Export and retention of dissolved inorganic nutrients in the Cachoeira ...
J. Limnol., 69(1): 138-145, 2010 DOI: 10.3274/JL10-69-1-13
Export and retention of dissolved inorganic nutrients in the Cachoeira River, Ilhéus, Bahia, Brazil Michelle C. LIMA1), Marcelo F.L. SOUZA1,2)*, Gilmara F. EÇA1,2) and Maria A.M. SILVA1,2) 1)
Laboratório de Oceanografia Química, Universidade Estadual de Santa Cruz, Departamento de Ciências Exatas e Tecnológicas, Rod. Ilhéus/Itabuna, km 16, CEP: 45.650-000; Ilhéus – Bahia – Brasil. 2) Programa de Pós-Graduação em Sistemas Aquáticos Tropicais, Universidade Estadual de Santa Cruz *e-mail corresponding author: [email protected]
ABSTRACT Dissolved inorganic nutrient concentrations and physical-chemical variables were determined in the lower reaches of the Cachoeira River watershed, from November 2003 to October 2004. Concentration of nutrients were high and highly variable. Mean concentrations and standard deviation of ammonium, nitrite, nitrate, phosphate and silicate were 25.4 ± 25.1; 3.9 ± 3.9; 62.2 ± 54.9; 15.8 ± 9.0 and 129.0 ± 5.6 (µmol L-1), respectively. Nutrient retention was observed mainly during the dry season. Chlorophyll-a concentrations were especially high in those periods. The Cachoeira River can be considered eutrophicated, and such condition becomes more intense with low fluvial flow during the dry months. Despite the spatial/temporal changes of the species of inorganic nitrogen, a removal of dissolved inorganic nitrogen was observed in relation to dissolved silicon and to phosphorus, with consequences for estuarine biogeochemistry. The basin exports annually about 3.5, 2.2 and 0.3 t y-1 of dissolved silicon, nitrogen, and phosphate to the estuary, respectively. The eutrophication and growth of macrophytes is responsible for most of these changes in nutrient fluxes to the estuary and coastal waters. Key words: eutrophication, macrophytes, nitrogen, phosphorus, silicon, fluvial inputs
1. INTRODUCTION Anthropogenic activities over drainage basins have been continuously increasing the point and diffuse sources of nitrogen and phosphorus to rivers around the world. Such activities include intensified use of fertilizers in crops, and deposition of domestic and industrial effluents. Determination of dissolved inorganic nutrients (such as nitrogen, phosphorus and silicon) has been used in the evaluation of eutrophication in aquatic ecosystems, since these substances are closely related to the system´s degree of pollution (Carmouze 1994; O'Donohue & Denisson 1997). Eutrophication implies changes of biogeochemical cycling of aquatic ecosystems, leading to several ecological consequences, such as increase of the biomass of certain kinds of algae and aquatic macrophytes. The occurrence of such flourishing harms the water quality and may thus reduce concentration of dissolved oxygen, changing the competitive balance among species, which results in loss of biodiversity (Cooper et al. 2002). In addition to the increase of algae density there are also other significant changes, such as the appearance of new species and the disappearance of others (O'Donohue & Denisson 1997). The growth of aquatic macrophytes may also influence the transportation of nutrients, favoring their retention and providing an additional substratum for peryphiton and its biochemical transformations (Kronvang et al. 1999; Peterson et al. 2001; Merriam et al. 2002).
The nutrient loading by the rivers influences the biotic activity in estuaries and coastal seas and may be an important indicator of changes in the condition of such waters (Sawidis & Tsekos 2004). The fluvial discharge also influences the capacity to dilute the excessive amount of nutrients and pollutants that reach the estuaries, minimizing the impact in these areas. Otherwise, a short residence time may not allow maximum removal through absorption, as usually occurs when there is low water flow (Sanders et al. 1997). The purpose of this study was to quantify nutrient (ammonium, nitrate, nitrite, phosphate and silicate) retention at the lower basin of Cachoeira River and its export to the estuary. 2. METHODS The hydrographic basin of the Cachoeira River covers a total area of around 4600 km2 in the south of Bahia State, Northeastern Brazil (BAHIA 2001; Fig. 1). The river crosses the cities of Itabuna and Ilhéus, from which receives input of domestic and industrial effluents. Average regional temperature is 24.6 ºC; precipitation is 1500 mm y-1 in Itabuna and 2000 mm y-1 in Ilhéus. The Atlantic Forest and shaded cacao crops (cabrucas) cover large areas of the river basin. This kind of culture has prevented soil erosion; however, the increase of coffee plantations and pasture has intensified this process (Klumpp et al. 2002). The Cachoeira River is characterized by extreme irregularity, with ill-defined dry and rainy seasons. Annual average pluvial discharge
Fluvial export and retention of nutrients
139
Fig. 1. Map with the localization of the sampling points in the Cachoeira River Basin, Bahia, Brazil: RI (Itabuna Reservoir); RCU (Cachoeira River near UESC, Universidade Estadual de Santa Cruz); VC (Cachoeira Village); BV (Vitoria Bank).
is 24.1 m3 s-1, although 0.2 and 1.460 m3 s-1 have already been recorded. During the dry season its rocky bed is mostly exposed, remaining only small residual streams of water that are completely covered by floating macrophytes, mainly Eicchornia crassipes (BAHIA 2001). A small clear overflow weir in Itabuna (~1.5 m height) increases stagnation and growth of macrophytes during the dry season. Plants are transported towards the estuary during rainy season, and large amounts can reach coastal waters and beaches 14 km southward during peak discharges (Souza 2005). Sampling was carried out between 2003 (November and December) and 2004 (January, February, March, April, August and October) in four stations (Fig. 1) between the small dam on Itabuna city (RI) and the upper estuary limit (BV). Sampling was also done about 5 km downstream the dam, after a stretch of rapids (RCU) and after a natural 6 m depth reservoir created by geological faults (VC). One sample was collected in each station, from the upstream station till the upper estuary boundary, comprising a total of 30 samples. Physical-chemical parameters such as pH, temperature, conductivity and dissolved oxygen (DO) were determined in the field with a WTW Multiline P4 and a Hanna HI 9143 portable meters, respectively. Samples were collected in polyethylene bottles previously washed with HCl 1:1 and distilled water, and kept under refrigeration during transportation. In the laboratory, samples were filtered in GF/C type fiberglass filters used further to chlorophyll-a analysis by the spectrophotometric trichromatic method on a 80% acetone extract (Parson et al. 1984). Aliquots of filtered samples
were frozen until analysis of dissolved inorganic nutrients. Concentration of nutrients was determined by spectrophotometric methods according Grasshoff et al. (1983) and dissolved silicate (orthosilicate) as described in Carmouze (1994). Nitrite was determined using the diazotation method (detection limit 0.02 μM); nitrate was determined by cadmium reduction into nitrite (nitrite limit plus a coefficient of variation ±3% in the range 0-10 μM); ammonium was measured spectrophotometrically by the indophenol blue method (detection limit 0.05 μM); orthophosphate by the ascorbic acid molybdate method (detection limit 0.01 μM); dissolved silicate was determined by the ammonium molybdate method (detection limit 0.1 μM). Daily fluvial discharge data (Q, m3 s-1) for a station located in Itabuna was obtained at ANA Hidroweb (Agencia Nacional de Águas, National Water Agency; www.hidroweb.ana.gov.br/hidroweb, station no. 53180000). This station is located upstream from our sampling stations, and the total increment of the area drained by the station is about 220 km2. A correction of the total monthly outflow for each drained area of the basin (QSt) was done according to (equation 1): QSt =(QANA × ASt)/AANA
(1
where QSt and ASt respectively mean total monthly fluvial outflow and drainage area of each station; and QANA and AANA represent fluvial discharge and drainage area at the ANA gauging station, respectively. Flux of dissolved inorganic nutrients was obtained by multiplying concentration values of each dissolved inorganic nutrient in the sample by the calculated fluvial
140
M. Lima et al. Tab. 1. Mean values of electrical conductivity, temperature, pH, dissolved oxygen (DO) concentration and saturation percentage in the Cachoeira River stations. Mean ± standard deviation; minimum and maximum in parenthesis. N = 4 except in * (N = 3) and ** (N = 1).
November 2003 December 2003 January 2004 February 2004 March 2004 April 2004 August 2004 October 2004
Conductivity (µS cm-1)
Temperature (°C)
pH
DO (mg L-1)
DO (% saturation)
0.553 ± 0.036 (0.505 – 0.582) 0.572 ± 0.024 (0.555 – 0.606) 0.603 ± 0.024 (0.585 – 0.638) 0.722**
29.5 ± 0.1 (29.4 – 29.7) 29.3 ± 0.8 (28.5 – 30.4) 27.9 ± 0.3 (22.4 – 28.2) 26.9**
7.3 ± 0.1 (7.2 – 7.4) 7.6 ± 0.5 (7.2 – 8.3) 7.2 ± 0.1 (7.1 – 7.3) 8.1**
4.1 ± 1.5 (2.3 – 5.6) 12.2 ± 13.4 (2.7 – 31.2) 2.3 ± 1.2 (1.2 – 3.7) -
54.2 ± 19.5 (30.2 - 73.1) 158.8 ± 172.4 (35.0 - 402.2) 29.4 ± 15.7 (15.4 - 47.3) -
0.208 ± 0.015 (0.196 – 0.228) 0.433 ± 0.039 * (0.389 – 0.462) 0.570 ± 0.042 (0.531 – 0.609) 0.609 ± 0.015 (0.595 – 0.630)
25.5 ± 0.3 (25.1 – 25.8) 26.2 ± 0.8 (25.3 – 27.0) 21.3 ± 3.1 (17.1 – 24.7) 23.8 ± 0.6 * (23.4 – 24.5)
7.2 ± 0.2 (7.0 – 7.5) 7.8 ± 0.2 (7.6 – 7.9) 7.9 ± 1.2 (7.1 – 9.7) 7.0 ± 0.3 (6.8 – 7.4)
7.2 ± 0.2 (6.9 – 7.4) 6.7 ± 0.5 (6.3 – 7.4) 6.2 **
88.0 ± 2.6 (84.5 - 90.1) 82.7 ± 5.6 (78.0 - 90.4) 75.1**
-
-
discharge values of each station on monthly basis. For nutrient concentration (Y) below detection limit, a value immediately below was used for monthly flow estimation. Fluxes were normalized dividing them by the area of each drainage basin, according to equation 2: FNY = FY / ASt
(2
where FNY represents normalized flow of dissolved inorganic nutrients, FY represents actual flow of nutrients and ASt. represents drainage area of each sampling station. We assumed that changes of nutrient concentrations in the river water during time scales lower than a month became insignificant when the actual concentration of the sample was multiplied by total monthly discharge values higher by 5-8 magnitude order, when fluxes were significant.
Fig. 2. Daily runoff at the ANA (Agência Nacional de Águas; National Water Agency) station No.53180000 during the study period.
3. RESULTS 3.1. Fluvial discharge
3.3. Dissolved inorganic nutrients
Runoff ranged from 0 to 826 m3 s-1 during the period studied (Fig. 2). Sampling during the dry season were represented by November and December 2003, and October 2004, with an average flow under 4 × 106 m3 month-1. December was the driest month, with the rainy season starting in January 2004. Maximum fluvial discharge occurred in March 2004.
The highest concentrations of ammoniacal nitrogen and phosphate were observed at station RI; nitrite and nitrate reached maximum concentrations at station RCU. Silicate presented small changes along sampling stations, without a marked spatial pattern (Fig. 3e). Maximum concentrations of ammoniacal nitrogen (Fig. 3a) and phosphate (Fig. 3d) occurred in January 2004 (beginning of the rainy season) while minimum values were observed in March 2004 (rainy season). For nitrite and nitrate (Figs 3b and 3c, respectively), the lowest concentrations were observed in November 2003 and March 2004 (rainy season). There was a decrease of the mean concentration of silicate (Fig. 3e) during January, February and March of 2004.
3.2. Physico-chemical variables Physico-chemical variables are shown in table 1. The conductivity varied from 722 μS cm-1 when the runoff begins to increase to
Export and retention of dissolved inorganic nutrients in the Cachoeira River, Ilhéus, Bahia, Brazil Michelle C. LIMA1), Marcelo F.L. SOUZA1,2)*, Gilmara F. EÇA1,2) and Maria A.M. SILVA1,2) 1)
Laboratório de Oceanografia Química, Universidade Estadual de Santa Cruz, Departamento de Ciências Exatas e Tecnológicas, Rod. Ilhéus/Itabuna, km 16, CEP: 45.650-000; Ilhéus – Bahia – Brasil. 2) Programa de Pós-Graduação em Sistemas Aquáticos Tropicais, Universidade Estadual de Santa Cruz *e-mail corresponding author: [email protected]
ABSTRACT Dissolved inorganic nutrient concentrations and physical-chemical variables were determined in the lower reaches of the Cachoeira River watershed, from November 2003 to October 2004. Concentration of nutrients were high and highly variable. Mean concentrations and standard deviation of ammonium, nitrite, nitrate, phosphate and silicate were 25.4 ± 25.1; 3.9 ± 3.9; 62.2 ± 54.9; 15.8 ± 9.0 and 129.0 ± 5.6 (µmol L-1), respectively. Nutrient retention was observed mainly during the dry season. Chlorophyll-a concentrations were especially high in those periods. The Cachoeira River can be considered eutrophicated, and such condition becomes more intense with low fluvial flow during the dry months. Despite the spatial/temporal changes of the species of inorganic nitrogen, a removal of dissolved inorganic nitrogen was observed in relation to dissolved silicon and to phosphorus, with consequences for estuarine biogeochemistry. The basin exports annually about 3.5, 2.2 and 0.3 t y-1 of dissolved silicon, nitrogen, and phosphate to the estuary, respectively. The eutrophication and growth of macrophytes is responsible for most of these changes in nutrient fluxes to the estuary and coastal waters. Key words: eutrophication, macrophytes, nitrogen, phosphorus, silicon, fluvial inputs
1. INTRODUCTION Anthropogenic activities over drainage basins have been continuously increasing the point and diffuse sources of nitrogen and phosphorus to rivers around the world. Such activities include intensified use of fertilizers in crops, and deposition of domestic and industrial effluents. Determination of dissolved inorganic nutrients (such as nitrogen, phosphorus and silicon) has been used in the evaluation of eutrophication in aquatic ecosystems, since these substances are closely related to the system´s degree of pollution (Carmouze 1994; O'Donohue & Denisson 1997). Eutrophication implies changes of biogeochemical cycling of aquatic ecosystems, leading to several ecological consequences, such as increase of the biomass of certain kinds of algae and aquatic macrophytes. The occurrence of such flourishing harms the water quality and may thus reduce concentration of dissolved oxygen, changing the competitive balance among species, which results in loss of biodiversity (Cooper et al. 2002). In addition to the increase of algae density there are also other significant changes, such as the appearance of new species and the disappearance of others (O'Donohue & Denisson 1997). The growth of aquatic macrophytes may also influence the transportation of nutrients, favoring their retention and providing an additional substratum for peryphiton and its biochemical transformations (Kronvang et al. 1999; Peterson et al. 2001; Merriam et al. 2002).
The nutrient loading by the rivers influences the biotic activity in estuaries and coastal seas and may be an important indicator of changes in the condition of such waters (Sawidis & Tsekos 2004). The fluvial discharge also influences the capacity to dilute the excessive amount of nutrients and pollutants that reach the estuaries, minimizing the impact in these areas. Otherwise, a short residence time may not allow maximum removal through absorption, as usually occurs when there is low water flow (Sanders et al. 1997). The purpose of this study was to quantify nutrient (ammonium, nitrate, nitrite, phosphate and silicate) retention at the lower basin of Cachoeira River and its export to the estuary. 2. METHODS The hydrographic basin of the Cachoeira River covers a total area of around 4600 km2 in the south of Bahia State, Northeastern Brazil (BAHIA 2001; Fig. 1). The river crosses the cities of Itabuna and Ilhéus, from which receives input of domestic and industrial effluents. Average regional temperature is 24.6 ºC; precipitation is 1500 mm y-1 in Itabuna and 2000 mm y-1 in Ilhéus. The Atlantic Forest and shaded cacao crops (cabrucas) cover large areas of the river basin. This kind of culture has prevented soil erosion; however, the increase of coffee plantations and pasture has intensified this process (Klumpp et al. 2002). The Cachoeira River is characterized by extreme irregularity, with ill-defined dry and rainy seasons. Annual average pluvial discharge
Fluvial export and retention of nutrients
139
Fig. 1. Map with the localization of the sampling points in the Cachoeira River Basin, Bahia, Brazil: RI (Itabuna Reservoir); RCU (Cachoeira River near UESC, Universidade Estadual de Santa Cruz); VC (Cachoeira Village); BV (Vitoria Bank).
is 24.1 m3 s-1, although 0.2 and 1.460 m3 s-1 have already been recorded. During the dry season its rocky bed is mostly exposed, remaining only small residual streams of water that are completely covered by floating macrophytes, mainly Eicchornia crassipes (BAHIA 2001). A small clear overflow weir in Itabuna (~1.5 m height) increases stagnation and growth of macrophytes during the dry season. Plants are transported towards the estuary during rainy season, and large amounts can reach coastal waters and beaches 14 km southward during peak discharges (Souza 2005). Sampling was carried out between 2003 (November and December) and 2004 (January, February, March, April, August and October) in four stations (Fig. 1) between the small dam on Itabuna city (RI) and the upper estuary limit (BV). Sampling was also done about 5 km downstream the dam, after a stretch of rapids (RCU) and after a natural 6 m depth reservoir created by geological faults (VC). One sample was collected in each station, from the upstream station till the upper estuary boundary, comprising a total of 30 samples. Physical-chemical parameters such as pH, temperature, conductivity and dissolved oxygen (DO) were determined in the field with a WTW Multiline P4 and a Hanna HI 9143 portable meters, respectively. Samples were collected in polyethylene bottles previously washed with HCl 1:1 and distilled water, and kept under refrigeration during transportation. In the laboratory, samples were filtered in GF/C type fiberglass filters used further to chlorophyll-a analysis by the spectrophotometric trichromatic method on a 80% acetone extract (Parson et al. 1984). Aliquots of filtered samples
were frozen until analysis of dissolved inorganic nutrients. Concentration of nutrients was determined by spectrophotometric methods according Grasshoff et al. (1983) and dissolved silicate (orthosilicate) as described in Carmouze (1994). Nitrite was determined using the diazotation method (detection limit 0.02 μM); nitrate was determined by cadmium reduction into nitrite (nitrite limit plus a coefficient of variation ±3% in the range 0-10 μM); ammonium was measured spectrophotometrically by the indophenol blue method (detection limit 0.05 μM); orthophosphate by the ascorbic acid molybdate method (detection limit 0.01 μM); dissolved silicate was determined by the ammonium molybdate method (detection limit 0.1 μM). Daily fluvial discharge data (Q, m3 s-1) for a station located in Itabuna was obtained at ANA Hidroweb (Agencia Nacional de Águas, National Water Agency; www.hidroweb.ana.gov.br/hidroweb, station no. 53180000). This station is located upstream from our sampling stations, and the total increment of the area drained by the station is about 220 km2. A correction of the total monthly outflow for each drained area of the basin (QSt) was done according to (equation 1): QSt =(QANA × ASt)/AANA
(1
where QSt and ASt respectively mean total monthly fluvial outflow and drainage area of each station; and QANA and AANA represent fluvial discharge and drainage area at the ANA gauging station, respectively. Flux of dissolved inorganic nutrients was obtained by multiplying concentration values of each dissolved inorganic nutrient in the sample by the calculated fluvial
140
M. Lima et al. Tab. 1. Mean values of electrical conductivity, temperature, pH, dissolved oxygen (DO) concentration and saturation percentage in the Cachoeira River stations. Mean ± standard deviation; minimum and maximum in parenthesis. N = 4 except in * (N = 3) and ** (N = 1).
November 2003 December 2003 January 2004 February 2004 March 2004 April 2004 August 2004 October 2004
Conductivity (µS cm-1)
Temperature (°C)
pH
DO (mg L-1)
DO (% saturation)
0.553 ± 0.036 (0.505 – 0.582) 0.572 ± 0.024 (0.555 – 0.606) 0.603 ± 0.024 (0.585 – 0.638) 0.722**
29.5 ± 0.1 (29.4 – 29.7) 29.3 ± 0.8 (28.5 – 30.4) 27.9 ± 0.3 (22.4 – 28.2) 26.9**
7.3 ± 0.1 (7.2 – 7.4) 7.6 ± 0.5 (7.2 – 8.3) 7.2 ± 0.1 (7.1 – 7.3) 8.1**
4.1 ± 1.5 (2.3 – 5.6) 12.2 ± 13.4 (2.7 – 31.2) 2.3 ± 1.2 (1.2 – 3.7) -
54.2 ± 19.5 (30.2 - 73.1) 158.8 ± 172.4 (35.0 - 402.2) 29.4 ± 15.7 (15.4 - 47.3) -
0.208 ± 0.015 (0.196 – 0.228) 0.433 ± 0.039 * (0.389 – 0.462) 0.570 ± 0.042 (0.531 – 0.609) 0.609 ± 0.015 (0.595 – 0.630)
25.5 ± 0.3 (25.1 – 25.8) 26.2 ± 0.8 (25.3 – 27.0) 21.3 ± 3.1 (17.1 – 24.7) 23.8 ± 0.6 * (23.4 – 24.5)
7.2 ± 0.2 (7.0 – 7.5) 7.8 ± 0.2 (7.6 – 7.9) 7.9 ± 1.2 (7.1 – 9.7) 7.0 ± 0.3 (6.8 – 7.4)
7.2 ± 0.2 (6.9 – 7.4) 6.7 ± 0.5 (6.3 – 7.4) 6.2 **
88.0 ± 2.6 (84.5 - 90.1) 82.7 ± 5.6 (78.0 - 90.4) 75.1**
-
-
discharge values of each station on monthly basis. For nutrient concentration (Y) below detection limit, a value immediately below was used for monthly flow estimation. Fluxes were normalized dividing them by the area of each drainage basin, according to equation 2: FNY = FY / ASt
(2
where FNY represents normalized flow of dissolved inorganic nutrients, FY represents actual flow of nutrients and ASt. represents drainage area of each sampling station. We assumed that changes of nutrient concentrations in the river water during time scales lower than a month became insignificant when the actual concentration of the sample was multiplied by total monthly discharge values higher by 5-8 magnitude order, when fluxes were significant.
Fig. 2. Daily runoff at the ANA (Agência Nacional de Águas; National Water Agency) station No.53180000 during the study period.
3. RESULTS 3.1. Fluvial discharge
3.3. Dissolved inorganic nutrients
Runoff ranged from 0 to 826 m3 s-1 during the period studied (Fig. 2). Sampling during the dry season were represented by November and December 2003, and October 2004, with an average flow under 4 × 106 m3 month-1. December was the driest month, with the rainy season starting in January 2004. Maximum fluvial discharge occurred in March 2004.
The highest concentrations of ammoniacal nitrogen and phosphate were observed at station RI; nitrite and nitrate reached maximum concentrations at station RCU. Silicate presented small changes along sampling stations, without a marked spatial pattern (Fig. 3e). Maximum concentrations of ammoniacal nitrogen (Fig. 3a) and phosphate (Fig. 3d) occurred in January 2004 (beginning of the rainy season) while minimum values were observed in March 2004 (rainy season). For nitrite and nitrate (Figs 3b and 3c, respectively), the lowest concentrations were observed in November 2003 and March 2004 (rainy season). There was a decrease of the mean concentration of silicate (Fig. 3e) during January, February and March of 2004.
3.2. Physico-chemical variables Physico-chemical variables are shown in table 1. The conductivity varied from 722 μS cm-1 when the runoff begins to increase to
