Bull Environ Contam Toxicol DOI 10.1007/s00128-015-1525-5
Nitrogen Loading and Nitrous Oxide Emissions from a River with Multiple Hydroelectric Reservoirs Jinsong Chen1 • Wenzhi Cao1 • Di Cao1 • Zheng Huang1 • Ying Liang1
Received: 26 July 2014 / Accepted: 17 March 2015 Ó Springer Science+Business Media New York 2015
Abstract River networks receive a large fraction of the anthropogenic nitrogen applied to river catchments. The different impacts of the stream nitrogen (N) loading on nitrous oxide (N2O) emissions from various of aquatic ecosystems are still unknown. In this study, direct measurements of water–air interface N2O exchange in different water bodies were conducted. Results showed that the water–air interface N2O exchange from tributaries, hydropower station reservoirs, a main stream, and its estuary were 10.14 ± 13.51, 15.64 ± 10.72, 27.59 ± 20.99, and 15.98 ± 12.26 lg N2O-N m-2 h-1, respectively, indicating the strong impacts of human activities on N2O emission rates. The water NO2--N values predicted the dissolved N2O concentrations better than did the NO3--N and NH4?-N values, indicating strong denitrification and nitrification processes. The dissolved inorganic N explained 36 % of the variations in the N2O emissions for the whole river network. Keywords Nutrients input Nitrous oxide saturation Water bodies Nitrous oxide emission rate Nitrous oxide (N2O) is a potent greenhouse gas and the leading cause of stratospheric ozone depletion with a global concern. N2O production is mainly through the process of denitrification and nitrification, both of which & Wenzhi Cao [email protected] 1
State Key Laboratory of Marine Environmental Science, Key Laboratory of the Ministry of Education for Coastal and Wetland Ecosystems, College of the Environment and Ecology, Xiamen University, South Xiang’an Road, Xiang’an District, Xiamen City 361102, Fujian Province, China
are controlled by many environmental factors such as nitrate concentrations, dissolved organic carbon concentrations, temperature and other factors (Stow et al. 2005). On the global scale, anthropogenic nitrogen (N) has increased sharply since the nineteenth century, when the Haber– Bosch process was created, and represented almost 65 % of the global annual newly fixed N in 2000 (Galloway and Cowling 2002). The Intergovernmental Panel on Climate Change (Parry 2007) reported that the agricultural fertilization-induced conversion of N to N2O in soils and aquatic ecosystems is the main source of anthropogenic N2O in the atmosphere. Increased stream NO3--N loading stimulates denitrification and concomitant N2O production, but does not increase the N2O yield (Beaulieu et al. 2008). Seitzinger and Kroeze (1998) estimated that the global N2O production from rivers alone is 1.8 Tg N year-1, accounting for more than 30 % of anthropogenic terrestrial N2O production. Although many studies have examined the magnitude of N2O emissions from part of rivers (Stow et al. 2005; Clough et al. 2007; Beauchamp 1997), N2O emissions and fluxes from a complete river system have not been carefully measured, and a large river often has high spatial heterogeneity in the mechanisms of N2O production (Beaulieu et al. 2010). The Jiulong River is the second largest river in Fujian Province, Southeast China. It is a typically subtropical agricultural watershed draining a catchment area of 14,741 km2. Of the land uses in the catchment, 12 % is arable land, 7 % horticultural, 66 % forest land, 1 % urban, and the remainder is others (Cao et al. 2005). Objectives of this integrated river-scale N2O survey were as follows: (1) to identify the behavior of nutrients in the Jiulong River during the summer; (2) to measure the N2O fluxes of various sections of the river to produce a fine spatial resolution estimate of N2O emissions; and (3) to explore the
123
Bull Environ Contam Toxicol
relationships among N2O emissions, nutrients, and other environmental variables.
Methods and Materials Total discharge from the two major tributaries (the North Stream and the West Stream) of the Jiulong River (Fig. 1) is 1.24 9 1010 m3 year-1 (Cao et al. 2003). The annual average precipitation is approximately 1365 mm, 75 % of which falls between April and November. Agricultural N from the Jiulong River catchment is the largest N source, contributing 60.87 % of the total nitrogen (TN) and 68.63 % of the dissolved inorganic N (DIN) to the estuary (Cao et al. 2005). In this study, the Jiulong River was classified into four types of water bodies: the main stream, the tributaries, the hydropower station reservoirs, and the estuary according to the large differences in hydrologic and biochemical conditions among them, such as water volume and velocity. The main stream consists of the middle and lower reaches of the North Stream (from Zhangping to the mouth of the
Fig. 1 Sampling sites along the Jiulong River
123
river). There are seven hydroelectric reservoirs along the trunk channel of the North Stream; therefore, the main channel is divided into seven cascade hydroelectric reservoirs and the other fragmented sections. These reservoirs change the hydraulic retention time of the stream water, alter the nutrient biogeochemical processes, and offer the potential for different biotic activities comparing to normal lotic ecosystems. In contrast, no hydroelectric reservoirs are located on the main stream of the West Stream. There are ten tributaries that flow into the middle and downstream reaches of the North Stream. The Jiulong River Estuary is a shallow estuary with a total water area of 100 km2 (21 9 6.5 km), and a depth of 3–16 m. During the summers of 2013 and 2014, we cruised the river from Zhangping to the mouth of the river, sampling at 27 normal sampling sites in the tributaries (T1-10), the reservoirs (R1-7), the main stream (M1-7), and the estuary (E1-3). 14 fixed-plot sampling sites (T1, T5, T8, T9, M2, M4, M5, R1, R3, R5, R7, E1, E2, E3) were chosen from the normal sampling sites for N2O emission rate measurements. From the upstream to the downstream, the Sampling SitesT1, T5, T8, T9 are located in tributaries, the M2, M4,
Bull Environ Contam Toxicol
M5 in main stream reaches, the R1, R3, R5, R7 in reservoirs, and the E1, E2, E3 in estuary. All the study sites were sampled four times in the summers of 2012 and 2013 (July and August in 2012; June and August in 2013). In addition to field environmental indicators, such as the water temperature (Temwater), pH, and water flow velocity, we collected stream water samples (0.3–0.5 m) from the thalweg of the river for nutrient chemistry and dissolved N2O concentration (N2Owater) tests. These samples were collected using a 5 L Niskin sampler. Water samples for nutrient measurements were filtered immediately through a 0.45 lm cellulose acetate membrane. Two duplicate samples (the sample as the previous one) in each sampling sites for the N2Owater measurements were sampled on the boat, and analyzed in lab with the headspace equilibration technique described by Hamilton and Ostrom (2007). The N2O flux rates through the water–air interface (FN2 O ) in each fixed-plot sampling sites were measured using the floating chamber method (Mimami 1987). In this method, a semi-closed chamber with a height-adjustable floating material for water–air interface measurement was placed above the surface of water. The upper chamber (0.36 9 0.31 9 0.15 m) was equipped with a thermometer to detect the inside air temperature. After the inner gas equilibrium reached in the chamber, the inner gas was taken at 0, 10, 20, 30, 60, and 90 min through a fine needle connected to the sampling bottle (18.5 mL vacuum flask), while the same volume of N2 was fed back into the chamber through the fine needle to equilibrate the inner atmospheric pressure. The water and air samples were analyzed within 1 week after the sample collections. The concentrations of nutrients, including the total nitrogen (TN), total phosphorus (TP), dissolved reactive phosphorus (DRP), ammonium-N (NH4?-N), nitrate-N (NO3--N), and nitrite-N (NO2--N), were measured in the lab using flow injection analysis (AA3 auto analyzer, BRAN-LUEBBE Co., Germany). The dissolved organic carbon (DOC) was analyzed by the standard colorimetric method. Purge and trap-gas chromatography (Chen et al. 2007) was used to determine the partial pressure of N2O in the water samples. Overlying atmospheric N2O (N2Oair) gas samples for N2O flux rate calculations were tested with a gas chromatograph (Agilent 4890D) equipped with an electron capture detector. The detection limit for N2O was \0.3 n mol L-1, and the CV of the repeated analysis was \5 %. The dissolved inorganic nitrogen (DIN) was calculated as the sum of NH4?-N, NO3--N, and NO2--N. The degree of N2O saturation (N2Osat.) for the sampling water was calculated with the equation of Weiss and Price (1980). The N2O emission rates were calculated from the linear change in the N2O concentrations in the floating chamber headspace as a function of time, base area, chamber
volume, and the molar volume of N2O at the chamber air temperature (Corredor et al. 1999).
Results Overall, the TN and TP concentrations in the target watershed averaged 2.78 ± 0.145 [mean ± standard deviation (SD)] mg N L-1 and 0.23 ± 0.08 mg P L-1, respectively. At all times during both summers and at all sampling sites along the 210 km stretch of the river, the water was persistently supersaturated in N2O, which ranged from 175.79 % to 530.16 % with a mean of 258.75 % relative to atmospheric equilibrium. In an average of the measurements of the four water bodies, the river contributed 19.39 ± 18.02 lg N2O-N m-2 h-1 to the atmosphere in summer, with a maximum flux rate of 35.7 lg N2O-N m-2 h-1and a minimum flux rate of 1.68 lg N2O-N m-2 h-1. The DOC concentrations in Jiulong River ranged from 0.84 to 6.39 lg C L-1, and the DOsat saturation relative to the oxygen concentration in air (100 %) ranged from 69.4 % to 104.7 %, with an average of 84.3 %. The nutrient and N2O emissions data, as well as other environmental conditions in the four types of water bodies, are summarized in Fig. 2 and Table 1. NO3--N was the dominant N form for the study section of the river, accounting for approximately 85 % of DIN and 64 % of TN, and spanned a wide range of concentrations within the four types of water bodies, particularly in the tributary and the estuary. The reservoirs had the largest mean NO3--N concentration (1.61 mg N L-1) among the four types of the water bodies, whereas the highest measured value of NO3--N (3.24 mg N L-1) was observed in a sample site (M7) in the
Fig. 2 Concentrations of nutrients in the four sections of the river
123
Bull Environ Contam Toxicol Table 1 Summary of the mean values of measurements in the four water bodies
Parameters
Units
Tributary
Main stream
FN2 O
lg N2O-N m-2 h-1
10.14 ± 13.51
15.64 ± 10.72
27.59 ± 20.99
15.98 ± 12.26
DIN
mg N L-1
1.41 ± 0.79
1.97 ± 0.31
2.24 ± 0.83
1.31 ± 1.3
DRP
mg N L-1
0.08 ± 0.03
0.09 ± 0.01
0.06 ± 0.02
0.21 ± 0.32
TN
mg N L-1
2.4 ± 1.01
2.99 ± 0.11
3.61 ± 1.22
2.51 ± 1.36
TP
mg N L-1
0.1 ± 0.07
0.08 ± 0.03
0.07 ± 0.03
0.1 ± 0.06
-1
Estuary
N2Owater
lg N2O-N L
N2Osat.
%
N2Oair
lg N2O-N L-1
0.21 ± 0.01
0.24 ± 0.02
0.21 ± 0.01
0.26 ± 0.08
Water tem.
°C
27.5 ± 9.13
31.33 ± 2.52
28.43 ± 4.38
29.37 ± 1.42
main stream of the West Stream. The tributaries draining primitive river watershed covered by natural forest (e.g., T2, T4, and T6) had consistently low DIN concentrations, and those draining intensive agricultural land or urban districts with a high population density (e.g., T9 and T10) had high values. The NH4?-N concentration ranged from 0.11 to 1.21 mg N L-1 in the estuary (a broader range than in the other three water bodies), and these values decreased gradually from the upper estuary to the lower part. Consistent with the DIN results, the N2Owater values varied among the tributary (0.26 ± 0.06 lg N2O-N L-1), the main stream (0.3 ± 0.05 lg N2O-N L-1), the reservoir (0.31 ± 0.08 lg N2O-N L-1), and the estuary (0.32 ± 0.04 lg N2O-N L-1). The corresponding N2O emission rates, which were derived from the14 fixed-plot assays, were also calculated for the tributaries (10.14 ± 13.51 lg N2ON m-2 h-1), the main stream (15.64 ± 10.72 lg N2ON m-2 h-1), the reservoirs (27.59 ± 20.99 lg N2ON m-2 h-1), and the estuary (15.98 ± 12.26 lg N2ON m-2 h-1). Statistical significance was determined using a one-way ANOVA test followed by Holm-Sidak test. The values of both N2Owater and FN2 O for the four water bodies passed the significant test (p \ 0.05). No statistically significant variation (t = 0.06, p [ 0.05) in N2Owater was found between the measurements of the main stream and those of the reservoirs. Comparing with the values of N2Owater concentration, the values of FN2 O showed higher spatial heterogeneity among those river reaches. The sampling sites in the trunk channel (M1–M6, R1– R7, E1–E3), from the city of Zhangping to the estuary (Figs. 1, 3), were combined to explore the average longitudinal variations in nutrients and dissolved N2O concentrations during the summer. The target river section reached from Zhangping to the mouth of the river and included the middle and lower sections of the North River as well as the estuary. Both the TN and TP concentrations gradually decreased in the section between Zhangping (M1) and R5, increased from R5 to E1 in the lower reaches of the river channel, and decreased again sharply in the estuary from E1 to E3. The longitudinal fluctuation in the
123
Reservoirs
0.26 ± 0.06
0.3 ± 0.05
210.1 ± 32.72
278.9 ± 49.61
0.31 ± 0.04 296.66 ± 47.6
0.32 ± 0.04 304.4 ± 58.38
N2O concentration was similar to that of the other nutrients. To further explore the relationships among N2Owater, the N2O emission rates, and other physical and chemical parameters, a Spearman correlation analysis was conducted (Table 2). The results showed that N2Owater was predicted well by NH4?-N, NO2--N, DIN, and N2Osat (p \ 0.05, n = 54) and weakly by NO3--N (p [ 0.05, n = 54). The correlation coefficients between N2Owater and NO3--N obviously improved (r = 0.71, p \ 0.05) when the data from T5, a shallow stream with high NO3--N and low NH4?-N concentrations, were excluded from the analysis. The FN2 O was strongly related to the DIN, N2Osat., and DOC but weakly predicted by NH4?-N, NO2--N, and NO3--N.
Discussion Large spatial variations in the nutrient and N2O concentrations were detected in the four water bodies and in the average longitudinal profile of the North Stream. The tributaries exhibited a broader range of NO3--N concentrations than did the main stream and the reservoirs. This pattern is consistent with the work of Beaulieu et al. (2009), which demonstrated that the concentration of NO3--N in small streams ranged across four orders of magnitude. This was mainly due to the similar land use composition (mostly agricultural land and forest) in the Jiulong River Watershed and the Kalamazoo River (Beaulieu et al. 2009). The nutrient concentrations in the sampling sites of the main stream were characterized by remarkable fluctuations due to the effects of anthropogenic activities (Cao et al. 2005). The relatively high nutrient concentrations of M1 and M2, for example, were found to be due to upstream runoff, which receives abundant N-rich sewage water from Zhangping as well as drainage water from a large agricultural region with intensive livestock farms near the city (Cao et al. 2005). High concentrations of TP and TN were also found in the upper portion of the
Bull Environ Contam Toxicol Fig. 3 Average longitudinal variation in the N2O and nutrient concentrations along the length of the river
Table 2 Relationship between FN2 O , N2Owater, and the physicochemical variables Parameters
DIN
TN
TP
NH4?-N
N2Owater
0.59*
0.48
-0.41
0.81**
FN2 O
0.61*
0.42
-0.28
0.43
Parameters
NO2--N
PO4--P
N2OAir
N2Osat.
N2Owater
0.83**
-0.28
-0.02
0.95**
FN2 O
0.43
-0.08
0.42
0.69**
Parameters
Temwater.
pH
DO
DOC
N2Owater
0.11
0.15
0.35
0.42
FN2 O
0.19
0.18
0.26
0.56*
* Correlation is significant at the 0.05 level ** Correlation is significant at the 0.01 level
estuary (E1), where the West Stream empties into the North Stream and delivers large quantities of pollutants from the large urban district of Zhangzhou. The dissolved N2O concentrations exhibited spatial variation patterns similar to those of the nutrients at the entire river scale. This pattern indicated a strong relationship between N2O concentrations and the amount of pollutants of urban sewage and agricultural drainage. Because N2O production is a complicated process involving many types of bacteria, such as nitrobacteria and denitrifying bacteria, N2O production can be influenced by many environmental variables. The results of the correlation analysis in Table 2 show that N2Owater levels were best predicted by NH4?-N and NO2--N. In addition, the DIN also had a high correlation coefficient with N2Owater. The NO3--N concentrations (Fig. 4a) exhibited lower correlation coefficients with the N2Owater levels in the
reservoirs (black triangles) than did the other three water bodies. The Wenshui stream, a shallow stream with a sandy bottom, was characterized by a high NO3--N concentration and oxygen saturation as well as a low DOC concentration. It is possible that the shallow water and sandy sediment provided a poor environment for N2O production, because it should be nitrification and denitrification rather than other processes to produce N2O. A relatively high correlation coefficient (r = 0.7, p \ 0.01) was found between N2O and NO3--N concentrations in the estuary (gray triangles), where the overall NO3--N concentrations were low and spanned a broad range. Therefore, we speculate that N2Owater is better predicted by NO3--N in sites where the NO3--N is more likely to be a limitation for denitrification. The NO2--N concentrations (Fig. 4b) predicted N2Owater well in all four water bodies, indicating that NO2--N was a better predictive factor at the watershed scale than was NO3--N or NH4?-N. As an intermediate product of both nitrification and denitrification, the NO2--N concentration may directly affect N2O production by controlling the substrate for nitrification and denitrification. The NH4?-N concentrations (Fig. 4c) were highly correlated with the N2O concentrations in various river sections, with the exception of the main stream (r = 0.35, p [ 0.05), which suffers from the strong influence of human activities. Previously described relationships between N2Owater and NH4?-N, NO3--N, and NO2--N have suggested that DIN strongly affects N2O production because it is the main substrate for nitrification and denitrification, which produce N2O. Existing studies (Beaulieu et al. 2011) have confirmed the finding that nitrogen nutrients are crucial control factors in water-based N2O production. And in this study, NO2--N was a better predictor for N2O emission.
123
Bull Environ Contam Toxicol
Fig. 5 Correlation analysis between N2O concentration, DIN, and N2O flux in fixed-plot sampling sites
Fig. 4 Relationships between DIN and N2Owater for the four types of water bodies
According to the data in Table 1, the rates of N2O emissions for the four water bodies varied and were similar to the values from a small agricultural river in Indiana (4.2 lg N2O-N m-2 h-1 (Smith et al. 2009)) and to those from two sites on the Potomac River (15.4 and 140 lg N2O-N m-2 h-1 (Seitzinger 1988)). The N2O emission rate in the tributary was 10.14 ± 13.51 lg N2ON m-2 h-1, the mean value of which was much lower than that in the other three water bodies, but there was no statistically distinguishable difference between tributary and other three water bodies for the wide range of FN2 O in different tributaries. Although the FN2 O was relatively low
123
in tributaries, the geographical areas of tributaries cover great part of the whole watershed and are potentially important source of atmospheric N2O (Peterson et al. 2001). Reservoirs are typical the ‘‘hotspots’’ of N2O production due to poor flow dynamics, long water retention time, and high DIN and DOC concentrations (Groffman et al. 2009). Summer could also be ‘‘hot moment’’ owing to high water temperatures in reservoirs (Groffman et al. 2009; Beaulieu et al. 2010). The correlation analysis in Table 2 and Fig. 4 suggests that NO3--N, NH4?-N, and NO2--N strongly affected the water N2Osat and the resulting water N2O emissions. Similar to the NH4?-N and NO3--N concentrations, the DIN concentrations ranged from 0.65 to 2.94 mg N L-1 and were relative to the water N2O concentrations (r = 0.59, p \ 0.05). The largest correlation coefficient between the direct measurements of FN2 O and the environmental variables was found between FN2 O and DIN (r = 0.61, p \ 0.05) (Fig. 5). The most likely explanation for this relationship is that the microbial N2O production rates in river ecosystems were partially controlled by the DIN in the water; in other words, DIN was a good predictor of N2O emission rates at the watershed scale. This notion is
Bull Environ Contam Toxicol
similar to the discovery by Corredor et al. (1999) that N2O concentrations vary in response to the available inorganic N; their model estimates showed that there was a direct connection between the anthropogenic distribution of N and the formation of N2O gas. In addition to DIN, a significant positive correlation was found between FN2 O and N2Owater (r = 0.76, p \ 0.001), suggesting that N2Owater was a reliable indicator of N2O emission rates. Relatively high correlation coefficients between FN2 O and DOC or DOsat were also found, as shown in Table 2, demonstrating that these two factors are also important factors that impact the N2O exchange at the water–air interface. No significant correlations were found between the N2O emission rates and the air temperature in summer, however, a whole year study from Beaulieu et al. (2010) showed 70 % and 36 % of the overall variation in water N2O saturation and emission rates could be explained by temperature alone. Acknowledgments We gratefully acknowledge the funding for this study from the Chinese Natural Science Foundation (41175130), the National Basic Research Program of China through Grant 2013CB956101, and the Ministry of Environmental Protection’s Public Welfare Projects (201309007). We thank the two anonymous reviewers for their valuable comments.
References Beauchamp E (1997) Nitrous oxide emission from agricultural soils. Can J Soil Sci 77:113–123 Beaulieu JJ, Arango CP, Hamilton SK, Tank JL (2008) The production and emission of nitrous oxide from headwater streams in the Midwestern USA. Glob Change Biol 14:878–894 Beaulieu JJ, Arango CP, Tank JL (2009) The effects of season and agriculture on nitrous oxide production in headwater streams. J Environ Qual 38:637–646 Beaulieu JJ, Shuster WD, Rebholz JA (2010) Nitrous oxide emissions from a large, impounded river: the Ohio River. Environ Sci Technol 44:7527–7533 Beaulieu JJ, Tanka JL, Hamiltonb SK (2011) Nitrous oxide emission from denitrification in stream and river networks. Proc Natl Acad Sci USA 108:214–219 Cao WZ, Hong HS, Yue SP, Zhang Y, Ding Y (2003) Nutrient loss from an agricultural catchment and land scape modeling in southeast China. Bull Environ Contam Toxicol 71:761–767
Cao WZ, Hong HS, Yue SP (2005) Modelling agricultural nitrogen contributions to the Jiulong River estuary and coastal water. Glob Planet Change 47:111–121 Chen Y, Yuan DX, Li QL (2007) Determination of nitrous oxide in seawater by room temperature purge and trap-gas chromatography. Chin J Anal Chem 35:897 Clough TJ, Buckthought LE, Kelliher FM, Sherlock RR (2007) Diurnal fluctuations of dissolved nitrous oxide (N2O) concentrations and estimates of N2O emissions from a spring-fed river: implications for IPCC methodology. Glob Change Biol 13:1016–1027 Corredor JE, Morell JM, Bauza J (1999) Atmospheric nitrous oxide fluxes from mangrove sediments. Mar Pollut Bull 38:473–478 Galloway JN, Cowling EB (2002) Reactive nitrogen and the world: 200 years of change. Ambio 31:64–71 Groffman PM, Butterbach-Bahl K, Fulweiler RW, Gold AJ, Morse JL, Stander EK, Tague C, Tonitto C, Vidon P (2009) Challenges to incorporating spatially and temporally explicit phenomena (hotspots and hot moments) in denitrification models. Biogeochemistry 93:49–77 Hamilton SK, Ostrom NE (2007) Measurement of the stable isotope ratio of dissolved N2 in 15N tracer experiments. Limnol Oceanogr Methods 5:233–240 Mimami K (1987) Emission of nitrous oxide (N2O) from agroecosystem. Jpn Agric Res Q 21:22–27 Parry ML (2007) Climate change 2007: impacts, adaptation and vulnerability: contribution of working group II to the fourth assessment report of the Intergovernmental panel on climate change, vol 4. Cambridge University Press, Cambridge Peterson BJ, Wollheim WM, Mulholland PJ, Webster JR, Meyer JL, Tank JL, Marti E, Bowden WB, Valett HM, Hershey AE, McDowell WH, Dodds WK, Hamilton SK, Gregory S, Morrall DD (2001) Control of nitrogen export from watersheds by headwater streams. Science 292:86–90 Seitzinger SP (1988) Denitrification in freshwater and coastal marine ecosystems: ecological and geochemical significance. Limnol Oceanogr 33:702–724 Seitzinger SP, Kroeze C (1998) Global distribution of nitrous oxide production and N inputs in freshwater and coastal marine ecosystems. Glob Biogeochem Cycles 12:93–113 Smith RL, Bo¨hlke JK, Repert DA, Hart CP (2009) Nitrification and denitrification in a mid-western stream containing high nitrate: in situ assessment using tracers in dome-shaped incubation chambers. Biogeochemistry 96:189–208 Stow CA, Walker JT, Cardoch L, Spence P, Geron C (2005) N2O emissions from streams in the Neuse river watershed, North Carolina. Environ Sci Technol 39:6999–7004 Weiss RF, Price BA (1980) Nitrous oxide solubility in water and seawater. Mar Chem 8:347–359
123
Nitrogen Loading and Nitrous Oxide Emissions from a River with Multiple Hydroelectric Reservoirs Jinsong Chen1 • Wenzhi Cao1 • Di Cao1 • Zheng Huang1 • Ying Liang1
Received: 26 July 2014 / Accepted: 17 March 2015 Ó Springer Science+Business Media New York 2015
Abstract River networks receive a large fraction of the anthropogenic nitrogen applied to river catchments. The different impacts of the stream nitrogen (N) loading on nitrous oxide (N2O) emissions from various of aquatic ecosystems are still unknown. In this study, direct measurements of water–air interface N2O exchange in different water bodies were conducted. Results showed that the water–air interface N2O exchange from tributaries, hydropower station reservoirs, a main stream, and its estuary were 10.14 ± 13.51, 15.64 ± 10.72, 27.59 ± 20.99, and 15.98 ± 12.26 lg N2O-N m-2 h-1, respectively, indicating the strong impacts of human activities on N2O emission rates. The water NO2--N values predicted the dissolved N2O concentrations better than did the NO3--N and NH4?-N values, indicating strong denitrification and nitrification processes. The dissolved inorganic N explained 36 % of the variations in the N2O emissions for the whole river network. Keywords Nutrients input Nitrous oxide saturation Water bodies Nitrous oxide emission rate Nitrous oxide (N2O) is a potent greenhouse gas and the leading cause of stratospheric ozone depletion with a global concern. N2O production is mainly through the process of denitrification and nitrification, both of which & Wenzhi Cao [email protected] 1
State Key Laboratory of Marine Environmental Science, Key Laboratory of the Ministry of Education for Coastal and Wetland Ecosystems, College of the Environment and Ecology, Xiamen University, South Xiang’an Road, Xiang’an District, Xiamen City 361102, Fujian Province, China
are controlled by many environmental factors such as nitrate concentrations, dissolved organic carbon concentrations, temperature and other factors (Stow et al. 2005). On the global scale, anthropogenic nitrogen (N) has increased sharply since the nineteenth century, when the Haber– Bosch process was created, and represented almost 65 % of the global annual newly fixed N in 2000 (Galloway and Cowling 2002). The Intergovernmental Panel on Climate Change (Parry 2007) reported that the agricultural fertilization-induced conversion of N to N2O in soils and aquatic ecosystems is the main source of anthropogenic N2O in the atmosphere. Increased stream NO3--N loading stimulates denitrification and concomitant N2O production, but does not increase the N2O yield (Beaulieu et al. 2008). Seitzinger and Kroeze (1998) estimated that the global N2O production from rivers alone is 1.8 Tg N year-1, accounting for more than 30 % of anthropogenic terrestrial N2O production. Although many studies have examined the magnitude of N2O emissions from part of rivers (Stow et al. 2005; Clough et al. 2007; Beauchamp 1997), N2O emissions and fluxes from a complete river system have not been carefully measured, and a large river often has high spatial heterogeneity in the mechanisms of N2O production (Beaulieu et al. 2010). The Jiulong River is the second largest river in Fujian Province, Southeast China. It is a typically subtropical agricultural watershed draining a catchment area of 14,741 km2. Of the land uses in the catchment, 12 % is arable land, 7 % horticultural, 66 % forest land, 1 % urban, and the remainder is others (Cao et al. 2005). Objectives of this integrated river-scale N2O survey were as follows: (1) to identify the behavior of nutrients in the Jiulong River during the summer; (2) to measure the N2O fluxes of various sections of the river to produce a fine spatial resolution estimate of N2O emissions; and (3) to explore the
123
Bull Environ Contam Toxicol
relationships among N2O emissions, nutrients, and other environmental variables.
Methods and Materials Total discharge from the two major tributaries (the North Stream and the West Stream) of the Jiulong River (Fig. 1) is 1.24 9 1010 m3 year-1 (Cao et al. 2003). The annual average precipitation is approximately 1365 mm, 75 % of which falls between April and November. Agricultural N from the Jiulong River catchment is the largest N source, contributing 60.87 % of the total nitrogen (TN) and 68.63 % of the dissolved inorganic N (DIN) to the estuary (Cao et al. 2005). In this study, the Jiulong River was classified into four types of water bodies: the main stream, the tributaries, the hydropower station reservoirs, and the estuary according to the large differences in hydrologic and biochemical conditions among them, such as water volume and velocity. The main stream consists of the middle and lower reaches of the North Stream (from Zhangping to the mouth of the
Fig. 1 Sampling sites along the Jiulong River
123
river). There are seven hydroelectric reservoirs along the trunk channel of the North Stream; therefore, the main channel is divided into seven cascade hydroelectric reservoirs and the other fragmented sections. These reservoirs change the hydraulic retention time of the stream water, alter the nutrient biogeochemical processes, and offer the potential for different biotic activities comparing to normal lotic ecosystems. In contrast, no hydroelectric reservoirs are located on the main stream of the West Stream. There are ten tributaries that flow into the middle and downstream reaches of the North Stream. The Jiulong River Estuary is a shallow estuary with a total water area of 100 km2 (21 9 6.5 km), and a depth of 3–16 m. During the summers of 2013 and 2014, we cruised the river from Zhangping to the mouth of the river, sampling at 27 normal sampling sites in the tributaries (T1-10), the reservoirs (R1-7), the main stream (M1-7), and the estuary (E1-3). 14 fixed-plot sampling sites (T1, T5, T8, T9, M2, M4, M5, R1, R3, R5, R7, E1, E2, E3) were chosen from the normal sampling sites for N2O emission rate measurements. From the upstream to the downstream, the Sampling SitesT1, T5, T8, T9 are located in tributaries, the M2, M4,
Bull Environ Contam Toxicol
M5 in main stream reaches, the R1, R3, R5, R7 in reservoirs, and the E1, E2, E3 in estuary. All the study sites were sampled four times in the summers of 2012 and 2013 (July and August in 2012; June and August in 2013). In addition to field environmental indicators, such as the water temperature (Temwater), pH, and water flow velocity, we collected stream water samples (0.3–0.5 m) from the thalweg of the river for nutrient chemistry and dissolved N2O concentration (N2Owater) tests. These samples were collected using a 5 L Niskin sampler. Water samples for nutrient measurements were filtered immediately through a 0.45 lm cellulose acetate membrane. Two duplicate samples (the sample as the previous one) in each sampling sites for the N2Owater measurements were sampled on the boat, and analyzed in lab with the headspace equilibration technique described by Hamilton and Ostrom (2007). The N2O flux rates through the water–air interface (FN2 O ) in each fixed-plot sampling sites were measured using the floating chamber method (Mimami 1987). In this method, a semi-closed chamber with a height-adjustable floating material for water–air interface measurement was placed above the surface of water. The upper chamber (0.36 9 0.31 9 0.15 m) was equipped with a thermometer to detect the inside air temperature. After the inner gas equilibrium reached in the chamber, the inner gas was taken at 0, 10, 20, 30, 60, and 90 min through a fine needle connected to the sampling bottle (18.5 mL vacuum flask), while the same volume of N2 was fed back into the chamber through the fine needle to equilibrate the inner atmospheric pressure. The water and air samples were analyzed within 1 week after the sample collections. The concentrations of nutrients, including the total nitrogen (TN), total phosphorus (TP), dissolved reactive phosphorus (DRP), ammonium-N (NH4?-N), nitrate-N (NO3--N), and nitrite-N (NO2--N), were measured in the lab using flow injection analysis (AA3 auto analyzer, BRAN-LUEBBE Co., Germany). The dissolved organic carbon (DOC) was analyzed by the standard colorimetric method. Purge and trap-gas chromatography (Chen et al. 2007) was used to determine the partial pressure of N2O in the water samples. Overlying atmospheric N2O (N2Oair) gas samples for N2O flux rate calculations were tested with a gas chromatograph (Agilent 4890D) equipped with an electron capture detector. The detection limit for N2O was \0.3 n mol L-1, and the CV of the repeated analysis was \5 %. The dissolved inorganic nitrogen (DIN) was calculated as the sum of NH4?-N, NO3--N, and NO2--N. The degree of N2O saturation (N2Osat.) for the sampling water was calculated with the equation of Weiss and Price (1980). The N2O emission rates were calculated from the linear change in the N2O concentrations in the floating chamber headspace as a function of time, base area, chamber
volume, and the molar volume of N2O at the chamber air temperature (Corredor et al. 1999).
Results Overall, the TN and TP concentrations in the target watershed averaged 2.78 ± 0.145 [mean ± standard deviation (SD)] mg N L-1 and 0.23 ± 0.08 mg P L-1, respectively. At all times during both summers and at all sampling sites along the 210 km stretch of the river, the water was persistently supersaturated in N2O, which ranged from 175.79 % to 530.16 % with a mean of 258.75 % relative to atmospheric equilibrium. In an average of the measurements of the four water bodies, the river contributed 19.39 ± 18.02 lg N2O-N m-2 h-1 to the atmosphere in summer, with a maximum flux rate of 35.7 lg N2O-N m-2 h-1and a minimum flux rate of 1.68 lg N2O-N m-2 h-1. The DOC concentrations in Jiulong River ranged from 0.84 to 6.39 lg C L-1, and the DOsat saturation relative to the oxygen concentration in air (100 %) ranged from 69.4 % to 104.7 %, with an average of 84.3 %. The nutrient and N2O emissions data, as well as other environmental conditions in the four types of water bodies, are summarized in Fig. 2 and Table 1. NO3--N was the dominant N form for the study section of the river, accounting for approximately 85 % of DIN and 64 % of TN, and spanned a wide range of concentrations within the four types of water bodies, particularly in the tributary and the estuary. The reservoirs had the largest mean NO3--N concentration (1.61 mg N L-1) among the four types of the water bodies, whereas the highest measured value of NO3--N (3.24 mg N L-1) was observed in a sample site (M7) in the
Fig. 2 Concentrations of nutrients in the four sections of the river
123
Bull Environ Contam Toxicol Table 1 Summary of the mean values of measurements in the four water bodies
Parameters
Units
Tributary
Main stream
FN2 O
lg N2O-N m-2 h-1
10.14 ± 13.51
15.64 ± 10.72
27.59 ± 20.99
15.98 ± 12.26
DIN
mg N L-1
1.41 ± 0.79
1.97 ± 0.31
2.24 ± 0.83
1.31 ± 1.3
DRP
mg N L-1
0.08 ± 0.03
0.09 ± 0.01
0.06 ± 0.02
0.21 ± 0.32
TN
mg N L-1
2.4 ± 1.01
2.99 ± 0.11
3.61 ± 1.22
2.51 ± 1.36
TP
mg N L-1
0.1 ± 0.07
0.08 ± 0.03
0.07 ± 0.03
0.1 ± 0.06
-1
Estuary
N2Owater
lg N2O-N L
N2Osat.
%
N2Oair
lg N2O-N L-1
0.21 ± 0.01
0.24 ± 0.02
0.21 ± 0.01
0.26 ± 0.08
Water tem.
°C
27.5 ± 9.13
31.33 ± 2.52
28.43 ± 4.38
29.37 ± 1.42
main stream of the West Stream. The tributaries draining primitive river watershed covered by natural forest (e.g., T2, T4, and T6) had consistently low DIN concentrations, and those draining intensive agricultural land or urban districts with a high population density (e.g., T9 and T10) had high values. The NH4?-N concentration ranged from 0.11 to 1.21 mg N L-1 in the estuary (a broader range than in the other three water bodies), and these values decreased gradually from the upper estuary to the lower part. Consistent with the DIN results, the N2Owater values varied among the tributary (0.26 ± 0.06 lg N2O-N L-1), the main stream (0.3 ± 0.05 lg N2O-N L-1), the reservoir (0.31 ± 0.08 lg N2O-N L-1), and the estuary (0.32 ± 0.04 lg N2O-N L-1). The corresponding N2O emission rates, which were derived from the14 fixed-plot assays, were also calculated for the tributaries (10.14 ± 13.51 lg N2ON m-2 h-1), the main stream (15.64 ± 10.72 lg N2ON m-2 h-1), the reservoirs (27.59 ± 20.99 lg N2ON m-2 h-1), and the estuary (15.98 ± 12.26 lg N2ON m-2 h-1). Statistical significance was determined using a one-way ANOVA test followed by Holm-Sidak test. The values of both N2Owater and FN2 O for the four water bodies passed the significant test (p \ 0.05). No statistically significant variation (t = 0.06, p [ 0.05) in N2Owater was found between the measurements of the main stream and those of the reservoirs. Comparing with the values of N2Owater concentration, the values of FN2 O showed higher spatial heterogeneity among those river reaches. The sampling sites in the trunk channel (M1–M6, R1– R7, E1–E3), from the city of Zhangping to the estuary (Figs. 1, 3), were combined to explore the average longitudinal variations in nutrients and dissolved N2O concentrations during the summer. The target river section reached from Zhangping to the mouth of the river and included the middle and lower sections of the North River as well as the estuary. Both the TN and TP concentrations gradually decreased in the section between Zhangping (M1) and R5, increased from R5 to E1 in the lower reaches of the river channel, and decreased again sharply in the estuary from E1 to E3. The longitudinal fluctuation in the
123
Reservoirs
0.26 ± 0.06
0.3 ± 0.05
210.1 ± 32.72
278.9 ± 49.61
0.31 ± 0.04 296.66 ± 47.6
0.32 ± 0.04 304.4 ± 58.38
N2O concentration was similar to that of the other nutrients. To further explore the relationships among N2Owater, the N2O emission rates, and other physical and chemical parameters, a Spearman correlation analysis was conducted (Table 2). The results showed that N2Owater was predicted well by NH4?-N, NO2--N, DIN, and N2Osat (p \ 0.05, n = 54) and weakly by NO3--N (p [ 0.05, n = 54). The correlation coefficients between N2Owater and NO3--N obviously improved (r = 0.71, p \ 0.05) when the data from T5, a shallow stream with high NO3--N and low NH4?-N concentrations, were excluded from the analysis. The FN2 O was strongly related to the DIN, N2Osat., and DOC but weakly predicted by NH4?-N, NO2--N, and NO3--N.
Discussion Large spatial variations in the nutrient and N2O concentrations were detected in the four water bodies and in the average longitudinal profile of the North Stream. The tributaries exhibited a broader range of NO3--N concentrations than did the main stream and the reservoirs. This pattern is consistent with the work of Beaulieu et al. (2009), which demonstrated that the concentration of NO3--N in small streams ranged across four orders of magnitude. This was mainly due to the similar land use composition (mostly agricultural land and forest) in the Jiulong River Watershed and the Kalamazoo River (Beaulieu et al. 2009). The nutrient concentrations in the sampling sites of the main stream were characterized by remarkable fluctuations due to the effects of anthropogenic activities (Cao et al. 2005). The relatively high nutrient concentrations of M1 and M2, for example, were found to be due to upstream runoff, which receives abundant N-rich sewage water from Zhangping as well as drainage water from a large agricultural region with intensive livestock farms near the city (Cao et al. 2005). High concentrations of TP and TN were also found in the upper portion of the
Bull Environ Contam Toxicol Fig. 3 Average longitudinal variation in the N2O and nutrient concentrations along the length of the river
Table 2 Relationship between FN2 O , N2Owater, and the physicochemical variables Parameters
DIN
TN
TP
NH4?-N
N2Owater
0.59*
0.48
-0.41
0.81**
FN2 O
0.61*
0.42
-0.28
0.43
Parameters
NO2--N
PO4--P
N2OAir
N2Osat.
N2Owater
0.83**
-0.28
-0.02
0.95**
FN2 O
0.43
-0.08
0.42
0.69**
Parameters
Temwater.
pH
DO
DOC
N2Owater
0.11
0.15
0.35
0.42
FN2 O
0.19
0.18
0.26
0.56*
* Correlation is significant at the 0.05 level ** Correlation is significant at the 0.01 level
estuary (E1), where the West Stream empties into the North Stream and delivers large quantities of pollutants from the large urban district of Zhangzhou. The dissolved N2O concentrations exhibited spatial variation patterns similar to those of the nutrients at the entire river scale. This pattern indicated a strong relationship between N2O concentrations and the amount of pollutants of urban sewage and agricultural drainage. Because N2O production is a complicated process involving many types of bacteria, such as nitrobacteria and denitrifying bacteria, N2O production can be influenced by many environmental variables. The results of the correlation analysis in Table 2 show that N2Owater levels were best predicted by NH4?-N and NO2--N. In addition, the DIN also had a high correlation coefficient with N2Owater. The NO3--N concentrations (Fig. 4a) exhibited lower correlation coefficients with the N2Owater levels in the
reservoirs (black triangles) than did the other three water bodies. The Wenshui stream, a shallow stream with a sandy bottom, was characterized by a high NO3--N concentration and oxygen saturation as well as a low DOC concentration. It is possible that the shallow water and sandy sediment provided a poor environment for N2O production, because it should be nitrification and denitrification rather than other processes to produce N2O. A relatively high correlation coefficient (r = 0.7, p \ 0.01) was found between N2O and NO3--N concentrations in the estuary (gray triangles), where the overall NO3--N concentrations were low and spanned a broad range. Therefore, we speculate that N2Owater is better predicted by NO3--N in sites where the NO3--N is more likely to be a limitation for denitrification. The NO2--N concentrations (Fig. 4b) predicted N2Owater well in all four water bodies, indicating that NO2--N was a better predictive factor at the watershed scale than was NO3--N or NH4?-N. As an intermediate product of both nitrification and denitrification, the NO2--N concentration may directly affect N2O production by controlling the substrate for nitrification and denitrification. The NH4?-N concentrations (Fig. 4c) were highly correlated with the N2O concentrations in various river sections, with the exception of the main stream (r = 0.35, p [ 0.05), which suffers from the strong influence of human activities. Previously described relationships between N2Owater and NH4?-N, NO3--N, and NO2--N have suggested that DIN strongly affects N2O production because it is the main substrate for nitrification and denitrification, which produce N2O. Existing studies (Beaulieu et al. 2011) have confirmed the finding that nitrogen nutrients are crucial control factors in water-based N2O production. And in this study, NO2--N was a better predictor for N2O emission.
123
Bull Environ Contam Toxicol
Fig. 5 Correlation analysis between N2O concentration, DIN, and N2O flux in fixed-plot sampling sites
Fig. 4 Relationships between DIN and N2Owater for the four types of water bodies
According to the data in Table 1, the rates of N2O emissions for the four water bodies varied and were similar to the values from a small agricultural river in Indiana (4.2 lg N2O-N m-2 h-1 (Smith et al. 2009)) and to those from two sites on the Potomac River (15.4 and 140 lg N2O-N m-2 h-1 (Seitzinger 1988)). The N2O emission rate in the tributary was 10.14 ± 13.51 lg N2ON m-2 h-1, the mean value of which was much lower than that in the other three water bodies, but there was no statistically distinguishable difference between tributary and other three water bodies for the wide range of FN2 O in different tributaries. Although the FN2 O was relatively low
123
in tributaries, the geographical areas of tributaries cover great part of the whole watershed and are potentially important source of atmospheric N2O (Peterson et al. 2001). Reservoirs are typical the ‘‘hotspots’’ of N2O production due to poor flow dynamics, long water retention time, and high DIN and DOC concentrations (Groffman et al. 2009). Summer could also be ‘‘hot moment’’ owing to high water temperatures in reservoirs (Groffman et al. 2009; Beaulieu et al. 2010). The correlation analysis in Table 2 and Fig. 4 suggests that NO3--N, NH4?-N, and NO2--N strongly affected the water N2Osat and the resulting water N2O emissions. Similar to the NH4?-N and NO3--N concentrations, the DIN concentrations ranged from 0.65 to 2.94 mg N L-1 and were relative to the water N2O concentrations (r = 0.59, p \ 0.05). The largest correlation coefficient between the direct measurements of FN2 O and the environmental variables was found between FN2 O and DIN (r = 0.61, p \ 0.05) (Fig. 5). The most likely explanation for this relationship is that the microbial N2O production rates in river ecosystems were partially controlled by the DIN in the water; in other words, DIN was a good predictor of N2O emission rates at the watershed scale. This notion is
Bull Environ Contam Toxicol
similar to the discovery by Corredor et al. (1999) that N2O concentrations vary in response to the available inorganic N; their model estimates showed that there was a direct connection between the anthropogenic distribution of N and the formation of N2O gas. In addition to DIN, a significant positive correlation was found between FN2 O and N2Owater (r = 0.76, p \ 0.001), suggesting that N2Owater was a reliable indicator of N2O emission rates. Relatively high correlation coefficients between FN2 O and DOC or DOsat were also found, as shown in Table 2, demonstrating that these two factors are also important factors that impact the N2O exchange at the water–air interface. No significant correlations were found between the N2O emission rates and the air temperature in summer, however, a whole year study from Beaulieu et al. (2010) showed 70 % and 36 % of the overall variation in water N2O saturation and emission rates could be explained by temperature alone. Acknowledgments We gratefully acknowledge the funding for this study from the Chinese Natural Science Foundation (41175130), the National Basic Research Program of China through Grant 2013CB956101, and the Ministry of Environmental Protection’s Public Welfare Projects (201309007). We thank the two anonymous reviewers for their valuable comments.
References Beauchamp E (1997) Nitrous oxide emission from agricultural soils. Can J Soil Sci 77:113–123 Beaulieu JJ, Arango CP, Hamilton SK, Tank JL (2008) The production and emission of nitrous oxide from headwater streams in the Midwestern USA. Glob Change Biol 14:878–894 Beaulieu JJ, Arango CP, Tank JL (2009) The effects of season and agriculture on nitrous oxide production in headwater streams. J Environ Qual 38:637–646 Beaulieu JJ, Shuster WD, Rebholz JA (2010) Nitrous oxide emissions from a large, impounded river: the Ohio River. Environ Sci Technol 44:7527–7533 Beaulieu JJ, Tanka JL, Hamiltonb SK (2011) Nitrous oxide emission from denitrification in stream and river networks. Proc Natl Acad Sci USA 108:214–219 Cao WZ, Hong HS, Yue SP, Zhang Y, Ding Y (2003) Nutrient loss from an agricultural catchment and land scape modeling in southeast China. Bull Environ Contam Toxicol 71:761–767
Cao WZ, Hong HS, Yue SP (2005) Modelling agricultural nitrogen contributions to the Jiulong River estuary and coastal water. Glob Planet Change 47:111–121 Chen Y, Yuan DX, Li QL (2007) Determination of nitrous oxide in seawater by room temperature purge and trap-gas chromatography. Chin J Anal Chem 35:897 Clough TJ, Buckthought LE, Kelliher FM, Sherlock RR (2007) Diurnal fluctuations of dissolved nitrous oxide (N2O) concentrations and estimates of N2O emissions from a spring-fed river: implications for IPCC methodology. Glob Change Biol 13:1016–1027 Corredor JE, Morell JM, Bauza J (1999) Atmospheric nitrous oxide fluxes from mangrove sediments. Mar Pollut Bull 38:473–478 Galloway JN, Cowling EB (2002) Reactive nitrogen and the world: 200 years of change. Ambio 31:64–71 Groffman PM, Butterbach-Bahl K, Fulweiler RW, Gold AJ, Morse JL, Stander EK, Tague C, Tonitto C, Vidon P (2009) Challenges to incorporating spatially and temporally explicit phenomena (hotspots and hot moments) in denitrification models. Biogeochemistry 93:49–77 Hamilton SK, Ostrom NE (2007) Measurement of the stable isotope ratio of dissolved N2 in 15N tracer experiments. Limnol Oceanogr Methods 5:233–240 Mimami K (1987) Emission of nitrous oxide (N2O) from agroecosystem. Jpn Agric Res Q 21:22–27 Parry ML (2007) Climate change 2007: impacts, adaptation and vulnerability: contribution of working group II to the fourth assessment report of the Intergovernmental panel on climate change, vol 4. Cambridge University Press, Cambridge Peterson BJ, Wollheim WM, Mulholland PJ, Webster JR, Meyer JL, Tank JL, Marti E, Bowden WB, Valett HM, Hershey AE, McDowell WH, Dodds WK, Hamilton SK, Gregory S, Morrall DD (2001) Control of nitrogen export from watersheds by headwater streams. Science 292:86–90 Seitzinger SP (1988) Denitrification in freshwater and coastal marine ecosystems: ecological and geochemical significance. Limnol Oceanogr 33:702–724 Seitzinger SP, Kroeze C (1998) Global distribution of nitrous oxide production and N inputs in freshwater and coastal marine ecosystems. Glob Biogeochem Cycles 12:93–113 Smith RL, Bo¨hlke JK, Repert DA, Hart CP (2009) Nitrification and denitrification in a mid-western stream containing high nitrate: in situ assessment using tracers in dome-shaped incubation chambers. Biogeochemistry 96:189–208 Stow CA, Walker JT, Cardoch L, Spence P, Geron C (2005) N2O emissions from streams in the Neuse river watershed, North Carolina. Environ Sci Technol 39:6999–7004 Weiss RF, Price BA (1980) Nitrous oxide solubility in water and seawater. Mar Chem 8:347–359
123
