**2. Description of the Ottawa River watershed**

The Ottawa River, which is the main tributary of the St. Lawrence River, takes its source in Lac Capimitchigama (**Figures 1** and **2**). Stretching over approximately 1130 km, the Ottawa drains a 146,334 km<sup>2</sup> watershed. From a geological standpoint, the river flows mainly through the Canadian Shield, which comprises Archean and/or Proterozoic igneous rocks, as well as Proterozoic metasedimentary and intrusive rocks. Upstream of its confluence with the St. Lawrence River, the Ottawa River flows through the relatively flat St. Lawrence Lowlands, comprising carbonate and siliciclastic sedimentary rocks. Climate in the watershed is cool continental temperate, characterized by very cold and snowy winters and warm and relatively dry summers. Temperatures and precipitations decrease from south to north in the watershed. From a hydrographic standpoint, 19 main tributaries flow into the Ottawa River, of which the primary ones are the Gatineau, Lièvre, Kipawa, Rouge, Madawaska, Montreal, Blanche, and Petawawa. The Ottawa River watershed also includes over 90,000 small and large lakes.

**Figure 1.** Location of the Ottawa River watershed and the St. Lawrence River.

Seasonal Variation of the Physico-chemical Composition of Ottawa River Waters in the St. Lawrence River

http://dx.doi.org/10.5772/intechopen.74122

5

**Figure 2.** Location of sampling stations along the St. Lawrence River.

The Ottawa River and most of its tributaries are heavily regulated, the watershed comprising more than 1000 small and large dams, in addition to 30 large reservoirs built to control flood flows. Reservoirs built in the upper reaches of the watershed have inverted the annual cycle of flows such that maximum flows occur in winter and minimum flows in springtime during snowmelt, contrary to the annual flow regime in natural rivers. This inversion, however, fades gradually in the lower reaches of the watershed due to input from natural tributaries (**Figure 3**). Annual mean discharge in the Ottawa River at its confluence with the St. Lawrence River is roughly 1980 m<sup>3</sup> /s [12], and the watershed is almost completely covered by forests Seasonal Variation of the Physico-chemical Composition of Ottawa River Waters in the St. Lawrence River http://dx.doi.org/10.5772/intechopen.74122 5

**Figure 1.** Location of the Ottawa River watershed and the St. Lawrence River.

and areas of agriculture, pasture, forests, and wetland in the mid- and lower reaches of the stream system [2, 3]. Water intrusions from tributaries contribute to the formation of several parallel water masses with distinct physical and chemical properties. Among these, the Ottawa River plays a significant role in structuring the biogeochemical properties of the brown water

Many studies have analyzed the physicochemical and biological characteristics of these waters (e.g., [5–24]) and of related sediments [25–28], while other studies focused on optical characterization of these waters (e.g., [27–32]). Most of these studies analyzed the spatial variability of these characteristics in the St. Lawrence River, but very few looked at their seasonal and interannual variability. One notable exception [12] compared the interannual variability of these characteristics measured upstream and downstream from the confluence of the Ottawa and St. Lawrence Rivers from May through September, from 1994 to 1996. However, the changes in physicochemical characteristics of Ottawa River waters flowing through the St. Lawrence River were not specifically studied at the seasonal or decadal level. The main goal of this study is to analyze the seasonal variability in physical and chemical properties of the Ottawa River water mass flowing in the St. Lawrence River along 80 km further downstream from the confluence. Water characteristics were measured at four stations in the spring (May), summer (August), and fall (October) of 2006, something that has never been analyzed. The secondary goal of the

river considering its strong discharge rate and largely human impacted watershed [4].

study is to compare these characteristics with those measured 10 years earlier by [12].

The Ottawa River, which is the main tributary of the St. Lawrence River, takes its source in Lac Capimitchigama (**Figures 1** and **2**). Stretching over approximately 1130 km, the Ottawa

through the Canadian Shield, which comprises Archean and/or Proterozoic igneous rocks, as well as Proterozoic metasedimentary and intrusive rocks. Upstream of its confluence with the St. Lawrence River, the Ottawa River flows through the relatively flat St. Lawrence Lowlands, comprising carbonate and siliciclastic sedimentary rocks. Climate in the watershed is cool continental temperate, characterized by very cold and snowy winters and warm and relatively dry summers. Temperatures and precipitations decrease from south to north in the watershed. From a hydrographic standpoint, 19 main tributaries flow into the Ottawa River, of which the primary ones are the Gatineau, Lièvre, Kipawa, Rouge, Madawaska, Montreal, Blanche, and Petawawa. The Ottawa River watershed also includes over 90,000

The Ottawa River and most of its tributaries are heavily regulated, the watershed comprising more than 1000 small and large dams, in addition to 30 large reservoirs built to control flood flows. Reservoirs built in the upper reaches of the watershed have inverted the annual cycle of flows such that maximum flows occur in winter and minimum flows in springtime during snowmelt, contrary to the annual flow regime in natural rivers. This inversion, however, fades gradually in the lower reaches of the watershed due to input from natural tributaries (**Figure 3**). Annual mean discharge in the Ottawa River at its confluence with the St. Lawrence

watershed. From a geological standpoint, the river flows mainly

/s [12], and the watershed is almost completely covered by forests

**2. Description of the Ottawa River watershed**

4 Achievements and Challenges of Integrated River Basin Management

drains a 146,334 km<sup>2</sup>

small and large lakes.

River is roughly 1980 m<sup>3</sup>

**Figure 2.** Location of sampling stations along the St. Lawrence River.

**Figure 3.** Monthly flow coefficients downstream from the Dozois reservoir (blue bars, 8210 km<sup>2</sup> ) and the carillon dam (red bars, 143,000 km<sup>2</sup> ) built on the main branch of the Ottawa River.

(deciduous, mixed, and boreal). Farming is only practiced in the lower part of the watershed and accounts for only 3% of its total surface area. The main urban areas in the watershed are the Ottawa-Gatineau and Laval areas.

#### **3. Analysis of the chemical and physical variables of waters**

Three 8-day sampling cruises were conducted in the St. Lawrence River (SLR) during spring (23–30 May), summer (9–15 August), and fall (11–17 October) 2006 aboard the RV "Lampsilis" from the Université du Québec à Trois-Rivières. We studied the SLR along a 450 km distance from its source at the outlet of the Great Lakes, until the interface with marine waters at the estuarine transition zone (ETZ), 50 km downstream from the marine intrusion (**Figure 1**). Water samples were collected at the surface (0.5–1.3 m) for all stations using a Go-Flow bottle (8 L) and immediately processed in the wet laboratory after collection.

As far as *a*CDOM measurements, water samples for the absorption coefficient of chromophoric dissolved organic matter (*a*CDOM) and DOC were filtered through Milli-Q-rinsed 0.22 μm Isopore membrane (millipore) and stored them in the dark at 4°C until analysis. We measured CDOM absorption spectra in a 10 mm quartz cell at 1 nm intervals between 190 and 900 nm using a spectrophotometer (Shimadzu UV-2401PC) referenced against Milli-Q water. We used absorbance at 690 nm (where the temperature dependency is near zero) to correct the UV absorption values. We converted absorbance values at 340 nm to absorption coefficients

(M1= M2): mean values are not significantly different at the 5% level (all statistical tests used). (0.78) = standard deviation.

**Table 1.** Comparison of seasonal mean values of physicochemical variables measured at four stations in St. Lawrence

**Variables May (M1) August (M2) October (M3) Results of comparison of** 

Seasonal Variation of the Physico-chemical Composition of Ottawa River Waters in the St. Lawrence River

, mg/L) 9.19 (2.93)\* 0.008 (0.003) 1.79 (0.53)\* M1 ≠ (M2 = M3)

, mg/L) 0.53 (0.43)\* 0.73 (0.21)\* 0.55 (0.40)\* M2 ≠ (M1 = M3)

, μg/L) 5.24 (1.03) 10.29 (1.89) 2.20 (1.99) M2 ≠ (M1 = M2)

2.46 (0.38) 1.81 (0.23) 0.97 (0.103) (M1 = M2 = M3)

Temperature (°C) 12.9 (0.78) 23.3 (0.21) 13.9 (0.17) M1 ≠ M2 ≠ M3 Total nitrogen (TN, mg/L) 4.52 (2.47)\* 3.05 (0.70)\* 0.70 (0.33)\* M1 ≠ M2 ≠ M3

Total phosphorous (TP, μg/L) 61.60 (30.89)\* 23.47 (12.21) 8.65 (3.17) M1 ≠ (M2 = M3) *aCDOM340nm* (m−1) 15.79 (5.75) 7.92 (1.65) 3.11 (1.35) M1 ≠ (M2 = M3)

Transmittance (trans, %) 24.43 (19.19) 63.28 (8.88) 79.04 (1.11) M2 ≠ (M1 = M3) Conductivity (cond, μS/cm) 0.062 (0.019) 0.072 (0.009) 269.0 (4.65) M3 ≠ (M1 = M2)

Turbidity (TURB, NTU) 9.41 (4.13)\* 9.39 (6.80)\* 4.47 (1.37)\* (M1 = M2 = M3)

(M1≠ M2≠ M3): mean values are significantly different at the 5% level (all statistical tests used).

**mean values**

http://dx.doi.org/10.5772/intechopen.74122

7

As far as spectral radiation and beam attenuation measurements are concerned, photosynthetically available radiation (PAR) (400–700 nm) in the water column was measured at each station as in [1]. Briefly, downward irradiance was measured at every 0.02 m with a spectroradiometer (Model Hyperpro, Satlantic Instruments), which was slowly lowered through the water column to measure depth profiles of the cosine-corrected downwelling under-

were corrected automatically for "dark irradiance" values obtained from the shutter darks.

) at every 3 nm between 351 and 750 nm (100 wavebands). Light data

(1)

(*a*CDOM340nm) using the following equation [8]:

The mean concentration exceeds the provincial standard limit value.

River water influences by the Ottawa River in 2006.

where *L* is the cuvette path length (0.01 m).

water irradiance (*Ed*

Nitrite (NO<sup>2</sup>

Nitrate (NO<sup>3</sup>

((*K*d(PAR)) m−1)

\*

Phosphate (PO4

Light extinction coefficient

Sampling was carried out at four stations in the water mass entering from the Ottawa River along 80 km downstream transect (**Figure 2**). At each site, water was subsampled directly in acid-washed bottles for total phosphorus (TP) and total nitrogen (TN) measurements. For soluble reactive phosphorous (PO4 ) and for nitrites (NO<sup>2</sup> ) and nitrates (NO<sup>3</sup> ), samples were filtrated on 45 mm diameter, 0.7 μm poresize GFF filters (Millipore). PO4 was analysed using the acid molybdate technique. NO<sup>3</sup> was first reduced into NO<sup>2</sup> by cadmium, and the nitrite concentrations were determined by the sulfanilamide method.). TP and TN concentrations (check) were obtained using the spectrophotometric determination of phosphates and nitrates after digestion by potassium persulfate. All phosphorus and nitrogen analyses were performed according to the American Public Health Association protocols [33]. We used a multiprobe depth profiler (YSI, model 6600EDS-M, Yellowspring Inc.) to measure the conductivity, temperature, and turbidity of the water column. Values for the surface of the water column were averaged between 0.5 and 1.5 m. Physicochemical variables or characteristics of St. Lawrence River waters analyzed as part of this study are presented in **Table 1**.

Seasonal Variation of the Physico-chemical Composition of Ottawa River Waters in the St. Lawrence River http://dx.doi.org/10.5772/intechopen.74122 7


(M1≠ M2≠ M3): mean values are significantly different at the 5% level (all statistical tests used).

(M1= M2): mean values are not significantly different at the 5% level (all statistical tests used). (0.78) = standard deviation. \* The mean concentration exceeds the provincial standard limit value.

**Table 1.** Comparison of seasonal mean values of physicochemical variables measured at four stations in St. Lawrence River water influences by the Ottawa River in 2006.

As far as *a*CDOM measurements, water samples for the absorption coefficient of chromophoric dissolved organic matter (*a*CDOM) and DOC were filtered through Milli-Q-rinsed 0.22 μm Isopore membrane (millipore) and stored them in the dark at 4°C until analysis. We measured CDOM absorption spectra in a 10 mm quartz cell at 1 nm intervals between 190 and 900 nm using a spectrophotometer (Shimadzu UV-2401PC) referenced against Milli-Q water. We used absorbance at 690 nm (where the temperature dependency is near zero) to correct the UV absorption values. We converted absorbance values at 340 nm to absorption coefficients (*a*CDOM340nm) using the following equation [8]:

$$aCDOM\_{340sm} = \frac{2.303A\_{340sm}}{L} \tag{1}$$

where *L* is the cuvette path length (0.01 m).

(deciduous, mixed, and boreal). Farming is only practiced in the lower part of the watershed and accounts for only 3% of its total surface area. The main urban areas in the watershed are

Three 8-day sampling cruises were conducted in the St. Lawrence River (SLR) during spring (23–30 May), summer (9–15 August), and fall (11–17 October) 2006 aboard the RV "Lampsilis" from the Université du Québec à Trois-Rivières. We studied the SLR along a 450 km distance from its source at the outlet of the Great Lakes, until the interface with marine waters at the estuarine transition zone (ETZ), 50 km downstream from the marine intrusion (**Figure 1**). Water samples were collected at the surface (0.5–1.3 m) for all stations using a Go-Flow bottle (8 L) and immediately processed in the wet laboratory after

Sampling was carried out at four stations in the water mass entering from the Ottawa River along 80 km downstream transect (**Figure 2**). At each site, water was subsampled directly in acid-washed bottles for total phosphorus (TP) and total nitrogen (TN) measurements. For solu-

tions were determined by the sulfanilamide method.). TP and TN concentrations (check) were obtained using the spectrophotometric determination of phosphates and nitrates after digestion by potassium persulfate. All phosphorus and nitrogen analyses were performed according to the American Public Health Association protocols [33]. We used a multiprobe depth profiler (YSI, model 6600EDS-M, Yellowspring Inc.) to measure the conductivity, temperature, and turbidity of the water column. Values for the surface of the water column were averaged between 0.5 and 1.5 m. Physicochemical variables or characteristics of St. Lawrence River waters analyzed as part

) and nitrates (NO<sup>3</sup>

), samples were filtrated

) and the carillon dam

was analysed using the acid

by cadmium, and the nitrite concentra-

) and for nitrites (NO<sup>2</sup>

was first reduced into NO<sup>2</sup>

on 45 mm diameter, 0.7 μm poresize GFF filters (Millipore). PO4

**3. Analysis of the chemical and physical variables of waters**

**Figure 3.** Monthly flow coefficients downstream from the Dozois reservoir (blue bars, 8210 km<sup>2</sup>

) built on the main branch of the Ottawa River.

the Ottawa-Gatineau and Laval areas.

6 Achievements and Challenges of Integrated River Basin Management

(red bars, 143,000 km<sup>2</sup>

collection.

ble reactive phosphorous (PO4

of this study are presented in **Table 1**.

molybdate technique. NO<sup>3</sup>

As far as spectral radiation and beam attenuation measurements are concerned, photosynthetically available radiation (PAR) (400–700 nm) in the water column was measured at each station as in [1]. Briefly, downward irradiance was measured at every 0.02 m with a spectroradiometer (Model Hyperpro, Satlantic Instruments), which was slowly lowered through the water column to measure depth profiles of the cosine-corrected downwelling underwater irradiance (*Ed* ) at every 3 nm between 351 and 750 nm (100 wavebands). Light data were corrected automatically for "dark irradiance" values obtained from the shutter darks. Diffuse attenuation coefficients (*K*d(PAR)) were calculated by linear regression of the natural logarithm of *Ed* versus depth. *Ed* values correspond to PAR. The Hyperpro was equipped with a C-star transmissometer (Wet Labs Inc., 25 cm path length, λ = 660 nm) to measure depth profiles of the scattering of underwater particles (trans) such as sediments.

(**Table 1** and **Figure 5**). As far as temperature is concerned, it is higher in August (summer) than in May (influence of snowmelt water) and October (effect of fall cooling). In summer, water temperature is roughly twice as high as in the spring or fall due to low water levels (low flow) and the increase in solar energy. TN, for its part, which is mainly derived from farming in Quebec, decreases from spring to fall. In springtime, there is widespread runoff on slopes due to snowmelt, which accounts for the increase in TN concentration in rivers. This concentration decreases in summer as runoff decreases. However, because of the relatively low water levels, the total nitrogen concentration remains higher than in the fall due to limited dilution. In any case, mean TN concentrations during the three seasons are

Seasonal Variation of the Physico-chemical Composition of Ottawa River Waters in the St. Lawrence River

http://dx.doi.org/10.5772/intechopen.74122

9

higher than the provincial standard limit value (0.5 mg/L).

**Figure 5.** Comparison of seasonal mean values of temperature and TN.

**Figure 6.** Comparison of seasonal mean values of NO<sup>2</sup>

, TP and *<sup>a</sup>*

CDOM340nm.

Statistical analysis consisted of comparing seasonal mean values of physicochemical variables measured at the four stations using the analysis of variance approach when the data were normal and the Kruskal-Wallis test when the data were not. The same statistical tests were used to compare mean values of certain characteristics at the decadal scale and those of seasonal water levels. Water level data for the St. Lawrence River, taken from the Environment Canada website (https://eau.ec.gc.ca/download/index\_f.html?results\_type=historical, viewed on September 20, 2017) and measured at the Lanoraie station (ID: 02OB011; 45°57′33" N, 73°15′52"W) since 1990, are strongly influenced by water masses entering from the Ottawa River.
