**Ecological Impacts of Pesticides in Agricultural Ecosystem**

Khalil Talebi1, Vahid Hosseininaveh1 and Mohammad Ghadamyari2

*1Department of Plant Protection, College of Agriculture and Natural Science, University of Tehran, 2Department of Plant Protection, College of Agriculture, University of Guilan, Iran* 

#### **1. Introduction**

142 Pesticides in the Modern World - Risks and Benefits

Mills, L. S. Et al (1993) The Keystone-Species Concept in Ecology and Conservation,

Miller, G. T. (2004), Sustaining the Earth, 6th edition. Thompson Learning, Inc. Pacific

Fox, J. E, et al (2007). "Pesticides reduce symbiotic efficiency of nitrogen-fixing rhizobia and

Johnston, A. E. (1986). "Soil organic-matter, effects on soils and crops". *Soil Use Management*

Lotter DW, Seidel R, and Liebhardt W (2003). "The performance of organic and conventional

Gilliom, R.J. *et al* (2007). The Quality of our nation's waters: Pesticides in the nation's

Heather, B. *et al* (1997). Movement of Pesticides and Best Management Practices, Ground

Henry, L. (2003) Levels of some Pesticides in Environmental Samples from Southern Lake

Hayes *et al*, (2010) The cause of global amphibians decline: a developmental endocrinologist

Muckenfuss A. E. *et al*, (1990). Natural mortality of diamondback moth in coastal South Carolina Clemson University, Coastal Research and Education Center Daly, H. *et al* (1998). Introduction to insect biology and diversity, 2nd edition, Oxford

Sacramento, C. A. (2008). Department of Pesticide Regulation "What are the Potential

Lorenz, E. S. (2009) "Potential Health Effects of Pesticides." Ag Communications and

Health Effects of Pesticides?" Community Guide to Recognizing and Reporting

cropping systems in an extreme climate year". *American Journal of Alternative* 

Victoria and its Catchments and their Chemodynamics in Tilapia Species, Water

Rockets, R. (2007). Down On the Farm, Yields, Nutrients And Soil Quality

host plants". *Proceedings of the National Academy of Sciences, USA*

streams and ground water, 1992–2001. US Geological Survey Bingham, S (2007), Pesticides in rivers and groundwater. Environment Agency, UK.

and Sediments Under Experimental Conditions, Tanzania

perspective, Journal of experimental biology, 213 (6) 921

BioScience 43, (4), 219-224

Grove, California, USA

*Agriculture* 18: 146–154

Water Pollution Primer, Virginia, USA

University Press, New York, USA

Pesticide Problems, 27-29

Marketing. 1-8

2, 97–105

Pesticides are essential tools in integrated pest management (IPM) programs which can have the great influence if they are used properly. However, the adverse impacts of these compounds on the environment and ecosystem should not be ignored. The ecological effects of pesticides can be discussed from different points of view. Some of the significant consequences of use of pesticides are side effects of the pesticides on non-target organisms, sub-lethal effects of the pesticides on target and non-target organisms, emergence of resistant populations and pesticide residue and their entry into the trophic network. Side effect of the pesticides is a controversial issue in pesticides applications. They kill natural enemies present in the field and ecosystem and destroy the natural equilibrium between the hosts and their natural enemies. In the absence of natural enemies, pest populations increase rapidly and makes more controlling efforts, usually pesticides, necessary. In spite of pests, pesticide resistance in natural enemies is not common due to lower exposure to pesticides. Sub-lethal deposits of pesticides can change some biological traits of the organisms exposed to low and highly low concentrations of the toxicants. Sublethal impacts of pesticides are mostly ignored in ecological pesticide assessment because most pesticide assessments are performed as individual-level bioassays and population-level of toxicants has not been considered. Insects (pests and natural enemies) exposed to sub-lethal concentrations of pesticides show some changes in their life history's traits. Resistant populations emerge due to the misuse of pesticides. The populations with high ecological potential are gradually selected generation by generation and subsequent populations are remarkably or completely insensitive to pesticides. Resistant populations are usually different from natural population in their fertility life table characteristics. Nowadays, the existence of pesticides residue in agricultural crops and their entrance into the trophic network has endangered human health and environment, and it has also necessitated the correct use of the pesticides. In the current chapter, the most significant ecological impacts of pesticides in agricultural ecosystems have been discussed.

#### **2. Impacts of pesticides on natural enemies**

The concept of Integrated Pest Management (IPM) was initially defined as the combined use of natural enemies and pesticides to manage pests (Stern et a1., 1959). The IPM

Ecological Impacts of Pesticides in Agricultural Ecosystem 145

insects and high validity of statistical analyses (Robertson & Preisler, 1992). On the contrary, insecticide bioassays with field-collected insect subjects reduce reliability on the real effects of insecticides because of heterogeneity among tested individuals (Robertson & Worner, 1990). In fact, the main research aim in ecotoxicological studies is predicting of insect field populations faced with sublethal concentrations/doses of pesticides (Ferson et al., 1996). However, most studies deal with the impact of insecticides on individuals or some components of individuals (Stark & Banks, 2003). In laboratory tests, individual responses to chronic toxicity may be evaluated from morphological, biochemical, physiological, molecular and ecological point of view. Some responses such as reduction of growth or fecundity or increase in mortality rates or development times (Grant, 1998) are more easily measured in most insecticide bioassays. Although, mortality is the endpoint of interest for many acute studies and it may be used as an endpoint criterion in chronic exposure bioassays, reproductive inhibition or growth retardation are generally considered more sensitive measurements, particularly for the estimation of sublethal responses (Villarroel, 1999; Stark & Banks 2003). Sublethal doses/concentrations of toxicants may change life span, development rates, fecundity, egg viability, sex ratio, consumption rate and behavior of subjects (Dempster 1968; Ruberson et al., 1998; Stark & Banks, 2003, Stark & Rangus 1994; Stark et al. 1992a, 1992b; Vinson, 1974). Individual (sub-population) level responses and population level consequences can be related using life table response experiments (Caswell, 1989). Data generated within life table response experiments give valuable information to the assessment of population-level consequences of toxicant sublethal effects (Caswell,

1989b; 1996a, 1996b; Ferson et al., 1996; Grant, 1998; Stark & Banks 2003).

Lethal dose/concentration of an insecticide that kills 50% of a population (LD50 or LC50) is commonly used as a simplistic criterion for determining and comparing the effects of toxicants (Stark et al., 2007). This approach relies on the death of individuals and ignores many consequent impacts of a toxicant on survivors. In addition to death, exposure to a toxicant may result in simultaneous manifestation of multiple sublethal effects (Stark & Banks, 2003; Stark et al., 2004). Under the phenomenon population compensation, if sublethal effects do not occur but the population density is reduced, survivors may have more resources available and actually produce more offspring than untreated populations (Stark et al., 2007). Effective concentration/dose of a toxicant that affect x% of a population (ECx or EDx) is also used when sublethal effects are scored. (Kammenga & Laskowski, 2000). Demographic toxicological analyses or life table response experiments is another approach, which takes into account total effects that a toxicant might have at the levels of organization higher than the individual (Stark et al., 2004). The advantage of this approach is that a total measure of the effect is determined that incorporates lethal and sublethal effects into a single endpoint, the intrinsic rate of natural increase or *rm* (Kammenga & Laskowski, 2000; Stark & Banks, 2003), which can detect subtle, individual-level effects of contaminants that alter the growth of populations at rates below the lethal concentration limits (Bechmann, 1994, as cited in Rezaei et al., 2007). The first life table response experiment was performed by Birch (1953) to study the impacts of temperature, moisture and food sources on flour beetles (as cited in Kammenga & Laskowski, 2000). The approach has been widely used in ecotoxicological studies; however few studies have been published on the use of demography and similar measures of the population growth rate for evaluating the effect of pesticides on insects, especially insect natural enemies (Kammenga & Laskowski, 2000;

**3.1 Life table response experiments** 

concept later includes coordinated use of all possible tactics to suppress pest damage (Smith et al., 1976, as cited in Ruberson et al., 1998). Use of selective pesticides or rates, temporal separation of pesticides and natural enemies, and spatial separation of pesticides and natural enemies are three main area of integrating natural enemies with pesticides in pest management programs (Ruberson et al., 1998). Conventional use of insecticides can have deleterious effects on natural enemy populations because beneficial arthropods can have greater susceptibility to low concentrations of insecticides than their prey or host (Ruberson et al., 1998; Torres & Ruberson, 2004). Pesticide compatibility with biological control agents is a major concern to practitioners of IPM, and knowledge about the activity of insecticides toward pests, non-target insects and the environment is a necessity (Stark et al., 2004).

Pesticides exert a wide range of lethal (acute and chronic) and sublethal (often chronic) impacts on natural enemies (Rezaei et al., 2007; Ruberson et al., 1998; Stark et al., 2004). Talebi et al. (2008) have published a comprehensive reviewed on the impacts of pesticides on arthropod biological control agents. Sublethal effects are expressed as some changes in the insect's life history attributes (Ruberson et al., 1998). Many studies have been performed on the evaluation of the toxicity of various pesticides to benecial organisms (Kavousi & Talebi, 2003; Lucas et al., 2004; Medina et al., 2003; Oomen et al., 1991; Paine et al., 2011; Rezaei et al., 2007; Steiner et al., 2011; Urbaneja et al., 2008; Van de Veire et al., 2002; Van den Bosch et al., 1956; Walker et al., 1998). Some important issues including natural enemy species, life stages/sexes, routes of pesticide entry, life history parameters, plot size for field screenings and pesticide formulations and rates must be considered for designing bioassays evaluating the effects of pesticides on natural enemies (Ruberson et al., 1998). One of the commonly used methods in testing the side effects of pesticides on natural enemies, recommended by the International Organization of Biological Control (IOBC), is a tiered approach whereby initial pesticide screening is done in the laboratory, and, depending on the results obtained, semi-eld or eld tests may be conducted (Dohmen, 1998; Hassan, 1998). This method has been designed to evaluate the acute residual toxicity as well as sublethal effects of the pesticides on the reproductive performance (Vogt et al., 1992). In this method, dead subjects are recorded (often daily) and the total mortality is calculated. The value of mortality (M) for the treated series is determined as the corrected mortality according to Abbott (1925). The average number of progenies (R) is measured as fecundity affected by exposing to a pesticide. The total effect of a pesticide (E) is calculated by the formula E = 100% - (100% - M) ×R proposed by Overmeer & Van Zon (1982). Based on the total effects, a pesticide is classified using IOBC evaluation categories (Sterk et al., 1999). Rezaei el al. (2007) investigated the effects of imidacloprid, propargite and pymetrozine in laboratory experiments using IOBC-system on the common green lacewing, *Chrysoperla carnea* (Stephens). All three tested pesticides produced significant adverse effects on preimmaginal survival (p<0.01). Imidacloprid had no significant effect on fecundity, but propargite and pymetrozin caused significant reductions (p<0.05). According to IOBC classification, imidacloprid was found to be harmless (E=27.44%), propargite (E=49.78%) and pymetrozine (E=66.9%) were determined as slightly harmful.

#### **3. Population-level impacts of pesticides**

Sublethal effects of pesticides on the fitness of individuals are usually assessed using laboratory bioassays with insects (Grant, 1998) due to reduced variation among subject

concept later includes coordinated use of all possible tactics to suppress pest damage (Smith et al., 1976, as cited in Ruberson et al., 1998). Use of selective pesticides or rates, temporal separation of pesticides and natural enemies, and spatial separation of pesticides and natural enemies are three main area of integrating natural enemies with pesticides in pest management programs (Ruberson et al., 1998). Conventional use of insecticides can have deleterious effects on natural enemy populations because beneficial arthropods can have greater susceptibility to low concentrations of insecticides than their prey or host (Ruberson et al., 1998; Torres & Ruberson, 2004). Pesticide compatibility with biological control agents is a major concern to practitioners of IPM, and knowledge about the activity of insecticides toward pests, non-target insects and the environment is a

Pesticides exert a wide range of lethal (acute and chronic) and sublethal (often chronic) impacts on natural enemies (Rezaei et al., 2007; Ruberson et al., 1998; Stark et al., 2004). Talebi et al. (2008) have published a comprehensive reviewed on the impacts of pesticides on arthropod biological control agents. Sublethal effects are expressed as some changes in the insect's life history attributes (Ruberson et al., 1998). Many studies have been performed on the evaluation of the toxicity of various pesticides to benecial organisms (Kavousi & Talebi, 2003; Lucas et al., 2004; Medina et al., 2003; Oomen et al., 1991; Paine et al., 2011; Rezaei et al., 2007; Steiner et al., 2011; Urbaneja et al., 2008; Van de Veire et al., 2002; Van den Bosch et al., 1956; Walker et al., 1998). Some important issues including natural enemy species, life stages/sexes, routes of pesticide entry, life history parameters, plot size for field screenings and pesticide formulations and rates must be considered for designing bioassays evaluating the effects of pesticides on natural enemies (Ruberson et al., 1998). One of the commonly used methods in testing the side effects of pesticides on natural enemies, recommended by the International Organization of Biological Control (IOBC), is a tiered approach whereby initial pesticide screening is done in the laboratory, and, depending on the results obtained, semi-eld or eld tests may be conducted (Dohmen, 1998; Hassan, 1998). This method has been designed to evaluate the acute residual toxicity as well as sublethal effects of the pesticides on the reproductive performance (Vogt et al., 1992). In this method, dead subjects are recorded (often daily) and the total mortality is calculated. The value of mortality (M) for the treated series is determined as the corrected mortality according to Abbott (1925). The average number of progenies (R) is measured as fecundity affected by exposing to a pesticide. The total effect of a pesticide (E) is calculated by the formula E = 100% - (100% - M) ×R proposed by Overmeer & Van Zon (1982). Based on the total effects, a pesticide is classified using IOBC evaluation categories (Sterk et al., 1999). Rezaei el al. (2007) investigated the effects of imidacloprid, propargite and pymetrozine in laboratory experiments using IOBC-system on the common green lacewing, *Chrysoperla carnea* (Stephens). All three tested pesticides produced significant adverse effects on preimmaginal survival (p<0.01). Imidacloprid had no significant effect on fecundity, but propargite and pymetrozin caused significant reductions (p<0.05). According to IOBC classification, imidacloprid was found to be harmless (E=27.44%), propargite (E=49.78%)

and pymetrozine (E=66.9%) were determined as slightly harmful.

Sublethal effects of pesticides on the fitness of individuals are usually assessed using laboratory bioassays with insects (Grant, 1998) due to reduced variation among subject

**3. Population-level impacts of pesticides** 

necessity (Stark et al., 2004).

insects and high validity of statistical analyses (Robertson & Preisler, 1992). On the contrary, insecticide bioassays with field-collected insect subjects reduce reliability on the real effects of insecticides because of heterogeneity among tested individuals (Robertson & Worner, 1990). In fact, the main research aim in ecotoxicological studies is predicting of insect field populations faced with sublethal concentrations/doses of pesticides (Ferson et al., 1996). However, most studies deal with the impact of insecticides on individuals or some components of individuals (Stark & Banks, 2003). In laboratory tests, individual responses to chronic toxicity may be evaluated from morphological, biochemical, physiological, molecular and ecological point of view. Some responses such as reduction of growth or fecundity or increase in mortality rates or development times (Grant, 1998) are more easily measured in most insecticide bioassays. Although, mortality is the endpoint of interest for many acute studies and it may be used as an endpoint criterion in chronic exposure bioassays, reproductive inhibition or growth retardation are generally considered more sensitive measurements, particularly for the estimation of sublethal responses (Villarroel, 1999; Stark & Banks 2003). Sublethal doses/concentrations of toxicants may change life span, development rates, fecundity, egg viability, sex ratio, consumption rate and behavior of subjects (Dempster 1968; Ruberson et al., 1998; Stark & Banks, 2003, Stark & Rangus 1994; Stark et al. 1992a, 1992b; Vinson, 1974). Individual (sub-population) level responses and population level consequences can be related using life table response experiments (Caswell, 1989). Data generated within life table response experiments give valuable information to the assessment of population-level consequences of toxicant sublethal effects (Caswell, 1989b; 1996a, 1996b; Ferson et al., 1996; Grant, 1998; Stark & Banks 2003).

#### **3.1 Life table response experiments**

Lethal dose/concentration of an insecticide that kills 50% of a population (LD50 or LC50) is commonly used as a simplistic criterion for determining and comparing the effects of toxicants (Stark et al., 2007). This approach relies on the death of individuals and ignores many consequent impacts of a toxicant on survivors. In addition to death, exposure to a toxicant may result in simultaneous manifestation of multiple sublethal effects (Stark & Banks, 2003; Stark et al., 2004). Under the phenomenon population compensation, if sublethal effects do not occur but the population density is reduced, survivors may have more resources available and actually produce more offspring than untreated populations (Stark et al., 2007). Effective concentration/dose of a toxicant that affect x% of a population (ECx or EDx) is also used when sublethal effects are scored. (Kammenga & Laskowski, 2000). Demographic toxicological analyses or life table response experiments is another approach, which takes into account total effects that a toxicant might have at the levels of organization higher than the individual (Stark et al., 2004). The advantage of this approach is that a total measure of the effect is determined that incorporates lethal and sublethal effects into a single endpoint, the intrinsic rate of natural increase or *rm* (Kammenga & Laskowski, 2000; Stark & Banks, 2003), which can detect subtle, individual-level effects of contaminants that alter the growth of populations at rates below the lethal concentration limits (Bechmann, 1994, as cited in Rezaei et al., 2007). The first life table response experiment was performed by Birch (1953) to study the impacts of temperature, moisture and food sources on flour beetles (as cited in Kammenga & Laskowski, 2000). The approach has been widely used in ecotoxicological studies; however few studies have been published on the use of demography and similar measures of the population growth rate for evaluating the effect of pesticides on insects, especially insect natural enemies (Kammenga & Laskowski, 2000;

Ecological Impacts of Pesticides in Agricultural Ecosystem 147

more popular and nearly all estimations are performed according to this method. The jackknife technique is used for ease of statistical comparisons among life table parameters related to each treatment and for estimating the standard errors (SE) associated with the parameters. First, the precise value of *rm* is calculated for all of the raw data (*rtotal*). Then, one of the insect subjects is omitted and an *rm* is computed for the remaining insects (*n-*1). Based on the suggested equation by Meyer et al. (1986) the jackknife pseudo-values were

The value of *n* is the number of insects needed to construct a fertility life table. This process is repeated until pseudo-values were calculated for all *n* possible omissions of one insect from the original data set. Finally *n* number of calculated *ir* are provided to calculate the

1

*is* is the variance of the *n* jackknife pseudo-values. This algorithm is used

1 *<sup>n</sup> j i i r r n*

<sup>2</sup> ( ) *SE r s n j i*

for estimating uncertainties associated with the four other parameters. All jackknife pseudovalues for each treatment are usually subjected to analysis of variance (ANOVA) followed by a convenient mean comparison test. The nonparametric tests are also used for

Intrinsic rate of natural increase, *rm*, is the main and the best estimator for growth rate of insect populations. When values of *rm* are positive, a population is increasing exponentially; when *rm* is equal to zero, a population is stable and when *rm* is negative, a population is declining exponentially and headed toward extinction (Kammenga & Laskowski, 2000). In toxicological studies, values for r*m* are statistically compared among different cohorts (toxicant-treated and control). Rezaei et al. (2007) in life table response experiments of *C. carnea* with some pesticides revealed that imidacloprid and propargite had no signicant effects on the intrinsic rate of natural increase, while pymetrozine caused a 34% reduction in *rm* value (p<0.05). Propargite was non-toxic to *C. carnea* under the tested conditions. The life table assay showed more adverse effects of pymetrozine than a non-life table response experiment method (IOBC method). Lashkari et al. (2007) studied the efficiency of imidacloprid and pymetrozine on population growth parameters of cabbage aphid, *Brevicoryne brassicae* L. (Homoptera: Aphididae). They revealed that *rm* were lower in imidacloprid and pymetrozine treatments than in controls. In such investigations, simple statistical comparisons of *rm* values among cohorts determine efficiency of toxicants. However, a more precise and complicated method is estimating of a concentration/dose of a toxicant at which *rm* value is reduced by 50% (population-level EC50 or ED50) or specific proportions (population-level ECx or EDx) under laboratory conditions. (Suter & Glenn,

some pseudovalues which are not meet ANOVA perquisites (Rezaei et al., 2007).

(6)

(7)

(8)

calculated for this subset of the original data according to:

( 1) *i total <sup>i</sup> r nr n r*

mean (*rj*) and its SE.

In the equation 8, <sup>2</sup>

**3.3 Life table parameters** 

1993; Tanaka & Nakanishi, 2001).

Rezaei et al., 2007; Stark & Banks, 2003). Life table response experiments are being increased to measure multiple endpoints of effects and have been recommended as a superior laboratory toxicological endpoint (Stark et al., 1997). In general, the main reason to use life table response experiments in toxicological studies is revealing of total effect (lethal, sublethal and too subtle impacts) of a toxicant on an insect at the population level. In a few investigations, especially in pesticide side effect studies, total effect of a pesticide is measured using the index E which incorporates mortality and fecundity (Overmeer & Van Zon, 1982; Rezaei et al., 2007). However, the index E is not like the demographic parameters (such as *rm*) which measure the impact of a toxicant at population level.

#### **3.2 Construction of a life table**

Demography has been used in a small number of toxicological studies to evaluate lethal and sublethal effects of toxicants on insect populations (Stark & Banks, 2003; Stark et al., 2007). The basic principal in insect toxicological demography is construction a fertility table. The construction of a number of life tables is an important component in the understanding of the population dynamics of a species (Carey, 1993).

A life table, for each treatment (toxicant concentration or dose), is constructed by following an insect cohort (egg, larva or adult), till the death of all individual members of a cohort, individually, and recording the age of each female (x), the probability that a new individual is alive at age x (Lx), and the number of female offspring produced by a female with attributed x (mx) were recorded. Each individual from the initial cohort is treated according to a convenient procedure depends on test subject, toxicant and purpose. The survived individuals from the treated individuals are maintained and monitored individually to collect necessary data for construction life tables.

The precise value of the intrinsic rate of increase (*rm*) is obtained by solving the Euler equation (Andrewartha & Birch, 1954):

$$\sum\_{\mathbf{x}=\mathbf{0}}^{\mathbf{y}} \mathbf{L}\_{\mathbf{x}} \mathbf{m}\_{\mathbf{x}} \mathbf{e}^{-\mathbf{r}\mathbf{x}} = \mathbf{1} \tag{1}$$

In this equation, y is the oldest age class, Lx is the survival of a newborn female to the midpoint of an age interval, and x is the age of each female at each age interval. In addition to rm, the other main fertility life table parameters including net reproductive rate (R0), generation time (T), doubling time (DT), and finite rate of increase (λ) are also computed using the following formulas, respectively:

$$\mathbf{R}\_0 = \sum \mathbf{L}\_\mathbf{x} \mathbf{m}\_\mathbf{x} \tag{2}$$

$$\mathbf{T} = \sum \mathbf{x} \mathbf{L\_x m\_x} \sqrt{\sum \mathbf{L\_x m\_x}} \tag{3}$$

$$\text{DT} = \ln(\text{2}) / \text{r}\_{\text{m}} \tag{4}$$

$$
\lambda = \mathbf{e}^{\mathbf{r}\_m} \tag{5}
$$

In fact, these parameters are estimations for a given population; therefore, the uncertainty associated with them must be estimated. Uncertainty associated with the parameters can be estimated using two techniques; jackknife and bootstrap. However, jackknife technique is

Rezaei et al., 2007; Stark & Banks, 2003). Life table response experiments are being increased to measure multiple endpoints of effects and have been recommended as a superior laboratory toxicological endpoint (Stark et al., 1997). In general, the main reason to use life table response experiments in toxicological studies is revealing of total effect (lethal, sublethal and too subtle impacts) of a toxicant on an insect at the population level. In a few investigations, especially in pesticide side effect studies, total effect of a pesticide is measured using the index E which incorporates mortality and fecundity (Overmeer & Van Zon, 1982; Rezaei et al., 2007). However, the index E is not like the demographic parameters

Demography has been used in a small number of toxicological studies to evaluate lethal and sublethal effects of toxicants on insect populations (Stark & Banks, 2003; Stark et al., 2007). The basic principal in insect toxicological demography is construction a fertility table. The construction of a number of life tables is an important component in the understanding of

A life table, for each treatment (toxicant concentration or dose), is constructed by following an insect cohort (egg, larva or adult), till the death of all individual members of a cohort, individually, and recording the age of each female (x), the probability that a new individual is alive at age x (Lx), and the number of female offspring produced by a female with attributed x (mx) were recorded. Each individual from the initial cohort is treated according to a convenient procedure depends on test subject, toxicant and purpose. The survived individuals from the treated individuals are maintained and monitored individually to

The precise value of the intrinsic rate of increase (*rm*) is obtained by solving the Euler

Lme 1

In this equation, y is the oldest age class, Lx is the survival of a newborn female to the midpoint of an age interval, and x is the age of each female at each age interval. In addition to rm, the other main fertility life table parameters including net reproductive rate (R0), generation time (T), doubling time (DT), and finite rate of increase (λ) are also computed

In fact, these parameters are estimations for a given population; therefore, the uncertainty associated with them must be estimated. Uncertainty associated with the parameters can be estimated using two techniques; jackknife and bootstrap. However, jackknife technique is

(1)

R Lm 0 xx (2)

DT ln(2) r m (4)

rm <sup>λ</sup> e (5)

T xL m L m xx xx (3)

<sup>y</sup> rx x x

x 0

(such as *rm*) which measure the impact of a toxicant at population level.

**3.2 Construction of a life table** 

the population dynamics of a species (Carey, 1993).

collect necessary data for construction life tables.

equation (Andrewartha & Birch, 1954):

using the following formulas, respectively:

more popular and nearly all estimations are performed according to this method. The jackknife technique is used for ease of statistical comparisons among life table parameters related to each treatment and for estimating the standard errors (SE) associated with the parameters. First, the precise value of *rm* is calculated for all of the raw data (*rtotal*). Then, one of the insect subjects is omitted and an *rm* is computed for the remaining insects (*n-*1). Based on the suggested equation by Meyer et al. (1986) the jackknife pseudo-values were calculated for this subset of the original data according to:

$$
\tilde{r}\_i = nr\_{\text{total}} - (n-1)\overset{\wedge}{r\_i} \tag{6}
$$

The value of *n* is the number of insects needed to construct a fertility life table. This process is repeated until pseudo-values were calculated for all *n* possible omissions of one insect from the original data set. Finally *n* number of calculated *ir* are provided to calculate the mean (*rj*) and its SE.

$$r\_{\tilde{j}} = \frac{1}{n} \sum\_{i=1}^{n} \tilde{r}\_i \tag{7}$$

$$
\hat{SE}(r\_j) = \sqrt{s\_i^2 \/ m} \tag{8}
$$

In the equation 8, <sup>2</sup> *is* is the variance of the *n* jackknife pseudo-values. This algorithm is used for estimating uncertainties associated with the four other parameters. All jackknife pseudovalues for each treatment are usually subjected to analysis of variance (ANOVA) followed by a convenient mean comparison test. The nonparametric tests are also used for some pseudovalues which are not meet ANOVA perquisites (Rezaei et al., 2007).

#### **3.3 Life table parameters**

Intrinsic rate of natural increase, *rm*, is the main and the best estimator for growth rate of insect populations. When values of *rm* are positive, a population is increasing exponentially; when *rm* is equal to zero, a population is stable and when *rm* is negative, a population is declining exponentially and headed toward extinction (Kammenga & Laskowski, 2000). In toxicological studies, values for r*m* are statistically compared among different cohorts (toxicant-treated and control). Rezaei et al. (2007) in life table response experiments of *C. carnea* with some pesticides revealed that imidacloprid and propargite had no signicant effects on the intrinsic rate of natural increase, while pymetrozine caused a 34% reduction in *rm* value (p<0.05). Propargite was non-toxic to *C. carnea* under the tested conditions. The life table assay showed more adverse effects of pymetrozine than a non-life table response experiment method (IOBC method). Lashkari et al. (2007) studied the efficiency of imidacloprid and pymetrozine on population growth parameters of cabbage aphid, *Brevicoryne brassicae* L. (Homoptera: Aphididae). They revealed that *rm* were lower in imidacloprid and pymetrozine treatments than in controls. In such investigations, simple statistical comparisons of *rm* values among cohorts determine efficiency of toxicants. However, a more precise and complicated method is estimating of a concentration/dose of a toxicant at which *rm* value is reduced by 50% (population-level EC50 or ED50) or specific proportions (population-level ECx or EDx) under laboratory conditions. (Suter & Glenn, 1993; Tanaka & Nakanishi, 2001).

Ecological Impacts of Pesticides in Agricultural Ecosystem 149

United State to sulphur and sulphur–lime (as inorganic pesticide) (as cited in Stenersen, 2004). The organochlorine and synthetic insecticides were commercialized for chemical control of pests in the 1940's. The first case of DDT resistance in insect was reported in *Musca domestica* few years after introduction. After that, new insecticides such as cyclodienes, pyrethroids, organophosphates (OP), carbamates, formamidines, *Bacillus thuringiensis*, avermectins, spinosyns, insect growth regulators (IGR) and neonicotinoids were introduced for pest control and the cases of resistance to these compounds appeared a few years after their application. Now, more than 504 key pest species were resistant to pesticides and the resistance to pesticides has become a major contemporary problem in pest management programs (IRM) worldwide. Stuart (2003) reported resistance of 520 insect and

Pesticides resistance reduces the ability control of pesticides on pests and leads to higher application rates to achieving satisfactory pest control. Pimentel (2003) estimated the major economic and environmental losses due to the application of pesticides on crops and veterinary purpose in the USA and showed the following costs: "public health, \$1.1 billion year-1; pesticide resistance in pests, \$1.5 billion; crop losses caused by pesticides, \$1.1 billion; bird losses due to pesticides, \$2.2 billion; and ground water contamination, \$2.0

Reduced pesticide selection pressure for each resistance mechanism is necessary for avoiding and delaying control failure prior to occurrence of resistance. For achieving this purpose, successful detection techniques are required for avoiding resistance developing and a control failure. "These techniques could be able to detect of resistant individual at low frequency in natural population" (Scott, 1995). Techniques for monitoring resistance to different pesticides in pest population gathering valuable information for insecticide

Detection and identification of resistance mechanisms to pesticides require monitoring approach with appropriate bioassay method. Monitoring of resistance is required in order to sustainable management of pesticide resistance and to know the status of resistance. Therefore, developing precision and reliable susceptibility test must be developed. These tests must be also accurate, chip, easy to perform in variety of conditions in laboratory and on farm site. So far, many standardized susceptibility test method were presented by Food Agriculture Organization (FAO) and World Health Organization (WHO) such as exposure to standard residue treatment on glass scintillation vial or filter paper, plastic bags, topical application, spray application of standard solutions, and resistance detection kits and strip. Resistance frequencies can be detected and monitored by bioassays using diagnostic (discriminating) dose (LD99) and estimating resistance factor (Rf= LD50 of resistant population/LD50 of susceptible population). The diagnostic dose (ie. LD99) can be calculated from regression line of log dose probit-mortality data using appropriate software such as POLO-PC. This dose discriminate the tested population as susceptible and resistant and the pests that die after exposures with LD99 of pesticide are classified as susceptible and those individual that survive from exposure considered as resistant. The discriminating-dose assay is a chip, less time consuming approach for monitoring resistance in natural pest populations. These bioassay procedures provide valuable data for monitoring of resistance but this method is not practical for detection of resistance in low frequencies in field

acari species, 150 plant pathogen species and 273 weed species to pesticides.

billion" (Pimentel, 2005).

**4.1 Detection and monitoring of resistance** 

resistance management (IRM) employer.

population of pests (Roush and Miller, 1986).

#### **3.4 Age-stage two-sex life table**

In construction of a fertility life table, raw data is commonly collected from survival and reproduction of female individuals. In this method, males are completely ignored and only used for fertilizing females in a cohort. Ignoring the sex of individuals can result in errors (Chi, 1988). Chi & Liu (1985) and Chi (1988) developed a new method, age-stage two-sex life table, for construction a life table with taking into consideration both female and male sexes. In a small number of investigations, two-sex life table theory have been used for construction of the life table and data analysis (Chi & Su, 2006; Kavousi et al., 2009; Refaat et al., 2005; Schneider et al., 2009; Yang et al., 2006; Yu et al., 2005). As far as the authors aware, there is only one investigation, Schneider et al. (2009), on the effect of a toxicant on an insect according to the age-stage two-sex life table theory. Schneider et al. (2009) determined the side-effects of glyphosate (a herbicide) on development, fertility and demographic parameters of *C. externa* (Neuroptera: Chrysopidae) in the laboratory. They revealed that glyphosate will decrease arthropod population performance and the major detrimental effect observed on *C. externa* was on fecundity and fertility.

#### **3.5 Drawbacks to the use of life table response experiments**

Although life table response experiments may provide the most complete data for the impacts of a pesticide on an animal subject at population-level, there are some disadvantages associated with this method. The most important one is that life table response experiments are expensive and time consuming. Construction of a life table is difficult or impossible for some species (long-lived species) when exposed to a pesticide because of the low rate of reproduction. The other major disadvantage is unrealistic conditions under which a life table is constructed. These conditions are far from the natural conditions in field (Kammenga & Laskowski, 2000; Stark & banks, 2003).

## **4. Resistance of pests to pesticides**

Pesticides are used extensively for control of invertebrate pests, plant pathogens, weeds and rodents and other pests in a wide range of crops and for veterinary purpose. Resistant to pesticides develop in insects, mites, fungi, weeds, bacteria and rodents. Repeated applications and extensive use of the synthetic pesticides has toxicity toward natural enemies and cause resistance development in pest species against major classes of pesticides throughout the world. The repeated and extensive application of pesticides caused majority on susceptible individuals in population and only some resistant individuals survive from pesticide exposure. The offspring genotype of survival individual is homozygous or heterozygous that depends on history of pesticide application and type of pesticides. The offspring inherit the resistant genes and survival ability from the exposure to the pesticides. The surviving individuals multiply in absence of their natural enemies and finally replace the non-resistant population. The development of pesticide resistance is a Darwinian evolutionary process at a rate that rare genes conferring resistance to pesticides are selected by the high selection of pesticides. Resistance to pesticides is defined as "the development of an ability in a population of a pest to tolerate doses of pesticides that would prove lethal to the majority of individuals in a normal population of the same species" (Stenersen, 2004).

The first case of resistance occurrence in insect pests was reported in 1908. This document reported the failure in the control of *Quadraspidiotus perniciosus* (Hem.: Diaspididae) by sulphur. After this report, Melander (1914) reported resistance of three scale strains in

In construction of a fertility life table, raw data is commonly collected from survival and reproduction of female individuals. In this method, males are completely ignored and only used for fertilizing females in a cohort. Ignoring the sex of individuals can result in errors (Chi, 1988). Chi & Liu (1985) and Chi (1988) developed a new method, age-stage two-sex life table, for construction a life table with taking into consideration both female and male sexes. In a small number of investigations, two-sex life table theory have been used for construction of the life table and data analysis (Chi & Su, 2006; Kavousi et al., 2009; Refaat et al., 2005; Schneider et al., 2009; Yang et al., 2006; Yu et al., 2005). As far as the authors aware, there is only one investigation, Schneider et al. (2009), on the effect of a toxicant on an insect according to the age-stage two-sex life table theory. Schneider et al. (2009) determined the side-effects of glyphosate (a herbicide) on development, fertility and demographic parameters of *C. externa* (Neuroptera: Chrysopidae) in the laboratory. They revealed that glyphosate will decrease arthropod population performance and the major detrimental

Although life table response experiments may provide the most complete data for the impacts of a pesticide on an animal subject at population-level, there are some disadvantages associated with this method. The most important one is that life table response experiments are expensive and time consuming. Construction of a life table is difficult or impossible for some species (long-lived species) when exposed to a pesticide because of the low rate of reproduction. The other major disadvantage is unrealistic conditions under which a life table is constructed. These conditions are far from the natural

Pesticides are used extensively for control of invertebrate pests, plant pathogens, weeds and rodents and other pests in a wide range of crops and for veterinary purpose. Resistant to pesticides develop in insects, mites, fungi, weeds, bacteria and rodents. Repeated applications and extensive use of the synthetic pesticides has toxicity toward natural enemies and cause resistance development in pest species against major classes of pesticides throughout the world. The repeated and extensive application of pesticides caused majority on susceptible individuals in population and only some resistant individuals survive from pesticide exposure. The offspring genotype of survival individual is homozygous or heterozygous that depends on history of pesticide application and type of pesticides. The offspring inherit the resistant genes and survival ability from the exposure to the pesticides. The surviving individuals multiply in absence of their natural enemies and finally replace the non-resistant population. The development of pesticide resistance is a Darwinian evolutionary process at a rate that rare genes conferring resistance to pesticides are selected by the high selection of pesticides. Resistance to pesticides is defined as "the development of an ability in a population of a pest to tolerate doses of pesticides that would prove lethal to the majority of individuals in a normal population of the same species" (Stenersen, 2004). The first case of resistance occurrence in insect pests was reported in 1908. This document reported the failure in the control of *Quadraspidiotus perniciosus* (Hem.: Diaspididae) by sulphur. After this report, Melander (1914) reported resistance of three scale strains in

**3.4 Age-stage two-sex life table** 

effect observed on *C. externa* was on fecundity and fertility.

**4. Resistance of pests to pesticides** 

**3.5 Drawbacks to the use of life table response experiments** 

conditions in field (Kammenga & Laskowski, 2000; Stark & banks, 2003).

United State to sulphur and sulphur–lime (as inorganic pesticide) (as cited in Stenersen, 2004). The organochlorine and synthetic insecticides were commercialized for chemical control of pests in the 1940's. The first case of DDT resistance in insect was reported in *Musca domestica* few years after introduction. After that, new insecticides such as cyclodienes, pyrethroids, organophosphates (OP), carbamates, formamidines, *Bacillus thuringiensis*, avermectins, spinosyns, insect growth regulators (IGR) and neonicotinoids were introduced for pest control and the cases of resistance to these compounds appeared a few years after their application. Now, more than 504 key pest species were resistant to pesticides and the resistance to pesticides has become a major contemporary problem in pest management programs (IRM) worldwide. Stuart (2003) reported resistance of 520 insect and acari species, 150 plant pathogen species and 273 weed species to pesticides.

Pesticides resistance reduces the ability control of pesticides on pests and leads to higher application rates to achieving satisfactory pest control. Pimentel (2003) estimated the major economic and environmental losses due to the application of pesticides on crops and veterinary purpose in the USA and showed the following costs: "public health, \$1.1 billion year-1; pesticide resistance in pests, \$1.5 billion; crop losses caused by pesticides, \$1.1 billion; bird losses due to pesticides, \$2.2 billion; and ground water contamination, \$2.0 billion" (Pimentel, 2005).

#### **4.1 Detection and monitoring of resistance**

Reduced pesticide selection pressure for each resistance mechanism is necessary for avoiding and delaying control failure prior to occurrence of resistance. For achieving this purpose, successful detection techniques are required for avoiding resistance developing and a control failure. "These techniques could be able to detect of resistant individual at low frequency in natural population" (Scott, 1995). Techniques for monitoring resistance to different pesticides in pest population gathering valuable information for insecticide resistance management (IRM) employer.

Detection and identification of resistance mechanisms to pesticides require monitoring approach with appropriate bioassay method. Monitoring of resistance is required in order to sustainable management of pesticide resistance and to know the status of resistance. Therefore, developing precision and reliable susceptibility test must be developed. These tests must be also accurate, chip, easy to perform in variety of conditions in laboratory and on farm site. So far, many standardized susceptibility test method were presented by Food Agriculture Organization (FAO) and World Health Organization (WHO) such as exposure to standard residue treatment on glass scintillation vial or filter paper, plastic bags, topical application, spray application of standard solutions, and resistance detection kits and strip.

Resistance frequencies can be detected and monitored by bioassays using diagnostic (discriminating) dose (LD99) and estimating resistance factor (Rf= LD50 of resistant population/LD50 of susceptible population). The diagnostic dose (ie. LD99) can be calculated from regression line of log dose probit-mortality data using appropriate software such as POLO-PC. This dose discriminate the tested population as susceptible and resistant and the pests that die after exposures with LD99 of pesticide are classified as susceptible and those individual that survive from exposure considered as resistant. The discriminating-dose assay is a chip, less time consuming approach for monitoring resistance in natural pest populations. These bioassay procedures provide valuable data for monitoring of resistance but this method is not practical for detection of resistance in low frequencies in field population of pests (Roush and Miller, 1986).

Ecological Impacts of Pesticides in Agricultural Ecosystem 151

mechanism. The up-regulation of a cytochrome P450 enzyme led to resistance when an insecticide is used in its toxic form on *M. domestica*. If pro-insecticide, i.e. a chemical must be converted in pest through metabolism to the active form, used against *M. doimestica*, downregulation of cytochrome P450 or other metabolizing enzymes will increase resistance (Scott,

AChE, the gamma-aminobutyric acid (GABA) receptor, Voltage-gated sodium channels, nicotinic acetylcholine receptor, octopamine receptor and the juvenile hormone (JH) receptor are known as targets of pesticides and substitution of amino acid residues in these sites led to insensitivity of structural protein toward pesticides (Kono and Tomita, 2006).

"Behavioral mechanisms, defined as evolved behaviors that reduce an insect's exposure to toxic compounds or that allow an insect to survive in what would otherwise be a toxic and fatal environment" (Sparks et al.,1989). There is a little literature on behavioural resistance mechanisms in insect due to difficulties in detection (as cited in Jensen, 2000). It seems the significance of this mechanism for resistance is less than other resistance mechanisms.

Reduced penetration of insecticide as a resistance mechanism has been studied in few insect species such as *Leptinotarsa decemlineata*. Reduced insecticide penetration via cuticle led to decrease the amount of dose in action site. The resistance ratio by this mechanism was lower than 3-fold (Scott, 1990), but because several different mechanisms are responsible for resistance to an insecticide and multiple resistance mechanisms may

Patil & Guthrie, 1979 compared the composition of the cuticular lipids of two resistant strains of *M. domestica* and their results showed that "total lipids, monoglycerides, diglycerides and sterol esters, sterols, fatty acids and phospholipid phosphorus were higher

Three methods for detecting of this mechanism include: Wash-off, diffusion cell and disk technique. In wash-off radiolabelled insecticide was topically applied to the insects and then, at fixed times after application, un-penetrated insecticide was washed off with an

Three enzyme groups involved in metabolic resistance to pesticides: esterases, glutathione S-transferases (GST) and mixed function oxidaes (MFO). The following technique can be

Esterases metabolize a variety of pesticides such as OP, carbamate, pyrethroids with ester linkages. "These enzymes confer resistance to pesticides in over 50 species of insects, ticks and mitesa" (Devorshak and Roe 1998; as cited in van Leeuwen et al., 2009). Detection and investigation of esterases-based mechanism can be achieved from synergistic bioassays and biochemical assays. For synergistic bioassays, some synergists such as DEF (S,S,Stributylphosphorotrithioate), TPP (O,O,O-triphenylphosphate), and IBP (O,O-bis[1-

**4.2.1.3 Structural change in insecticide- target molecules** 

co‐exist in an insect and act either additively or synergistically.

appropriate solvent and quantified (as cited in Jensen, 2000).

in resistant strains than in the susceptible strain".

1995).

**4.2.2 Behavioral resistance** 

**4.2.3 Reduced penetration** 

**4.2.4 Metabolism of toxicants** 

**4.2.4.1 Esterase** 

used for detection of these mechanisms.

The mechanisms of resistance are behavioral, reduced penetration, metabolism of toxicant to inactive product and target site insensitivity. These mechanisms can be detected using biochemical assay techniques (spectrophotometric and fluorometric methods) and molecular assays (base on DNA diagnostic) in one individual or small number of insect. Identification of resistance mechanisms is critical for determining of the cross resistance spectrum (Brogdon and McAllister, 1998).

"Molecular methods and traditional assays (ie. bioassay) used for distinguish heterozygotes (SR), homozygous susceptible (SS) and homozygous resistant (RR) genotypes" (Scott, 1995). The environmental conditions such as temperatures, humidity, pH and light increase errors in biochemical and bioassay results but these conditions can not affect the results of molecular methods (Scott, 1995). Now, PCR-based techniques have been designed for field detection of modified acetylcholinesterase (AChE) and *knock down (Kdr)* in individual *Myzus persicae* (Field et al., 1996). "The amplified E4 or FE4 genes can be identified by restriction enzyme analysis or polymerase chain reaction (PCR)-based methods" (Field et al., 1996).

#### **4.2 Mechanisms of resistance to pesticides**

Biochemical and molecular basis of resistance mechanisms to pesticides in insects, acari, fungi, bacteria, weeds and vertebrate pests are similar. An exhaustive knowledge on biochemical and molecular resistance mechanisms in pests are useful for designing insecticide resistance management (IRM) strategies. Also, identification of resistance mechanisms is necessary for developing discriminating techniques for detecting and monitoring resistance genes and cross resistance spectrum in the field populations of pests (Hammock and Soderlund, 1986). The factors affecting pesticides effectiveness were distinguished in two classes: The first class decreases the amount of pesticide dose in action site including behavioural resistance, reduced penetration or adsorption, sequestration and detoxification. The second class is decreased target site sensitivity to pesticides that reduce the affinity of target protein toward activated pesticide (van leeuwen, et al., 2009). "In practice, probably more than 90% of all resistance cases in insects and mites are caused by a less sensitive target site and/or an enhanced pesticide detoxification" (Roush and Tabashnik, 1990 as cited in van leeuwen, et al., 2009). The relative importance of these mechanisms depends on pest species and history of chemical application.

#### **4.2.1 Genetic mechanisms**

Genetic mechanisms of pesticide resistance involve some point mutations in genes and their over expression. These mechanisms were elucidated as follow:

#### **4.2.1.1 Gene amplification**

Devonshire and Moores, 1982 showed that the gene amplification of one of two closely related carboxylesterases (E4 and FE4) in *M. persicae* were associated with resistance to OP, carbamates and pyrethroids. Carboxylesterases sequester or degrade carbamate and OP insecticides before they reach to AChE in the nervous system. E4 and EF4 overproduction in resistant strains of *M. persicae* is due to amplification of structural genes encoding these enzymes (Field et al., 1988).

#### **4.2.1.2 Up- and down-regulation**

The research showed that cytochrome P450 enzyme were over expressed in some resistance strain of *M. domestica* through the increase of gene transcription by up-regulated mechanism. The up-regulation of a cytochrome P450 enzyme led to resistance when an insecticide is used in its toxic form on *M. domestica*. If pro-insecticide, i.e. a chemical must be converted in pest through metabolism to the active form, used against *M. doimestica*, downregulation of cytochrome P450 or other metabolizing enzymes will increase resistance (Scott, 1995).

#### **4.2.1.3 Structural change in insecticide- target molecules**

AChE, the gamma-aminobutyric acid (GABA) receptor, Voltage-gated sodium channels, nicotinic acetylcholine receptor, octopamine receptor and the juvenile hormone (JH) receptor are known as targets of pesticides and substitution of amino acid residues in these sites led to insensitivity of structural protein toward pesticides (Kono and Tomita, 2006).
