by volume) even when there is overlapping of spectrum lines from different molecules. It is known that for molecules, absorption spectra in the near infrared (NIR) and the infrared (IR) ranges work as fingerprints, that is, they are unique for each molecular species. Currently, there are commercially available lasers that emit in these regions, for example, semiconductor lasers (DFB diode lasers) and gas lasers (CO and CO2), respectively. The great advantage of using these light sources is the resulting high spectral resolution.

Promising new sources of light are the quantum cascade lasers (QCL). These lasers emit in the mid-IR range, where the molecules have higher absorption coefficients, and operate at room temperature. Since the photoacoustic signal intensity is also a function of the molecular absorption coefficients, the low power these lasers emit is hence compensated. Another important feature and advantage of these lasers is the high spectral resolution. Furthermore, the emission wavelength can be selected to match two, or more, absorption lines of a given molecule. Third generations of extremely promising lasers are the Optical Parametric Oscillator lasers (OPO).

Detection of Greenhouse Gases Using the Photoacoustic Spectroscopy 221

The QCL laser coupled with a photoacoustic detector has been successfully used to measure the concentration of different gases which absorb radiation in the mid- infrared spectral region, such as ozone (Silva et al., 2004), ammonia (Baptista-Filho et al., 2006), and NO2 and

This type of laser is part of the family of semiconductor lasers, with the particularity of using quantum transitions within the same band. A quantum cascade laser comprises a series of alternate thin layers of two different materials. This configuration enables different electrical potentials to be established across the device, so that electrons can get trapped in these sites, called quantum wells. Thus, a series of sub-bands with different energies is

When stimulated, the electrons undergo transitions and quantum tunneling to a lower energy sub-band, and consequently photons are emitted. A single electron can perform several transitions, that is, issue multiple photons. (Gmachl et al., 2001; Beyer et al., 2003)

Another important characteristic of quantum cascade lasers is that the wavelength can be determined by the thickness of the layers, rather than being determined by the energy difference between bands. The thickness and refractive index change by setting different temperatures turning the wavelength (Kosterev et al., 2002; Curl et al., 2010; Kosterev et al.,

An interesting optical process that has been used to produce near- and mid-infrared radiation is the optical parametric generation. In 1962, Armstrong et al and Kroll described the fundamental theory of the optical parametric generator and three years later, Giordmaine and Miller demonstrated the operation of an optic parametric oscillator (Armstrong et al., 1962; Giordmaine & Miller, 1965). When an optically nonlinear crystal is submitted to electromagnetic fields of high density of energy such as those present in pulsed lasers, the electrons respond with significant displacement that gives rise to the contribution of the second-order nonlinear component for the electric polarization of a nonlinear medium. The mixing of two electromagnetic waves under condition of nonlinear polarization produces parametric effects such as second harmonic generation (frequency doubling), sum frequency generation and difference frequency generation. In the latter case (figure 2), when the nonlinear crystal is pumped by two input photons with wavelength at λp (pump photon) and λs (signal photon, λs > λp), the signal photon stimulates the conversion

of a pump photo into a new signal and an idler photon with wavelength at λ<sup>i</sup>

(fulfilling the energy conservation principle). Of course, the efficiency of such conversion is limited to phase match between the pump and signal waves. This process of increasing the number of signal photons is known as optical parametric amplification (OPA) and thus it fundamentally differs from the amplification mechanism in laser since no population inversion and excited states take place. Theoretically, the remarkable advantage of OPA is the infinity possibility of combining two waves that generates a third wave with different wavelength. This fulfills the expected desirable feature of an excitation source for analytical spectroscopy application that is the broad wavelength tunability. Therefore, this makes the optical parametric phenomena of great significance to spectroscopy application in the sense



N2O gases. (Lima et al, 2006; Grossel et al., 2007)

**2.2.3 The Optical Parametric Oscillator laser (OPO)** 

created inside the conduction band.

2002).

#### **2.2.1 CO2 laser**

The use of tunable CO2 lasers (Patel, 1964), through the scanning of its emission wavelengths (9-11μm) ensures the exploration of the so-called molecular fingerprint, enabling the identification and simultaneous monitoring of several gaseous compounds with a single instrument. Figure 1 shows the absorption spectra of ozone (O3), ammonia (NH3) and ethylene (C2H4), for the emission range of a CO2 laser. As gas lasers have high power (> 10 watts) and the signal intensity is directly proportional to the emitted light power, its use results in sensitivities in the range of pptv. Nevertheless, these lasers have two disadvantages, they are large and expensive and thus its use is limited to laboratories.

Fig. 1. Fingerprints of ozone (O3), ammonia (NH3) and ethylene (C2H4) corresponding to the emission lines of a CO2 laser.

#### **2.2.2 Quantum Cascade Laser (QCL)**

The quantum cascade laser (QCL), introduced by Faist (Faist, et al., 1994,1997), represents an excellent source of radiation for trace gas monitoring. Although providing output power well below the CO2 laser (a few mW), this source has some advantages such as compact size, continuous emission, high spectral resolution and the possibility of being operated near room temperatures.

In addition, they can be manufactured to operate in a wide range of wavelengths, from 3 to 24 μm. Among its applications, we highlight the environmental monitoring, industrial process control, besides chemical and biomedical applications. (Beck et al., 2002)

The use of tunable CO2 lasers (Patel, 1964), through the scanning of its emission wavelengths (9-11μm) ensures the exploration of the so-called molecular fingerprint, enabling the identification and simultaneous monitoring of several gaseous compounds with a single instrument. Figure 1 shows the absorption spectra of ozone (O3), ammonia (NH3) and ethylene (C2H4), for the emission range of a CO2 laser. As gas lasers have high power (> 10 watts) and the signal intensity is directly proportional to the emitted light power, its use results in sensitivities in the range of pptv. Nevertheless, these lasers have two disadvantages, they are large and expensive and thus its use is limited to laboratories.

Fig. 1. Fingerprints of ozone (O3), ammonia (NH3) and ethylene (C2H4) corresponding to the

The quantum cascade laser (QCL), introduced by Faist (Faist, et al., 1994,1997), represents an excellent source of radiation for trace gas monitoring. Although providing output power well below the CO2 laser (a few mW), this source has some advantages such as compact size, continuous emission, high spectral resolution and the possibility of being operated near

In addition, they can be manufactured to operate in a wide range of wavelengths, from 3 to 24 μm. Among its applications, we highlight the environmental monitoring, industrial

process control, besides chemical and biomedical applications. (Beck et al., 2002)

**2.2.1 CO2 laser** 

emission lines of a CO2 laser.

room temperatures.

**2.2.2 Quantum Cascade Laser (QCL)** 

The QCL laser coupled with a photoacoustic detector has been successfully used to measure the concentration of different gases which absorb radiation in the mid- infrared spectral region, such as ozone (Silva et al., 2004), ammonia (Baptista-Filho et al., 2006), and NO2 and N2O gases. (Lima et al, 2006; Grossel et al., 2007)

This type of laser is part of the family of semiconductor lasers, with the particularity of using quantum transitions within the same band. A quantum cascade laser comprises a series of alternate thin layers of two different materials. This configuration enables different electrical potentials to be established across the device, so that electrons can get trapped in these sites, called quantum wells. Thus, a series of sub-bands with different energies is created inside the conduction band.

When stimulated, the electrons undergo transitions and quantum tunneling to a lower energy sub-band, and consequently photons are emitted. A single electron can perform several transitions, that is, issue multiple photons. (Gmachl et al., 2001; Beyer et al., 2003)

Another important characteristic of quantum cascade lasers is that the wavelength can be determined by the thickness of the layers, rather than being determined by the energy difference between bands. The thickness and refractive index change by setting different temperatures turning the wavelength (Kosterev et al., 2002; Curl et al., 2010; Kosterev et al., 2002).

## **2.2.3 The Optical Parametric Oscillator laser (OPO)**

An interesting optical process that has been used to produce near- and mid-infrared radiation is the optical parametric generation. In 1962, Armstrong et al and Kroll described the fundamental theory of the optical parametric generator and three years later, Giordmaine and Miller demonstrated the operation of an optic parametric oscillator (Armstrong et al., 1962; Giordmaine & Miller, 1965). When an optically nonlinear crystal is submitted to electromagnetic fields of high density of energy such as those present in pulsed lasers, the electrons respond with significant displacement that gives rise to the contribution of the second-order nonlinear component for the electric polarization of a nonlinear medium. The mixing of two electromagnetic waves under condition of nonlinear polarization produces parametric effects such as second harmonic generation (frequency doubling), sum frequency generation and difference frequency generation. In the latter case (figure 2), when the nonlinear crystal is pumped by two input photons with wavelength at λp (pump photon) and λs (signal photon, λs > λp), the signal photon stimulates the conversion of a pump photo into a new signal and an idler photon with wavelength at λ<sup>i</sup> -1 = λ<sup>p</sup> -1 - λ<sup>s</sup> -1 (fulfilling the energy conservation principle). Of course, the efficiency of such conversion is limited to phase match between the pump and signal waves. This process of increasing the number of signal photons is known as optical parametric amplification (OPA) and thus it fundamentally differs from the amplification mechanism in laser since no population inversion and excited states take place. Theoretically, the remarkable advantage of OPA is the infinity possibility of combining two waves that generates a third wave with different wavelength. This fulfills the expected desirable feature of an excitation source for analytical spectroscopy application that is the broad wavelength tunability. Therefore, this makes the optical parametric phenomena of great significance to spectroscopy application in the sense

Detection of Greenhouse Gases Using the Photoacoustic Spectroscopy 223

The methodologies based on photothermal techniques, mainly CO2 laser photoacoustic spectroscopy, have suitable characteristics to detect trace gas, as high sensitivity and selectivity and possibility of in situ measurements. A CO2 laser based photoacoustic spectrometer (Figure 3) can be used to detect volatile organic compounds (VOCs) emissions, such as ethylene (Demtröder, 2003; Fejer et al., 1992, Tang et al., 1992, Zhang et al., 1993). Ethylene is a reactive pollutant, since it is an unsaturated organic compound (Zhou et al., 1998). For this reason, this chemical species is a precursor for the generation of the tropospheric ozone (Da Silva et al., 2006), which is present in photochemical smog and directly affects human health. Besides, ozone is a powerful greenhouse gas, whose formation is greatly potentiated by the incidence of sun radiation and the presence of nitrogen oxides (NOx) (Yu & Kung, 1999; McCulloch et al., 2005; Teodoro et al., 2010). According to the Intergovernmental Panel of Climatic Changes (IPCC), ozone has a positive radiative forcing of about 0.35 W/m2, being, therefore, an important source of global

**3. Detection of greenhouse gases: Experimental setups** 

Fig. 3. Scheme of the photoacoustic experimental setup

To guarantee a refined detection of ethylene, the photoacoustic spectrometer is daily calibrated by submitting the cell to a flow of a certified mixture. This measurement was carried out using a certified gas mixture of 1.1 ppmV ethylene in N2 flowing into the cell at a rate of 83.3 sccm (standard cubic centimeter). The acoustic signal is detected by a microphone that generates an electric signal. This electric signal is pre-amplified and then detected by a lock-in amplifier (Stanford SR850) with a time constant of 300 ms. The lock-in response is registered in a microcomputer. A continuous wave CO2 infrared laser ( LTG, model LTG150 626G), tuneable over about 80 different lines between 9.2 and 10.6 μm, with a

**3.1 CO2 laser experimental setup** 

warming.

of allowing selective trace gas detection and thus analysis of multicomponents samples, such as the air. Although the finite width of the emission line may reduce the selectivity, for atmosphere application the broadening of the line width of detected specie due to the atmospheric pressure (Demtröder, 2003; Sigrist a, 1994) for itself reduces the significance of an ultra-narrow line width of an applied source.

Fig. 2. Optical parametric process. The incident pump radiation with wavelength λp and circular frequency at ωp is converted into signal and idler radiations with wavelength λs and λi and circular frequency at ωs and ωi, respectively. c is the velocity of the light in the medium.

The efficiency of the nonlinear conversion depends on the phase matching of the pump and signal waves. The dispersion of electromagnetic wave propagating through a crystal is directly related to the refractive index of the media that is different for each wavelength. Therefore the pump and signal wave move in and out of phase relatively to each other, limiting the quantity of generated signal photons. Consequently, initially the first issue was to find materials that provide phase matching for at least two wavelengths. The use of birefrigent crystalline materials was the first key to fix the relative phase of the pump and signal waves. However, the available range of wavelengths that satisfies the phase-matching condition is limited to the variability of birefrigent crystalline materials. More recently, the use of periodically poled lithium niobate (PPLN) has overcome this restriction. PPLN chips display an engineered inverted orientation of lithium niobate crystals that promotes a quasi-phase-matched combination of the pump and signal waves compensating the phase mismatch present in parametric interaction (Fejer et al., 1992;Tang et al., 1992).

For the optic parametric oscillator (OPO), the crystal is initially pumped by photons of only one wavelength λp. Based on the fundamental quantum uncertainty in the electric field (quantum noise), a pump photon (λp) propagating in a nonlinear optical crystal spontaneously breaks up spontaneously into two lower-energy photons with wavelength at λs (signal photon) and λi (idler photon) (Zhang et al., 1993; Zhou et al., 1998). This optic parametric process is called optic parametric generation (OPG). Afterwards, the created signal wave mixes with the pump wave under condition of nonlinearity resulting in new signal and idler waves (stimulated generation). To increase the number of signal and idler photons, the crystal is placed within an optic cavity formed by two mirrors (optical resonator). Single resonance is achieved when the signal wave is reflected back and forth in the optic cavity.

## **3. Detection of greenhouse gases: Experimental setups**

## **3.1 CO2 laser experimental setup**

222 Greenhouse Gases – Emission, Measurement and Management

of allowing selective trace gas detection and thus analysis of multicomponents samples, such as the air. Although the finite width of the emission line may reduce the selectivity, for atmosphere application the broadening of the line width of detected specie due to the atmospheric pressure (Demtröder, 2003; Sigrist a, 1994) for itself reduces the significance of

Fig. 2. Optical parametric process. The incident pump radiation with wavelength λp and circular frequency at ωp is converted into signal and idler radiations with wavelength λs and

The efficiency of the nonlinear conversion depends on the phase matching of the pump and signal waves. The dispersion of electromagnetic wave propagating through a crystal is directly related to the refractive index of the media that is different for each wavelength. Therefore the pump and signal wave move in and out of phase relatively to each other, limiting the quantity of generated signal photons. Consequently, initially the first issue was to find materials that provide phase matching for at least two wavelengths. The use of birefrigent crystalline materials was the first key to fix the relative phase of the pump and signal waves. However, the available range of wavelengths that satisfies the phase-matching condition is limited to the variability of birefrigent crystalline materials. More recently, the use of periodically poled lithium niobate (PPLN) has overcome this restriction. PPLN chips display an engineered inverted orientation of lithium niobate crystals that promotes a quasi-phase-matched combination of the pump and signal waves compensating the phase mismatch present in parametric interaction (Fejer et al.,

For the optic parametric oscillator (OPO), the crystal is initially pumped by photons of only one wavelength λp. Based on the fundamental quantum uncertainty in the electric field (quantum noise), a pump photon (λp) propagating in a nonlinear optical crystal spontaneously breaks up spontaneously into two lower-energy photons with wavelength at λs (signal photon) and λi (idler photon) (Zhang et al., 1993; Zhou et al., 1998). This optic parametric process is called optic parametric generation (OPG). Afterwards, the created signal wave mixes with the pump wave under condition of nonlinearity resulting in new signal and idler waves (stimulated generation). To increase the number of signal and idler photons, the crystal is placed within an optic cavity formed by two mirrors (optical resonator). Single resonance is achieved when the signal wave is reflected back and forth in

λi and circular frequency at ωs and ωi, respectively. c is the velocity of the light in the

an ultra-narrow line width of an applied source.

medium.

1992;Tang et al., 1992).

the optic cavity.

The methodologies based on photothermal techniques, mainly CO2 laser photoacoustic spectroscopy, have suitable characteristics to detect trace gas, as high sensitivity and selectivity and possibility of in situ measurements. A CO2 laser based photoacoustic spectrometer (Figure 3) can be used to detect volatile organic compounds (VOCs) emissions, such as ethylene (Demtröder, 2003; Fejer et al., 1992, Tang et al., 1992, Zhang et al., 1993). Ethylene is a reactive pollutant, since it is an unsaturated organic compound (Zhou et al., 1998). For this reason, this chemical species is a precursor for the generation of the tropospheric ozone (Da Silva et al., 2006), which is present in photochemical smog and directly affects human health. Besides, ozone is a powerful greenhouse gas, whose formation is greatly potentiated by the incidence of sun radiation and the presence of nitrogen oxides (NOx) (Yu & Kung, 1999; McCulloch et al., 2005; Teodoro et al., 2010). According to the Intergovernmental Panel of Climatic Changes (IPCC), ozone has a positive radiative forcing of about 0.35 W/m2, being, therefore, an important source of global warming.

Fig. 3. Scheme of the photoacoustic experimental setup

To guarantee a refined detection of ethylene, the photoacoustic spectrometer is daily calibrated by submitting the cell to a flow of a certified mixture. This measurement was carried out using a certified gas mixture of 1.1 ppmV ethylene in N2 flowing into the cell at a rate of 83.3 sccm (standard cubic centimeter). The acoustic signal is detected by a microphone that generates an electric signal. This electric signal is pre-amplified and then detected by a lock-in amplifier (Stanford SR850) with a time constant of 300 ms. The lock-in response is registered in a microcomputer. A continuous wave CO2 infrared laser ( LTG, model LTG150 626G), tuneable over about 80 different lines between 9.2 and 10.6 μm, with a

Detection of Greenhouse Gases Using the Photoacoustic Spectroscopy 225

For this reason, QCLs are ideal for the development of compact trace gas analyzers that are also suitable for field measurements. In recent years the detection of a series of important trace gases has been demonstrated with these devices.(Faist, et al., 1994,1997; Beck et al., 2002; Silva et al., 2004; Baptista-Filho et al., 2006; Lima et al, 2006; Grossel et al., 2007; Gmachl et al., 2001; Beyer et al., 2003; Kosterev et al., 2002; Curl et al., 2010; Kosterev et al., 2002). By way of illustration we report on measurements of sulphur hexafluoride and methane with a homemade Laser Photoacoustic Spectrometer equipped with QC lasers and a Differential Photoacoustic Cell. The motivation of our research comes from the need for simple, sensitive, and spectrally selective devices for measuring traces of greenhouse gases in agriculture, automobile exhaust monitoring, power distribution facilities, cattle breeding and chemistry industries. The experimental set up employed in the detection limit

As radiation source, a pulsed quantum cascade is normally used. In this experiment two quantum cascade lasers were employed separately, each laser emission band matching the absorption lines of one of the specified molecules. The laser used in the detection of CH4, emits in the range of 7.71 - 7.88 μm and can reach a power of 5.6 mW (at lowest operating temperature of the laser), the one employed in the detection of SF6, emits in the range of

determination of the analyzed gases is illustrated in Figure 4.

10.51 - 10.56 μm and can reach a power of 3.7 mW.

Fig. 4. Experimental set up using quantum cascade laser.

power of 1.9W at the emission line 10P(14) (10.53 μm), by internal PZT (Piezoelectric Transducer), is employed as the excitation source. At this power level, no saturation effects of the photoacoustic signal were observed. These lines can be swept by a step motor controlled by a microcomputer. Within this spectral region, many small molecules show a unique fingerprint. The photoacoustic instrument used has been developed for the detection of small concentrations of gases. All the measurements and the sample collection are made at room temperature. Therefore, the analysis of these samples is made for a number of n different species, rather than just one. This was accomplished by measuring the photoacoustic (PA) signal S(λi) at a set of wavelengths λi (i = 1, 2… m) chosen on the basis of the absorption spectra of the individual components to be detected. These individual absorption spectra were obtained from the HITRAN-PC database, which calculates the absorption cross sections (σ) of a given molecule at different wave numbers ki = 1/λi in a given interval. Thus, the expression used to determine the concentrations of a given component in the multicomponent gas mixture is the eq. (5). The absorption cross section σij is related to the photoacoustic generation efficiency of each gas component for each CO2 laser line. The sum is taken over the n components present in the sample.

#### **3.1.1 Photoacoustic cell calibration and sensitivity measurements**

The calibration and sensitivity measurements of the photoacoustic cell were performed by obtaining the cell coupling constant C in the eq (5). This was performed by taking a 1.1 ppmV certified mixture of ethylene in N2 and diluting it in nitrogen until the least concentration achieved (about 16 ppbv) (Mothé et al, 2010). A linear dependence of the photoacoustic signal on the ethylene concentration could be proven and this linearity could be extended to ppmV levels. The absorption cross section σ of ethylene is well known at the 10P(14) (949.51cm−1) CO2 laser line (σ = 170 × 10−20 cm2). Hence, the C constant value was then obtained from the eq. (3), which yielded 40.2 V.cm/W. The unity of the cell coupling constant was furnished by the manufacturer of our photoacoustic cell (Prof. Markus W. Sigrist). Recent measurements made in greenhouse gas SF6, indicated that using the CO2 laser, it was possible to achieve a detection limit of 20 ppbv. (Rocha et al., 2010)

#### **3.2 Quantum Cascade Laser (QCL) experimental setup**

With the recent development of quantum-cascade lasers (QCLs), compact, low-cost, solidstate radiation sources are available, covering the important infrared (IR) region with specific molecular absorption lines. In addition, spectral regions can be selected in which water vapor has a very low absorption coefficient, known as atmospheric windows. Another important advantage of QCLs in practical applications is that they work near room temperature, whereas diode lasers such as lead salt lasers, which emit in the fundamental IR region, have to be cryogenically cooled. Recent applications of QCLs clearly indicate their potential as tunable light sources in the mid-infrared, especially between 3 - 13 μm, with strong fundamental absorption bands. Current interest is based on the lack of other convenient coherent laser sources. In fact, it can be expected that QCLs will open new possibilities for real-time diagnostics of various molecular species in the 3-5 μm and 8-13 μm atmospheric windows.(Kosterev et al., 2002) Pulsed quantum-cascade distributed-feedback (QC-DFB) lasers provide quasi room temperature operation, combined with a high spectral selectivity and sensitivity, real time measurement capabilities, robustness, and compactness.

power of 1.9W at the emission line 10P(14) (10.53 μm), by internal PZT (Piezoelectric Transducer), is employed as the excitation source. At this power level, no saturation effects of the photoacoustic signal were observed. These lines can be swept by a step motor controlled by a microcomputer. Within this spectral region, many small molecules show a unique fingerprint. The photoacoustic instrument used has been developed for the detection of small concentrations of gases. All the measurements and the sample collection are made at room temperature. Therefore, the analysis of these samples is made for a number of n different species, rather than just one. This was accomplished by measuring the photoacoustic (PA) signal S(λi) at a set of wavelengths λi (i = 1, 2… m) chosen on the basis of the absorption spectra of the individual components to be detected. These individual absorption spectra were obtained from the HITRAN-PC database, which calculates the absorption cross sections (σ) of a given molecule at different wave numbers ki = 1/λi in a given interval. Thus, the expression used to determine the concentrations of a given component in the multicomponent gas mixture is the eq. (5). The absorption cross section σij is related to the photoacoustic generation efficiency of each gas component for each CO2

The calibration and sensitivity measurements of the photoacoustic cell were performed by obtaining the cell coupling constant C in the eq (5). This was performed by taking a 1.1 ppmV certified mixture of ethylene in N2 and diluting it in nitrogen until the least concentration achieved (about 16 ppbv) (Mothé et al, 2010). A linear dependence of the photoacoustic signal on the ethylene concentration could be proven and this linearity could be extended to ppmV levels. The absorption cross section σ of ethylene is well known at the 10P(14) (949.51cm−1) CO2 laser line (σ = 170 × 10−20 cm2). Hence, the C constant value was then obtained from the eq. (3), which yielded 40.2 V.cm/W. The unity of the cell coupling constant was furnished by the manufacturer of our photoacoustic cell (Prof. Markus W. Sigrist). Recent measurements made in greenhouse gas SF6, indicated that using the CO2

With the recent development of quantum-cascade lasers (QCLs), compact, low-cost, solidstate radiation sources are available, covering the important infrared (IR) region with specific molecular absorption lines. In addition, spectral regions can be selected in which water vapor has a very low absorption coefficient, known as atmospheric windows. Another important advantage of QCLs in practical applications is that they work near room temperature, whereas diode lasers such as lead salt lasers, which emit in the fundamental IR region, have to be cryogenically cooled. Recent applications of QCLs clearly indicate their potential as tunable light sources in the mid-infrared, especially between 3 - 13 μm, with strong fundamental absorption bands. Current interest is based on the lack of other convenient coherent laser sources. In fact, it can be expected that QCLs will open new possibilities for real-time diagnostics of various molecular species in the 3-5 μm and 8-13 μm atmospheric windows.(Kosterev et al., 2002) Pulsed quantum-cascade distributed-feedback (QC-DFB) lasers provide quasi room temperature operation, combined with a high spectral selectivity and sensitivity, real time measurement capabilities, robustness, and compactness.

laser line. The sum is taken over the n components present in the sample.

**3.1.1 Photoacoustic cell calibration and sensitivity measurements** 

laser, it was possible to achieve a detection limit of 20 ppbv. (Rocha et al., 2010)

**3.2 Quantum Cascade Laser (QCL) experimental setup** 

For this reason, QCLs are ideal for the development of compact trace gas analyzers that are also suitable for field measurements. In recent years the detection of a series of important trace gases has been demonstrated with these devices.(Faist, et al., 1994,1997; Beck et al., 2002; Silva et al., 2004; Baptista-Filho et al., 2006; Lima et al, 2006; Grossel et al., 2007; Gmachl et al., 2001; Beyer et al., 2003; Kosterev et al., 2002; Curl et al., 2010; Kosterev et al., 2002). By way of illustration we report on measurements of sulphur hexafluoride and methane with a homemade Laser Photoacoustic Spectrometer equipped with QC lasers and a Differential Photoacoustic Cell. The motivation of our research comes from the need for simple, sensitive, and spectrally selective devices for measuring traces of greenhouse gases in agriculture, automobile exhaust monitoring, power distribution facilities, cattle breeding and chemistry industries. The experimental set up employed in the detection limit determination of the analyzed gases is illustrated in Figure 4.

As radiation source, a pulsed quantum cascade is normally used. In this experiment two quantum cascade lasers were employed separately, each laser emission band matching the absorption lines of one of the specified molecules. The laser used in the detection of CH4, emits in the range of 7.71 - 7.88 μm and can reach a power of 5.6 mW (at lowest operating temperature of the laser), the one employed in the detection of SF6, emits in the range of 10.51 - 10.56 μm and can reach a power of 3.7 mW.

Fig. 4. Experimental set up using quantum cascade laser.

Detection of Greenhouse Gases Using the Photoacoustic Spectroscopy 227

Fig. 5. Calibration curve for methane.

Fig. 6. Calibration curve for sulphur hexafluoride.

The photoacoustic spectroscopy with quantum cascade lasers has proved to be extremely efficient for the detection of greenhouse gases (Atkinson, 2000), being sensible and selective. In the case of the greenhouse gas methane, this methodology allows measurements in anthropogenic sources that emits methane in concentration higher than 1.5 ppmv and also atmospheric measurements, once it is estimated that the current average concentration of this gas in the atmosphere is of 1.7 ppmv. For sulphur hexafluoride the method is suitable

The laser emission lines are given according to the diode temperature, which is determined by a temperature control unit. The pulsed QCL light beam, with a repetition rate of 400kHz and a pulse duration of 50ns (duty cycle of 2%), is gated by an external transistor-totransistor logic (TTL) signal at 3.8kHz to excite the first longitudinal acoustic mode of the resonant differential photoacoustic cell. A germanium lens (focus ~30.7 mm and diameter ~10.35mm) is employed to focus the QCL radiation through the cell. The cell (Miklos, 2001) has two resonant cylindrical tubes (5.5 mm in diameter and 4 cm in length) on whose edges are arranged acoustic buffers which reduce noise caused by gas turbulence and background signal produced by the heating in the cell windows when these are exposed to the radiation. The gas flow streams through both pipes and noise and background are equally detected by the microphones placed on each of them, but only the microphones placed in the tube crossed by the laser beam detect the pressure change induced by the absorption of modulated radiation in the gaseous sample containing the molecules under consideration. Thus the photoacoustic signal is obtained by simple differentiation of the signal produced by the microphones in the two tubes.

The laser power is monitored by a power detector (OPHIR, 3A-SH-ROHS) and the gases flows are controlled by electronic mass flow controllers (model MKS, 247), one of de 50 sccm and one of 300sccm. The PA data analysis was performed by the lock-in technique using a lock-in amplifier (model Stanford SR\_850 DSP) with a set data acquisition time constant of 300ms.

#### **3.2.1 Calibration and sensitivity measurements**

The following concentration measurements were performed keeping the temperature of each laser constant. At these emission lines the lasers power was 0.8 mW, feed current of 26.2 mA, for the laser used to measure CH4, and 1.12 mW, feed current of 25.3 mA, for the laser used to measure SF6. In such type of measurement, a high stability is observed during the entire experiment. In order to determine the detection boundaries of the gases of interest, a dilution of standard mixtures were carried out. Dilution experiments are depicted in figure 5 and figure 6. Small concentrations of the investigated gases were synthesized by using two electronic mass-flow controllers, one for N2 (with full scale control of 200 sccm) and another for the investigated gas (CH4 or SF6) (with full scale control of 50 sccm). The electronic mass-flow controllers were connected in parallel to the gas inlet of the photoacoustic cell. The initial concentrations of 4.5 ppmv CH4 and 5 ppmv SF6 was diluted with pure nitrogen (zero gas) down to the lowest concentrations detected by the system.

The acoustic and electronic noise was determined by blocking the laser light while keeping all other devices running. The value of the noise signal was typically 0.300μV. As expected (Eq.3) , a linear dependence of the photoacoustic signal on the methane and sulphur hexafluoride concentration was found. The fitted straight line are also shown in Figures 5 and 6. The smallest measured concentrations were of 1.5 ppmv for methane and 49 ppbv for sulphur hexafluoride (Rocha et al., 2010 b). Although the smallest concentration of methane detected was of 1.5 ppmv, the strong linear slope of the fitted straight line allows us to estimate that the instrumentation has the sensitivity to detect concentration changes smaller than 1.5 ppmv. In recent measurements made with methane, it was possible to achieve an experimental detection limit of 50 ppbv, for this gas (Rocha et al, 2011). It is possible to estimate a detection limit of 30 ppbv methane, by extending the straight line until the noise limit, at a signal to noise ratio, in single pass (Grossel et al., 2007).

Fig. 5. Calibration curve for methane.

The laser emission lines are given according to the diode temperature, which is determined by a temperature control unit. The pulsed QCL light beam, with a repetition rate of 400kHz and a pulse duration of 50ns (duty cycle of 2%), is gated by an external transistor-totransistor logic (TTL) signal at 3.8kHz to excite the first longitudinal acoustic mode of the resonant differential photoacoustic cell. A germanium lens (focus ~30.7 mm and diameter ~10.35mm) is employed to focus the QCL radiation through the cell. The cell (Miklos, 2001) has two resonant cylindrical tubes (5.5 mm in diameter and 4 cm in length) on whose edges are arranged acoustic buffers which reduce noise caused by gas turbulence and background signal produced by the heating in the cell windows when these are exposed to the radiation. The gas flow streams through both pipes and noise and background are equally detected by the microphones placed on each of them, but only the microphones placed in the tube crossed by the laser beam detect the pressure change induced by the absorption of modulated radiation in the gaseous sample containing the molecules under consideration. Thus the photoacoustic signal is obtained by simple differentiation of the signal produced

The laser power is monitored by a power detector (OPHIR, 3A-SH-ROHS) and the gases flows are controlled by electronic mass flow controllers (model MKS, 247), one of de 50 sccm and one of 300sccm. The PA data analysis was performed by the lock-in technique using a lock-in amplifier (model Stanford SR\_850 DSP) with a set data acquisition time constant of 300ms.

The following concentration measurements were performed keeping the temperature of each laser constant. At these emission lines the lasers power was 0.8 mW, feed current of 26.2 mA, for the laser used to measure CH4, and 1.12 mW, feed current of 25.3 mA, for the laser used to measure SF6. In such type of measurement, a high stability is observed during the entire experiment. In order to determine the detection boundaries of the gases of interest, a dilution of standard mixtures were carried out. Dilution experiments are depicted in figure 5 and figure 6. Small concentrations of the investigated gases were synthesized by using two electronic mass-flow controllers, one for N2 (with full scale control of 200 sccm) and another for the investigated gas (CH4 or SF6) (with full scale control of 50 sccm). The electronic mass-flow controllers were connected in parallel to the gas inlet of the photoacoustic cell. The initial concentrations of 4.5 ppmv CH4 and 5 ppmv SF6 was diluted with pure nitrogen (zero gas) down to the lowest concentrations detected by the system.

The acoustic and electronic noise was determined by blocking the laser light while keeping all other devices running. The value of the noise signal was typically 0.300μV. As expected (Eq.3) , a linear dependence of the photoacoustic signal on the methane and sulphur hexafluoride concentration was found. The fitted straight line are also shown in Figures 5 and 6. The smallest measured concentrations were of 1.5 ppmv for methane and 49 ppbv for sulphur hexafluoride (Rocha et al., 2010 b). Although the smallest concentration of methane detected was of 1.5 ppmv, the strong linear slope of the fitted straight line allows us to estimate that the instrumentation has the sensitivity to detect concentration changes smaller than 1.5 ppmv. In recent measurements made with methane, it was possible to achieve an experimental detection limit of 50 ppbv, for this gas (Rocha et al, 2011). It is possible to estimate a detection limit of 30 ppbv methane, by extending the straight line until the noise

limit, at a signal to noise ratio, in single pass (Grossel et al., 2007).

by the microphones in the two tubes.

**3.2.1 Calibration and sensitivity measurements** 

Fig. 6. Calibration curve for sulphur hexafluoride.

The photoacoustic spectroscopy with quantum cascade lasers has proved to be extremely efficient for the detection of greenhouse gases (Atkinson, 2000), being sensible and selective. In the case of the greenhouse gas methane, this methodology allows measurements in anthropogenic sources that emits methane in concentration higher than 1.5 ppmv and also atmospheric measurements, once it is estimated that the current average concentration of this gas in the atmosphere is of 1.7 ppmv. For sulphur hexafluoride the method is suitable

Detection of Greenhouse Gases Using the Photoacoustic Spectroscopy 229

Owing to the wavelength tunability of the OPO, in recent years it has been shown the feasibility of using optical parametric oscillator as light source for trace gas detection of several chemical species of environmental appeal. Limits of part per billion for greenhouse gases have been demonstrated when OPO is combined with photoacoustic detection methods. Concentration limit of 60 ppbv (part per billion by volume) and 20 ppbv were estimated for nitrous oxides (N2O) and methane (C2H4), respectively, when the OPO radiation is in amplitude modulated (Costopoulos et al., 2002, Liang et al., 2000). The sensitivity of the detection for methane can be improved when the technique of wavelength modulation of the OPO radiation, combined with multipass configuration, is employed. This modulation mode has the advantage of the suppression of the baseline caused by hits of the cell windows and the reflection on the mirrors. According to Nd *et all* , an ultimate sensitivity of 136 parts in 1012 for methane was estimated when the technique of wavelength

Since the increase rate of greenhouse in the atmospheric air is higher than the sensitivity of an OPO setup, it gives rise in this way the possibility of using OPO devices to monitor the annual change of pollutants in atmospheric air. Detection of N2O in ambient air was already carried out using photoacoustic spectroscopy. Applying the photoacoustic spectroscopy in combination with a pulsed grazing-incident optical parametric oscillator, concentrations of 311 ppbv were found for ambient samples collected at nearby roads (Da Silva et al., 2006).

The Photoacoustic Spectroscopy has proved to be extremely sensitive and selective for gas detection. Several experimental arrangements with laser sources have been presented to the detection of greenhouse gases and its precursors in the ppmv and ppbv range. Recent advances in radiation sources for photoacoustic detection of gases has been presented, such as the quantum cascade laser (QCL) and the optical parametric oscillator (OPO) system as well as improvements in the photoacoustic detector, like the differential photoacoustic cell and the simple resonant photoacoustic cell. Such progresses encourage the development of

Abeles, F.B. & Heggestad, H.E. (1973) Ethylene: An urban air pollutant. *J. Air Pollut. Cont.* 

Allen, M. R.; Frame, D. J.; Huntingford, C.; Jones, C. D.; Meinshausen, J. A. L. M. &

Emissions Towards the Trillionth Tonne, *Nature*, Vol. 458. ISSN 0028-0836. Angelmahr, M.; Miklos, A. & Hess, P., (2006) Photoacoustic Spectroscopy of formaldehyde

Atkinson, R. (2000). Atmospheric chemistry of VOCs and NOx. *Atmosph. Environ. 34*, 2063-

Baird, C. (2002). *Environmental Chemistry*, 2nd ed.; Brookman: Porto Alegre, Brazil. ISBN

ISSN: 0946-2171 (print version) ISSN: 1432-0649 (electronic version). Armstrong, J. A.; Bloembergen, N.; Ducuing, J. & Pershan, P. S.(1962). Phys. Rev. 127, 1918.

Meinshausen, N (April 30 2009). Warning Caused by Cumulative Carbon

with tunable laser radiation at the part per billion level. *Appl. Phys. B* , *85*, 285-288.

modulation is used (Kung et al., 2004) [92].

new researches in the field of atmospheric pollution.

*Assoc.* , *23*, 517-521. ISSN 0002-2470.

ISSN 1943-2879.

9788536300023.

2101. ISSN 1352-2310.

**4. Conclusion** 

**5. References** 

for concentrations greater than 49 ppbv, a concentration already detected by conventional equipments. Another important greenhouse gas is the nitrous oxide (N2O) which can also be detected by quantum cascade laser (QCL), whose detection limit of 14 ppbv was obtained for this gas (Grossel et al, 2007).

#### **3.3 OPO Laser experimental setup**

Figure 7 shows a schematic diagram of an experimental setup for an OPO in a simple grazing-incidence grating configuration (GIOPO) (Yu & Kung, 1999). Two gold mirrors (M1 and M2) are used to produce an optical resonator. The pump light (1064 nm) is put into the resonator by 45o incidence on a third mirror (M3) coated for high reflection at 1064 nm and high transmission for signal and idler waves. A coated highly transmitting at 1064 nm lens (L1) is used to focus the pump beam at the middle of the PPLN crystal (C1) length. In the GIOPO configuration a grating (600/mm groove density) (G1) is placed at grazing incidence relative to the cavity axis. The grating serves as dispersing element. The diffracted first order off the grazing is reflected back into the cavity by M2 and used as injection seed. The zero order is out coupled from the resonator to be used as exciting light for photoacoustic spectroscopy. Before the beam reaches the photoacoustic detector, a highly reflecting mirror at 1064 nm is employed to eliminate the pump wave and a germanium element (Ge) is used to filtering the signal wave from the beam. This OPO configuration results in a typical linewidth of about 0.1 cm-1 and the idler wave covers a wavelength range between 2.4 and 4 μm with power average of some hundreds of miliwatts by either tuning the mirror M2 or changing the temperature of the crystal.

Fig. 7. Schematic setup for OPO with in a grazing-incidence grating configuration.

Owing to the wavelength tunability of the OPO, in recent years it has been shown the feasibility of using optical parametric oscillator as light source for trace gas detection of several chemical species of environmental appeal. Limits of part per billion for greenhouse gases have been demonstrated when OPO is combined with photoacoustic detection methods. Concentration limit of 60 ppbv (part per billion by volume) and 20 ppbv were estimated for nitrous oxides (N2O) and methane (C2H4), respectively, when the OPO radiation is in amplitude modulated (Costopoulos et al., 2002, Liang et al., 2000). The sensitivity of the detection for methane can be improved when the technique of wavelength modulation of the OPO radiation, combined with multipass configuration, is employed. This modulation mode has the advantage of the suppression of the baseline caused by hits of the cell windows and the reflection on the mirrors. According to Nd *et all* , an ultimate sensitivity of 136 parts in 1012 for methane was estimated when the technique of wavelength modulation is used (Kung et al., 2004) [92].

Since the increase rate of greenhouse in the atmospheric air is higher than the sensitivity of an OPO setup, it gives rise in this way the possibility of using OPO devices to monitor the annual change of pollutants in atmospheric air. Detection of N2O in ambient air was already carried out using photoacoustic spectroscopy. Applying the photoacoustic spectroscopy in combination with a pulsed grazing-incident optical parametric oscillator, concentrations of 311 ppbv were found for ambient samples collected at nearby roads (Da Silva et al., 2006).

## **4. Conclusion**

228 Greenhouse Gases – Emission, Measurement and Management

for concentrations greater than 49 ppbv, a concentration already detected by conventional equipments. Another important greenhouse gas is the nitrous oxide (N2O) which can also be detected by quantum cascade laser (QCL), whose detection limit of 14 ppbv was obtained

Figure 7 shows a schematic diagram of an experimental setup for an OPO in a simple grazing-incidence grating configuration (GIOPO) (Yu & Kung, 1999). Two gold mirrors (M1 and M2) are used to produce an optical resonator. The pump light (1064 nm) is put into the resonator by 45o incidence on a third mirror (M3) coated for high reflection at 1064 nm and high transmission for signal and idler waves. A coated highly transmitting at 1064 nm lens (L1) is used to focus the pump beam at the middle of the PPLN crystal (C1) length. In the GIOPO configuration a grating (600/mm groove density) (G1) is placed at grazing incidence relative to the cavity axis. The grating serves as dispersing element. The diffracted first order off the grazing is reflected back into the cavity by M2 and used as injection seed. The zero order is out coupled from the resonator to be used as exciting light for photoacoustic spectroscopy. Before the beam reaches the photoacoustic detector, a highly reflecting mirror at 1064 nm is employed to eliminate the pump wave and a germanium element (Ge) is used to filtering the signal wave from the beam. This OPO configuration results in a typical linewidth of about 0.1 cm-1 and the idler wave covers a wavelength range between 2.4 and 4 μm with power average of some hundreds of miliwatts by either tuning the mirror M2 or

Fig. 7. Schematic setup for OPO with in a grazing-incidence grating configuration.

for this gas (Grossel et al, 2007).

**3.3 OPO Laser experimental setup** 

changing the temperature of the crystal.

The Photoacoustic Spectroscopy has proved to be extremely sensitive and selective for gas detection. Several experimental arrangements with laser sources have been presented to the detection of greenhouse gases and its precursors in the ppmv and ppbv range. Recent advances in radiation sources for photoacoustic detection of gases has been presented, such as the quantum cascade laser (QCL) and the optical parametric oscillator (OPO) system as well as improvements in the photoacoustic detector, like the differential photoacoustic cell and the simple resonant photoacoustic cell. Such progresses encourage the development of new researches in the field of atmospheric pollution.

## **5. References**


Detection of Greenhouse Gases Using the Photoacoustic Spectroscopy 231

Gmachl, C.; Capasso, F; Sivco, D.L. & Cho, A., (2001) Recent progress in quantum cascade

Gondal, M.A., (1997). Laser photoacoustic spectrometer for remote monitoring of

Greene, C. H.; Monger, B. C.; McGarry, L. P., (2009*).* Some like it cold, northern shrimp

Grossel, A.; Z´eninari, V.; Parvitte, B.; Joly, L. & D. Courtois (2007). Optimization of a

Hansen, J. et al. (2008) Target atmosphere CO2: Where Should Humanity Aim . *The Open* 

Harren, F. & Reuss, J. R., (1997). Photoacoustic spectroscopy 'Encyclopedia of Applied Physics' Vol. 19 (Ed.) G.L. Trigg (VCH, Weinheim, 413-435 . ISBN 3–527–28134–7 . Harren, F. J.M.; Cotti, G.; Oomens, J. & Hekkert, S. te L., (2000). Photoacoustic Spectroscopy

Harren, F.J. M. et al, (2008). Laser-based systems for trace gas detection in life sciences,

Hess, P. (1983). Resonant photoacoustic spectroscopy. *Top. Curr. Chem*. 111, 1-32. ISSN 0340-

Hess, P., (1992). Principles of photoacoustic and photothermal detection in gases. In

IPCC. (February 2007). *Climate Change: The Physical Science Basis*; Report of Working Group I;

Janssen, M. (1998). Use of complex adaptive systems for modeling global Change, *Ecosystems*, 1, 457-463 ISSN 1432-9840 (print) ISSN 1435-0629 (electronic version). Kerr, E. L., & Atwood J. G., (1968). The laser illuminated absorptivity spectrophone: A

Kosterev, A. A. & Tittel, F. K., ( 2002). Chemical sensors based on Quantum Cascade Lasers. IEEE Journal of Quantum Eletronics, V.38, N.6, 582-591. ISSN 0018–9197. Kosterev, A. A.; Curl, R. F.; Tittel, F. K. ; Rochat, M.; Beck, M.; Hofstetter, D. & Faist, J.

*Opt.* 7, 915-92 ISSN: 1559-128X (print) ISSN 2155-3165 (online).

4885(print) ISSN 1361-6633 (electronic version).

ISSN 0036-8075 (print), 1095-9203 (online).

*Atmospheric Science journal,* 2, 217-231. ISSN: 1874-2823.

Wiley & Sons Ltd, Chichester. ISBN 0471851981.

1022(print) ISSN 1436-5049(electronic version).

IPCC: Paris, France. ISBN 9291691224.

Ed.), Chapter 4. Elsevier, New York. ISSN: 0301-4797

ISSN: 2155-3165 (online).

1091-6490.

version).

lasers and applications,. Reports on Progress in Physics, 64: 1533-1601. ISSN 0034-

atmospheric pollutants, Applied Optics, Vol. 36, No. 15ISSN: 1559-128X (print)

stocks thrive when climatic conditions lead to cold bottom waters. *Science,* Vol. 324.

compact photoacoustic quantum cascade laser spectrometer for atmospheric flux measurements: application to the detection of methane and nitrous oxide. Appl. Phys. B 88, 483–492. ISSN: 0946-2171 (print) ISSN: 1432-0649 (electronic version). Hansen, J. & Makiko, (2004). Greenhouse gas growth rates, *PNAS,* vol.101, no. 46 ISSN:

in Trace Gas Monitoring, *Encyclopedia of Analytical Chemistry,* pp. 2203–2226, John

Applied Physics B, 92, 343-349. ISSN: 0946-2171 (print) ISSN: 1432-0649 (electronic

*Principles and Perspectives of Photothermal and Photoacoustic Phenoma* (A. Mandelis,

method for measurement of weak absorptivity in gases at laser wavelengths. *Appl.* 

(2002). Chemical sensing with pulsed QC-DFB lasers operating at 15.6 \_m, Appl. Phys. B,75, 351–357 ISSN: 0946-2171 (print) ISSN: 1432-0649 (electronic version). Kreuzer, L. B., (1971). Ultralow gas concentration infrared absorption spectroscopy *J. App/. Phys.* 42. ISSN 2934-2943. (print) ISSN 0021-8979. (electronic version).


Baptista-Filho, M.; Da Silva, M.G.; Sthel, M.S.; Schramm, D.U.; Vargas, H.; Miklos, A. &

Beck, M., et al. (2002) Continuous Wave Operation of a Mid-Infrared Semiconductor Laser at Room Temperature. Science Vol. 295. ISSN 0036-8075 (print), 1095-9203 (online). Bell, A. G., (1880). On the production and reproduction of sound. *Am. J. Sci.* 20, 305-324.

Berrou, A.; Raybaut, M.; Godard, A. & Lefebvre, M., (2010). High-resolution photoacoustic

Beyer, T.; Braun, M. ; Lambrecht, & A. Fast (2003) .Gas spectroscopy using pulsed quantum

Costopoulos, D.; Miklos, A. & P. Hess, Appl. Phys. B 75, 385 (2002). ISSN: 0946-2171 (print)

Curl, R. F.; Capasso, F.; Gmachl, C.; Kosterev, A. A.; McManus, B.; Lewicki, R. ; Pusharsky,

Da Silva,M.G.Da Silva, M.G.; Miklós, A.; Falkenroth, A. & Hess, P., (2006) Photoacoustic

Demtröder, W. (2003) *Laser Spectroscopy: basic concepts and instrumentation*, pag. 75-81, Third

Dewey, C. F., Jr., Kamm, R. D., & Hackett, C. E., (1973). Acoustic amplifier for detection of

Elia, A.; Lugarà, P. M.; Di Franco, C. & Spagnolo, V., (2009). Photoacoustic Techniques for

Emanuel, K.; (2005) Increasing destructiveness of tropical cyclones over the past 30 years.

Faist, J et al, (1994). Quantum cascade laser. Science 264, 553. ISSN 0036-8075 (print) 1095-

Faist J. et al, (1997). "Distributed feedback quantum cascade lasers," Appl. Phys. Lett. 70,

Fejer, M. M.; Magle, G. A.; Jundt, D. H. & Byer, R. L. (1992) IEEE J. Quantum Electron. 18,

Filho, M.B.; Silva, M.G.; Sthel, M.S.; Schramm, D.U.;Vargas, H.; Miklos, A. & Hess, P., (2006).

*Appl. Opt.*,*45*, 4966-4971. ISSN: 1559-128X (print) ISSN: 2155-3165 (online). Giordmaine, J. A. & Miller, R. C. (1965). Phys. Rev. Lett. 14, 973 . ISSN. 0031-9007 (print)

Ammonia detection by using quantum cascade laser Photoacoustic Spectroscopy.

2670–2672. ISSN 0003-6951 (print) ISSN 1077-3118 (online).

Edition (Springer-Verlag, Berlin. ISBN 3-540-67838-7.

ISSN: 1945-452X (online) ISSN: 0002-9599 (print).

(print) ISSN: 1432-0649 (electronic version).

ISSN: 1432-0649 (electronic version).

1432-0649 (electronic version).

*Nature*, Vol. 436/4. ISSN 0028-0836.

3165 (online).

(electronic version).

1077-3118 (online).

ISSN 1424-8220

9203 (online)

907. ISSN: 0018-9197.

1079-7114 (online).

Hess, P., (2006) Ammonia detection using quantum-cascade laser photoacoustic spectroscopy. Applied Optics, 45: 4966-4971 ISSN: 1559-128X (print) ISSN: 2155-

and direct absorption spectroscopy of main greenhouse gases by use of a pulsed entangled cavity doubly resonant OPO. *Appl. Phys. B* , *98*, 217-230. ISSN: 0946-2171

cascade lasers, J. Appl. Phys, 93, 3158–3160 ISSN: 0947-8396 (print) ISSN: 1432-0630

M.; Wysocki, G. & Tittel, F. K. (2010). Quantum cascade lasers in chemical physics. Chemical Physics Letters, 487, 1–18 ISSN 0003-6951 (print) ISSN 1077-3118 (online).

measurement of N2O concentrations in ambient air with a pulsed optical parametric oscillator, Appl. Phys. B 82, 329–336. ). ISSN: 0946-2171 (print) ISSN:

atmospheric pollutants. *Appl. Phys. Lett.* 23, 633-635. ISSN 0003-6951 (print) ISSN

Trace Gas Sensing Based on Semiconductor Laser Sources, *Sensors*, *9*, 9616-9628.


Detection of Greenhouse Gases Using the Photoacoustic Spectroscopy 233

Repond, P. & Sigrist, M.W., (1996). Photoacoustic Spectroscopy on trace gases with

Rocha, M.; Sthel, M ; Lima, G.; Da Silva, M. , Schramm, D; Miklós, A. & Vargas,H, (2010) a.

EXHIBITION",Calgary, Canadá. ISSN: 1052-6102. ISBN: 978-1-617-82093-9. Rocha, M. V.; Sthel, M. S.; Silva, M. G.; Paiva, L. B.; Pinheiro, F. W. ; Miklòs, A. & Vargas, H.

Rosenzweig, C. et al, (2008)*.* Attributing physical and biological impacts to antropogenic

Röntgen, W. C., (1881). Ueber Toene, welche durch intermittierende Bestrahlung eines Gases

Rothman, L.S.; Gamache, R.R.; Tipping, R.H.; Rinsland, C.P.; Smith, M.A.H.; Benner, D.C.;

Sander et al, B. D., (2006). Forced and unforced ocean temperature changes in Atlantic and Pacific tropical cyclogenesis regions, *PNAS,* vol. 103, no. 38. ISSN: 1091-6490. Schramm, D. U.; Sthel, M. S.; da Silva, M. G. ; Carneiro, L. O.; Junior, A. J. S. & H. Vargas,

Seinfeld, J.H. (1986). *Atmospheric Chemistry and Physics of Air Pollution*; John Wiley & Sons:

Seinfeld, J.H. (1989). Urban air pollution: State of the science. *Science* , *243*, 745-752. ISSN

Siddall, M.; Stocker, T. F. & Clark, P. U., *(2009).* Constrains on future sea-level from past sea-

Sigrist, M.W., (1994) a. *Air Monitoring by Spectroscopic Techniques. Chemical Analysis*; John Wiley & Sons, Inc.; New York, NY, USA, Volume 12 ISBN 0-8247-8946-6. Sigrist, M. W., (1994)b. Laser Photoacoustic Spectrometry for Trace Gas Monitoring, *Analyst,,* 

Sigrist, M.W.; Bartlome, R.; Marinov, D.; Rey, J.M.; Vogler, D.E.; Wächter, H. (2008) Trace

Silva, M. G.; Vargas, H.; Miklós, A. & Hess, P. (2004) Photoacoustic detection of ozone using

Solomon, S. et al., *(*2009*)*, Irreversible climate change due to carbon dioxide emissions,

300. . ISSN 0946-2171 (print) ISSN 1432-0649 (electronic version).

gas monitoring with infrared laser-based detection schemes. *Appl. Phys. B* ,, *90*, 289-

quantum cascade laser, Appl. Phys. B , 78, 1513–1516. . ISSN 0946-2171 (print) ISSN

level change, *Nature Geoscience,* Vol. 2. ISSN : 1752-0894.

entstehen.*Ann. Phys.* (Leipzig) [3] 12, 155-159. ISSN: 0003-4916.

Photoacoustic Spectroscopy, *Sensors* , *10*, 9359-9368 ISSN 1424-8220. Rocha M.V.,Sthel M.S., Gomes M.S. ,Schramm, D. U. Vargas, H. (2010) b .Greenhouse gases

(print) ISSN: 2155-3165 (online version).

climate change, *Nature,* Vol 453. ISSN 0028-0836.

48, 469–507 ISSN: 0022-4073.

Hoboken, NJ, USA. ISBN: 0-471-65128-1.

0036-8075 (print), 1095-9203 (online).

*Vol. 119 ,* 525 ISSN 0003-2654.

1432-0649 (electronic version).

*PNAS,* Vol. 106, no. 6, ISSN 1091-6490.

2171

8220.

continuously tunable CO2 laser, Applied Optics, Vol. 35, No 21 .ISSN: 1559-128X

A Sulfur Hexafluoride Sensor Using Quantum Cascade and CO2 Laser-Based

detection with quantum cascade laser photoacoustic spectroscopy, Air & Waste Management Association's 103 rd ANNUAL CONFERENCE &

(2011). Quantum Cascade laser photoacoustic detection of methane emitted from natural gas powered engines, accepted for Applied Optics B (in press). ISSN 0946-

Devi, V.M.; Flaud, J.M.; Camy-Peyret, C.; Perrin, A.; Goldman, A.; Massie, S.T.; Brown, L.R. & Toth, R.A. (1992)Quantitative Spectroscopy and Radiative Transfer

*(2003).* Application of the Laser Photoacoustic Spectroscopy on the Analysis of Gas Samples Emitted by Diesel Engines. *Inf. Phys. Technol* ,44, p 263-269. ISSN 1424-


Kreuzer, L. B., Kenyon, N. P., & Patel, C. K. N., (1972). Air pollution: Sensitive detection of

Kurz, W. A. et al, (2008). Mountain pine beetle and forest carbon feedback to climate change,

Liang, G.C.; Liu, H.H.; Kung, A.H. ; Mihacsi, A.; Miklos, A. & Hess, P. (2000) J. Phys. Chem.

Lima J.P. et al, (2006). Photoacoustic Detection of NO2 and N2O using quantum cascade

Luft, K. F., (1943). Ueber eine neue Methode der registrierenden Gasanalyse mit Hilfe der

Mann, M. E.; Bradley, R. S. & Hughest, M. K., (1998.) Global temperature patterns and climate forcing over the past six centuries. *Nature,* Vol. 392. ISSN : 0028-0836. McCulloch, M.T.; Langford, N. & Duxbury, G. (2005) Real- time trace- level detection of

Meinshausen, M. et al, (2009). Greenhouse – Gas Emission Targets for Limiting Global

Meyer, P. L., Bernegger, S., & Sigrist, M. W., (1988). On-line monitoring of air pollutants

Meyer, P. L., & Sigrist, M. W., (1990). Atmospheric pollution monitoring using CO2-laser

Miklos, A.; Hess, P. & Bozoki, P., (2001) Application of acoustic resonators in photoacoustic

Moeckli, M.A.; Hilbes C. & Sigrist, M.W. (1998) . Photoacoustic Multicomponent Gas

Mothé, G.; Castro , M.; Sthel,M; Lima, G.; Brasil,L; Campos, L.; Rocha, A& Vargas, H,

Nathan P, et al, (2008)*.* Attribution of polar warming to human influence Nature *Geoscience,* 

Ng, J.; Kung, A.H.; Miklos, A. & Hess, P.(2004). Opt. Lett. 29, 1206 . ISSN: 0146-9592 (print)

Patel, C. K. N. (1964). "Continuous-Wave Laser Action on Vibrational-Rotational Transitions

Peza, A. B. & Simmonds, I., (2005). The first South Atlantic hurricane: Unprecedented

Rayleigh,J. W. (Lord), (1881). The photophone. *Nature (London)* 23, 274-275. ISSN 0028-0836.

ISSN: 0946-2171 (print) ISSN: 1432-0649 (electronic version).

ISSN: 1063-7842 (print) ISSN: 1090-6525 (electronic version).

349. ISSN 0036-8075 (print), 1095-9203 (online).

1559-128X (print) ISSN: 2155-3165 (online).

*Opt. Sci.* 58, 127-130 . ISSN: 0342-4111.

0034-6748. (electronic version).

*Vol. 1* ISSN : 1752-0894.

ISSN: 1539-4794 (online).

5712. ISSN 0094-8276.

(Online)

Warning to 2 o C. *Nature*, 458 ISSN : 0028-0836.

1807 (print) ISSN 0034-6748. (electronic version).

*Nature,* Vol. 452*.* ISSN : 0028-0836.

A, 104, 10179 . ISSN 1089-5639.

(electronic version).

ten pollutant gases by carbon monoxide and carbon dioxide lasers. *Science* 177, 347-

lasers. Appl. Phys. B, 85, 279-284. ISSN: 0946-2171 (print) ISSN: 1432-0649

Absorption ultraroter Strahlen ohne spektrale Zerlegung. Z. *Tech.. Phys.* 5, 97- 104.

carbon dioxide and ethylene in car exhaust gases. *Appl. Opt. 44*, 2887-2884. ISSN:

with a mobile computer-controlled CO2-laser photoacoustic system.*Springer Ser.* 

photoacoustic spectroscopy and other techniques. *Rev. Sci. Instrum.* 61, ISSN 1779-

trace gas analysis and metrology. *Rev. Sci. Instrum.*, *72*, ISSN 1937-1955. (print) ISSN

Analysis Using a Levenberg-Marquardt Fit Algorithm *Appl. Phys*. B . 67, 449–458 . .

(2010).Detection of Greenhouse Gas Precursors from Diesel Engines Using Electrochemical and Photoacoustic Sensors, *Sensors* , *10*, 9726-9741. ISSN 1424-8220.

of CO2". *Physical Review* 136,1187-1193 ISSN 1050-2947 (Print) ISSN 1094-1622

blocking, how shear and climate change, *Geophysical Research Letters,* Vol. 32, Li


**11**

*Japan* 

*Osaka University* 

**Miniaturized Mass Spectrometer in**

**The Performance and Possibilities** 

It is no doubt that mass spectrometry (MS) is widely applied to several research fields. In recent years, design and development of novel miniaturized mass spectrometers have been at the forefront of research in MS. A lot of miniaturized mass spectrometers are reported and commercialized from a lot of institutes and industries. The instruments have widespread applications, for example, detection and identification of chemical and biological hazards for the homeland security (Contreras et al., 2008; Smith et al., 2011), the food safety (Garcia-Reyes et al., 2009) and so on. Because of their small size and lightweight, miniaturized mass spectrometers have a potential for field use. These features are absolutely suitable for on-site environmental analyses, especially in greenhouse gases analyses. However, there are few reports on application of miniaturized mass spectrometers in this research field. This chapter describes instrumentation and application feasibility of miniaturized mass spectrometers to the greenhouse gases analysis. This chapter is consists

1. New analytical concept of "On-site mass spectrometry" and field usable mass

2. Introduction of our technology: multi-turn time-of-flight (TOF) mass spectrometer

3. Miniaturized ultra-high mass resolution multi-turn TOF mass spectrometer

In this section, a novel analytical concept "On-site mass spectrometry", overview of several reported miniaturized mass spectrometers and issues for the field use are mentioned.

The basis of science is to measure phenomena of nature in real-time and with strict accuracy. However, we know that it is the most difficult to perform such measurements using high

**2. "On-site mass spectrometry" using miniaturized mass spectrometers** 

4. Application of greenhouse gases detection using MULTUM-S II.

**1. Introduction** 

of following four sections.

spectrometers,

(MULTUM),

"MULTUM-S II",

**2.1 On-site mass spectrometry** 

**Analysis of Greenhouse Gases:**

Shuichi Shimma and Michisato Toyoda


## **Miniaturized Mass Spectrometer in Analysis of Greenhouse Gases: The Performance and Possibilities**

Shuichi Shimma and Michisato Toyoda *Osaka University Japan* 

## **1. Introduction**

234 Greenhouse Gases – Emission, Measurement and Management

Sorvajärvi, T.; Manninen, A. ; Toivonen, J.; Saarela, J. & Hernberg, R. ( 2009) Resonant

Sthel, M.S.; Tavares, J.R.; Lima, G.R.; Mothé, G.; Schramm, D.U.S.; Da Silva, M.G. & Castro,

Sthel, M.S.; Schramm, D.U.; Lima, G.R., Carneiro, L.; Faria Jr. , R.T.; Castro,

Tang, C. L. ; Bosenber, W. R.; Ukachi, T. ; Lane, R.J. & Cheng, L.K. (1992). Proc. IEEE, 80, 365.

Teodoro, C.G.; Schramm, D.U.; Sthel, M.S.; Lima, G.; Rocha, M.V.; Tavares, J. & Vargas, H.

Trenberth, K. E. & Shea, D. J., (2006). Atlantic hurricanes and natural variability in 2005,

Tyndall,J., (1881). Action of an intermittent beam of radiant heat llpon gaseous matter. *Proc.* 

Weidmann, D.; Kosterev, A.A.; Roller, C.; Curtel, R.F.; Fraser, M.P.; & Tittel, F.K. (2004).

West, G. A., (1983). Photoacoustic spectroscopy. *Rev. Sci. Instrum.* 54, 797-817. ISSN

Wolf, G.T. & Korsog, P.E. (1992) Ozone control strategies based on the ratio of volatile

Wright, S.; Duxbury, G. & Langford, N. (2006). A compact quantum-cascade laser based

Yu,C.S. & Kung, A.H. (1999). Grazing-incidence periodically poled LiNbO3 optical parametric oscillator, Optical Society of America, 16, 2233-2238. ISSN 1943-0620.

Zhang, J. Y.; Huang, J. Y. & Shen, J. Y. R. (1993) Opt. Soc. Am. B 10, 1758. ISSN 1520-8532. Zhou, J. X.; Hou, X.; Yang, K. X.; Tsai, S. J. J. &Michel, R. G. (1998) Appl. Spec. 52, 176 ISSN

Monitoring of ethylene by a pulsed quantun cascade laser. *Appl. Opt. 43*, 151-155.

organic compound to nitrogen oxides. *J. Air Waste Manage. Assoc*. *42*, 1173-1177.

spectrometer for monitoring the concentrations of methane and nitrous oxide in the troposphere, Appl. Phys. B 85, 243-249 . ISSN: 0946- 2171 (print) ISSN: 1432-0649

*Geophysical Research Letter,* Vol. 33, 1029 ISSN: 0094-8276.

Viengerov, M., (1938). *Dokl. Akad. Nauk SSSR* 19,687. ISSN 0002-3264.

ISSN 1559-128X (print) ISSN 2155-3165 (online).

fuels wide scale. *Int. Rev. Chem. Eng. 1*. 564-570. ISSN: 0020-6318.

Germany. ISBN: 3-540-42400-8.

ISSN 0018-9219.

1071-1086. ISSN: 1618-2650.

(printed): 0034-6748.

ISSN: 1047-3289.

(electronic version).

0021-9037.

*Acta* Part A 78, 458–462. ISSN 1386-1425.

*R. Soc. London* 31, 307-317. ISSN 1364-5021.

photoacoustic cell for pulsed laser analysis of gases at high temperature *Rev. Sci. Instrum* , 80, 123103. ISSN 0034-6748. (print) ISSN 1089-7623(electronic version). Steffen, W.; Sanderson, A.; Jäger, J.; Tyson, P.D.; Moore, B., III; Matson, P.A.; Richardson, K.;

Oldfield, F.; Schellnhuber, H.-J.; Turner, B.L., II &Wasson, R.J (2004). *Global Change and the Earth System: A Planet under Pressure*; Springer-Verlag: Heidelberg,

M.P.P. (2010)Atmospheric pollution: Global warming and a possible use of bio-

M.P.P.;Alexandre, J.; Toledo, R. ; Silva, M.G. & H. Vargas, 2011). CO2 laser photoacoustic detection of ammonia emitted by ceramic industries, *Spectrochimica* 

(2010). CO2 laser photoacoustic detection of ethylene emitted by diesel engines used in urban public transports. *Infr. Phys. Technol. 53*, 151-155. ISSN 1424-8220. Thomas, S., (2006). Photoacoustic Spectroscopy for process analysis *Anal. Bioanal. Chem*., *384*, It is no doubt that mass spectrometry (MS) is widely applied to several research fields. In recent years, design and development of novel miniaturized mass spectrometers have been at the forefront of research in MS. A lot of miniaturized mass spectrometers are reported and commercialized from a lot of institutes and industries. The instruments have widespread applications, for example, detection and identification of chemical and biological hazards for the homeland security (Contreras et al., 2008; Smith et al., 2011), the food safety (Garcia-Reyes et al., 2009) and so on. Because of their small size and lightweight, miniaturized mass spectrometers have a potential for field use. These features are absolutely suitable for on-site environmental analyses, especially in greenhouse gases analyses. However, there are few reports on application of miniaturized mass spectrometers in this research field. This chapter describes instrumentation and application feasibility of miniaturized mass spectrometers to the greenhouse gases analysis. This chapter is consists of following four sections.


## **2. "On-site mass spectrometry" using miniaturized mass spectrometers**

In this section, a novel analytical concept "On-site mass spectrometry", overview of several reported miniaturized mass spectrometers and issues for the field use are mentioned.

#### **2.1 On-site mass spectrometry**

The basis of science is to measure phenomena of nature in real-time and with strict accuracy. However, we know that it is the most difficult to perform such measurements using high

Miniaturized Mass Spectrometer in Analysis of

**Instrument Ionization method Mass (***m/z***)** 

MALDI, matrix-assisted laser desorption ionization.

**2.3 A few issues for miniaturized mass spectrometers** 

spectrometers is less than a few hundreds as shown in Table 1.

to avoid false positive and false negative from contaminant peaks.

**Mini 11** MIMS, direct leak,

**HAPSITE** 

**Suitcase** 

MULTUM-S

**HAPSITE** 

**Suitcase** 

**MULTUM-S** 

Greenhouse Gases: The Performance and Possibilities 237

**Instrument Weight (kg) Power (W) Size (***H* **x** *L* **x** *W* **mm) Analyzer type Mini 11** 4 30 180 x 220 x 120 Rectilinear ion trap **Guardion-7** 14.5 120 380 x 390 x 230 Toroidal ion trap

**system** 19 30 180 x 460 x 430 QMF **Griffin 450** 8.6 600 488 x 488 x 536 Cylindrical ion trap

**TOF** - - - TOF

II 36 600 450 x 640 x 230 TOF

**Guardion-7** EI 40 – 500 540 Torion

**system** EI 41 – 300 - INFICON **Griffin 450** EI 40 – 425 400 FLIR

**TOF** MALDI 50 000 30 Johns Hopkins

**IonCam** EI 7 – 250 250 O. I. Analytical

Table 1. Specifications of miniaturized mass spectrometers. MIMS, membrane inlet MS; EI, electron ionization; DESI, desorption electrospray ionization; ESI, electro spray ionization;

Miniaturized instruments, especially ion traps or QMF described above, appear to have usable performance in the field use. However, sensitivity and mass resolution in these physically smaller devices are lower compared to laboratory instruments. To overcome the loss of sensitivity due to lower transmission of ions into and out of the analyzer, ion traps using array geometry were proposed. On the other hand, high mass resolution cannot in principal be obtained in QMF or ion traps. The typical mass resolution in miniaturized mass

We consider that miniaturized instruments would be more commonly utilized in the future. In particular, high mass resolution is important when it is difficult to perform sufficient sample preparations as described in section 1. In this case, high mass resolution is important

How can we realize miniaturized high mass resolution mass spectrometers? In general, mass spectrometers for high mass resolution mass spectrometry are magnetic sector mass analyzers, Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometers (Marshall

**II** EI 1 - 1000 30 000 Osaka University

**IonCam** 19 150 438 x 432 x 254 Mattauch-Herzog sector

**range Resolution Vendor/institute** 

Technologies Inc.

Univ.

ESI, DESI 2000 100 Purdue University

performance analyzers. We must often prepare a lot of appropriate instruments to detect each target. In this sense, the features of mass spectrometers, for example variations of detectable targets, sensitivity and throughput, have advantages over other analytical instruments. Based on this point, "On-site mass spectrometry" is to bring high specifications mass spectrometers into field sites and to provide high quality mass spectrometric data in real-time. This concept is challenging, but on-site mass spectrometry attracts rising attention.

Considering conventional analytical methodologies using mass spectrometers, samples are usually taken from field sites. Then, the samples are brought into laboratories and purified before introducing into high performance mass spectrometers. Here, high performance mass spectrometers equipped in the laboratory have generally large size. To bring mass spectrometers into the on-site is required in several research fields, however, can we bring such instruments into the on-site? The answer is probably impossible. It is essential to miniaturize mass spectrometers with keeping performances ideally. Furthermore, we have to require not only simple and high performance miniaturized mass spectrometers to measure poorly purified samples but also simple sampling methods/sample preparations.

#### **2.2 Overview of miniaturized mass spectrometers**

Methods to reduce weight and size were attempted by various research groups. According to the reported papers on miniaturized mass spectrometers, a wide variety of instruments type including ion traps, quadrupole mass filters (QMF) (Geear et al., 2005), magnetic sector mass spectrometers (Diaz et al., 2001a; Diaz et al., 2001b), and TOF mass spectrometers (Cornish&Cotter, 1997; Cotter et al., 1999; Berkout et al., 2001; Ecelberger et al., 2004) were described. The main specifications are summarized in Table 1. It is considered that ion traps or QMF are more favourable than other instruments for miniaturization. In fact, almost all commercialized miniature mass spectrometers, for example *Guardion-7* (Lammert et al., 2006) and *Griffin Analytical 600* in Table 1, have adopted ion traps and QMF. In addition, it can be noted that there is a large variety of ion traps: (1) three dimensional hyperbolic ion traps, (2) rectilinear ion traps (Liang et al., 2008; Fico et al., 2009; Li et al., 2009; Ouyang et al., 2009), (3) toroidal ion traps (Lammert et al., 2006), (4) planar electrode ion traps (Austin et al., 2007; Austin et al., 2008; Yang et al., 2008), and (5) cylindrical ion traps (Van Amerom et al., 2008; Wells et al., 2008; Chaudhary et al., 2009).

Why were a lot of miniaturized mass spectrometers developed using ion trap techniques? The main reasons for choosing ion traps for portable instruments are that ion traps provide a relaxed vacuum condition, simple structures for easily miniaturized geometry with weight saving. A portable ion trap mass spectrometer system "Mini 11" was reported in 2009 (Gao et al., 2009), whose total weight with batteries is 5.0 kg, power consumption is 35 W, and dimensions are 22 cm x 12 cm x 18 cm. In addition to the miniaturized characteristics, the instrument has been coupled with wide varieties of ambient ionization sources, for example desorption electrospray ionization, electrospray ionization and paper spray ionization (Li et al., 2011; Soparawalla et al., 2011).

performance analyzers. We must often prepare a lot of appropriate instruments to detect each target. In this sense, the features of mass spectrometers, for example variations of detectable targets, sensitivity and throughput, have advantages over other analytical instruments. Based on this point, "On-site mass spectrometry" is to bring high specifications mass spectrometers into field sites and to provide high quality mass spectrometric data in real-time. This concept is challenging, but on-site mass spectrometry attracts rising

Considering conventional analytical methodologies using mass spectrometers, samples are usually taken from field sites. Then, the samples are brought into laboratories and purified before introducing into high performance mass spectrometers. Here, high performance mass spectrometers equipped in the laboratory have generally large size. To bring mass spectrometers into the on-site is required in several research fields, however, can we bring such instruments into the on-site? The answer is probably impossible. It is essential to miniaturize mass spectrometers with keeping performances ideally. Furthermore, we have to require not only simple and high performance miniaturized mass spectrometers to measure poorly purified samples but also simple sampling

Methods to reduce weight and size were attempted by various research groups. According to the reported papers on miniaturized mass spectrometers, a wide variety of instruments type including ion traps, quadrupole mass filters (QMF) (Geear et al., 2005), magnetic sector mass spectrometers (Diaz et al., 2001a; Diaz et al., 2001b), and TOF mass spectrometers (Cornish&Cotter, 1997; Cotter et al., 1999; Berkout et al., 2001; Ecelberger et al., 2004) were described. The main specifications are summarized in Table 1. It is considered that ion traps or QMF are more favourable than other instruments for miniaturization. In fact, almost all commercialized miniature mass spectrometers, for example *Guardion-7* (Lammert et al., 2006) and *Griffin Analytical 600* in Table 1, have adopted ion traps and QMF. In addition, it can be noted that there is a large variety of ion traps: (1) three dimensional hyperbolic ion traps, (2) rectilinear ion traps (Liang et al., 2008; Fico et al., 2009; Li et al., 2009; Ouyang et al., 2009), (3) toroidal ion traps (Lammert et al., 2006), (4) planar electrode ion traps (Austin et al., 2007; Austin et al., 2008; Yang et al., 2008), and (5) cylindrical ion traps (Van Amerom et al., 2008; Wells et al., 2008;

Why were a lot of miniaturized mass spectrometers developed using ion trap techniques? The main reasons for choosing ion traps for portable instruments are that ion traps provide a relaxed vacuum condition, simple structures for easily miniaturized geometry with weight saving. A portable ion trap mass spectrometer system "Mini 11" was reported in 2009 (Gao et al., 2009), whose total weight with batteries is 5.0 kg, power consumption is 35 W, and dimensions are 22 cm x 12 cm x 18 cm. In addition to the miniaturized characteristics, the instrument has been coupled with wide varieties of ambient ionization sources, for example desorption electrospray ionization, electrospray ionization and paper spray ionization (Li et

attention.

methods/sample preparations.

Chaudhary et al., 2009).

al., 2011; Soparawalla et al., 2011).

**2.2 Overview of miniaturized mass spectrometers** 


Table 1. Specifications of miniaturized mass spectrometers. MIMS, membrane inlet MS; EI, electron ionization; DESI, desorption electrospray ionization; ESI, electro spray ionization; MALDI, matrix-assisted laser desorption ionization.

## **2.3 A few issues for miniaturized mass spectrometers**

Miniaturized instruments, especially ion traps or QMF described above, appear to have usable performance in the field use. However, sensitivity and mass resolution in these physically smaller devices are lower compared to laboratory instruments. To overcome the loss of sensitivity due to lower transmission of ions into and out of the analyzer, ion traps using array geometry were proposed. On the other hand, high mass resolution cannot in principal be obtained in QMF or ion traps. The typical mass resolution in miniaturized mass spectrometers is less than a few hundreds as shown in Table 1.

We consider that miniaturized instruments would be more commonly utilized in the future. In particular, high mass resolution is important when it is difficult to perform sufficient sample preparations as described in section 1. In this case, high mass resolution is important to avoid false positive and false negative from contaminant peaks.

How can we realize miniaturized high mass resolution mass spectrometers? In general, mass spectrometers for high mass resolution mass spectrometry are magnetic sector mass analyzers, Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometers (Marshall

Miniaturized Mass Spectrometer in Analysis of

reduced. The total path length of one cycle was 1.308 m.

Fig. 1. Development history of MULTUM series.

deca peptide of angiotensin I.

Greenhouse Gases: The Performance and Possibilities 239

was fixed on a base plate of 40 cm x 40 cm. Using the electron ionization (EI) for a gas analysis, a maximum mass resolution using this instrument of 350 000 (*m/z* 28 of N2+) was achieved after 500 cycles (approximately 645 m in flight length). This system was not simple for operation; 28 electrostatic quadrupole lenses were used. As a next generation multi-turn mass spectrometer, we studied more simplified optical geometries. As a result, a new geometry of "MULTUM II" was designed and constructed (Okumura et al., 2004a; Okumura et al., 2004b). In the MULTUM II geometry, no quadrupole lenses were used. MULTUM II consisted of only four toroidal electrostatic sectors. The components were dramatically

The maximum mass resolution of 250 000 (*m/z* 28 of N2+) was achieved after 1200 cycles (approximately 1500 m in flight length). MULTUM II was equipped with not only the EI ion source but a matrix-assisted laser desorption/ionization (MALDI) ion source for biological applications. Using the MALDI ion source, a mass resolution of 61 000 was achieved for a

Recently, we have been developing various types of mass spectrometers based on the MULTUM II technology. The first instrument was a tandem TOF mass spectrometer "MULTUM-TOF/TOF" for the structural analysis of biomolecules (Toyoda et al., 2007). The second instruments were used for imaging mass spectrometry. Here, imaging mass

et al., 1998; Hu et al., 2005), and TOF mass spectrometers. However, if high mass resolution is to be achieved in these instruments, the size of instruments will become large and heavy. Especially, magnetic sector mass analyzers and FT-ICR mass spectrometers are required large electrical magnets and superconductive magnets. Therefore, it is difficult for simply miniaturized these instruments to improve mass resolution. This point is supported by specifications of IonCam listed in Table 1 (Hadjar et al., 2011).

According to the structural simplicity and weight, TOF mass spectrometers are more favorable for reduction in size. Here, the mass resolution (*m/m*) of TOF can be written as

$$\frac{m}{\Delta m} = \frac{T}{2\Delta T} \tag{1}$$

*T* and *T* are TOF and a peak width (FWHM: full-width at half maximum), respectively. We can easily find that mass resolution is directly proportional to TOF (i. e., the size of the instrument). Therefore, simply shortening of the flight length to miniaturize the instrumentation size decreases mass resolution.

To overcome this fundamental problem, the flight length is extended by various methods. The proposed systems are listed as follows:


In our laboratory, the multi-turn type TOF mass spectrometers, which have a figure of eight flight path, are mainly designed and constructed. The first multi-turn TOF mass spectrometer "MULTUM-Linear plus" was constructed. In the next section, we will introduce the overview and the developing history.

## **3. History of multi-turn time-of-flight (TOF) mass spectrometers in Osaka University**

Detailed discussions about the ion optical conditions which need to be met for operation of the multi-turn TOF mass spectrometer have been given elsewhere (Ishihara et al., 2000). Here, we simply explain the overview of our developed system and features of the ion optical system of the multi-turn TOF mass spectrometer.

### **3.1 Development history of MULTUM at Osaka University**

Figure 1 shows the development history of multi-turn TOF mass spectrometers at Osaka University. We designed and constructed the first multi-turn TOF mass spectrometer "MULTUM Linear plus" as a laboratory model for cometary exploration (Matsuo et al., 1999; Toyoda et al., 2000; Toyoda et al., 2003). The system consists of four discrete units, each comprised of an electrostatic quadrupole lens, a cylindrical electrostatic sector and an electrostatic quadrupole lens. The total path length of one cycle is 1.284 m. The entire system

et al., 1998; Hu et al., 2005), and TOF mass spectrometers. However, if high mass resolution is to be achieved in these instruments, the size of instruments will become large and heavy. Especially, magnetic sector mass analyzers and FT-ICR mass spectrometers are required large electrical magnets and superconductive magnets. Therefore, it is difficult for simply miniaturized these instruments to improve mass resolution. This point is supported by

According to the structural simplicity and weight, TOF mass spectrometers are more

<sup>2</sup> *m T*

can easily find that mass resolution is directly proportional to TOF (i. e., the size of the instrument). Therefore, simply shortening of the flight length to miniaturize the

To overcome this fundamental problem, the flight length is extended by various methods.

1. Electrostatic multi-pass mirror systems (Wollnik&Przewloka, 1990; Casares et al., 2001;

2. Helical or jig-saw type systems (Matsuda, 2000; Satoh et al., 2005; Satoh et al., 2007;

In our laboratory, the multi-turn type TOF mass spectrometers, which have a figure of eight flight path, are mainly designed and constructed. The first multi-turn TOF mass spectrometer "MULTUM-Linear plus" was constructed. In the next section, we will

3. Multi-turn ion optical geometries using electrostatic sectors (Poschenrieder, 1972).

**3. History of multi-turn time-of-flight (TOF) mass spectrometers in Osaka** 

Detailed discussions about the ion optical conditions which need to be met for operation of the multi-turn TOF mass spectrometer have been given elsewhere (Ishihara et al., 2000). Here, we simply explain the overview of our developed system and features of the ion

Figure 1 shows the development history of multi-turn TOF mass spectrometers at Osaka University. We designed and constructed the first multi-turn TOF mass spectrometer "MULTUM Linear plus" as a laboratory model for cometary exploration (Matsuo et al., 1999; Toyoda et al., 2000; Toyoda et al., 2003). The system consists of four discrete units, each comprised of an electrostatic quadrupole lens, a cylindrical electrostatic sector and an electrostatic quadrupole lens. The total path length of one cycle is 1.284 m. The entire system

*T* are TOF and a peak width (FWHM: full-width at half maximum), respectively. We

*m T* (1)

*m*) of TOF can be written as

specifications of IonCam listed in Table 1 (Hadjar et al., 2011).

favorable for reduction in size. Here, the mass resolution (*m/*

instrumentation size decreases mass resolution.

The proposed systems are listed as follows:

Yavor et al., 2008; Satoh et al., 2011).

introduce the overview and the developing history.

optical system of the multi-turn TOF mass spectrometer.

**3.1 Development history of MULTUM at Osaka University** 

Wollnik&Casares, 2003).

*T* and 

**University** 

was fixed on a base plate of 40 cm x 40 cm. Using the electron ionization (EI) for a gas analysis, a maximum mass resolution using this instrument of 350 000 (*m/z* 28 of N2+) was achieved after 500 cycles (approximately 645 m in flight length). This system was not simple for operation; 28 electrostatic quadrupole lenses were used. As a next generation multi-turn mass spectrometer, we studied more simplified optical geometries. As a result, a new geometry of "MULTUM II" was designed and constructed (Okumura et al., 2004a; Okumura et al., 2004b). In the MULTUM II geometry, no quadrupole lenses were used. MULTUM II consisted of only four toroidal electrostatic sectors. The components were dramatically reduced. The total path length of one cycle was 1.308 m.

Fig. 1. Development history of MULTUM series.

The maximum mass resolution of 250 000 (*m/z* 28 of N2+) was achieved after 1200 cycles (approximately 1500 m in flight length). MULTUM II was equipped with not only the EI ion source but a matrix-assisted laser desorption/ionization (MALDI) ion source for biological applications. Using the MALDI ion source, a mass resolution of 61 000 was achieved for a deca peptide of angiotensin I.

Recently, we have been developing various types of mass spectrometers based on the MULTUM II technology. The first instrument was a tandem TOF mass spectrometer "MULTUM-TOF/TOF" for the structural analysis of biomolecules (Toyoda et al., 2007). The second instruments were used for imaging mass spectrometry. Here, imaging mass

Miniaturized Mass Spectrometer in Analysis of

**4.1 Overview of MULTUM-S II** 

the multi-turn TOF mass spectrometer.

of the analyzer, (b) the outside of the system.

miniaturized TOF mass spectrometer "MULTUM-S II"

Greenhouse Gases: The Performance and Possibilities 241

In this section, we explain the system and novel characteristics of the recently developed

Photograph of the developed MULTUM-S II system are shown in Fig. 2. The size of the analyzer is less than 20 cm x 20 cm, which is the half size of MULTUM II (Fig. 2a). The photograph of the whole system is shown in Fig. 2b. The developed system consists of the following: the ion source, multi-turn mass analyzer, vacuum system, and high voltage circuit unit. The complete mass spectrometer weighs 35 kg. The total size of the instrument is 45 cm x 25 cm x 64 cm. The equipped ionization source is a two-stage acceleration ion source of EI type introduced by W. C. Wiley and I. H. McLaren (Wiley&McLaren, 1955). The accelerated ions are focused using the Einzel lens. After focusing, the ions are injected into

Fig. 2. Desktop high-mass resolution TOF mass spectrometer "MULTUM-S II"; (a) the inside

The main geometry of the analyzer was the same as MULTUM II. In the miniaturized instruments, it is an important issue on how ions are introduced with high efficiency into the mass analyzer from the ion source. Additionally, introduced ions need to travel with stability in the closed orbit to obtain high mass resolution. In the previous MULTUM system described in section 3, the ion beam passed through small holes in the outer electrodes of two of the electric sectors. When ions were injected or ejected, the voltages applied to the sector electrodes were switched. To prevent the reduction of resolution due to instability of the power supply for switching, we offer higher stability (< 50 ppm) (Toyoda et al., 2007). In order to overcome the problem of ion injection/ejection and stable traveling, MULTUM-S II has two additional sectors shown in Fig. 2a. Since these sectors specialized in ion injection/ejection, the static voltage is simply applied to the orbiting sectors. For this reason,

the electrical circuits for MULTUM-S II could be simplified and miniaturized.

**4. Miniaturized multi-turn TOF mass spectrometer "MULTUM-S II"** 

spectrometry is a novel visualization method using mass spectrometry (Amstalden et al., 2010). The stigmatic imaging was performed with "MULTUM-IMG" (Hazama et al., 2008a; Hazama et al., 2008b). Another instrument was equipped with an ionization source for secondary ion mass spectrometry (SIMS), so that the instrument was used for high spatial resolution (approximately sub micrometer) imaging mass spectrometry. The third instruments were miniaturized multi-turn TOF mass spectrometers "MULTUM-S" and "MULTUM-S II" (Shimma et al., 2010). MULTUM-S was the first prototype. This instrument was manufactured using a wide-use lathe and milling machine, resulting in a lack of manufacturing precision and assembly accuracy. The "MULTUM-S II" was optimized the design of ion optics and manufacturing precision. The detailed descriptions of MULTUM-S II are found in section 4.

#### **3.2 Ion optics of MULTUM**

There are two conditions that are required for multi-turn systems. The first is the geometrical conditions, namely the necessity to close the ion optical orbit. Multi-turn TOF mass spectrometer geometries with such an orbit have previously been proposed (Poschenrieder, 1972; Sakurai et al., 1999). They did not, however, satisfy the second condition, namely the "perfect focusing" condition (Ishihara et al., 2000). Therefore, in these cases, it is expected that the ion beam will diverge in both time and space. Then, both the mass resolution and the ion transmission (sensitivity) are compromised as the number of cycles around the instrument increases. To avoid this problem, ions should return to the point of origin in the system. In other words, the absolute value of the position and angle at the final position (which is identical with the initial position) should be the same as at the initial position in both the horizontal and vertical directions. Such conditions can be expressed using the transfer matrix method in the first order approximation as

$$
\begin{pmatrix} x\_{\prime} \\ a\_{\prime} \\ y\_{\prime} \\ \beta\_{\prime} \\ \beta\_{\prime} \\ \gamma \\ \delta \\ l\_{\prime} \end{pmatrix} = \begin{pmatrix} \pm 1 & \underline{0} & 0 & 0 & 0 & \underline{0} & 0 \\ \underline{0} & \pm 1 & 0 & 0 & 0 & \underline{0} & 0 \\ 0 & 0 & \pm 1 & \underline{0} & 0 & 0 & 0 \\ 0 & 0 & \underline{0} & \pm 1 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 1 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 0 & 0 & 1 & 0 \\ \underline{0} & \underline{0} & 0 & 0 & 0 & R(l|\gamma) & \underline{0} & 1 \end{pmatrix} \times \begin{pmatrix} x\_{\prime} \\ a\_{\prime} \\ y\_{\prime} \\ \beta\_{\prime} \\ \gamma \\ \delta \\ l\_{\prime} \end{pmatrix} \tag{2}
$$

Here, the ion optical position vectors (*x,, y, ,,* ), where *x, y* and , denote the lateral and angular deviations of the ion under consideration from a reference ion at the object. The mass and energy deviations and pathlength deviation are described as , and . The subscripts *i* and *f* represent initial and final position, respectively. It should be noted that the character *0* (zero with underline) means the matrix element which should be forced to be zero. Accordingly, we require the "nine-fold focusing", i.e. the nine *0* elements should be zero. By introducing symmetry in the arrangement of the ion optical components (i. e. the figure of eight geometry), multiple focusing conditions are easily achieved. We found ion optical systems ("MULTUM", "MULTUM II") for a multi-turn TOF mass spectrometer which satisfy perfect focusing.

## **4. Miniaturized multi-turn TOF mass spectrometer "MULTUM-S II"**

In this section, we explain the system and novel characteristics of the recently developed miniaturized TOF mass spectrometer "MULTUM-S II"

## **4.1 Overview of MULTUM-S II**

240 Greenhouse Gases – Emission, Measurement and Management

spectrometry is a novel visualization method using mass spectrometry (Amstalden et al., 2010). The stigmatic imaging was performed with "MULTUM-IMG" (Hazama et al., 2008a; Hazama et al., 2008b). Another instrument was equipped with an ionization source for secondary ion mass spectrometry (SIMS), so that the instrument was used for high spatial resolution (approximately sub micrometer) imaging mass spectrometry. The third instruments were miniaturized multi-turn TOF mass spectrometers "MULTUM-S" and "MULTUM-S II" (Shimma et al., 2010). MULTUM-S was the first prototype. This instrument was manufactured using a wide-use lathe and milling machine, resulting in a lack of manufacturing precision and assembly accuracy. The "MULTUM-S II" was optimized the design of ion optics and manufacturing precision. The detailed descriptions of MULTUM-S

There are two conditions that are required for multi-turn systems. The first is the geometrical conditions, namely the necessity to close the ion optical orbit. Multi-turn TOF mass spectrometer geometries with such an orbit have previously been proposed (Poschenrieder, 1972; Sakurai et al., 1999). They did not, however, satisfy the second condition, namely the "perfect focusing" condition (Ishihara et al., 2000). Therefore, in these cases, it is expected that the ion beam will diverge in both time and space. Then, both the mass resolution and the ion transmission (sensitivity) are compromised as the number of cycles around the instrument increases. To avoid this problem, ions should return to the point of origin in the system. In other words, the absolute value of the position and angle at the final position (which is identical with the initial position) should be the same as at the initial position in both the horizontal and vertical directions. Such conditions can be

> 1 0 0 0 0 00 0 1 0 0 0 00 0 0 1 0 0 00 0 0 0 1 0 00 0 0 0 0 1 00 0 0 0 0 0 10 0 0 0 0 ( )01

(2)

denote the lateral

, and . The

, 

), where *x, y* and

*f i f i f i f i*

*x x*

*y y*

*f i*

and angular deviations of the ion under consideration from a reference ion at the object. The

subscripts *i* and *f* represent initial and final position, respectively. It should be noted that the character *0* (zero with underline) means the matrix element which should be forced to be zero. Accordingly, we require the "nine-fold focusing", i.e. the nine *0* elements should be zero. By introducing symmetry in the arrangement of the ion optical components (i. e. the figure of eight geometry), multiple focusing conditions are easily achieved. We found ion optical systems ("MULTUM", "MULTUM II") for a multi-turn TOF mass spectrometer

*l R l l*

*, y, ,,* 

mass and energy deviations and pathlength deviation are described as

expressed using the transfer matrix method in the first order approximation as

Here, the ion optical position vectors (*x,*

which satisfy perfect focusing.

II are found in section 4.

**3.2 Ion optics of MULTUM** 

Photograph of the developed MULTUM-S II system are shown in Fig. 2. The size of the analyzer is less than 20 cm x 20 cm, which is the half size of MULTUM II (Fig. 2a). The photograph of the whole system is shown in Fig. 2b. The developed system consists of the following: the ion source, multi-turn mass analyzer, vacuum system, and high voltage circuit unit. The complete mass spectrometer weighs 35 kg. The total size of the instrument is 45 cm x 25 cm x 64 cm. The equipped ionization source is a two-stage acceleration ion source of EI type introduced by W. C. Wiley and I. H. McLaren (Wiley&McLaren, 1955). The accelerated ions are focused using the Einzel lens. After focusing, the ions are injected into the multi-turn TOF mass spectrometer.

Fig. 2. Desktop high-mass resolution TOF mass spectrometer "MULTUM-S II"; (a) the inside of the analyzer, (b) the outside of the system.

The main geometry of the analyzer was the same as MULTUM II. In the miniaturized instruments, it is an important issue on how ions are introduced with high efficiency into the mass analyzer from the ion source. Additionally, introduced ions need to travel with stability in the closed orbit to obtain high mass resolution. In the previous MULTUM system described in section 3, the ion beam passed through small holes in the outer electrodes of two of the electric sectors. When ions were injected or ejected, the voltages applied to the sector electrodes were switched. To prevent the reduction of resolution due to instability of the power supply for switching, we offer higher stability (< 50 ppm) (Toyoda et al., 2007).

In order to overcome the problem of ion injection/ejection and stable traveling, MULTUM-S II has two additional sectors shown in Fig. 2a. Since these sectors specialized in ion injection/ejection, the static voltage is simply applied to the orbiting sectors. For this reason, the electrical circuits for MULTUM-S II could be simplified and miniaturized.

Miniaturized Mass Spectrometer in Analysis of

merged in one spectrum as shown in Fig. 3.

MULTUM-S II was very wide.

low electrical noise.

miniaturized instrument.

**4.3 Obtained mass spectra of low mass ions using MULTUM-S II** 

Greenhouse Gases: The Performance and Possibilities 243

the timing of Ion Gate and Ejection Electrode. Users can only input the target mass value and the number of cycles. After spectra acquisition, obtained segment mass spectra are

In the specifications of mass spectrometer, the detectable mass range is important. The wide mass range is obviously preferred. The upper mass is restricted to the ionization methods, so that the upper mass value of MULTUM-S II is approximately *m/z* 1000. The lower limit is also important for the simultaneous gas analysis. However, high resolution MS in the low mass region, which is less than *m/z* 50, is generally difficult due to the electric noise from the instruments. As shown in Table 1, the lower mass limits of almost all instruments are *m/z* 40. In our instrument, the electric circuits are very simple. This feature becomes a strong advantage in detection of low mass molecules. Figure 4 is detection of hydrogen. In this experiment, standard gas of hydrogen was directly introduced into the ionization chamber via a needle valve. The peak at *m/z* 2 derived from hydrogen molecules were clearly detected. The detected peaks around *m/z* 15 were derived from residual gases. This result demonstrated that the detectable mass range in

Fig. 4. Detection of molecules in the low mass region. Hydrogen molecule was detected with

The next result is a separation of helium atom (He) and deuterium molecules (D2). Since both He and D2 had the nominal mass of *m/z* 4, He and D2 were detected as one peak in the low mass resolution mass spectrum (Fig. 5a). To obtain this spectrum, the standard gases of He and D2 were directly introduced into the ionization chamber via the needle valve. Before the introduction, these two gases were mixed in the gas-sampling bag. Figure 5b is the high mass resolution mass spectrum at 10 cycles. The values of the accurate mass were 4.0026 in He and 4.0282 in D2. Although the mass difference between He and D2 was 0.025 u, MULTUM-S II was able to easily separate He and D2 even in the

#### **4.2 Data acquisition methods**

Fig. 3. Operation diagram of MULTUM-S II.

Block diagrams of the timing control are shown in Fig. 3. In the basic operation, the ions were extracted from the EI ion source by applying the pulsed voltage to the push electrode. The voltage applied to Injection Electrode was turned on when ions were injected into the analyzer part. After the ions had been injected into the analyzer, the voltage applied to Injection Electrode was turned off. On the other hand, the voltage applied to Ejection Electrode was initially switched off; thus, the ion packets flew into the closed orbit composed of Orbiting Electrodes. After a preset number of cycles, the voltage applied to Ejection Electrode was switched on, and the ion packets were ejected from the closed orbit. Then, the ions were detected. Therefore, the switching timing of Ejection Electrode controlled the number of cycles. Due to the performance of the digitizer equipped in our system, the repetition time of the cycle in Fig. 3 was 1 kHz. Therefore, acquisition time for one spectrum was 1 millisecond.

This feature becomes powerful tool for specific ion measurement, however "Overtaking problem" will appear in detection of ions over the wide mass range. During orbiting, lighter ions take over the heavier ions, because the ion speed depends on the mass of the ion. To avoid the over taking problem, the measuring mass range is divided into some segments. In this multi-segment mode, Ion Gate equipped in the orbiting trajectory controls the mass range. Therefore, the switching timing of Ion Gate and Ejection Electrode are individually configured as shown in Fig. 3. In this operation mode, our software automatically calculates

Block diagrams of the timing control are shown in Fig. 3. In the basic operation, the ions were extracted from the EI ion source by applying the pulsed voltage to the push electrode. The voltage applied to Injection Electrode was turned on when ions were injected into the analyzer part. After the ions had been injected into the analyzer, the voltage applied to Injection Electrode was turned off. On the other hand, the voltage applied to Ejection Electrode was initially switched off; thus, the ion packets flew into the closed orbit composed of Orbiting Electrodes. After a preset number of cycles, the voltage applied to Ejection Electrode was switched on, and the ion packets were ejected from the closed orbit. Then, the ions were detected. Therefore, the switching timing of Ejection Electrode controlled the number of cycles. Due to the performance of the digitizer equipped in our system, the repetition time of the cycle in Fig. 3 was 1 kHz. Therefore, acquisition time for

This feature becomes powerful tool for specific ion measurement, however "Overtaking problem" will appear in detection of ions over the wide mass range. During orbiting, lighter ions take over the heavier ions, because the ion speed depends on the mass of the ion. To avoid the over taking problem, the measuring mass range is divided into some segments. In this multi-segment mode, Ion Gate equipped in the orbiting trajectory controls the mass range. Therefore, the switching timing of Ion Gate and Ejection Electrode are individually configured as shown in Fig. 3. In this operation mode, our software automatically calculates

**4.2 Data acquisition methods** 

Fig. 3. Operation diagram of MULTUM-S II.

one spectrum was 1 millisecond.

the timing of Ion Gate and Ejection Electrode. Users can only input the target mass value and the number of cycles. After spectra acquisition, obtained segment mass spectra are merged in one spectrum as shown in Fig. 3.

## **4.3 Obtained mass spectra of low mass ions using MULTUM-S II**

In the specifications of mass spectrometer, the detectable mass range is important. The wide mass range is obviously preferred. The upper mass is restricted to the ionization methods, so that the upper mass value of MULTUM-S II is approximately *m/z* 1000. The lower limit is also important for the simultaneous gas analysis. However, high resolution MS in the low mass region, which is less than *m/z* 50, is generally difficult due to the electric noise from the instruments. As shown in Table 1, the lower mass limits of almost all instruments are *m/z* 40. In our instrument, the electric circuits are very simple. This feature becomes a strong advantage in detection of low mass molecules. Figure 4 is detection of hydrogen. In this experiment, standard gas of hydrogen was directly introduced into the ionization chamber via a needle valve. The peak at *m/z* 2 derived from hydrogen molecules were clearly detected. The detected peaks around *m/z* 15 were derived from residual gases. This result demonstrated that the detectable mass range in MULTUM-S II was very wide.

Fig. 4. Detection of molecules in the low mass region. Hydrogen molecule was detected with low electrical noise.

The next result is a separation of helium atom (He) and deuterium molecules (D2). Since both He and D2 had the nominal mass of *m/z* 4, He and D2 were detected as one peak in the low mass resolution mass spectrum (Fig. 5a). To obtain this spectrum, the standard gases of He and D2 were directly introduced into the ionization chamber via the needle valve. Before the introduction, these two gases were mixed in the gas-sampling bag. Figure 5b is the high mass resolution mass spectrum at 10 cycles. The values of the accurate mass were 4.0026 in He and 4.0282 in D2. Although the mass difference between He and D2 was 0.025 u, MULTUM-S II was able to easily separate He and D2 even in the miniaturized instrument.

Miniaturized Mass Spectrometer in Analysis of

Greenhouse Gases: The Performance and Possibilities 245

Real-time monitoring of N2O is required to elucidate the generating mechanism and investigate its trend of spread. N2O is known as the greenhouse gas, and the warming effect is about 310 times larger than CO2. Furthermore, N2O is one of the ozone-depleting substances (Ravishankara et al., 2009). If we try to carry out real-time monitoring of N2O and CO2 simultaneously using mass spectrometry, a mass spectrometer with high mass resolution is required, because the nominal mass of N2O is the same as that of CO2. However, the difference of accurate mass between CO2 and N2O is 0.0113 u, therefore these two peaks were expected to be separated in MULTUM-S II as shown in Fig. 5 (b) and Fig. 6 . In this experiment, the mixture of ultrapure CO2 and N2O (49.4%:50.6%) was purchased from DAIHO SANGYO Inc. (Minato-ku, Tokyo, Japan). This standard gas was introduced into the EI ion source via the needle valve. Figure 7 shows the obtained mass spectra by changing the number of cycles. Figure 7a shows the mass spectrum of the CO2 and N2O doublet peak after 10 cycles. In this cycle, the doublet peak did not still separate due to the lack of mass resolution. After 20 cycles (Fig. 7b), the top of the peak began to separate. After 50 cycles (Fig. 7c), these two peaks were

Fig. 7. Separation of CO2 and N2O doublet; the spectrum at (a) 10 cycles, (b) 20 cycles and (c)

50 cycles.

Fig. 5. Separation of 4He and D2 doublet; (a) the low mass resolution mass spectrum and (b) high mass resolution mass spectrum. The number of cycles was 10 cycles.

## **5. Applications for greenhouse gases detection using "MULTUM-S II"**

As shown in section 4, MULTUM-S II have a powerful potential especially for the gas detection. The final section describes application of MULTUM-S II for high resolution and simultaneous greenhouse gases detection, especially carbon dioxide (CO2), nitrous oxide (N2O) and methane (CH4). In the conventional methodology in the field study, researchers collect samples in the field, bring them into laboratory, and then analyze using laboratory equipped instruments. As another procedure, researchers prepare and place several detectors to detect each species directly in the field. However, a great variety of detectable species and portability in MULTUM-S II will be the powerful merits for the field study in the future. This high quality gas analyzer will also help the reduction of systematic error attributed to different detectors and measurement conditions.

#### **5.1 High mass resolution mass spectra of greenhouse gases**

Figure 6 is a high-mass resolution mass spectrum of CH4 at 10 cycles. In this spectrum, the peak of CH4 is separated from the oxygen peak (both nominal mass is *m/z* 16) derived from the fragment ion of the oxygen molecule or di-charged ion of oxygen molecule. The obtained mass resolution was 3200. We consider that this feature is merit for monitoring of CH4 without oxygen contamination.

Fig. 6. High-resolution mass spectrum of oxygen and methane.

Fig. 5. Separation of 4He and D2 doublet; (a) the low mass resolution mass spectrum and (b)

As shown in section 4, MULTUM-S II have a powerful potential especially for the gas detection. The final section describes application of MULTUM-S II for high resolution and simultaneous greenhouse gases detection, especially carbon dioxide (CO2), nitrous oxide (N2O) and methane (CH4). In the conventional methodology in the field study, researchers collect samples in the field, bring them into laboratory, and then analyze using laboratory equipped instruments. As another procedure, researchers prepare and place several detectors to detect each species directly in the field. However, a great variety of detectable species and portability in MULTUM-S II will be the powerful merits for the field study in the future. This high quality gas analyzer will also help the reduction of systematic error

Figure 6 is a high-mass resolution mass spectrum of CH4 at 10 cycles. In this spectrum, the peak of CH4 is separated from the oxygen peak (both nominal mass is *m/z* 16) derived from the fragment ion of the oxygen molecule or di-charged ion of oxygen molecule. The obtained mass resolution was 3200. We consider that this feature is merit for monitoring of

high mass resolution mass spectrum. The number of cycles was 10 cycles.

attributed to different detectors and measurement conditions.

CH4 without oxygen contamination.

**5.1 High mass resolution mass spectra of greenhouse gases** 

Fig. 6. High-resolution mass spectrum of oxygen and methane.

**5. Applications for greenhouse gases detection using "MULTUM-S II"** 

Real-time monitoring of N2O is required to elucidate the generating mechanism and investigate its trend of spread. N2O is known as the greenhouse gas, and the warming effect is about 310 times larger than CO2. Furthermore, N2O is one of the ozone-depleting substances (Ravishankara et al., 2009). If we try to carry out real-time monitoring of N2O and CO2 simultaneously using mass spectrometry, a mass spectrometer with high mass resolution is required, because the nominal mass of N2O is the same as that of CO2. However, the difference of accurate mass between CO2 and N2O is 0.0113 u, therefore these two peaks were expected to be separated in MULTUM-S II as shown in Fig. 5 (b) and Fig. 6 . In this experiment, the mixture of ultrapure CO2 and N2O (49.4%:50.6%) was purchased from DAIHO SANGYO Inc. (Minato-ku, Tokyo, Japan). This standard gas was introduced into the EI ion source via the needle valve. Figure 7 shows the obtained mass spectra by changing the number of cycles. Figure 7a shows the mass spectrum of the CO2 and N2O doublet peak after 10 cycles. In this cycle, the doublet peak did not still separate due to the lack of mass resolution. After 20 cycles (Fig. 7b), the top of the peak began to separate. After 50 cycles (Fig. 7c), these two peaks were

Fig. 7. Separation of CO2 and N2O doublet; the spectrum at (a) 10 cycles, (b) 20 cycles and (c) 50 cycles.

Miniaturized Mass Spectrometer in Analysis of

was three minutes.

capillary column is reduced.

O2, Ar) and (CO2, N2O)

column.

Greenhouse Gases: The Performance and Possibilities 247

**5.3 Combination of rough separation and high resolution MS for real-time monitoring**  In the conventional gas chromatographic technique, the main contents of air (N2, O2, Ar) and (CO2, N2O) can be separated using Carbon PLOT column. We firstly performed the gas chromatography of their content using a gas chromatography coupled with MULTUM-S II (Fig. 9a). In this experiment, the GS-CarbonPLOT column (30 m x 0.32 mm i.d.; 3 m film thickness) was purchased from Agilent Technologies (Santa Clara, CA, USA). The column temperature was kept an isothermal condition of 40oC. The obtained the total ion chromatogram and mass chromatogram of *m/z* 44 are shown in Fig. 9b. In these chromatograms, the non-polar contents of N2, O2 and Ar were clearly separated from CO2 and N2O. Furthermore, due to the different polarity of CO2 and N2O, the enough separation of these peaks was also achieved. However, the duration of analysis

The generation of N2O via activities of microorganisms is known as a fast process (HAYATSU et al., 2008; Kool et al., 2010). Therefore, higher sampling rate is likely to be required. According to the GC techniques, one of the methodologies to shorten analysis time is truncation of the capillary column. Understandably, the separation capability of the

Fig. 9. (a) GC-MULTUM system; (b) the conventional chromatographic separation of (N2,

Here, we consider that higher separation capability for the GC system is unnecessary in this case. As shown in Fig. 7, the complete separation of CO2 and N2O can be available in our system. Therefore, we have only to perform the rough GC separation of (N2, O2, Ar) and (CO2, N2O). To demonstrate this idea, we performed gas chromatography with a 10 m PLOT

The obtained chromatograms are shown in Fig. 10. We found that the separation was completed within one minute. In this experiment, the ions of N2 and O2 were not detected, because their retention time was too fast. According to the value of the vacuum gauge just after sample injection, the large amount of gases seemed to be injected into the ionization source at a time. Under this condition, electrical discharge could occur. Therefore, we applied the high voltage to the ionization source after five seconds of sample injection.

Due to the short capillary column, the peaks derived from CO2 and N2O were not separated in the total ion chromatogram shown in Fig. 10. However, the averaged mass spectra from the retention time 0.3 min to 0.35 min, the high-mass resolution mass

completely separated, and consequently, the obtained mass resolution was 30 000. This ultra high mass resolution was firstly achieved in the miniaturized mass spectrometers. This mass resolution was comparable to those of laboratory-equipped instruments. We would like to note that the maximal time to acquire one spectrum was 1 millisecond as described in section 4.2. Therefore, high mass resolution accompanying the fast data acquisition will be helpful for the on-site real-time monitoring.

Achieving high mass resolution is advantageous when trying to determine accurate masses. In the previous study, the mass accuracy of 2.3 ppm was achieved (Shimma et al., 2010). Availability of accurate mass measurement in the miniaturized mass spectrometer is another advantage to MULTUM-S II.

#### **5.2 Direct sample injection method**

We could confirm the separation of CO2 and N2O doublet using MULTUM-S II. The next experiment was to perform the simultaneous gas detection. In the simultaneous gas detection, there was concern about a dynamic range. To confirm the capability of detection in both low and high concentration species, detection of 30 ppm N2O in the air was performed. The standard gas of 30 ppm N2O (N2 balanced) was purchased from DAIHO SANGYO Inc. In this experiment, the operation mode of multi-segment mode was used. Three segments were configured to detect N2, O2, CO2 and N2O. The obtained mass spectrum at 50 cycles is shown in Fig. 8.

Fig. 8. Simultaneous gas detection of N2, O2, CO2 and N2O in 50 cycles.

As shown in Fig. 8, the balanced N2 peak was predominantly detected. The 30 ppm N2O was detected with complete separation of CO2, however, the peak intensities of other species were extremely low. Considering the concentration of N2O in the nature, the concentration is much lower than 30 ppm. Therefore, the dominant species of N2 and O2 in the air should be removed to realize the higher sensitive N2O detection.

completely separated, and consequently, the obtained mass resolution was 30 000. This ultra high mass resolution was firstly achieved in the miniaturized mass spectrometers. This mass resolution was comparable to those of laboratory-equipped instruments. We would like to note that the maximal time to acquire one spectrum was 1 millisecond as described in section 4.2. Therefore, high mass resolution accompanying the fast data acquisition will be helpful for

Achieving high mass resolution is advantageous when trying to determine accurate masses. In the previous study, the mass accuracy of 2.3 ppm was achieved (Shimma et al., 2010). Availability of accurate mass measurement in the miniaturized mass spectrometer is

We could confirm the separation of CO2 and N2O doublet using MULTUM-S II. The next experiment was to perform the simultaneous gas detection. In the simultaneous gas detection, there was concern about a dynamic range. To confirm the capability of detection in both low and high concentration species, detection of 30 ppm N2O in the air was performed. The standard gas of 30 ppm N2O (N2 balanced) was purchased from DAIHO SANGYO Inc. In this experiment, the operation mode of multi-segment mode was used. Three segments were configured to detect N2, O2, CO2 and N2O. The obtained mass

Fig. 8. Simultaneous gas detection of N2, O2, CO2 and N2O in 50 cycles.

be removed to realize the higher sensitive N2O detection.

As shown in Fig. 8, the balanced N2 peak was predominantly detected. The 30 ppm N2O was detected with complete separation of CO2, however, the peak intensities of other species were extremely low. Considering the concentration of N2O in the nature, the concentration is much lower than 30 ppm. Therefore, the dominant species of N2 and O2 in the air should

the on-site real-time monitoring.

another advantage to MULTUM-S II.

**5.2 Direct sample injection method** 

spectrum at 50 cycles is shown in Fig. 8.

## **5.3 Combination of rough separation and high resolution MS for real-time monitoring**

In the conventional gas chromatographic technique, the main contents of air (N2, O2, Ar) and (CO2, N2O) can be separated using Carbon PLOT column. We firstly performed the gas chromatography of their content using a gas chromatography coupled with MULTUM-S II (Fig. 9a). In this experiment, the GS-CarbonPLOT column (30 m x 0.32 mm i.d.; 3 m film thickness) was purchased from Agilent Technologies (Santa Clara, CA, USA). The column temperature was kept an isothermal condition of 40oC. The obtained the total ion chromatogram and mass chromatogram of *m/z* 44 are shown in Fig. 9b. In these chromatograms, the non-polar contents of N2, O2 and Ar were clearly separated from CO2 and N2O. Furthermore, due to the different polarity of CO2 and N2O, the enough separation of these peaks was also achieved. However, the duration of analysis was three minutes.

The generation of N2O via activities of microorganisms is known as a fast process (HAYATSU et al., 2008; Kool et al., 2010). Therefore, higher sampling rate is likely to be required. According to the GC techniques, one of the methodologies to shorten analysis time is truncation of the capillary column. Understandably, the separation capability of the capillary column is reduced.

Fig. 9. (a) GC-MULTUM system; (b) the conventional chromatographic separation of (N2, O2, Ar) and (CO2, N2O)

Here, we consider that higher separation capability for the GC system is unnecessary in this case. As shown in Fig. 7, the complete separation of CO2 and N2O can be available in our system. Therefore, we have only to perform the rough GC separation of (N2, O2, Ar) and (CO2, N2O). To demonstrate this idea, we performed gas chromatography with a 10 m PLOT column.

The obtained chromatograms are shown in Fig. 10. We found that the separation was completed within one minute. In this experiment, the ions of N2 and O2 were not detected, because their retention time was too fast. According to the value of the vacuum gauge just after sample injection, the large amount of gases seemed to be injected into the ionization source at a time. Under this condition, electrical discharge could occur. Therefore, we applied the high voltage to the ionization source after five seconds of sample injection.

Due to the short capillary column, the peaks derived from CO2 and N2O were not separated in the total ion chromatogram shown in Fig. 10. However, the averaged mass spectra from the retention time 0.3 min to 0.35 min, the high-mass resolution mass

Miniaturized Mass Spectrometer in Analysis of

Fig. 11. Proposed automatic gas sampler.

manufactured in the near future.

**6. Conclusion** 

(*m/*

ongoing.

Greenhouse Gases: The Performance and Possibilities 249

We successfully confirmed the methodology to inject the main components of air and greenhouse gases separately into the ionization source. Considering the field use, to bring GC system to the field is unrealistic for the real-time monitoring. To overcome this problem, we designed and developed a simple automatic gas sampler (Fig. 11). This sampling system is consists of a six port valve, two solenoid valves, a mass flow controller, a sampling loop, a diaphragm pump for the gas sampling, and the short PLOT column. The operation of valves is controlled using a lab-built LabVIEW base software. In our laboratory, the performance evaluation of this sampling system including the optimization of sampling condition is

Another task to realize the high sensitive on-site greenhouse gas monitor is reduction of background pressure in the analyzer. The typical pressure value during measurement was between 8 x 10-5 Pa and 1 x 10-4 Pa. This pressure was insufficient for the sub ppm level N2O detection. Due to higher background pressure, contamination peaks derived from residual gases in analyzer were detected. Ionization space in the EI source is understandably finite, therefore the total number of ions stored in the ionization source is also finite. This is known as the space charge effect. If the contaminant ions are reduced, the number of ions of interest will be relatively increased. To realize the pressure below 10-5 Pa level, we designed a highly sealed ionization chamber specialized for the gas monitoring. The material of the chamber was titanium to reduce outgassing from the inner chamber wall. This chamber will be

In this chapter, we described the overview of novel miniaturized mass spectrometers and their applications. Our developed the unique miniaturized TOF mass spectrometer "MULTUM-S II" has comparable mass resolution to lab-equipped mass spectrometers

*m* > 30 000). At present, we performed proof-of-concept studies for real-time

**5.4 Future tasks for on-site simultaneous greenhouse gas detection** 

spectrometry was performed (inset of Fig. 10a). The obtained mass chromatograms are also shown in Fig. 10b and Fig. 10c. These results indicate that our idea worked correctly. We could establish the combination of rough gas chromatograph and high mass resolution mass spectrometry.

Fig. 10. The rough chromatographic separation of CO2 and N2O; (a) total ion chromatogram, (b) the mass chromatogram of CO2, and (c) the mass chromatogram of N2O.

spectrometry was performed (inset of Fig. 10a). The obtained mass chromatograms are also shown in Fig. 10b and Fig. 10c. These results indicate that our idea worked correctly. We could establish the combination of rough gas chromatograph and high mass

Fig. 10. The rough chromatographic separation of CO2 and N2O; (a) total ion chromatogram,

(b) the mass chromatogram of CO2, and (c) the mass chromatogram of N2O.

resolution mass spectrometry.

#### **5.4 Future tasks for on-site simultaneous greenhouse gas detection**

We successfully confirmed the methodology to inject the main components of air and greenhouse gases separately into the ionization source. Considering the field use, to bring GC system to the field is unrealistic for the real-time monitoring. To overcome this problem, we designed and developed a simple automatic gas sampler (Fig. 11). This sampling system is consists of a six port valve, two solenoid valves, a mass flow controller, a sampling loop, a diaphragm pump for the gas sampling, and the short PLOT column. The operation of valves is controlled using a lab-built LabVIEW base software. In our laboratory, the performance evaluation of this sampling system including the optimization of sampling condition is ongoing.

Fig. 11. Proposed automatic gas sampler.

Another task to realize the high sensitive on-site greenhouse gas monitor is reduction of background pressure in the analyzer. The typical pressure value during measurement was between 8 x 10-5 Pa and 1 x 10-4 Pa. This pressure was insufficient for the sub ppm level N2O detection. Due to higher background pressure, contamination peaks derived from residual gases in analyzer were detected. Ionization space in the EI source is understandably finite, therefore the total number of ions stored in the ionization source is also finite. This is known as the space charge effect. If the contaminant ions are reduced, the number of ions of interest will be relatively increased. To realize the pressure below 10-5 Pa level, we designed a highly sealed ionization chamber specialized for the gas monitoring. The material of the chamber was titanium to reduce outgassing from the inner chamber wall. This chamber will be manufactured in the near future.

## **6. Conclusion**

In this chapter, we described the overview of novel miniaturized mass spectrometers and their applications. Our developed the unique miniaturized TOF mass spectrometer "MULTUM-S II" has comparable mass resolution to lab-equipped mass spectrometers (*m/m* > 30 000). At present, we performed proof-of-concept studies for real-time

Miniaturized Mass Spectrometer in Analysis of

25, No. 1, pp. 14-19, ISSN 0270-5214.

No. 7, pp. 1338-1347, ISSN 0003-2654.

pp. 1156-1166, ISSN 1057-7157.

0305.

1387-3806.

ISSN 0951-4198.

pp. 179-189, ISSN 1387-3806.

5, pp. 759-772, ISSN 1365-2389.

5174.

Greenhouse Gases: The Performance and Possibilities 251

Diaz J.A., Giese C.F. & Gentry W.R. (2001b). Sub-miniature ExB sector-field mass

Ecelberger S.A., Cornish T.J., Collins B.F., Lewis D.L. & Bryden W.A. (2004). Suitcase TOF: A

Fico M., Maas J.D., Smith S.A., Costa A.B., Ouyang Z., Chappell W.J. & Cooks R.G. (2009).

Gao L., Li G.T., Nie Z.X., Duncan J., Ouyang Z. & Cooks R.G. (2009). Characterization of a

Garcia-Reyes J.F., Jackson A.U., Molina-Diaz A. & Cooks R.G. (2009). Desorption

Geear M., Syms R.R.A., Wright S. & Holmes A.S. (2005). Monolithic MEMS quadrupole

Hadjar O., Johnson G., Laskin J., Kibelka G., Shill S., Kuhn K., Cameron C. & Kassan S.

Hayatsu M., Tago K. & Saito M. (2008). Various players in the nitrogen cycle: Diversity and

Hazama H., Aoki J., Nagao H., Suzuki R., Tashima T., Fujii K., Masuda K., Awazu K.,

Hu Q.Z., Noll R.J., Li H.Y., Makarov A., Hardman M. & Cooks R.G. (2005). The orbitrap: A

Ishihara M., Toyoda M. & Matsuo T. (2000). Perfect space and time focusing ion optics for

Kool D.M., Wrage N., Zechmeister-Boltenstern S., Pfeffer M., Brus D., Oenema O. & Van

*Science & Plant Nutrition,* Vol. 54, No. 1, pp. 33-45, ISSN 1747-0765.

food. *Anal. Chem.,* Vol. 81, No. 2, pp. 820-829, ISSN 0003-2700.

*Spectrom.,* Vol. 22, No. 4, pp. 612-623, ISSN 1044-0305.

spectrometer. *J. Am. Soc. Mass Spectrom.,* Vol. 12, No. 6, pp. 619-632, ISSN 1044-

man-portable time-of-flight mass spectrometer. *Johns Hopkins APL Tech. Dig.,* Vol.

Circular arrays of polymer-based miniature rectilinear ion traps. *Analyst,* Vol. 134,

discontinuous atmospheric pressure interface. multiple ion introduction pulses for improved performance. *Int. J. Mass Spectrom.,* Vol. 283, No. 1-3, pp. 30-34, ISSN

electrospray ionization mass spectrometry for trace analysis of agrochemicals in

mass spectrometers by deep silicon etching. *J. Microelectromech. S.,* Vol. 14, No. 5,

(2011). IonCCD fordirect position-sensitive charged-particle detection: From electrons and keV ions to hyperthermal biomolecular ions. *J. Am. Soc. Mass* 

functions of the microorganisms involved in nitrification and denitrification. *Soil* 

Toyoda M. & Naito Y. (2008a). Construction of a novel stigmatic MALDI imaging mass spectrometer. *Appl. Surf. Sci.,* Vol. 255, No. 4, pp. 1257-1263, ISSN 0169-4332. Hazama H., Nagao H., Suzuki R., Toyoda M., Masuda K., Naito Y. & Awazu K. (2008b).

Comparison of mass spectra of peptides in different matrices using matrix-assisted laser desorption/ionization and a multi-turn time-of-flight mass spectrometer, MULTUM-IMG. *Rapid Commun. Mass Spectrom.,* Vol. 22, No. 10, pp. 1461-1466,

new mass spectrometer *J. Mass Spectrom.,* Vol. 40, No. 4, pp. 430-443, ISSN 1076-

multiturn time of flight mass spectrometers. *Int. J. Mass Spectrom.,* Vol. 197, No. 1-3,

Groenigen J.-. (2010). Nitrifier denitrification can be a source of N2O from soil: A revised approach to the dual-isotope labelling method. *Eur. J. Soil Sci.,* Vol. 61, No.

greenhouse gas monitoring system using MULTUM-S II. Based on issues from this study, MULTUM-S II is under improvement for the high sensitive gas detection.

### **7. Acknowledgement**

The authors thank Prof. Ryusuke Hatano of Hokkaido University for his fruitful discussion. This work was supported by a Supporting Program for Creating University Ventures, from the Japan Science and Technology Agency (JST). One of the authors (M. T.) was supported by Grant in Aid for Young Scientists (A) (21685010) from the Ministry of Education, Culture, Sports, Science and Technology, Japan.

## **8. References**


greenhouse gas monitoring system using MULTUM-S II. Based on issues from this study,

The authors thank Prof. Ryusuke Hatano of Hokkaido University for his fruitful discussion. This work was supported by a Supporting Program for Creating University Ventures, from the Japan Science and Technology Agency (JST). One of the authors (M. T.) was supported by Grant in Aid for Young Scientists (A) (21685010) from the Ministry of Education, Culture,

Amstalden v.H., Smith D.F. & Heeren R.M.A. (2010). A concise review of mass spectrometry

Austin D.E., Peng Y., Hansen B.J., Miller I.W., Rockwood A.L., Hawkins A.R. & Tolley S.E.

Austin D.E., Wang M., Tolley S.E., Maas J.D., Hawkins A.R., Rockwood A.L., Tolley H.D.,

Berkout V.D., Cotter R.J. & Segers D.P. (2001). Miniaturized EI/Q/oa TOF mass

Casares A., Kholomeev A. & Wollnik H. (2001). Multipass time-of-flight mass spectrometers

Chaudhary A., van Amerom F.H.W. & Short R.T. (2009). Development of microfabricated

Contreras J.A., Murray J.A., Tolley S.E., Oliphant J.L., Tolley H.D., Lammert S.A., Lee E.D.,

Cornish T.J. & Cotter R.J. (1997). High-order kinetic energy focusing in an end cap reflectron

Cotter R.J., Fancher C. & Cornish T.J. (1999). Miniaturized time-of-flight mass spectrometer

Diaz J.A., Giese C.F. & Gentry W.R. (2001a). Portable double-focusing mass-spectrometer

*Mass Spectrom.,* Vol. 19, No. 10, pp. 1425-1434, ISSN 1044-0305.

imaging. *Journal of Chromatography A,* Vol. 1217, No. 25, pp. 3946-3954, ISSN 0021-

(2008). Novel ion traps using planar resistive electrodes: Implications for miniaturized mass analyzers. *J. Am. Soc. Mass Spectrom.,* Vol. 19, No. 10, pp. 1435-

Lee E.D. & Lee M.L. (2007). Halo ion trap mass spectrometer. *Anal. Chem.,* Vol. 79,

spectrometer. *J. Am. Soc. Mass Spectrom.,* Vol. 12, No. 6, pp. 641-647, ISSN 1044-

with high resolving powers. *Int. J. Mass Spectrom.,* Vol. 206, No. 3, pp. 267-273, ISSN

cylindrical ion trap mass spectrometer arrays. *J. Microelectromech. S.,* Vol. 18, No. 2,

Later D.W. & Lee M.L. (2008). Hand-portable gas chromatograph-toroidal ion trap mass spectrometer (GC-TMS) for detection of hazardous compounds. *J. Am. Soc.* 

time-of-flight mass spectrometer. *Anal.Chem.,* Vol. 69, No. 22, pp. 4615-4618, ISSN

for peptide and oligonucleotide analysis. *J. Mass Spectrom.,* Vol. 34, No. 12, pp.

system for field gas monitoring. *Field Anal. Chem. Technol.,* Vol. 5, No. 3, pp. 156-

MULTUM-S II is under improvement for the high sensitive gas detection.

**7. Acknowledgement** 

**8. References** 

9673.

0305.

1387-3806.

0003-2700.

Sports, Science and Technology, Japan.

1441, ISSN 1044-0305.

No. 7, pp. 2927-2932, ISSN 0003-2700.

pp. 442-448, ISSN 1057-7157.

1368-1372, ISSN 1076-5174.

167, ISSN 1086-900X.


Miniaturized Mass Spectrometer in Analysis of

1044-0305.

1097-0231.

Greenhouse Gases: The Performance and Possibilities 253

Satoh T., Sato T., Kubo A. & Tamura J. (2011). Tandem time-of-flight mass spectrometer with

Shimma S., Nagao H., Aoki J., Takahashi K., Miki S. & Toyoda M. (2010). Miniaturized high-

Smith J.N., Noll R.J. & Cooks R.G. (2011). Facility monitoring of chemical warfare agent

Soparawalla S., Tadjimukhamedov F.K., Wiley J.S., Ouyang Z. & Cooks R.G. (2011). In situ

Toyoda M., Giannakopulos A.E., Colburn A.W. & Derrick P.J. (2007). High-energy collision

Toyoda M., Ishihara M., Yamaguchi S., Ito H., Matsuo T., Roll R. & Rosenbauer H. (2000).

Toyoda M., Okumura D., Ishihara M. & Katakuse I. (2003). Multi-turn time-of-flight mass

Van Amerom F.H.W., Chaudhary A., Cardenas M., Bumgarner J. & Short R.T. (2008).

Wells J.M., Roth M.J., Keil A.D., Grossenbacher J.W., Justes D.R., Patterson G.E. & Barket

Wiley W.C. & McLaren I.H. (1955). Time-of-flight mass spectrometer with improved

Wollnik H. & Casares A. (2003). An energy-isochronous multi-pass time-of-flight mass

Wollnik H. & Przewloka M. (1990). Time-of-flight mass spectrometers with multiply

Yang M., Kim T.Y., Hwang H.C., Yi S.K. & Kim D.H. (2008). Development of a palm

resolution. *Rev. Sci. Instrum.,* Vol. 26, No. 12, pp. 1150-1157,.

Vol. 227, No. 2, pp. 217-222, ISSN 1387-3806.

path. *Anal. Chem.,* Vol. 82, No. 20, pp. 8456-8463, ISSN 0003-2700.

mass spectrometer. *Analyst,* ISSN 0003-2654.

*Spectrom.,* Vol. 35, No. 2, pp. 163-167, ISSN 1076-5174.

*Instrum.,* Vol. 78, pp. 074101,.

1142, ISSN 1076-5174.

6445.

0305.

ISSN 0168-1176.

1448, ISSN 1044-0305.

high precursor ion selectivity employing spiral ion trajectory and improved offset parabolic reflectron. *J. Am. Soc. Mass Spectrom.,* Vol. 22, No. 5, pp. 797-803, ISSN

resolution time-of-flight mass spectrometer MULTUM-S II with an infinite flight

simulants in air using an automated, field-deployable, miniature mass spectrometer. *Rapid Commun. Mass Spectrom.,* Vol. 25, No. 10, pp. 1437-1444, ISSN

analysis of agrochemical residues on fruit using ambient ionization on a handheld

induced dissociation fragmentation pathways of peptides, probed using a multuitum tandem time-of-flight mass spectrometer "MULUM-TOF/FOF". *Rev. Sci.* 

Construction of a new multi-turn time-of-flight mass spectrometer. *J. Mass* 

spectrometers with electrostatic sectors. *J. Mass Spectrom.,* Vol. 38, No. 11, pp. 1125-

Microfabrication of cylindrical ion trap mass spectrometer arrays for handheld chemical analyzers. *Chem. Eng. Commun.,* Vol. 195, No. 2, pp. 98-114, ISSN 0098-

D.J. (2008). Implementation of DART and DESI ionization on a fieldable mass spectrometer. *J. Am. Soc. Mass Spectrom.,* Vol. 19, No. 10, pp. 1419-1424, ISSN 1044-

spectrometer consisting of two coaxial electrostatic mirrors. *Int. J. Mass Spectrom.,* 

reflected ion trajectories. *Int. J. Mass Spectrom. Ion Proc.,* Vol. 96, No. 3, pp. 267-274,

portable mass spectrometer. *J. Am. Soc. Mass Spectrom.,* Vol. 19, No. 10, pp. 1442-


Lammert S., Rockwood A., Wang M., Lee M., Lee E., Tolley S., Oliphant J., Jones J. & Waite

Li A., Wang H., Ouyang Z. & Cooks R.G. (2011). Paper spray ionization of polar analytes

Li X.X., Jiang G.Y., Luo C., Xu F.X., Wang Y.Y., Ding L. & Ding C.F. (2009). Ion trap array

Liang G., Sugiarto A., Harper J.D., Cooks R.G. & Zheng O.Y. (2008). Design and

Marshall A.G., Hendrickson C.L. & Jackson G.S. (1998). Fourier transform ion cyclotron

Matsuda H. (2000). Spiral orbit time of flight mass spectrometer. *J. Mass Spectrom. Soc. Jpn,* 

Matsuo T., Ishihara M., Toyoda M., Ito H., Yamaguchi S., Roll R. & Rosenbauer H. (1999). A

Okumura D., Toyoda M., Ishihara M. & Katakuse I. (2004a). Application of a multi-turn

Okumura D., Toyoda M., Ishihara M. & Katakuse I. (2004b). A compact sector-type multi-

Ouyang Z., Noll R.J. & Cooks R.G. (2009). Handheld miniature ion trap mass spectrometers.

Poschenrieder W.P. (1972). Multiple-focusing time-of-flight mass spectrometers part II.

Ravishankara A.R., Daniel J.S. & Portmann R.W. (2009). Nitrous oxide (N2O): The dominant

Sakurai T., Nakabushi H., Hiasa T. & Okanishi K. (1999). A new multi-passage time-of-flight

Satoh T., Sato T. & Tamura J. (2007). Development of a high-performance MALDI-TOF mass

Satoh T., Tsuno H., Iwanaga M. & Kammei Y. (2005). The design and characteristic features

*Mass Spectrom.,* Vol. 16, No. 12, pp. 1969-1975, ISSN 1044-0305.

*Mass Spectrom.,* Vol. 17, No. 7, pp. 916-922, ISSN 1044-0305.

*Chem.,* Vol. 80, No. 19, pp. 7198-7205, ISSN 0003-2700.

*Adv. Space Res.,* Vol. 23, No. 2, pp. 341-348, ISSN 0273-1177.

*Anal. Chem.,* Vol. 81, No. 7, pp. 2421-2425, ISSN 0003-2700.

Vol. 48, No. 5, pp. 303-305, ISSN 1340-8097.

No. 1-2, pp. 331-337, ISSN 0168-9002.

pp. 357-373, ISSN 0020-7381.

pp. 123-125, ISSN 0036-8075.

7, pp. 1318-1323, ISSN 1044-0305.

ISSN 0168-9002.

1359-7345.

4846, ISSN 0003-2700.

35, ISSN 0277-7037.

90, ISSN 1076-5174.

R. (2006). Miniature toroidal radio frequency ion trap mass analyzer. *J. Am. Soc.* 

using non-polar solvents. *Chem. Commun.,* Vol. 47, No. 10, pp. 2811-2813, ISSN

mass analyzer: Structure and performance. *Anal. Chem.,* Vol. 81, No. 12, pp. 4840-

characterization of a multisource hand-held tandem mass spectrometer. *Anal.* 

resonance mass spectrometry: A primer. *Mass Spectrom.Rev.,* Vol. 17, No. 1, pp. 1-

space time-of-flight mass spectrometer for exobiologically-oriented applications.

time-of-flight mass spectrometer, MULTUM II, to organic compounds ionized by matrix-assisted laser desorption/ionization. *J. Mass Spectrom.,* Vol. 39, No. 1, pp. 86-

turn time-of-flight mass spectrometer 'MULTUM II'. *Nucl. Instr. Meth. A,* Vol. 519,

TOFMS with equal energy acceleration. *Int. J. Mass Spectom. Ion Phys.,* Vol. 9, No. 4,

ozone-depleting substance emitted in the 21st century. *Science,* Vol. 326, No. 5949,

mass spectrometer at JAIST. *Nucl. Instr. Meth. A,* Vol. 427, No. 1,2, pp. 182-186,

spectrometer utilizing a spiral ion trajectory. *J. Am. Soc. Mass Spectrom.,* Vol. 18, No.

of a new time-of-flight mass spectrometer with a spiral ion trajectory. *J. Am. Soc.* 


**12**

*Argentina* 

**CO2 and CH4 Flux Measurements from Landfills** 

Municipal solid waste (MSW) landlls are used to dispose of household wastes: food and garden waste, paper, metal, glass, wood, textiles, rubber, leather, plastic, ash, dust and electronic waste (Meju 2000). Decomposition of landlled MSW by long-term physicochemical, chemical and biological processes causes dissolution or decay of landll materials and production of leachate and gases (Bjerg et al. 2005). In particular, bacterial decomposition of the biodegradable fraction of MSW generates mainly methane (CH4) and carbon dioxide (CO2) as well as a wide variety of minor and trace components: hydrogen, water vapor, hydrogen sulfide, ammonia, and volatile organic compounds (Scottish Environment Protection Agency [SEPA], 2004). In spite of efforts being made to control landfill gases (LG) - gas containment, collection and utilisation, flaring and treatment- , these are usually released into the atmosphere directly and indirectly, from the ground or

These LG emissions into the atmosphere represent potential hazards that are of concern locally, regionally and globally (UK Environment Agency, 2010). The trace components of LG pose an odour and toxicity risk (although CO2 is also toxic if it is present in high enough concentrations). Explosion and asphyxia risk is related to sub-surface migration and accumulations in enclosed spaces. Root zone displacement of oxygen by landfill gas is the most likely cause of local ecotoxicity. In addition, volatile organic compounds (VOC) have detrimental effects on human health and they participate in photochemical pollution as precursors of tropospheric ozone. CO2 and CH4 are greenhouse gases and contribute to global warming. Landfills have been implicated as being the largest anthropogenic sources of atmospheric CH4 in the world, comprising about 11% of the total anthropogenic global

Direct measurements of CH4 and CO2 emissions from ground to atmosphere are used as an effective tool to estimate the degassing rate of individual sources and to calibrate global Earth degassing estimates (Cardellini et al. 2003). Early studies of diffuse degassing, focus on the ow of CO2 out of the soil, commonly called CO2 ''ux'' or ''efux'' and expressed as [mass] [area] -1 [time] -1, they were developed in agricultural or ecological areas to measure soil respiration or the flux from soil of other gaseous species (Hanson et al. 1993; Kinzig and

**1. Introduction** 

via sub-surface gas migration, respectively.

CH4 contribution (Spokas et al., 2003).

**– A Case Study: Gualeguaychú Municipal** 

**Landfill, Entre Ríos Province, Argentina** 

*Instituto de Geocronología y Geología Isotópica (UBA-CONICET)* 

Romina Sanci and Héctor O. Panarello

Yavor M.I., Verentchikov A.N., Hasin J., Kozlov B., Gavrik M. & Trufanov A. (2008). Planar multi-reflecting time-of-flight mass analyzer with a jig-saw ion path. *Physics Procedia,* Vol. 1, No. 1, pp. 391-400, ISSN 1875-3892.

## **CO2 and CH4 Flux Measurements from Landfills – A Case Study: Gualeguaychú Municipal Landfill, Entre Ríos Province, Argentina**

Romina Sanci and Héctor O. Panarello *Instituto de Geocronología y Geología Isotópica (UBA-CONICET) Argentina* 

## **1. Introduction**

254 Greenhouse Gases – Emission, Measurement and Management

Yavor M.I., Verentchikov A.N., Hasin J., Kozlov B., Gavrik M. & Trufanov A. (2008). Planar

*Procedia,* Vol. 1, No. 1, pp. 391-400, ISSN 1875-3892.

multi-reflecting time-of-flight mass analyzer with a jig-saw ion path. *Physics* 

Municipal solid waste (MSW) landlls are used to dispose of household wastes: food and garden waste, paper, metal, glass, wood, textiles, rubber, leather, plastic, ash, dust and electronic waste (Meju 2000). Decomposition of landlled MSW by long-term physicochemical, chemical and biological processes causes dissolution or decay of landll materials and production of leachate and gases (Bjerg et al. 2005). In particular, bacterial decomposition of the biodegradable fraction of MSW generates mainly methane (CH4) and carbon dioxide (CO2) as well as a wide variety of minor and trace components: hydrogen, water vapor, hydrogen sulfide, ammonia, and volatile organic compounds (Scottish Environment Protection Agency [SEPA], 2004). In spite of efforts being made to control landfill gases (LG) - gas containment, collection and utilisation, flaring and treatment- , these are usually released into the atmosphere directly and indirectly, from the ground or via sub-surface gas migration, respectively.

These LG emissions into the atmosphere represent potential hazards that are of concern locally, regionally and globally (UK Environment Agency, 2010). The trace components of LG pose an odour and toxicity risk (although CO2 is also toxic if it is present in high enough concentrations). Explosion and asphyxia risk is related to sub-surface migration and accumulations in enclosed spaces. Root zone displacement of oxygen by landfill gas is the most likely cause of local ecotoxicity. In addition, volatile organic compounds (VOC) have detrimental effects on human health and they participate in photochemical pollution as precursors of tropospheric ozone. CO2 and CH4 are greenhouse gases and contribute to global warming. Landfills have been implicated as being the largest anthropogenic sources of atmospheric CH4 in the world, comprising about 11% of the total anthropogenic global CH4 contribution (Spokas et al., 2003).

Direct measurements of CH4 and CO2 emissions from ground to atmosphere are used as an effective tool to estimate the degassing rate of individual sources and to calibrate global Earth degassing estimates (Cardellini et al. 2003). Early studies of diffuse degassing, focus on the ow of CO2 out of the soil, commonly called CO2 ''ux'' or ''efux'' and expressed as [mass] [area] -1 [time] -1, they were developed in agricultural or ecological areas to measure soil respiration or the flux from soil of other gaseous species (Hanson et al. 1993; Kinzig and

CO2 and CH4 Flux Measurements from Landfills

– A Case Study: Gualeguaychú Municipal Landfill, Entre Ríos Province, Argentina 257

(non-steady-state non-ow-through chamber), closed dynamic chambers (non-steady-state chamber ow-through) and open dynamic chambers (steady-state ow-through chamber).

Fig. 1. Biodegradation stages during solid waste decomposition (Scottish Environment Protection Agency [SEPA], 2004): 1. aerobic, 2. non-methanogenic anaerobic, 3. unsteady

In closed chamber methods (static and dynamic), the CO2 flux is determinated from the rate of concentration increase in a isolated chamber, that has been placed on the soil surface for a known period of time. In static chambers, samples are taken using disposable syringes wich are transported and analized in laboratory. In dynamic chambers, the gas is extracted from the chamber, sent to an external infrared gas analyzer (IRGA), and then injected again into the chamber (Figure 2). In open chamber systems, described by Iritz et al. (1997), Moren and Lindroth (2000) and Rayment and Jarvis (1997), CO2 ux is calculated from the difference

None of these methods have yet been recognized as standard because experimental work has indicated differences in estimating uxes among chamber types (Jensen et al. 1996) and has demonstrated limitations related with the chamber design (Hutchison and Mosier 1981; Welles et al. 2001). Consequently, some researchers have begun to carry out studies to determine the accuracy of measurements in comparison with a true ux through calibration systems to calculate the actual CO2 ux and thus to estimate the real contribution to the global carbon cycle (Butnor and Johnsen 2004; Martin et al. 2004; Nay et al. 1994; Widen and Lindroth 2003). Likewise, Pumpanen et al. (2004) have determined correction factors for different chambers and specic soil types. The accumulation chamber method has also been

tested under controlled laboratory conditions (Chiodini et al. 1998; Evans et al. 2001).

To characterize the biodegradation stages during waste decomposition, soil gas probes at different depths, made of stainless steel, are used to measure soil CO2, CH4, N2 and O2 levels. Soil gases are brought to the vicinity of the tip of the probe by applying a vacuum and collected in TedlarTM bags (Figure 3). Samples are analysed in the laboratory by gas

methanogenic anaerobic, 4. steady methanogenic anaerobic, 5. mature.

between CO2 concentration at the inlet and the outlet of the chamber.

**3.2 Other techniques to measure landfill gases** 

Socolow, 1994; Norman et al. 1992; Parkinson 1981). Later, they were applied in active volcanic-geothermal environments where CO2 is derived from some geologic source at depth (Bergfeld et al. 2001; Chiodini et al. 1999, 2004; Gerlach et al. 2001) and also in landfills (Borjesson et al., 2000; Georgaki et al. 2008; Hedge et al., 2003; Jha et al., 2008; Mosher et al., 1999; Pier & Kelly, 1997).

In this chapter, we summarize all the steps involved in the process of the quantifying of CO2 and CH4 fluxes (background, measurement methods calibration, geostatistical treatment of results, presentation of data) from landfills, and its application in Gualeguaychú Municipal Landfill, Entre Ríos province, Argentina. The method of determining the biodegradation processes of solid wastes by extracting gases with a probe and analyzing carbon isotopes on those gases is also included in the text. In addition, the dissolution of these gases in shallow aquifers is evaluated since in the case study that we present the groundwater acts as a sink for the CO2 that is developing in the landfill.

## **2. Gases production during waste decomposition**

The composition of LG will vary from one site to another, from one cell of a landfill to another, and will change over time. Because of this, it is possible to find varying amounts CO2, CH4 and trace components, plus nitrogen and oxygen derived from air that has been drawn into the landfill. LG production is a function of the composition (organic content), density of and moisture content of wastes, climate variables, particle size and thickness of landll cover, air-lled porosity, pH, temperature, nutrient availability, methods of land lling (i.e. open dumping or sanitary landll) and structural features of the site (Barlaz et al. 2004; Kumar et al. 2004).

In addition, LG composition depends on the predominant form of microbial activity (e.g. aerobic/anaerobic) within the landll environment. Assuming that an anaerobic environment is achieved and maintained after waste placement, a pattern of ve sequential stages for LG production or biodegradation stages (Figure 1) is proposed: aerobic, non-methanogenic anaerobic, unsteady methanogenic anaerobic, steady methanogenic anaerobic and mature phases (Farquar and Rovers 1973; (Scottish Environment Protection Agency [SEPA], 2004). During the initial stage of organic degradation within a landll, CO2 is produced in molar equivalents to free O2 consumed. Once O2 concentration is low enough, anaerobic oxidation, hydrolysis and acidication reactions begin and CO2 concentration (up to 70%), hydrogen and organic acids such as acetic reach their peak. As anaerobic degradation continues, the concentrations of acetic and other organic acids decreases, associated with an increase in CH4 generation (methanogenesis). CO2 concentration declines and methanogenesis begins to prevail, establishing a phase of steady CH4 production: 50–70% CH4 (with 30 to 50% CO2). During the last stage (mature), there is not enough organic substrate required for microbial activity and the composition of interstitial gases becomes more similar to atmospheric air.

## **3. Landfill gases measurements**

## **3.1 Accumulation chamber methods**

To assess the impacts of anomalous emissions to the atmosphere, different accumulation chamber methods to measure CO2 and CH4 uxes from individual sources have been used over recent years. Norman et al. (1997) described these methods as closed static chambers

Socolow, 1994; Norman et al. 1992; Parkinson 1981). Later, they were applied in active volcanic-geothermal environments where CO2 is derived from some geologic source at depth (Bergfeld et al. 2001; Chiodini et al. 1999, 2004; Gerlach et al. 2001) and also in landfills (Borjesson et al., 2000; Georgaki et al. 2008; Hedge et al., 2003; Jha et al., 2008; Mosher et al.,

In this chapter, we summarize all the steps involved in the process of the quantifying of CO2 and CH4 fluxes (background, measurement methods calibration, geostatistical treatment of results, presentation of data) from landfills, and its application in Gualeguaychú Municipal Landfill, Entre Ríos province, Argentina. The method of determining the biodegradation processes of solid wastes by extracting gases with a probe and analyzing carbon isotopes on those gases is also included in the text. In addition, the dissolution of these gases in shallow aquifers is evaluated since in the case study that we present the groundwater acts as a sink

The composition of LG will vary from one site to another, from one cell of a landfill to another, and will change over time. Because of this, it is possible to find varying amounts CO2, CH4 and trace components, plus nitrogen and oxygen derived from air that has been drawn into the landfill. LG production is a function of the composition (organic content), density of and moisture content of wastes, climate variables, particle size and thickness of landll cover, air-lled porosity, pH, temperature, nutrient availability, methods of land lling (i.e. open dumping or sanitary landll) and structural features of the site (Barlaz et al.

In addition, LG composition depends on the predominant form of microbial activity (e.g. aerobic/anaerobic) within the landll environment. Assuming that an anaerobic environment is achieved and maintained after waste placement, a pattern of ve sequential stages for LG production or biodegradation stages (Figure 1) is proposed: aerobic, non-methanogenic anaerobic, unsteady methanogenic anaerobic, steady methanogenic anaerobic and mature phases (Farquar and Rovers 1973; (Scottish Environment Protection Agency [SEPA], 2004). During the initial stage of organic degradation within a landll, CO2 is produced in molar equivalents to free O2 consumed. Once O2 concentration is low enough, anaerobic oxidation, hydrolysis and acidication reactions begin and CO2 concentration (up to 70%), hydrogen and organic acids such as acetic reach their peak. As anaerobic degradation continues, the concentrations of acetic and other organic acids decreases, associated with an increase in CH4 generation (methanogenesis). CO2 concentration declines and methanogenesis begins to prevail, establishing a phase of steady CH4 production: 50–70% CH4 (with 30 to 50% CO2). During the last stage (mature), there is not enough organic substrate required for microbial activity and the composition of interstitial gases becomes more similar to atmospheric air.

To assess the impacts of anomalous emissions to the atmosphere, different accumulation chamber methods to measure CO2 and CH4 uxes from individual sources have been used over recent years. Norman et al. (1997) described these methods as closed static chambers

1999; Pier & Kelly, 1997).

2004; Kumar et al. 2004).

**3. Landfill gases measurements 3.1 Accumulation chamber methods** 

for the CO2 that is developing in the landfill.

**2. Gases production during waste decomposition** 

(non-steady-state non-ow-through chamber), closed dynamic chambers (non-steady-state chamber ow-through) and open dynamic chambers (steady-state ow-through chamber).

Fig. 1. Biodegradation stages during solid waste decomposition (Scottish Environment Protection Agency [SEPA], 2004): 1. aerobic, 2. non-methanogenic anaerobic, 3. unsteady methanogenic anaerobic, 4. steady methanogenic anaerobic, 5. mature.

In closed chamber methods (static and dynamic), the CO2 flux is determinated from the rate of concentration increase in a isolated chamber, that has been placed on the soil surface for a known period of time. In static chambers, samples are taken using disposable syringes wich are transported and analized in laboratory. In dynamic chambers, the gas is extracted from the chamber, sent to an external infrared gas analyzer (IRGA), and then injected again into the chamber (Figure 2). In open chamber systems, described by Iritz et al. (1997), Moren and Lindroth (2000) and Rayment and Jarvis (1997), CO2 ux is calculated from the difference between CO2 concentration at the inlet and the outlet of the chamber.

None of these methods have yet been recognized as standard because experimental work has indicated differences in estimating uxes among chamber types (Jensen et al. 1996) and has demonstrated limitations related with the chamber design (Hutchison and Mosier 1981; Welles et al. 2001). Consequently, some researchers have begun to carry out studies to determine the accuracy of measurements in comparison with a true ux through calibration systems to calculate the actual CO2 ux and thus to estimate the real contribution to the global carbon cycle (Butnor and Johnsen 2004; Martin et al. 2004; Nay et al. 1994; Widen and Lindroth 2003). Likewise, Pumpanen et al. (2004) have determined correction factors for different chambers and specic soil types. The accumulation chamber method has also been tested under controlled laboratory conditions (Chiodini et al. 1998; Evans et al. 2001).

#### **3.2 Other techniques to measure landfill gases**

To characterize the biodegradation stages during waste decomposition, soil gas probes at different depths, made of stainless steel, are used to measure soil CO2, CH4, N2 and O2 levels. Soil gases are brought to the vicinity of the tip of the probe by applying a vacuum and collected in TedlarTM bags (Figure 3). Samples are analysed in the laboratory by gas

CO2 and CH4 Flux Measurements from Landfills

Chiodini and Frondini 2001; Gerlach et al. 2001).

cases (Börjesson et al. 2000; Spokas et al. 2003).

**5. Landfill gases effects on groundwater** 

carbonic acid (H2CO3), bicarbonate (HCO3

CO2 (aq) + H2O ↔ H2CO 3 ↔ HCO3-

H2CO3 (aq) + CaCO3 (s) ↔ Ca2+ (aq) + 2HCO3

by microbes and the rest is released by outgassing (Baedecker and Back 1979).

CO2 interacts with water as follows:

– A Case Study: Gualeguaychú Municipal Landfill, Entre Ríos Province, Argentina 259

stratied sampling) are widely used in environmental studies, for a given sampling density, the regular grid provides better estimations than the other two. Histograms are plotted for the uxes measured in the eld and are log-transformed (ln) and tted on a cumulative probability curve to verify the lognormal distribution and to identify different ux populations through changes in graph slope (Bergfeld et al. 2001; Cardellini et al. 2003;

Since geospatial techniques are advisable to estimate the whole landll surface ux from the spatially distributed chamber sites (Spokas et al. 2003), variograms are used to determine the level of spatial dependence of different sites. In experimental variograms (omnidirectional and directional) the semivariance (h) is plotted against the lag(h)—i.e. distance between sample sites (Webster & Oliver 1992). Since variograms may take several forms, different theoretical models are tted to the data using the VARIOWIN software (Eddy & Paninatier 1996). Kriging is applied to generate contour maps, using the parameters of the variogram model that have been derived from the experimental variogram. Kriging is an interpolation method that takes advantage of the spatial dependence of a given variable. A number of papers compare spatial interpolation methods under different conditions, and kriging has proven to give the best estimations in numerous

LG contain a range of components that can dissolve in aqueous media, e.g. CO2, CH4 and some traces. In landlls, CO2 is the most water-soluble constituent of LG. The dissolution of this gas is partially responsible for observed variations in LG emissions to the atmosphere. CO2 can dissolve in groundwater as described by Henry's law and react with water to form a balance of several ionic and nonionic species, collectively known as dissolved inorganic carbon –DIC- (Stumm & Morgan 1996). These species are free carbon dioxide (CO2(aq)),

species, which ultimately affects CO2 solubility, depends on the pH, amongst other things.

Moreover, adding CO2 to groundwater changes the pH in the absence of interaction with aquifer solids. Kerfoot et al. (2004) calculated that landll CO2 may cause pH to drop to 4.7 in the absence of buffering reactions. However, carbonic acid can react with carbonate minerals (such as calcite) in the aquifer to buffer pH changes, according to the following reaction:

Although a rise in alkalinity suggests that groundwater is affected by the CO2 generated in the landll, it should be noted, that it might also be caused by leachates. Large amounts of CO2 are produced in the landll beneath the water table by organic matter decay into groundwater. Some of the CO2 is retained as bicarbonate, part of it may be converted to CH4

CH4 in water has the potential to act as a reducing agent, chemically reducing species and thereby potentially dissolving metal ions from aquifer solids (Kerfoot et al., 2004). An


+ H + ↔ CO32 - + 2H+ (1)


Fig. 2. Sketch of the instrument used to measure fluxes. A. Taken and modified from Kinzig and Socolow (1994). B. Taken and modified from Chiodini *et al.* (1998)

Fig. 3. Diagram of soil probe. A: filter, B: vacuum pump; C: Tedlar bags

chromatography. In addition, the determination of stable isotopes of carbon in CH4 and CO2 is an effective way to identify the different phases of biodegradation in a landll (Coleman et al. 1993; Hackley et al. 1996). According to these authors, CO2 is isotopically light during the initial aerobic and anaerobic oxidation phases of biodegradation with δ13C values that range from -35 to -10‰, which covers the range of most terrestrial plants. The initial input of isotopically light CO2 associated with the earlier biodegradation phases is soon overcome during the methanogenesis phase by the constant input of isotopically heavy CO2 associated with acetate fermentation and microbial CO2 reduction (the two primary metabolic pathways by which microbial CH4 is produced). During methanogenesis, CH4 is enriched in the lighter carbon isotope (12C) and the CO2 associated with microbial CH4 production is enriched in the heavier isotope (13C). Thus, in a semiclosed environment such as a landfill, the δ13C of CO2 is strongly affected by methanogenesis reactions with reported values between -10 and +20‰.

#### **4. Data analysis**

Flux measurements are distributed over regular grids, following Wang and Qi's (1998) statement that although three sampling patterns (regular grid, simple random and cellular

Fig. 2. Sketch of the instrument used to measure fluxes. A. Taken and modified from Kinzig

and Socolow (1994). B. Taken and modified from Chiodini *et al.* (1998)

Fig. 3. Diagram of soil probe. A: filter, B: vacuum pump; C: Tedlar bags

**4. Data analysis** 

chromatography. In addition, the determination of stable isotopes of carbon in CH4 and CO2 is an effective way to identify the different phases of biodegradation in a landll (Coleman et al. 1993; Hackley et al. 1996). According to these authors, CO2 is isotopically light during the initial aerobic and anaerobic oxidation phases of biodegradation with δ13C values that range from -35 to -10‰, which covers the range of most terrestrial plants. The initial input of isotopically light CO2 associated with the earlier biodegradation phases is soon overcome during the methanogenesis phase by the constant input of isotopically heavy CO2 associated with acetate fermentation and microbial CO2 reduction (the two primary metabolic pathways by which microbial CH4 is produced). During methanogenesis, CH4 is enriched in the lighter carbon isotope (12C) and the CO2 associated with microbial CH4 production is enriched in the heavier isotope (13C). Thus, in a semiclosed environment such as a landfill, the δ13C of CO2 is strongly affected by methanogenesis reactions with reported values between -10 and +20‰.

Flux measurements are distributed over regular grids, following Wang and Qi's (1998) statement that although three sampling patterns (regular grid, simple random and cellular stratied sampling) are widely used in environmental studies, for a given sampling density, the regular grid provides better estimations than the other two. Histograms are plotted for the uxes measured in the eld and are log-transformed (ln) and tted on a cumulative probability curve to verify the lognormal distribution and to identify different ux populations through changes in graph slope (Bergfeld et al. 2001; Cardellini et al. 2003; Chiodini and Frondini 2001; Gerlach et al. 2001).

Since geospatial techniques are advisable to estimate the whole landll surface ux from the spatially distributed chamber sites (Spokas et al. 2003), variograms are used to determine the level of spatial dependence of different sites. In experimental variograms (omnidirectional and directional) the semivariance (h) is plotted against the lag(h)—i.e. distance between sample sites (Webster & Oliver 1992). Since variograms may take several forms, different theoretical models are tted to the data using the VARIOWIN software (Eddy & Paninatier 1996). Kriging is applied to generate contour maps, using the parameters of the variogram model that have been derived from the experimental variogram. Kriging is an interpolation method that takes advantage of the spatial dependence of a given variable. A number of papers compare spatial interpolation methods under different conditions, and kriging has proven to give the best estimations in numerous cases (Börjesson et al. 2000; Spokas et al. 2003).

## **5. Landfill gases effects on groundwater**

LG contain a range of components that can dissolve in aqueous media, e.g. CO2, CH4 and some traces. In landlls, CO2 is the most water-soluble constituent of LG. The dissolution of this gas is partially responsible for observed variations in LG emissions to the atmosphere. CO2 can dissolve in groundwater as described by Henry's law and react with water to form a balance of several ionic and nonionic species, collectively known as dissolved inorganic carbon –DIC- (Stumm & Morgan 1996). These species are free carbon dioxide (CO2(aq)), carbonic acid (H2CO3), bicarbonate (HCO3-) and carbonate (CO32-). The balance of these species, which ultimately affects CO2 solubility, depends on the pH, amongst other things. CO2 interacts with water as follows:

$$\rm CO\_2(aq) + H\_2O \leftrightarrow H\_2CO\_3 \leftrightarrow HCO\_3^- + H^+ \leftrightarrow CO\_3^{2-} + 2H^+ \tag{1}$$

Moreover, adding CO2 to groundwater changes the pH in the absence of interaction with aquifer solids. Kerfoot et al. (2004) calculated that landll CO2 may cause pH to drop to 4.7 in the absence of buffering reactions. However, carbonic acid can react with carbonate minerals (such as calcite) in the aquifer to buffer pH changes, according to the following reaction:

$$\rm H\_2CO\_3 \text{ (aq)} + \rm CaCO\_3 \text{ (s)} \leftrightarrow Ca^{2+} \text{ (aq)} + 2HCO\_3 \text{(aq)}\tag{2}$$

Although a rise in alkalinity suggests that groundwater is affected by the CO2 generated in the landll, it should be noted, that it might also be caused by leachates. Large amounts of CO2 are produced in the landll beneath the water table by organic matter decay into groundwater. Some of the CO2 is retained as bicarbonate, part of it may be converted to CH4 by microbes and the rest is released by outgassing (Baedecker and Back 1979).

CH4 in water has the potential to act as a reducing agent, chemically reducing species and thereby potentially dissolving metal ions from aquifer solids (Kerfoot et al., 2004). An

CO2 and CH4 Flux Measurements from Landfills

Fig. 4. Localization and description of MWSFDS

**7.2 Measurement methods calibration** 

– A Case Study: Gualeguaychú Municipal Landfill, Entre Ríos Province, Argentina 261

geophysical studies (Pomposiello et al. 2009; Prezzi et al. 2005) in the filled-up disposal area, are of household origin and the filling depth does not exceed 2 m. Another part of the

The Gualeguaychu municipal solid waste final disposal site (MSWDS) is located in the lower section of the Gualeguaychu River basin, which has a subhumid–humid climate (Sanci et al., 2009a). The Punta Gorda Group underlying the MSFDSW is the unit outcropping in the study area (Iriondo, 1980). Most of the sedimentary cover in the province of Entre Rios belongs to this group, which is composed mainly of brown, yellow and greenish silts (loess), clays and calcareous levels called ''tosca'', composed of calcite. It was formed in lacustrine and aeolian environments and was assigned to Middle–Upper Pleistocene age (Iriondo, 1996). In addition, this group is intercalated by uvial facies and silty levels, with abundant marine fossils from the marine ingression in the upper Pleistocene (Pereyra et al. 2002). The Punta Gorda Group contains a low-productivity phreatic aquifer, whose water is used for livestock farming and agriculture. This unit acts like a semi-conning layer for the underlying aquifer (Salto Chico) that is highly productive

The accumulation chamber (0.26-m high and 0.30-m diameter) was placed on a collar that had been previously installed on the ground. The chamber also had outlet and inlet manifolds connected to the external pump in order to mix and distribute the air inside. A flow-meter was intercalated in order to regulate the mixing rate. There was also a port to test the temperature within the chamber. A pressure-relief vent was connected at the top of the chamber to

facility is currently being landfilled. There are no gas vents or recovery systems.

and has high quality of water, used for human consumption and irrigation.

example is the reaction with pyrolusite to produce soluble manganese (II) from insoluble insoluble manganse (IV):

$$4\text{MnO}\_2\text{(s)} + \text{CH}\_4 + 8\text{H}^\* \leftrightarrow 4\text{Mn}\_2\text{\*} + \text{CO}\_2 \text{ +6H}\_2\text{O} \tag{3}$$

In addition, carbon isotope analysis of DIC in groundwater is a useful tool to determine the sources of CO2 (soil gas, dissolution of calcareous material, CO2 produced by solid waste organic decomposition). Values of δ13C-DIC between -15‰ and -12‰ can be explained by isotope fractionation from the fixation of CO2 from soil respiration in the form of HCO3 within a process of calcite dissolution (Mook, 2000). In agreement with the explanation in section 3.2, δ13C-DIC values richer than -12‰ can be explained by the input of enriched CO2 from solid waste degradation which dissolves calcite (Kerfoot et al., 2003).

## **6. Isotopic analysis**

As has been previously stated, the determination of stable isotopes of carbon in gases and groundwater is an effective way to identify solid waste biodegradation processes in a landll. Samples of soil gas and groundwater are analyzed to determine the 13C/12C ratio on BaCO3 obtained from precipitating CO2 and HCO3 - with alkaline BaCl2. In our laboratory experiments, the resulting BaCO3 is collected on an acid-washed glass bre lter (GF/F) under a nitrogen atmosphere, rinsed with distilled water and dried to 60ºC. Then, samples are reacted with H3PO4 (100%) in vacuo, according to MacCrea (1950). The resulting CO2 is cryogenically puried; transferred with liquid N2 to a glass vial and measured against a working standard (CO2 from Carrara marble) in a dual inlet, triple collector mass spectrometer, Finnigan MAT Delta S. Carbon isotope composition is expressed as 13C, according to:

$$\delta^{\rm 33}\text{C} = 1000 \left( \frac{\left[ ^{13}\text{C} / \, ^{12}\text{C} \right]\_{\text{S}} - \left[ ^{13}\text{C} / \, ^{12}\text{C} \right]\_{\text{R}}}{\left[ ^{13}\text{C} / \, ^{12}\text{C} \right]\_{\text{R}}} \right) \, ^{\rm \%} \text{o} \tag{4}$$

where 13C/12C is the carbon isotope ratio, sufx S corresponds to the sample and sufx R to the reference standard, Pee Dee belemnitella (PDB), redened in function of the NBS 19, TS-Limestone standard as V-PDB (Gonantini et al. 1995). Analytical uncertainty (2) was ±0.2%.

#### **7. Case Study: Gualeguaychú Municipal Landfill, Entre Ríos Province, Argentina**

#### **7.1 Site description**

The Gualeguaychú municipal landfill is located 3 km south of that city, in the southeast of the province of Entre Rios, Argentina (Figure 4). The municipal facility was first exploited for mineral extraction, and waste was later disposed of in the depressions. One part of this facility was used for final disposal of MSW and closed in 2000 when it had filled up. Gas emissions were sampled there. There is no information regarding any environmental protection actions undertaken during the operation of the site, but visual observation of the site revealed that the topsoil cover is permeable and not compacted. So far, according to

example is the reaction with pyrolusite to produce soluble manganese (II) from insoluble

 4MnO2 (s) + CH4 + 8H+ ↔ 4Mn2+ + CO2 +6H2O (3) In addition, carbon isotope analysis of DIC in groundwater is a useful tool to determine the sources of CO2 (soil gas, dissolution of calcareous material, CO2 produced by solid waste organic decomposition). Values of δ13C-DIC between -15‰ and -12‰ can be explained by isotope fractionation from the fixation of CO2 from soil respiration in the form of HCO3 within a process of calcite dissolution (Mook, 2000). In agreement with the explanation in section 3.2, δ13C-DIC values richer than -12‰ can be explained by the input of enriched CO2

As has been previously stated, the determination of stable isotopes of carbon in gases and groundwater is an effective way to identify solid waste biodegradation processes in a landll. Samples of soil gas and groundwater are analyzed to determine the 13C/12C ratio on BaCO3 obtained from precipitating CO2 and HCO3- with alkaline BaCl2. In our laboratory experiments, the resulting BaCO3 is collected on an acid-washed glass bre lter (GF/F) under a nitrogen atmosphere, rinsed with distilled water and dried to 60ºC. Then, samples are reacted with H3PO4 (100%) in vacuo, according to MacCrea (1950). The resulting CO2 is cryogenically puried; transferred with liquid N2 to a glass vial and measured against a working standard (CO2 from Carrara marble) in a dual inlet, triple collector mass spectrometer, Finnigan MAT Delta S. Carbon isotope composition is expressed as 13C,

*δ*13*C* = 1000

**7. Case Study: Gualeguaychú Municipal Landfill, Entre Ríos Province,** 

where 13C/12C is the carbon isotope ratio, sufx S corresponds to the sample and sufx R to the reference standard, Pee Dee belemnitella (PDB), redened in function of the NBS 19, TS-Limestone standard as V-PDB (Gonantini et al. 1995). Analytical uncertainty (2) was

The Gualeguaychú municipal landfill is located 3 km south of that city, in the southeast of the province of Entre Rios, Argentina (Figure 4). The municipal facility was first exploited for mineral extraction, and waste was later disposed of in the depressions. One part of this facility was used for final disposal of MSW and closed in 2000 when it had filled up. Gas emissions were sampled there. There is no information regarding any environmental protection actions undertaken during the operation of the site, but visual observation of the site revealed that the topsoil cover is permeable and not compacted. So far, according to

13 12 13 12 13 12 / / /

*CC CC C C* 

*S R R*

‰ (4)

from solid waste degradation which dissolves calcite (Kerfoot et al., 2003).

insoluble manganse (IV):

**6. Isotopic analysis** 

according to:

±0.2%.

**Argentina** 

**7.1 Site description** 

geophysical studies (Pomposiello et al. 2009; Prezzi et al. 2005) in the filled-up disposal area, are of household origin and the filling depth does not exceed 2 m. Another part of the facility is currently being landfilled. There are no gas vents or recovery systems.

The Gualeguaychu municipal solid waste final disposal site (MSWDS) is located in the lower section of the Gualeguaychu River basin, which has a subhumid–humid climate (Sanci et al., 2009a). The Punta Gorda Group underlying the MSFDSW is the unit outcropping in the study area (Iriondo, 1980). Most of the sedimentary cover in the province of Entre Rios belongs to this group, which is composed mainly of brown, yellow and greenish silts (loess), clays and calcareous levels called ''tosca'', composed of calcite. It was formed in lacustrine and aeolian environments and was assigned to Middle–Upper Pleistocene age (Iriondo, 1996). In addition, this group is intercalated by uvial facies and silty levels, with abundant marine fossils from the marine ingression in the upper Pleistocene (Pereyra et al. 2002). The Punta Gorda Group contains a low-productivity phreatic aquifer, whose water is used for livestock farming and agriculture. This unit acts like a semi-conning layer for the underlying aquifer (Salto Chico) that is highly productive and has high quality of water, used for human consumption and irrigation.

Fig. 4. Localization and description of MWSFDS

#### **7.2 Measurement methods calibration**

The accumulation chamber (0.26-m high and 0.30-m diameter) was placed on a collar that had been previously installed on the ground. The chamber also had outlet and inlet manifolds connected to the external pump in order to mix and distribute the air inside. A flow-meter was intercalated in order to regulate the mixing rate. There was also a port to test the temperature within the chamber. A pressure-relief vent was connected at the top of the chamber to

CO2 and CH4 Flux Measurements from Landfills

total population x total revealed area).

– A Case Study: Gualeguaychú Municipal Landfill, Entre Ríos Province, Argentina 263

CO2 flux data sets of the three campaigns are plotted in histograms (Figure 5). Distribution of measured CO2 fluxes in the longest data set (107 sites) is log-normal, as can be seen in the linear cumulative probability plot of ln(CO2) flux of Figure 6a. Changes in slope indicate the presence of different populations within the data set, as follows: below 62 g m-2 day-1 (ln < 4.2) coinciding with measured background values beyond the MSWFDS; intermediate, between 67 g m-2 day-1 and 191 g m-2 day-1 (4.3 < ln < 5.3); and high values, above 219 g m-2 day-1 (ln>5.3). Mean CO2 flux in the first population (23% of the data) at the 95% confidence level is (46 4) g m-2 day-1 (or 13 g C m-2 day-1). Mean CO2 flux in the intermediate population (63% of the data) is (110 9) g m-2 day-1 (or 30 g C m-2 day-1). Mean flux in the third population (14% of the data) is (270 22) g m-2 day-1 (or 74 g C m-2 day-1). Extreme populations (high and low fluxes) correspond to a smaller number of sampling sites than the intermediate flux population. In addition, an estimation of CO2 released to the atmosphere in tn day-1 (Table 1) from MSWFDS was calculated taking into account the mean of each sub-population and their respective areas (sub-population mean x percentage of

Since variogram reliability increases with the number of sites used in the model, the different regular sampling grids (14, 50 and 107 site surveys) were analysed until the geostatistical analysis indicated that an adequate sampling density had been achieved (Sanci et al. 2009a). Omni-directional and directional variograms were plotted, and spatial dependence was only observed between log-transformed data of CO2 fluxes in the 107-site

Fig. 5. Histograms of the CO2 flux data from the three field campaigns: a. 14 stations

(March); b. 50 stations (July); c. 107 stations (October).

maintain the outside-inside pressure equilibrium. Flux rates were calculated by fitting linear regression to the variation of concentration (C) vs. time and adjusting for chamber volume (0.018 m3) and covered area (0.070 m2), according to the following equation:

$$F = \left(\frac{V}{A}\right) \left(\frac{d\mathbf{C}}{dt}\right) \tag{5}$$

where F is the surface emission rate (g m-2 day-1), V is the chamber volume and A is the soil area under the chamber and dC/dt is the variation of C with t within the chamber.

The dynamic and static closed chamber methods were applied during fieldwork, after calibration in the laboratory (Sanci et al. 2009b). This calibration consisted of a system where known CO2 concentrations flowed through different porous materials, simulating CO2 diffusion through the soil. This system allowed the determination of the differences between reference CO2 flux values and experimental measurements under different sampling conditions. In the closed dynamic chamber method, soil gases are pumped for analysis from the accumulation chamber to a portable IRGA (PP Systems EGM-4) and subsequently returned to the chamber. The best fit (deviation smaller than 10%) was obtained taking short readings every 3 min during 12 min and mixing 25 s prior to CO2 extraction (R2 = 0.99). The best mixing rate was 250 ml min-1. The portable IRGA has an internal pump and a scale of (0– 20,000) mol mol-1. It allows the determining of CO2 concentrations within an analytical uncertainty of ±1% of the reading. In the closed static chamber method, soil gases are extracted with syringes and analysed by gas chromatography (GC-TCD HP 5890 Series II). Although both methods allow measuring CO2 fluxes directly, the static method allows the detection of another greenhouse gas, CH4, which proved to be useful in the exploratory surveys. The best fit (deviation 10%) was obtained taking three samples every 10 min during 20 min (R2 = 0.99) and mixing 35 s prior to CO2 sampling. The best mixing rate was 250 mL min-1.

#### **7.3 Gas samples**

The surveyed area covers about 150,000 m2. Measurements were distributed over regular grids: spacing was 100 m in the exploratory survey (March 2007, 14 sites), in the detailed surveys spacing was 50 m (50 sites) and 25 m (107 sites) respectively. Detailed surveys were carried out in July and October 2007. No measurements were made in the inaccessible part of the MSWFDS (Figure 4). Fieldwork was undertaken in dry and stable periods to avoid the inuence of rainfall, soil humidity and atmospheric pressure on surface emissions. During the exploratory survey, CH4 and CO2 fluxes were measured with the static chamber method, whereas the dynamic chamber method was used during the detailed surveys for CO2 flux measurements. Sampling density was increased taking into consideration the original location of the sites of the exploratory fieldwork.

In the exploratory survey, CO2 fluxes ranged from 25 g m-2 day-1 to 194 g m-2 day-1. CH4 fluxes were not detected. In the detailed fieldwork, CO2 fluxes in 50 and 107 stations ranged from 5 g m-2 day-1 to 214 g m-2 day-1 and from 31 g m-2day-1 to 331 g m-2 day-1, respectively. Irregularly spaced stations upstream of the MSWFDS were added to the study in October to measure the background values of soil respiration, which ranged from 29 g m-2 day-1 to 59 g m-2 day-1. At all surveyed sites, soil temperature ranged from 21ºC to 30ºC (March), 12ºC to 17ºC (July) and 20ºC to 30ºC (October).

maintain the outside-inside pressure equilibrium. Flux rates were calculated by fitting linear regression to the variation of concentration (C) vs. time and adjusting for chamber volume

where F is the surface emission rate (g m-2 day-1), V is the chamber volume and A is the soil

The dynamic and static closed chamber methods were applied during fieldwork, after calibration in the laboratory (Sanci et al. 2009b). This calibration consisted of a system where known CO2 concentrations flowed through different porous materials, simulating CO2 diffusion through the soil. This system allowed the determination of the differences between reference CO2 flux values and experimental measurements under different sampling conditions. In the closed dynamic chamber method, soil gases are pumped for analysis from the accumulation chamber to a portable IRGA (PP Systems EGM-4) and subsequently returned to the chamber. The best fit (deviation smaller than 10%) was obtained taking short readings every 3 min during 12 min and mixing 25 s prior to CO2 extraction (R2 = 0.99). The best mixing rate was 250 ml min-1. The portable IRGA has an internal pump and a scale of (0– 20,000) mol mol-1. It allows the determining of CO2 concentrations within an analytical uncertainty of ±1% of the reading. In the closed static chamber method, soil gases are extracted with syringes and analysed by gas chromatography (GC-TCD HP 5890 Series II). Although both methods allow measuring CO2 fluxes directly, the static method allows the detection of another greenhouse gas, CH4, which proved to be useful in the exploratory surveys. The best fit (deviation 10%) was obtained taking three samples every 10 min during 20 min (R2 = 0.99)

*dC dt* 

(5)

(0.018 m3) and covered area (0.070 m2), according to the following equation:

*<sup>F</sup>* = *<sup>V</sup> A* 

area under the chamber and dC/dt is the variation of C with t within the chamber.

and mixing 35 s prior to CO2 sampling. The best mixing rate was 250 mL min-1.

The surveyed area covers about 150,000 m2. Measurements were distributed over regular grids: spacing was 100 m in the exploratory survey (March 2007, 14 sites), in the detailed surveys spacing was 50 m (50 sites) and 25 m (107 sites) respectively. Detailed surveys were carried out in July and October 2007. No measurements were made in the inaccessible part of the MSWFDS (Figure 4). Fieldwork was undertaken in dry and stable periods to avoid the inuence of rainfall, soil humidity and atmospheric pressure on surface emissions. During the exploratory survey, CH4 and CO2 fluxes were measured with the static chamber method, whereas the dynamic chamber method was used during the detailed surveys for CO2 flux measurements. Sampling density was increased taking into consideration the original

In the exploratory survey, CO2 fluxes ranged from 25 g m-2 day-1 to 194 g m-2 day-1. CH4 fluxes were not detected. In the detailed fieldwork, CO2 fluxes in 50 and 107 stations ranged from 5 g m-2 day-1 to 214 g m-2 day-1 and from 31 g m-2day-1 to 331 g m-2 day-1, respectively. Irregularly spaced stations upstream of the MSWFDS were added to the study in October to measure the background values of soil respiration, which ranged from 29 g m-2 day-1 to 59 g m-2 day-1. At all surveyed sites, soil temperature ranged from 21ºC to 30ºC (March), 12ºC to

**7.3 Gas samples** 

location of the sites of the exploratory fieldwork.

17ºC (July) and 20ºC to 30ºC (October).

CO2 flux data sets of the three campaigns are plotted in histograms (Figure 5). Distribution of measured CO2 fluxes in the longest data set (107 sites) is log-normal, as can be seen in the linear cumulative probability plot of ln(CO2) flux of Figure 6a. Changes in slope indicate the presence of different populations within the data set, as follows: below 62 g m-2 day-1 (ln < 4.2) coinciding with measured background values beyond the MSWFDS; intermediate, between 67 g m-2 day-1 and 191 g m-2 day-1 (4.3 < ln < 5.3); and high values, above 219 g m-2 day-1 (ln>5.3). Mean CO2 flux in the first population (23% of the data) at the 95% confidence level is (46 4) g m-2 day-1 (or 13 g C m-2 day-1). Mean CO2 flux in the intermediate population (63% of the data) is (110 9) g m-2 day-1 (or 30 g C m-2 day-1). Mean flux in the third population (14% of the data) is (270 22) g m-2 day-1 (or 74 g C m-2 day-1). Extreme populations (high and low fluxes) correspond to a smaller number of sampling sites than the intermediate flux population. In addition, an estimation of CO2 released to the atmosphere in tn day-1 (Table 1) from MSWFDS was calculated taking into account the mean of each sub-population and their respective areas (sub-population mean x percentage of total population x total revealed area).

Since variogram reliability increases with the number of sites used in the model, the different regular sampling grids (14, 50 and 107 site surveys) were analysed until the geostatistical analysis indicated that an adequate sampling density had been achieved (Sanci et al. 2009a). Omni-directional and directional variograms were plotted, and spatial dependence was only observed between log-transformed data of CO2 fluxes in the 107-site

Fig. 5. Histograms of the CO2 flux data from the three field campaigns: a. 14 stations (March); b. 50 stations (July); c. 107 stations (October).

CO2 and CH4 Flux Measurements from Landfills

spatially homogeneous: values grow towards the southeast.

from -25.4to -17.6‰, and in the second from -34.2 to -17.8‰.

Fig. 7. 13C–CO2 versus CO2 concentrations (Sanci et al., 2011).

Considering that MSW were buried at the same depth as the water table (0.10 to 1.70 m below surface) groundwater samples for hydrochemical and isotopic analyses were taken in October 2007 to check whether the phreatic aquifer was acting as a sink for the CO2 generated by waste biodegradation. Groundwater samples were ltered and stored in 1000 mL plastic bottles and cooled until analysis. pH, temperature, electrical conductivity and alkalinity were determined in the eld. The latter was determined by titration with H2SO4. Major ion concentrations were measured in laboratory: Na+ ,K+, Ca2+ and Mg2+ by atomic absorption spectrometry (Buck Scientic 200 A); SO42-, as S, was quantied by inductively coupled plasma-atomic emission spectroscopy (BAIRD-ICP 2070) and Cl- by titration with AgNO3. Carbon isotopic analyses on groundwater were done according the procedure explained in section 6. Piezometers for groundwater observation were installed before

**7.3 Groundwater samples** 

– A Case Study: Gualeguaychú Municipal Landfill, Entre Ríos Province, Argentina 265

parameters were used to elaborate the contour map of ln(CO2) flux by kriging (Figure 6b). The distribution of ln(CO2) flux indicates that diffuse gas fluxes from the MSWFDS are not

To characterize the biodegradation stages within MSFDSW, relative concentrations were analysed of CO2, CH4, O2 and N2, and the isotopic composition of CO2 (13C/12C) was determined. A stainless steel probe was inserted to 20 cm and soil gases brought to the vicinity of the tip of the probe by applying a vacuum. Soil gas samples were taken at 28 locations spaced at 50 m, which coincide with sites where CO2 uxes were measured. Samples were collected in TedlarTM bags and analysed in laboratory with a GC-TCD HP 5890 Series II. Volume fraction of CO2 in soil gas samples ranged from ca. 0.01 to 0.103. In all cases, CH4 fractions were 0.01, while those of O2 and N2 ranged from 0.165 to 0.209 and from 0.701 to 0.780, respectively, close to atmospheric concentrations. Isotope ratio 13C/12C for CO2 was determined by bubbling the sampled CO2 into an alkaline BaCl2 solution which precipitated as BaCO3. The 13C–CO2 ranged from -34.2 to -17.6‰. The plot of CO2 concentration versus 13C–CO2 of Figure 7 (Sanci et al., 2011) shows two areas where different processes take place. One is related to normal soil respiration, about 0.01 of CO2 (Welles et al. 2001). The second has values above 0.01, which may indicate the presence of anomalous concentrations of CO2 from the biodegradation of urban solid waste (Pier & Kelly 1997) with a possible contribution of soil respiration. In the rst group, 13C ranges


Table 1. Statistical parameters of the sub-populations determined for the 107 stations sampled in October 2007.

Fig. 6. A. Linear cumulative probability plot of ln(CO2) flux of 107 stations; B. Omnidirectional variogram; C. Contours map of ln(CO2) flux (Sanci et al., 2011).

survey (Sanci et al., 2011). The semivariance (h) plotted against a 30 m lag (h) between sample points is shown in Figure 6a. This variogram is linear with nugget effect, and its parameters were used to elaborate the contour map of ln(CO2) flux by kriging (Figure 6b). The distribution of ln(CO2) flux indicates that diffuse gas fluxes from the MSWFDS are not spatially homogeneous: values grow towards the southeast.

To characterize the biodegradation stages within MSFDSW, relative concentrations were analysed of CO2, CH4, O2 and N2, and the isotopic composition of CO2 (13C/12C) was determined. A stainless steel probe was inserted to 20 cm and soil gases brought to the vicinity of the tip of the probe by applying a vacuum. Soil gas samples were taken at 28 locations spaced at 50 m, which coincide with sites where CO2 uxes were measured. Samples were collected in TedlarTM bags and analysed in laboratory with a GC-TCD HP 5890 Series II. Volume fraction of CO2 in soil gas samples ranged from ca. 0.01 to 0.103. In all cases, CH4 fractions were 0.01, while those of O2 and N2 ranged from 0.165 to 0.209 and from 0.701 to 0.780, respectively, close to atmospheric concentrations. Isotope ratio 13C/12C for CO2 was determined by bubbling the sampled CO2 into an alkaline BaCl2 solution which precipitated as BaCO3. The 13C–CO2 ranged from -34.2 to -17.6‰. The plot of CO2 concentration versus 13C–CO2 of Figure 7 (Sanci et al., 2011) shows two areas where different processes take place. One is related to normal soil respiration, about 0.01 of CO2 (Welles et al. 2001). The second has values above 0.01, which may indicate the presence of anomalous concentrations of CO2 from the biodegradation of urban solid waste (Pier & Kelly 1997) with a possible contribution of soil respiration. In the rst group, 13C ranges from -25.4to -17.6‰, and in the second from -34.2 to -17.8‰.

Fig. 7. 13C–CO2 versus CO2 concentrations (Sanci et al., 2011).

#### **7.3 Groundwater samples**

264 Greenhouse Gases – Emission, Measurement and Management

**Number points** 

Background 23 25 46 (42-50) 1,59 Intermediate 63 67 110 (101-119) 10,39 High 14 15 270 (248-292) 5,67 Table 1. Statistical parameters of the sub-populations determined for the 107 stations

Fig. 6. A. Linear cumulative probability plot of ln(CO2) flux of 107 stations; B. Omnidirectional variogram; C. Contours map of ln(CO2) flux (Sanci et al., 2011).

survey (Sanci et al., 2011). The semivariance (h) plotted against a 30 m lag (h) between sample points is shown in Figure 6a. This variogram is linear with nugget effect, and its

**Mean CO2 Flux (g/m2 día) with 95% confidence level** 

**CO2 released to atmosphere (tn day-1 )** 

**Subpopulation Percentage of total** 

sampled in October 2007.

**population (%)** 

Considering that MSW were buried at the same depth as the water table (0.10 to 1.70 m below surface) groundwater samples for hydrochemical and isotopic analyses were taken in October 2007 to check whether the phreatic aquifer was acting as a sink for the CO2 generated by waste biodegradation. Groundwater samples were ltered and stored in 1000 mL plastic bottles and cooled until analysis. pH, temperature, electrical conductivity and alkalinity were determined in the eld. The latter was determined by titration with H2SO4. Major ion concentrations were measured in laboratory: Na+ ,K+, Ca2+ and Mg2+ by atomic absorption spectrometry (Buck Scientic 200 A); SO42-, as S, was quantied by inductively coupled plasma-atomic emission spectroscopy (BAIRD-ICP 2070) and Cl- by titration with AgNO3. Carbon isotopic analyses on groundwater were done according the procedure explained in section 6. Piezometers for groundwater observation were installed before

CO2 and CH4 Flux Measurements from Landfills

concentrations.

**7.4 Interpretation of results** 

– A Case Study: Gualeguaychú Municipal Landfill, Entre Ríos Province, Argentina 267

as background (pristine water), and the evolution of groundwater ow along the MSFDSW, calcium concentrations downstream are similar to background values. On the other hand, concentrations of magnesium, sulphate, chloride and sodium are greater than background

The measurement of CO2 uxes in the MSWFDS with previous calibration in laboratory made it possible to reliably measure the spatial variability of the emissions in the eld. CO2 uxes obtained (31–331 g m-2 day-1) revealed the skewed distribution of the data (Figure 5c). This type of distribution (log-normal) was also observed in other soil CO2 ux studies of natural and anthropic emission sources (Bergfeld et al. 2001; Cardellini et al. 2003; Chiodini and Frondini 2001; Gerlach et al. 2001). Logarithmic probability plots (Fig. 6b) show the polymodal distribution of CO2 uxes. They are a combination of three log-normal populations, which indicate that different processes of CO2 generation take place. Low CO2 uxes are similar to the background values found around the MSFDSW and derive from soil respiration (Welles et al. 2001). The remaining values can be grouped into moderate and high uxes. Values for both populations agree with those reported for biodegradation of solid waste in other sanitary landlls (Cardellini et al. 2003; Georgaki et al. 2008; Hedge et al. 2003; Jha et al. 2008; Pier & Kelly 1997). Spatial variations observed in surface CO2 ux distribution (Figure 6c) are due to MSFDSW inhomogeneities. This variability may indicate that waste was buried in a NW–SE direction. Younger parts of the MSFDSW where

biodegradation started later are more likely to have the highest CO2 ux values.

MSWFDS is in an aerobic phase of biodegradation.

no generation of CH4, arriving at the current maturation stage.

Quantied anomalous CO2 uxes within the MSFDSW show that the site is still undergoing MSW biodegradation. Measured CO2, CH4, N2 and O2 concentrations are similar to those described for an initial phase of aerobic oxidation or a posthumous stage of biodegradation (mature), and they are different from those described for anaerobic phases. Considering the time since the end of operations at the MSFDSW (about 10 years), the values obtained for these gases may indicate that the MSFDSW is in a mature stage. Moreover, the results obtained for C isotopes in the probe-sampled CO2 (-34.2 to -17.6‰) suggest that the

Although concentrations of CO2, CH4, N2 and O2 and C isotope are indicators of the degree of maturity of sanitary landlls, in practice, the factors affecting gas generation need to be considered. These factors affect the duration of each particular biodegradation stage, as well as the degrees of phase overlap and phase omission. In this case, the shallow burial of waste (2 m), the absence of CH4 and mainly the permeability of the top cover due to little compaction and inadequate materials, make it possible to assume that anaerobic conditions necessary for methanogenic reactions have not been achieved in the MSFDSW. Therefore, since MSW were rst disposed, biodegradation was completely aerobic or the initial phase of aerobic oxidation might have been followed by another phase of anaerobic oxidation with

Data show that groundwater alkalinity grows across the MSFDSW in the direction of underground ow (Fig. 8). This suggests an input of C generated by anomalous CO2, which dissolves calcite from calcareous levels such as ''tosca'' in the MSFDSW. The interaction between carbonic acid and mineral carbonates might even buffer pH variations through

October 2007 to compare the electric conductivity measured in situ with that estimated by geoelectrical studies (Pomposiello et al. 2009). Wells were drilled to a depth of ≤2.6 m, along the main direction of the local groundwater ow (S–N): upstream (P3–P10), downstream (P9) and within the landll (P7–P5).

Measured physical and chemical parameters are shown in Table 2 and the chemical classication of water in Figure 8a (sodium chloride and bicarbonate waters). Hydrochemical and isotope values varied in the different ow paths: P3 to P9; P10 to P9; P7 to P5 (Sanci et al., 2011). Alkalinity and 13C-DIC tended to increase along the ow paths previously mentioned (Figure 8b). Values changed from 2.98 mmol L-1/-12.1‰to 8.29 mmol L-1/4.4‰ (P3–P9), 2.44 mmol L-1/-15.0‰ to 8.29 mmol L-1/4.4‰ (P10–P9) and 8.18 mmol L-1/-8.1‰ to 41.45 mmol L-1/0.8‰ (P7–P5). Considering the results obtained for P3 and P10


Table 2. Groundwater composition of piezometers

Fig. 8. A. Piper diagram showing the chemical classification of wells. P3, P7, P5: sodium bicarbonate groundwater; P10, P9: sodium chloride groundwaterchemical classication of groundwater; B. Alkalinity versus δ13C-DIC (Sanci et al., 2011).

as background (pristine water), and the evolution of groundwater ow along the MSFDSW, calcium concentrations downstream are similar to background values. On the other hand, concentrations of magnesium, sulphate, chloride and sodium are greater than background concentrations.

## **7.4 Interpretation of results**

266 Greenhouse Gases – Emission, Measurement and Management

October 2007 to compare the electric conductivity measured in situ with that estimated by geoelectrical studies (Pomposiello et al. 2009). Wells were drilled to a depth of ≤2.6 m, along the main direction of the local groundwater ow (S–N): upstream (P3–P10), downstream

Measured physical and chemical parameters are shown in Table 2 and the chemical classication of water in Figure 8a (sodium chloride and bicarbonate waters). Hydrochemical and isotope values varied in the different ow paths: P3 to P9; P10 to P9; P7 to P5 (Sanci et al., 2011). Alkalinity and 13C-DIC tended to increase along the ow paths previously mentioned (Figure 8b). Values changed from 2.98 mmol L-1/-12.1‰to 8.29 mmol L-1/4.4‰ (P3–P9), 2.44 mmol L-1/-15.0‰ to 8.29 mmol L-1/4.4‰ (P10–P9) and 8.18 mmol L-1/-8.1‰ to 41.45 mmol L-1/0.8‰ (P7–P5). Considering the results obtained for P3 and P10

*Parameter* **P3 P10 P9 P7 P5**  Temperature (ºC) 18.5 19 23.2 21.3 20.4 pH 6.9 6.9 7.1 6.9 7.0 Conductivity (S/cm) 610 720 2440 1320 5450 Alkalinity (mmol/L) 2.98 2.44 8.29 8.18 41.45 Sulphate (mmol/L) 0.05 0.05 0.08 0.30 5.00 Chloride (mmol/L) 2.20 3.75 13.68 5.42 13.15 Sodium (mmol/L) 4.52 4.65 20.75 14.53 70.90 Potassium (mmol/L) 0.06 1.46 0.16 0.02 0.07 Calcium (mmol/L) 0.67 0.45 0.47 0.39 0.77 Magnesium (mmol/L) 0.23 0.30 1.60 0.33 1.81 13C (‰ vs. PDB) -12.1 -15.0 4.4 -8.1 0.8

100\*(Cat-|An|)/(Cat+|An|) 8.86 8.17 6.17 5.97 8.42

Fig. 8. A. Piper diagram showing the chemical classification of wells. P3, P7, P5: sodium bicarbonate groundwater; P10, P9: sodium chloride groundwaterchemical classication of

groundwater; B. Alkalinity versus δ13C-DIC (Sanci et al., 2011).

(P9) and within the landll (P7–P5).

Analytical error (%)

Table 2. Groundwater composition of piezometers

The measurement of CO2 uxes in the MSWFDS with previous calibration in laboratory made it possible to reliably measure the spatial variability of the emissions in the eld. CO2 uxes obtained (31–331 g m-2 day-1) revealed the skewed distribution of the data (Figure 5c). This type of distribution (log-normal) was also observed in other soil CO2 ux studies of natural and anthropic emission sources (Bergfeld et al. 2001; Cardellini et al. 2003; Chiodini and Frondini 2001; Gerlach et al. 2001). Logarithmic probability plots (Fig. 6b) show the polymodal distribution of CO2 uxes. They are a combination of three log-normal populations, which indicate that different processes of CO2 generation take place. Low CO2 uxes are similar to the background values found around the MSFDSW and derive from soil respiration (Welles et al. 2001). The remaining values can be grouped into moderate and high uxes. Values for both populations agree with those reported for biodegradation of solid waste in other sanitary landlls (Cardellini et al. 2003; Georgaki et al. 2008; Hedge et al. 2003; Jha et al. 2008; Pier & Kelly 1997). Spatial variations observed in surface CO2 ux distribution (Figure 6c) are due to MSFDSW inhomogeneities. This variability may indicate that waste was buried in a NW–SE direction. Younger parts of the MSFDSW where biodegradation started later are more likely to have the highest CO2 ux values.

Quantied anomalous CO2 uxes within the MSFDSW show that the site is still undergoing MSW biodegradation. Measured CO2, CH4, N2 and O2 concentrations are similar to those described for an initial phase of aerobic oxidation or a posthumous stage of biodegradation (mature), and they are different from those described for anaerobic phases. Considering the time since the end of operations at the MSFDSW (about 10 years), the values obtained for these gases may indicate that the MSFDSW is in a mature stage. Moreover, the results obtained for C isotopes in the probe-sampled CO2 (-34.2 to -17.6‰) suggest that the MSWFDS is in an aerobic phase of biodegradation.

Although concentrations of CO2, CH4, N2 and O2 and C isotope are indicators of the degree of maturity of sanitary landlls, in practice, the factors affecting gas generation need to be considered. These factors affect the duration of each particular biodegradation stage, as well as the degrees of phase overlap and phase omission. In this case, the shallow burial of waste (2 m), the absence of CH4 and mainly the permeability of the top cover due to little compaction and inadequate materials, make it possible to assume that anaerobic conditions necessary for methanogenic reactions have not been achieved in the MSFDSW. Therefore, since MSW were rst disposed, biodegradation was completely aerobic or the initial phase of aerobic oxidation might have been followed by another phase of anaerobic oxidation with no generation of CH4, arriving at the current maturation stage.

Data show that groundwater alkalinity grows across the MSFDSW in the direction of underground ow (Fig. 8). This suggests an input of C generated by anomalous CO2, which dissolves calcite from calcareous levels such as ''tosca'' in the MSFDSW. The interaction between carbonic acid and mineral carbonates might even buffer pH variations through

CO2 and CH4 Flux Measurements from Landfills

without developing methanogenic processes.

these possibilities best explains the process.

Vol. 38, pp. 4891–4899

**9. Acknowledgment** 

**10. References** 

– A Case Study: Gualeguaychú Municipal Landfill, Entre Ríos Province, Argentina 269

completely aerobic or that it may have gone through a period of anaerobic oxidation,

Based on the increase in groundwater alkalinity as it ows across the MSWFDS and the DIC isotope composition, two different situations are possible: either CO2 derived from MSW biodegradation is dissolving, or dissolved organic matter is decaying within the free aquifer due to the presence of leachates. Future research will be devoted to determining which of

This research was supported by the Instituto de Geocronología y Geología Isotópica (UBA-CONICET) and PICT 2002 Nº 12243. The authors are grateful to Eduardo Llambas, Anibal

Baedecker, M. & Back, W. (1979) Hydrogeological processes and chemical reactions at a

Barlaz, M.; Green, R.; Chanton, J.; Goldsmith, C. & Hater, G. (2004) Evaluation of a

Bergfeld, D.; Goff, F. & Janik, C. (2001) Elevated carbon dioxide flux at the Dixie Valley

Bjerg, P.; Albrechtsen, H.; Kjeldsen, P. & Christensen, T. (2005) The groundwater

Borjesson, G.; Danielson, A. & Svensson, B. (2000) Methane fluxes from a Swedish landfill

Butnor, J. & Johnsen, K. (2004) Calibrating soil respiration measures with a dynamic flux

Cardellini, C.; Chiodini, G.; Frondini, F.; Granieri, D.; Lewicki, J. & Peruzzi, L. (2003)

Chiodini, G.; Cioni, R.; Guidi, M.; Raco, B. & Marini, L. (1998) Soil CO2 flux measurement in volcanic and geothermal areas. *Appl Geochem*, Vol. 13, Nº 5, pp. 543–552 Chiodini, G.; Frondini, F.; Kerrick, D.; Rogie, J.; Parello, F.; Peruzzi, L. & Zanzari, A. (1999)

Chiodini, G. ; Avino, R.; Brombach, T.; Caliro, S.; Cardellini, C. ; De Vita, S.; Frondini, F.;

Sherwood Lollar (Ed.), pp. 579–612, Elsevier–Pergamon, Oxford

geothermal areas and landfills. *Appl Geochem*, Vol. 18, pp. 45–54

region, Central Italy. *Chem Geol*, Vol 177, pp. 67–83

biologically active cover for mitigation of landfill gas emissions. *Environ Sci Technol*,

geothermal field, Nevada; relations between surface phenomena and geothermal

geochemistry of waste disposal facilities. In: *Treatise on Geochemistry,* Volume 9*,* B.

determined by geostatistical treatment of static chamber measurements. *Environ Sci* 

apparatus using artificial soil media of varying porosity. *Eur J Soil Sci* , Vol. 55, pp.

Accumulation chamber measurement of methane fluxes: application to volcanic-

Quantification of deep CO2 fluxes from Central Italy. Examples of carbon balance for regional aquifers and soil diffuse degassing. *Chem Geol*, Vol. 159, pp. 205–222 Chiodini, G. & Frondini, F. (2001) Carbon dioxide degassing from the Albani Hills volcanic

Granieri, D.; Marotta, E. & Ventura, G. (2004) Fumarolic and diffuse soil degassing west of Mount Epomeo, Ischia, Italy. *J Volcanol Geotherm Res*, Vol. 133, pp. 291–309

Tricarico and Gabriel Giordarengo for their collaboration in the eld.

landfill. *Ground Water* , Vol. 17, Nº 5, pp. 429–437

reservoir. *Chem Geol*, Vol. 177, pp. 43–66

*Technol* , Vol. 34, pp. 4044–4050

639–647

increased CO2 (no changes are observed in the pH of the samples). Moreover, the expected effect of the MSWFDS CO2 gas on groundwater chemistry could be an increase not only in alkalinity, but also in calcium. Calcium concentrations obtained in this study were similar to the background values. The presence of clays promoting the Ca/Na exchange reaction would explain this fact. In addition, the results of applying stable C isotopes on DIC (δ13C of -15.0 to 4.4 ‰) conrmed the changes observed in water alkalinity due to the anomalous ingression of CO2. 13C-DIC values richer than -12‰ can be explained by the input of CO2 from MSW degradation which dissolves calcite (a geologically feasible process). Another possibility is the decomposition of dissolved organic matter within the phreatic aquifer due to inltration of leachates, given observable signs, such as previous geophysical studies (Pomposiello et al. 2009), subhumid/humid climate with annual rainfall of 1,077 mm, permeable cover, shallow phreatic aquifer and increased concentration of leachate associated constituents such as chlorides and sodium. Enrichment of 13C-DIC in groundwater affected by leachates reached +13‰ (van Breukelen et al. 2003) and +20‰ (North et al. 2004). Future studies will examine and compare the hydrogeochemical changes caused by the dissolution of landll gases with those produced by the presence of leachates.

## **8. Conclusions**

According to the experiences obtained in the process of quantification of CO2 and CH4 fluxes, the two most important factors that had to be considered were the sampling methods and the sampling sites. In the first case, none of the accumulation chamber methods are nowadays considered to be standard because of differences in flux estimations between chamber types or chamber-specific limitations. Therefore, laboratory experiments were needed to achieve accurate flux measurements before applying those methods in the field. In addition, to assess a reliable spatial variability of fluxes, it was necessary to test different regular sampling grids until the geostatistical analysis indicated that an adequate sampling density had been achieved. The degree of spatial dependence between chamber sites was analyzed in experimental variograms.

In this way, direct measurements of CO2 diffuse degassing from surface, with the accumulation chamber methodology tested in laboratory, allowed the detection of the spatial variability of CO2 uxes in the MSWFDS and the assessment of CO2 released to atmosphere from this source. Different subpopulations were identied by the statistical and geostatistical analyses of CO2 uxes. Processes giving rise to the subpopulations are background values attributable to plant respiration and different anomalous values related to biodegradation of urban solid waste disposed in the MSWFDS.

Analysis of probe-sampled concentrations of CO2, CH4, N2 and O2, as well as carbon isotope composition of the CO2, showed that the current process is an aerobic phase of biodegradation. However, other factors that affect the gas generation were also considered to determine the phase of biodegradation of MSWFDS. They were: the time since the end of operations in the MSWFDS, the characteristics of the environment where MSW were disposed (mainly permeability of capping due to little compaction and inappropriate cover material), the shallow depth of SW burial and the absence of CH4. All of these factors made it possible to assume that we are in the presence of a mature or posthumous stage of biodegradation, and allowed us to assume that biodegradation in the MSWFDS was completely aerobic or that it may have gone through a period of anaerobic oxidation, without developing methanogenic processes.

Based on the increase in groundwater alkalinity as it ows across the MSWFDS and the DIC isotope composition, two different situations are possible: either CO2 derived from MSW biodegradation is dissolving, or dissolved organic matter is decaying within the free aquifer due to the presence of leachates. Future research will be devoted to determining which of these possibilities best explains the process.

## **9. Acknowledgment**

268 Greenhouse Gases – Emission, Measurement and Management

increased CO2 (no changes are observed in the pH of the samples). Moreover, the expected effect of the MSWFDS CO2 gas on groundwater chemistry could be an increase not only in alkalinity, but also in calcium. Calcium concentrations obtained in this study were similar to the background values. The presence of clays promoting the Ca/Na exchange reaction would explain this fact. In addition, the results of applying stable C isotopes on DIC (δ13C of -15.0 to 4.4 ‰) conrmed the changes observed in water alkalinity due to the anomalous ingression of CO2. 13C-DIC values richer than -12‰ can be explained by the input of CO2 from MSW degradation which dissolves calcite (a geologically feasible process). Another possibility is the decomposition of dissolved organic matter within the phreatic aquifer due to inltration of leachates, given observable signs, such as previous geophysical studies (Pomposiello et al. 2009), subhumid/humid climate with annual rainfall of 1,077 mm, permeable cover, shallow phreatic aquifer and increased concentration of leachate associated constituents such as chlorides and sodium. Enrichment of 13C-DIC in groundwater affected by leachates reached +13‰ (van Breukelen et al. 2003) and +20‰ (North et al. 2004). Future studies will examine and compare the hydrogeochemical changes caused by the dissolution of landll gases with those produced by the presence of

According to the experiences obtained in the process of quantification of CO2 and CH4 fluxes, the two most important factors that had to be considered were the sampling methods and the sampling sites. In the first case, none of the accumulation chamber methods are nowadays considered to be standard because of differences in flux estimations between chamber types or chamber-specific limitations. Therefore, laboratory experiments were needed to achieve accurate flux measurements before applying those methods in the field. In addition, to assess a reliable spatial variability of fluxes, it was necessary to test different regular sampling grids until the geostatistical analysis indicated that an adequate sampling density had been achieved. The degree of spatial dependence between chamber sites was

In this way, direct measurements of CO2 diffuse degassing from surface, with the accumulation chamber methodology tested in laboratory, allowed the detection of the spatial variability of CO2 uxes in the MSWFDS and the assessment of CO2 released to atmosphere from this source. Different subpopulations were identied by the statistical and geostatistical analyses of CO2 uxes. Processes giving rise to the subpopulations are background values attributable to plant respiration and different anomalous values related

Analysis of probe-sampled concentrations of CO2, CH4, N2 and O2, as well as carbon isotope composition of the CO2, showed that the current process is an aerobic phase of biodegradation. However, other factors that affect the gas generation were also considered to determine the phase of biodegradation of MSWFDS. They were: the time since the end of operations in the MSWFDS, the characteristics of the environment where MSW were disposed (mainly permeability of capping due to little compaction and inappropriate cover material), the shallow depth of SW burial and the absence of CH4. All of these factors made it possible to assume that we are in the presence of a mature or posthumous stage of biodegradation, and allowed us to assume that biodegradation in the MSWFDS was

to biodegradation of urban solid waste disposed in the MSWFDS.

leachates.

**8. Conclusions** 

analyzed in experimental variograms.

This research was supported by the Instituto de Geocronología y Geología Isotópica (UBA-CONICET) and PICT 2002 Nº 12243. The authors are grateful to Eduardo Llambas, Anibal Tricarico and Gabriel Giordarengo for their collaboration in the eld.

## **10. References**


CO2 and CH4 Flux Measurements from Landfills

47, Nº 11, pp. 24-31

514–518

2463

44, pp. 115–150

120, UNESCO, Paris

*Meteorol*, Vol. 101, pp. 1–14

*Atmos Environ*, Vol. 38, pp. 4921–4929

*Chem Phys*, Vol. 18, Nº 6, pp. 849–857

– A Case Study: Gualeguaychú Municipal Landfill, Entre Ríos Province, Argentina 271

Kerfoot, H.; Baker, J. & Burt, D. (2004) Geochemical changes on ground water due to landfill

Kinzig, A. P. y Socolow, R. H. (1994) Human impacts on nitrogen cycle. *Physics Today,* Vol.

Kumar, S.; Mondal, A.; Gaikwad, S.; Devotta, S. & Singh, R. (2004) Qualitative assessment of

MacCrea, J. (1950) On the isotopic chemistry of carbonates and paleotemperature scale. J

Martin, J.; Bolstad, P. & Norman, J. (2004) A carbon dioxide flux generator for testing

Meju, M. (2000) Geoelectrical investigation of old/abandoned, covered landll sites in urban

Mook, W. (2000) Natural abundance of the stable isotopes of C, H and O, In: *Environmental* 

Morën, A. & Lindroth, A. (2000) CO2 exchange at the floor of a boreal forest. *Agric For* 

Mosher, B.; Czepiel, P. & Harriss, R. (1999) Methane emissions at nine landfill sites in the northeastearn United States. *Environ Sci Technol*, Vol. 33, pp. 2088–2094 Nay, S.; Matsson, K. & Bormann, B. (1994) Biases of chamber methods for measuring soil

Norman, J.; Garcia, R. & Verma, S. (1992) Soil surface CO2 fluxes and the carbon budget of a

Norman, J.; Kucharik, C.; Gower, S.; Baldocchi, D.; Crill, P.; Rayment, M.; Savage, K. &

North, J.; Rusell, D. & Barrie, M. (2004) The use of carbon and nitrogen isotope ratios to

Parkinson, K. (1981) An improved method for measuring soil respiration in the field. *J Appl* 

Pereyra, F.; Tchilinguirian, P. & Baumann, V. (2002) Hoja Geológica 3360-IV Gualeguaychú. Boletín No 335, Servicio Geológico Minero Argentino, Buenos Aires Pier, P. & Kelly, J. (1997) Measured and estimated methane and carbon dioxide emissions

Pomposiello, C.; Dapeña, C.; Boujon, P. & Favetto, A. (2009) Tomografías eléctricas en el

Prezzi, C.; Orgeira, M.; Ostera, H. & Vazquez, C. (2005) Ground magnetic survey of a

grassland. *J Geophys Res*, Vol. 97, Nº D17, pp. 18845–18853

dioxide fluxes. *J Geophys Res,* Vol. 102, pp. 28771–28777

Zealand. *Environ Int*, Vol. 30, pp. 631–637

strategies. *Bioresour Technol*, Vol. 61, pp. 213–220

de contaminación. *Rev Asoc Geol Arg*, Vol. 64, pp. 603–614

*Ecol*, Vol. 18, pp. 221–228

pp. 889–897

methane emission inventory from municipal solid waste disposal: a case study.

infrared gas analyser based soil respiration systems. *Soil Sci Soc Am J*, Vol. 68, pp.

areas: mode development with a genetic diagnosis approach. *J Appl Geophys*, Vol.

*isotopes in the hydrological cycle: principles and applications*, W., Mook (Ed.), pp. 89–

CO2 efflux demonstrated with a laboratory apparatus. *Ecology*, Vol. 75, pp. 2460–

Striegl, R. (1997) A comparison of six methods for measuring soil-surface carbon

identify landfill leachate contamination: Green Island Landfill, Dunedin, New

from sawdust waste in the Tennessee Valley under alternative management

Basurero Municipal Ciudad de Gualeguaychú, provincia de Entre Ríos. Evidencias

municipal solid waste landfill: pilot study in Argentina. *Environ Geol*, Vol. 47, Nº 7,

gas effects. *Ground Water Monit Rem*, Vol. 24, Nº 1, pp. 60–65


Coleman, D.; Liu, C.; Hacley, K. & Benson, L. (1993) Identification of landfill methane using

Engineering Professional Development, University of Wisconsin, Madison. Eddy, W. & Paninatier, Y. (1996) *Variowin: software for Spatial Data Analysis in 2-D*, Springer,

Evans, C.; Sorey, M.; Kennedy, B.; Stonnestrom, D.; Rogie, J. & Shuster, D. (2001) High CO2

Farquar, G. & Rovers, F. (1973) Gas production during refuse decomposition. *Water Air Soil* 

Georgaki, I.; Soupios, P.; Sacas, N.; Ververidis, F.; Trantas, E.; Vallianatos, F. & Manios, T.

Gerlach, T.; Doukas, M.; McGee, K. & Klesser, R. (2001) Soil efflux and total emission rates of

Gonfiantini, R.; Stichler, W. & Rozanski, K. (1995) Standards and intercomparison materials

Hackley, K.; Liu, C. & Coleman, D. (1996) Environmental isotope characteristics of landfill

Hanson, P.; Wullschleger, S.; Bohlman, S. & Todd, D. (1993) Seasonal and topographic

Hedge, U.; Chang, T. & Yang, S. (2003) Methane and carbon dioxide emissions from Shan-Chu-Ku landfill site in northern Taiwan. *Chemosphere*, Vol. 52, pp. 1275–1285 Hutchison, J. & Mosier, A. (1981) Improved soil cover method for field measurements of

Iriondo, M. (1980) El Cuaternario de Entre Ríos. *Rev Asoc Cienc Nat Litoral*, Vol. 11, pp. 125–

Iriondo, M. (1996) Estratigrafía del Cuaternario de la cuenca del río Uruguay, *XIII Congreso* 

Iritz, Z.; Lindroth, A. & Gardenas, A. (1997) Open ventilated chamber for measurements of H2O and CO2 fluxes from the soil surface. *Soil Technol*, Vol. 10, pp. 169–184 Jensen, L.; Mueller, T.; Tate, K.; Ross, D.; Magid, J & Nielsen, N. (1996) Soil surface CO2 flux

Jha, A.; Sharma, C.; Singh, N.; Ramesh, R.; Purvaja, R. & Gupta, P. (2008) Greenhouse gas

Kerfoot, H.; Baker, J. & Burt, D. (2003) The use of isotopes to identify landfill gas effects on

study of Chennai landfill sites. *Chemosphere*, Vol. 71, pp. 750–758

*Geológico Argentino y III Congreso de Exploración de Hidrocarburos*, pp. 15–25,

as an index of soil respiration in situ: a comparison of two chamber methods. *Soil* 

emissions from municipal solid waste management in Indian mega-cities: a case

*materials for stable isotopes of light elements*, pp. 13–29, IAEA, Vienna.

leachates and gases. *Ground Water*, Vol. 34, Nº 5, pp. 827–836

nitrous oxide fluxes. *Soil Sci Soc Am J*, Vol. 45, pp. 311–316

*Biol Biochem*, Vol. 28, Nº 10–11, pp. 1297–1306

groundwater. *J Environ Monit,* Vol. 5, pp. 896-901

measurement and fractionation. *Chem Geol*, Vol. 177, pp. 15–29

New York

522–531

pp. 1–15

141

AGAIAPG, Buenos Aires.

*Pollut*, Vol. 2, pp. 483–495

1995–1999. *Chem Geol*, Vol. 177, pp. 85–99

carbon and hydrogen isotope analysis, *Proceedings of 16th International Madison Waste Conference, Municipal & Industrial Waste*, pp. 303–314, Department of

emissions through porous media: transport mechanisms and implications for flux

(2008) Evaluating the use of electrical resistivity imaging technique for improving CH4 and CO2 emission rate estimations in landfills. *Sci Total Environ*, Vol. 389, pp.

magmatic CO2 at the Horseshoe Lake tree kill, Mammoth Mountain California,

distributed by the International Atomic Energy Agency for stable isotope measurements, *Proceedings of a consultants meeting of reference and intercomparison* 

patterns of forest floor CO2 efflux from an upland oak forest. *Tree Physiol*, Vol. 13,


**Part 2** 

**Greenhouse Gases Effect and Management** 


http://www.sepa.org.uk/waste/waste\_regulation/landfill.aspx


http://www.environment-agency.gov.uk/business/sectors/108918.aspx

