SEARCH

SEARCH BY CITATION

Keywords:

  • IPCC;
  • greenhouse gas;
  • nitrate;
  • nitrogen;
  • nitrous oxide;
  • stream

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

[1] Nitrous oxide (N2O) is a potent greenhouse gas produced during nitrogen cycling. Global nitrogen enrichment has resulted in increased atmospheric N2O concentrations due in large part to increased soil emissions. There is also a potentially important flux from streams, rivers and estuaries; although measurements of these emissions are sparse, and role of aquatic ecosystems in global N2O budgets remains highly uncertain. Using the longest-term measurements of N2O fluxes from streams to date, we found annual fluxes from 14 sites in five streams of south-central Ontario, Canada varied widely–from net uptake of 3.2 ± 0.2 (standard deviation)μmol N2O m−2 d−1 to net release of 776 ± 61 μmol N2O m−2 d−1. N2O consumption was associated with very low nitrate concentrations (<2.7 μM). Mean annual (log-transformed) N2O emissions from our study streams (across sites and years) were positively related to nitrate concentrations (r2= 0.59).This nitrate-N2O relationship can be generalized across all 20 streams (in Canada, Japan, Mexico, and the midwestern United States) for which published data now exist and could provide a new basis for the IPCC to calculate agricultural emissions from streams. In addition to predicting annual emissions, we present the first measurements of N2O concentrations under ice in streams. Nitrate was a strong predictor of N2O % saturation during periods of ice cover (r2 = 0.89), when gas exchange is negligible. Given the small surface area of streams within a catchment and the fact our measured areal fluxes are comparable to reported fluxes from agricultural soils, this suggests streams are a small regional N2O source.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

[2] Human alterations of the nitrogen (N) cycle have led to increased atmospheric concentrations of nitrous oxide (N2O), an important greenhouse gas, and contributor to stratospheric ozone depletion [Ravishankara et al., 2009]. Much work has been dedicated to quantifying N2O fluxes from soils, but fluxes from aquatic sources are still poorly characterized. This poor knowledge of aquatic fluxes continues despite the twofold to twentyfold increase in riverine N concentrations in numerous regions [Vitousek et al., 1997], the prediction that N-loading will continue to increase [Tilman et al., 2001], observed relationships between N concentrations and N2O production [Firestone and Davidson, 1989], and models of the global N2O budget that suggest rivers, streams and estuaries are large emissions sources [Intergovernmental Panel on Climate Change (IPCC), 2006; Beaulieu et al., 2010a] particularly in the northern hemisphere [Seitzinger and Kroeze, 1998].

[3] N2O is produced as a result of nitrification and denitrification. Denitrification is the sequential reduction of N oxides to N gases (NO3 [RIGHTWARDS ARROW] NO2 [RIGHTWARDS ARROW] NO [RIGHTWARDS ARROW] N2O [RIGHTWARDS ARROW] N2). N2O is a free intermediate in the denitrification pathway, and can be both produced and consumed in the process [Zumft, 1997]. N2O is also produced via nitrification as a consequence of the oxidation of the intermediate hydroxylamine, or, in the pathway termed nitrifier-denitrification, where organisms that oxidize ammonium to nitrate may also carry out the reduction of nitrite to N2O or N2 [Stein and Yung, 2003; Wrage et al., 2001]. N2O emissions are a function both of rates of nitrification and denitrification, and the N2O yields of these processes (N2O:NO3 and N2O:N2 respectively) [Firestone and Davidson, 1989].

[4] Simple relationships between rates of N cycling and rates of N2O emissions may not exist [Beaulieu et al., 2008] due to variable N2O yields and variable contributions of nitrification and denitrification to emissions. However, current methods to estimate global emissions are based upon such assumptions [IPCC, 2006; Seitzinger and Kroeze, 1998]. While controls on rates of nitrification and denitrification are well characterized in aquatic ecosystems (Table 1), factors affecting the N2O yields of these processes are not well known, and much of our knowledge of these yields comes from terrestrial ecosystems. In streams, denitrification is typically limited by nitrate (NO3) concentrations [Garcia-Ruiz et al., 1998a]. Increased concentrations of NO3 may contribute to higher N2O:N2, although recent evidence suggests this ratio is smaller in aquatic ecosystems than in terrestrial ecosystems (Table 1) [Beaulieu et al., 2010a]. Nitrification rates are often stimulated by the addition of ammonium [Strauss et al., 2002], although N2O yields may not be affected [Jiang and Bakken, 1999]. A suite of other factors can also affect rates of N2O production via effects on N cycling, N2O yields, or both (Table 1). Importantly, many of these factors vary over annual periods. As a result, the assessment of the importance of streams in N2O emissions requires study over relatively long (annual or multiannual) timescales.

Table 1. Possible Effects of Changes in Environmental Conditions on Rates of N Cycling, N2O Yields, and N2O Productiona
Environmental ChangeRate AffectedChange in RateEffects on N2O YieldbResulting Effects on N2O Production
  • a

    Where indicated, we cite results from studies performed in ecosystems other than rivers or streams.

  • b

    Nitrification: N2O:NO3, denitrification N2O:N2.

  • c

    Indicates an ecosystem other than a river or stream.

  • d

    If the DOC affects nitrification rates by increasing competition for NH4+ [Strauss et al., 2002], no effects on N2O:NO3 may be expected.

Increased nitrateDenitrificationIncrease [Beaulieu et al., 2010a; Garcia-Ruiz et al., 1998b; Inwood et al., 2005; Kemp and Dodds, 2002; Martin et al., 2001; Richardson et al., 2004; Silvennoinen et al., 2008b]Increase [Silvennoinen et al., 2008b]Increase
   No effect [Beaulieu et al., 2010a] 
 
Increased ammoniumNitrificationIncrease [Kemp and Dodds, 2002; Strauss and Lamberti, 2000; Strauss et al., 2002]Decrease, or no effect [Jiang and Bakken, 1999]cPossible increase
  No effect [Strauss et al., 2004]  
 
Increased dissolved organic carbonDenitrificationIncrease [Richardson et al., 2004]Decrease [Weier et al., 1993]cUnknown
 NitrificationDecrease [Bernhardt and Likens, 2002; Strauss and Lamberti, 2000; Strauss et al., 2002]Unknown/ no effectdUnknown
  No effect [Strauss et al., 2002; Strauss et al., 2004]  
 
Increased temperatureDenitrificationIncrease to maximum [Silvennoinen et al., 2008a]Decrease [Silvennoinen et al., 2008a]Unknown
  Increase [Garcia-Ruiz et al., 1998b; Martin et al., 2001]  
  No effect [Martin et al., 2001]  
 NitrificationIncrease [Starry et al., 2005]Decrease [Maag and Vinther, 1996]cUnknown
  No effect [Kemp and Dodds, 2002]  
 
Increased oxygenDenitrificationDecrease [Kemp and Dodds, 2002; Silvennoinen et al., 2008a]Depends on temperature, range of O2 [Silvennoinen et al., 2008a]Unknown
   Increase [Cavigelli and Robertson, 2000]c, [Betlach and Tiedje, 1981]c, [Weier et al., 1993]c 
 NitrificationIncrease [Kemp and Dodds, 2002]Decrease [Goreau et al., 1980]c; [Khalil et al., 2004]cUnknown
 
Increased pHDenitrificationIncrease [Müller et al., 1980]cDecrease [Cavigelli and Robertson, 2000]c,[Martikainen, 1985]cUnknown
   Max at pH 6.5 [Stevens et al., 1998]c 
 NitrificationMax at pH 7.5 [Strauss et al., 2002]Decrease [Jiang and Bakken, 1999]cUnknown
  Max at pH 8.5 ([Warwick, 1986]– note reviewc)  

[5] In this study, we report rates of N2O emissions from five small streams. Small streams are thought to be hot spots of nitrogen cycling [Alexander et al., 2000] and can remove a large proportion of nutrient inputs [Wollheim et al., 2006], hence may be important sites of N2O emissions. We assess predictors of daily and annual N2O fluxes, and then broaden our analysis using annual flux estimates from published data. We used data from periods of ice cover to assess N2O dynamics during periods of negligible gas exchange and determine whether predictors of N2O concentration under ice differ from predictors of flux in the open water season. Finally, although streams are typically thought to be sources of N2O, we identify conditions under which net N2O consumption occurs.

2. Materials and Methods

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

2.1. Study Sites

[6] We studied five streams in southern Ontario, Canada (Figure 1). We sampled 2–4 sites per stream to help account for high spatial variability that has been reported in past studies [e.g., Reay et al., 2003]. Our sampling started in June 2006, and spans a 29-month period, but with reduced sampling in winter and at some sites. All samples were obtained during daylight. The shortest sampling period (June 2006 to November 2007) was in Stouffville Creek. This is a third order stream (Strahler) in a mixed urban-agricultural catchment. A flood control reservoir is located upstream of our sampling sites. The remaining four streams were sampled from June 2006 through the winter of 2007–2008. In a subset of sites, we extended our sampling until October 2008. The 2008 data, while not reflecting our full suite of sites, allow us to contrast between conditions in unusually dry years (2006, 2007) and a year of record precipitation (2008).

image

Figure 1. Study catchments in Ontario, Canada (area of detailed map is shown as inset). Clockwise from bottom catchments are Stouffville Creek, Black River, Layton Creek (smaller adjacent to Mariposa), Mariposa Brook (larger adjacent to Layton), and Jackson Creek. Sampling sites are shown as white dots with the most upstream site as 1, and increasing numbers downstream. The approximate overall direction of flow is indicated with an arrow.

Download figure to PowerPoint

[7] In addition to the urban-agricultural stream which has a mixture of sand and rocky substrates and a mean depth of 0.2 m, our study sites include Layton Creek (stream-wetland complex with a highly organic substrate, second to third order at our sampling sites, depth = 0.4 m), the Black River (sandy substrate, fourth to fifth order, depth = 0.3 m), Mariposa Brook (silt-rocky substrate fourth order, depth = 0.4 m) and Jackson Creek (sandy substrate upstream, rocky mid and downstream, third to fourth order, depth = 0.4 m). The Jackson Creek sites include one site upstream of a major wetland complex, and two that are located at different distances downstream of the wetland.

2.2. Physicochemical Parameters

[8] We measured a variety of chemical parameters including ammonium (NH4+), organic N, dissolved organic carbon (DOC) and sulphate (SO42−) concentrations (Table 2). In all cases, we have a measurement of NO3 + nitrite (NO2) which we abbreviate as NO2+3. Starting in May 2007 we also measured NO2 separately. N2O samples were obtained in serum bottles sealed with a butyl-rubber stopper, and preserved with mercuric chloride (Table 2). Concentrations were determined using GC-ECD on a Varian CP-3800 gas chromatograph following headspace equilibration. Headspace concentrations were standardized using a series of certified standards (100, 500, 1000 and 10,000 ppbv; Matheson Tri-gas, Praxair), then concentrations in water and percent saturation were determined using solubility equations [Weiss and Price, 1980]. During the ice-free season, discharge was measured using the velocity-area method and a Swoffer 2100 velocity meter. Periods of ice cover were based on direct observation, and are estimated to be accurate within one week.

Table 2. Analytical Methods
 Summary of MethodSample Bottles, Treatment
NO2+3 (Nitrate + nitrite)Reduction to nitrite (using heated hydrazine in an alkaline solution catalyzed by Cu2+) and subsequent analysis using the red azo dye method [Ministry of the Environment, 2001b]Filtered (pre-ashed, B-pure leached GFFs) 500 mL PET bottles
Nitrite (NO2)As for NO2+3, but without prior reduction.As above
Nitrate (NO3)NO2+3 minus NO2As above
Ammonium (NH4+)Indophenol blue method (in buffered alkaline solution, [Ministry of the Environment, 2001b]As above
Organic nitrogenAnalysis as total Kjeldahl nitrogen (TKN) minus ammonium nitrogen. TKN involves high temperature digestion with sulphuric acid, hydrochloric acid and potassium sulphate followed by neutralization and analysis using the phenate-hypochlorite method [Ministry of the Environment, 2001a]As above
Dissolved organic carbon (DOC)Oxidative combustion-infrared analysis (Shimadzu TOC-V analyzer)As above
N2OTriplicate equilibrations of subsamples in air-filled Exetainer vials (Labco Ltd., Buckinghamshire England) with known volumes. Samples were shaken for 90 min at 120 rpm (VWR orbital shaker), weighed, and the headspace was analyzed using GC (Varian CP-3800)-ECD following separation on a Hayesep D column.125 mL serum bottles sealed with pre-baked butyl rubber stoppers, preserved with 0.4mL saturated mercuric chloride solution
Sulphate (SO42−)Ion chromatography (Dionex)Unfiltered, 500 mL PET bottles
Chloride (Cl)Ion chromatography (Dionex)Unfiltered, 500 mL PET bottles
Total phosphorus (TP)Ammonium-molybdate-stannous chloride method [Ministry of the Environment, 1994]Glass vials

2.3. Dissolved Gas Fluxes

[9] We estimated gas fluxes using the two-layer model of diffusive gas exchange [Liss and Slater, 1974]. Positive numbers indicate a net flux from the stream to the atmosphere:

  • display math

where k is the gas exchange velocity. This is equal to the gas transfer coefficient (K, in units of d−1) times stream depth. CS is saturation concentration of N2O, and CL is the measured N2O concentration.

[10] Saturation N2O concentrations were calculated using the solubility equations of Weiss and Price [1980], assuming constant atmospheric mixing ratios of 320 ppbv. N2O is well mixed in the atmosphere due to its long lifetime [Stein and Yung, 2003]. The use of a single CL is dependent upon the assumption that the streams are well mixed, and variation in N2O concentrations across the stream channel is negligible. At low flow, variation in N2O concentrations across our study channels (6%) was approximately equal to analytical error (5%).

[11] We measured rates of gas transfer via the addition of a gas tracer (sulphur hexafluoride; analysis by GC-ECD as for N2O) and conservative tracer [Kilpatrick et al., 1987], as well as diel oxygen (O2) curves [Venkiteswaran et al., 2007, 2008]. Briefly the O2 method uses an O2 mass balance model, and iteratively fits the data (in Matlab©) using a defined range of possible rates of photosynthesis, respiration (both 0–5000 mg O2 m−2 h−1) and gas exchange velocities (0.01–0.50 m h−1) to obtain the lowest sums of squares between measured and modeled data. The model was re-run 15 times with different randomly selected starting points for each diel time series. Model results wherer2 between modeled and observed data was >0.80 were averaged to obtain the k value. These multiple results varied by an average of 0.006% (coefficient of variation (CV); maximum CV was 0.02%). Model results constrained by O2 isotopes, and where the model was run without these data varied by <1% to ∼4%. Rates of gas transfer were normalized to stream temperature [Thomann and Mueller, 1987] and N2O using Schmidt number scaling [Wanninkhof, 1992] assuming an exponent of 2/3 which reflects smooth surfaces.

[12] We have at least one measurement of gas transfer velocity per site, with the exception of the downstream site on the Black River, which we assume (based on similar substrate, depth and velocity relationships) to parallel the upstream site. The downstream site on Jackson Creek has different substrate and is shallower than either of the other study sites, so lacking a reliable estimate of gas transfer velocity we exclude this site from flux estimates.

[13] Seasonal changes in stream depth and velocity can contribute to variability in gas exchange. Numerous approaches to address this issue have been used. Beaulieu et al. [2008] developed a gas transfer model in one stream, and applied this model across several streams with similar slope and substrate. In a group of Mexican streams, a measurement of gas transfer velocity in one stream was applied across several streams, and comparisons were made with chamber methods [Harrison and Matson, 2003]. In this study we calibrate gas transfer models to our study streams using an approach based on the ratio of measured to model-predicted gas transfer coefficients [Moog and Jirka, 1998]. The calculation is presented in its original form (equation (2)) using the gas transfer coefficient (K, which is equal to gas exchange velocity divided by depth) [Moog and Jirka, 1998]:

  • display math

where KPC is the calibrated prediction, Km is the measured gas transfer coefficient, Kp is the predicted gas transfer coefficient under measurement conditions and n reflects the number of measurements. We selected the models of O'Connor and Dobbins [1958], Owens et al. [1964], and Bennett and Rathburn [1972], which were developed for conditions similar to those of our study streams [Cox, 2003]. Our flux estimates are based on equation (1), using the three estimated KPC values based on these three models, and 1–4 measurements of gas transfer velocity per stream (Figure 2).

image

Figure 2. Solute concentrations, flow and KPC. Sites are shown from (left) the most upstream site to (right) the most downstream site in a stream. Box shows median ice-free season data bounded by 25th and 75th percentiles. Whiskers show the 10th and 90th percentiles. Dots indicate outliers. White triangles show concentrations from periods of complete or partial ice cover. Data plotted are (a) N2O % saturation (the horizontal dashed line indicates 100% saturation), (b) discharge, (c) N2O concentration, (d) KPC (mean of 3 models, values for N2O at stream temperature); white circles on KPC plot show measured KN2O values (at 20°C), (e–k) nutrients and other solutes. The Mariposa Brook data are truncated in the plot of % saturation. Outliers at the most upstream site (1700% saturation) and most downstream site (1630% saturation) are not shown.

Download figure to PowerPoint

[14] We assessed whether tile drains present in the catchments of three of our study streams (Layton Creek, Black River, and Mariposa Brook) were likely to affect gas concentrations using a first-order gas loss equation [Chapra and di Toro, 1991] to define the length of reach over which 95% of the gases input by a tile drain will be dissipated:

  • display math

where v is stream velocity, KPC is the calibrated gas transfer coefficient, Dist5% is the distance at which 5% of the gases will remain in solution.

[15] Based on this relationship, tile drains in Layton Creek and the Black River are sufficiently far from our sampling sites that they will not affect gas dynamics. Mariposa Brook has numerous tile drains which affect our measured fluxes at all sites. Gas dynamics at the most upstream sampling site of Stouffville Creek are affected by an upstream flood control reservoir. Impacts are also consistently evident at the second site downstream, but not at the two sites located further downstream.

[16] Distance between sampling sites varied, but on average sites were separated by 5 km. Based on criteria described in equation (3), the distance between sites was sufficiently long that the vast majority (>95%) of gases were emitted prior to passage to adjacent downstream sites. However, in Stouffville Creek, gas concentrations at the second site were periodically influenced by the adjacent upstream site.

[17] We estimated time-weighted yearly fluxes for each site and each year. Time-weighted fluxes were calculated by averaging the instantaneous flux at the start and end of a sampling interval, multiplying this measurement by the duration of the sampling interval, then summing results for a 365 day period. 13–22 flux measurements were used in each annual flux estimate. We assumed no emissions occurred during periods of complete ice cover [Macdonald et al., 1991].

[18] We then calculated the mean flux for a stream across years and sites. Annual fluxes were calculated as described for 365 day periods ranging from June 2006–2007, June 2007–2008 and October 2006–2007 and October 2007–2008 for all sites where we had data. Recognizing the overlap in some annual estimates, we took the mean of any overlapping annual periods, and then the mean within a site (across years), then finally, the mean of sites within a stream.

[19] We excluded the third site on Mariposa Brook from our mean stream flux estimates because of concern that high fluxes of N2O are not indicative of the study reach. Due to access limitations at this site we measured gas transfer velocity downstream of our sampling site. While this reflects a real flux, it appears to reflect N2O accumulated in a deeper, slow-flowing upstream section, then degassed as it passes through this more turbulent downstream riffle.

2.4. Statistical Analyses

[20] We used best-subset multiple linear regression (SYSTAT 13.0) to develop models of annual N2O flux based on stream chemistry (NO2+3, NH4+, total phosphorus [TP], DOC), accepting models where ΔAIC was ≤2 (from the minimum AIC).

[21] Next, we compiled N2O flux and NO3 concentration data from all published studies of streams where annual values were reported. Although discharge, depth, velocity and stream order were not reported, all of the streams assessed were small streams. Discharge in these systems (where reported) was typically <2 m3 s−1 [Baulch, 2009]. Where necessary, N2O flux and NO3 concentration data were digitized from figures (using Engauge Digitizer 3.0). Literature data were combined with the results of this study and simple linear regression was used to determine whether N2O fluxes were related to concentrations of NO3. Data were log-transformed if necessary to meet the assumptions of statistical analyses.

[22] In addition to assessing predictors on an annual scale, we report simple correlations between daily fluxes and each variable across study period within our five study streams. As well, to identify predictors of under-ice N2O concentrations, we used best subset multiple linear regression, (Systat 13.0; using NO2+3, NH4+, TP, organic N, SO42− and DOC data) and thresholds noted previously (ΔAIC ≤ 2). Data were restricted to periods where upstream ice coverage was 90–100% complete (average cover was 99%) to restrict the data to periods where gas exchange was zero, or extremely low.

[23] To further explore the issue of N2O consumption in Jackson Creek, we used a regression tree to analyze predictors of N2O % saturation among sites and over time. A regression tree is similar to stepwise regression procedures, but is a nonparametric method that does not assume linear relationships and is well-suited to analyses where interaction effects are anticipated, or relationships may vary through sample space. Data are repeatedly partitioned into two groups, with each split based on the explanatory variable which makes the resulting groups as homogenous as possible [De'ath and Fabricius, 2000]. The analysis results in a graph which can be interpreted in a similar fashion to a dichotomous key. We applied k-fold cross validation (k= 10), and based the selection of tree size on minimum cost criteria (lowest cross-validated relative error), regardless of tree size. We report two metrics of model fit, the cross validated relative error, where 0 indicates perfect fit, and 1 is equivalent to random guessing, and, the re-substitution relative error, which is akin to the coefficient of non-determination (i.e., 1 −r2). The predictive variables available for entry into the model were: NO2+3, NH4+, organic N, DOC, SO42−, and temperature. Statistical analyses were performed using JMP 7.0.2, CART Pro V6.0, Systat 8.0 or Systat 13.0. A level of significance of α = 0.10 was applied.

3. Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

3.1. Daily and Annual N2O Fluxes and Water Chemistry

[24] N2O fluxes varied considerably over time. Consistent seasonal trends were not apparent among streams (Figure 3). Annual N2O fluxes also varied among sites and over years (Figure 4). Jackson Creek showed the lowest annual emissions. Net uptake of N2O as high as 1160 μmol m−2 y−1 was observed for one site in this stream. Mariposa Brook consistently showed the highest annual emissions, averaging 20,000 μmol m−2 y−1 (over all annual windows, excluding third site).

image

Figure 3. (a–f) N2O flux across all study streams and study sites. Error bars show standard deviation based on three models of gas flux. Periods of ice cover are indicated by the white squares at the top of the plots. (Note the upstream site on Stouffville Creek was never ice covered.)

Download figure to PowerPoint

image

Figure 4. Mean flux (tabulated over annual windows) as a function of nitrate concentration during different annual periods. Error bars represent standard deviation based on predictions of three calibrated gas transfer models. In some cases error bars are smaller than symbols, hence are not visible. Data from the third site on Mariposa Brook are not plotted. These data would lie at 136 μM NO2+3, 776 μmol N2O m−2d−1 (spring 06–07) and 122 μM NO2+3, 679 μmol N2O m−2d−1 (fall 06–07).

Download figure to PowerPoint

[25] Interestingly, N2O fluxes from some streams in the extremely wet year (2008) appeared to differ from those in 2006–2007. One stream (Layton Creek) transitioned from a N2O source in 2006–2007 to a net sink of N2O during periods of 2008 (Figure 3). Jackson Creek became a stronger N2O sink than it had been at similar times in previous years (Figure 3). In contrast, another stream (Mariposa Brook) showed somewhat higher N2O fluxes in the wet year (2008).

[26] Nitrate concentrations varied broadly both within and among streams, and over time (Figure 2). Nitrate (NO3) and NO3 + NO2 were highly correlated (r > 0.999), and NO2 never composed a large portion of the NO2+3 pool. Ammonium concentrations were typically lower than nitrate concentrations and varied less among sites. DOC concentrations were typically higher at sites downstream of wetlands. SO42− concentrations followed the reverse pattern.

[27] Best subset regression results for annual N2O flux data suggest several alternative models where the ΔAIC is ≤2 (from the minimum AIC; Table 3). We report linear regression results for the simplest model which uses only NO2+3 as a predictor (Table 4), to allow comparison to global data, where available data do not include all parameters across all supported models. The predictive capability of this model (r2adjusted = 0.59) is equal to the lowest AIC model which is more complex (incorporating NO2+3, TP and DOC; r2adjusted = 0.59), and similar to another alternative model which incorporates NH4+ and NO2+3 (r2adjusted = 0.58; Table 3). The relationship between NO2+3and (log-transformed) N2O flux is also supported if we include the site from Mariposa Brook where we believe fluxes exceed true values at the reach scale (r2adjusted = 0.80 for 5 streams), and exclude Stouffville Creek sites where gas concentrations are influenced by the reservoir (r2adjusted= 0.87 for 5 streams). If we extend our analysis to include all streams for which there are published annual data, we find a significant positive relationship between log-transformed NO3concentrations and log-transformed N2O flux (Figure 5, Table 4, r2adjusted = 0.39). Exclusion of an outlier noted by Beaulieu et al. [2008] yields a much stronger relationship (Table 4, r2adjusted = 0.61).

image

Figure 5. Relationship between annual average nitrate concentration and N2O flux from all low order streams and canals for which there are long-term estimates. Note log-log scale. Some data reflect NO2+3 rather than NO3 alone. Data exclude periods of ice cover. Data from Hasegawa et al. [2000], Harrison [2002], Harrison and Matson [2003], Beaulieu et al. [2008], and this study.

Download figure to PowerPoint

Table 3. Results of Best Subset Regression for Variable Selection to Predict Under-Ice N2O % Saturation and Annual N2O Fluxes
VariablesAICr2adjusted
Annual Flux Data
TP, DOC, NO2+310.20.59
NO2+311.70.59
NH4+, NO2+311.80.58
DOC, NO2+312.20.54
TP, NH4+, NO2+312.20.38
 
Under-Ice Data
DOC, NO2+3159.30.91
DOC, SO42−, NO2+3159.80.91
TP, DOC, SO42−, NO2+3160.20.91
SO42−, NO2+3160.50.90
Organic N, SO42−, NO2+3160.80.90
NO2+3161.20.89
Table 4. Results of Regression Analyses Assessing Predictors of Annual N2O Flux Among Streams, and Under-Ice N2O Concentrationsa
AnalysisResulting EquationModel F-Ratior2adjustedNp
  • a

    Results of a simple regression analysis for all small streams for which there are published annual flux data are also presented. NO2+3 is in μM, flux is μmol N2O m−2 y−1. Data are shown in Figures 5 and 6.

This Study
Annual N2O fluxes, all streams (this study)Log (N2O flux) = 2.40 + 0.02[NO2+3]6.70.5950.08
Under-ice N2O % saturationN2O % saturation = 136 + 1.43[NO2+3]1120.8915<0.0001
 
20 Low Order Streams With Annual Flux Estimates
All streamsLog (N2O flux) = 2.34 + 0.948 ⋅ log [NO2+3]13.00.39200.002
All streams excluding outlier noted by Beaulieu et al. [2008]Log (N2O flux) = 1.38 + 1.42 ⋅ log [NO2+3]28.80.6119<0.0001

[28] None of the variables we tested showed consistent correlations with daily N2O flux. NO2+3 (or NO3) was the strongest correlate of daily N2O flux across most streams while NO2, NH4+, DOC and organic N showed significant correlations in some cases (Table 5).

Table 5. Correlation Coefficient, r, Between Water Chemistry, Temperature and N2Oa
 NO2+3NO2NO3NH4+Organic NDOCTPSO42−Temperature
  • a

    The direction of the relationship is indicated first. Flux data are log or log + 25 transformed. Where correlates are transformed this is indicated by a L. Blank cells indicate r2 < 0.10. This is higher than the threshold for statistical significance (at p = 0.05), but indicative of very weak associations. NM means this relationship was not measured. (Under ice NO2 was consistently unmeasurable and temperature was 0°C.)

Flux Data (Ice-Free Season)
N2O areal flux all streams+0.52 L +0.57 L      
N2O areal flux Black River  +0.49      
N2O areal flux Jackson Creek     −0.50   
N2O areal flux Layton Creek+0.38L +0.49 L      
N2O areal flux Mariposa Brook+0.44 L +0.68 L +0.38+0.45   
N2O areal flux Stouffville Creek +0.57 L +0.41 L     
 
Percent Saturation Data
N2O % saturation all streams during ice cover+0.95NM+0.95 +0.75  +0.61NM
N2O % saturation Jackson Creek ice-free season+0.73 L +0.83 L+0.45 L−0.33−0.42 +0.46L−0.61

3.2. Under Ice

[29] The highest N2O % saturation was frequently observed during periods of partial or complete ice cover, and the highest NO3 concentrations were almost uniformly observed under ice (Figures 2 and 6). Winter NH4+ concentrations were also in excess of median values. Patterns of other solutes were less distinct, with the exception of SO42−, which was frequently near its maximum during periods of ice cover (Figure 2). Organic N and SO42−showed significant positive correlations with under-ice N2O % saturation (Table 5). Similar to the annual flux data, best subset regression results for the under-ice data also showed several alternative models were supported (ΔAIC ≤ 2), all of which included NO2+3 (Table 3). We present the simplest model, which has an r2adjusted of 0.89 (Tables 3 and 4) and uses NO2+3 as the sole predictor (Table 4).

image

Figure 6. Relationship between N2O % saturation and nitrate concentrations under ice (upstream ice coverage of 90% or greater). Error bars show ±1 standard deviation.

Download figure to PowerPoint

3.3. Conditions Favoring N2O Consumption

[30] Results of our regression tree analysis suggest that Jackson Creek functions as a N2O sink at NO2+3 concentrations of less than 2.7 μM (Figure 7). While temperatures above 3.1°C are also indicated, low temperatures and low NO2+3concentrations never co-occurred. The regression tree also indicates that the degree of under-saturation is greatest at high concentrations of DOC. NO2+3 enters into the regression tree a second time, predicting lower saturation at NO2+3 concentrations of less than or equal to 1 μM, with the model showing good fit (Figure 7).

image

Figure 7. Regression tree describing predictors of N2O % saturation within Jackson Creek (cross-validated relative error [CVRE] 0.46 ± 0.12; re-substitution relative error [RRE] 0.18). The leaves or terminal nodes indicate mean % saturation ± standard deviation followed by the number of data points in the node. Nitrate and DOC are reported asμM, temperature as °C.

Download figure to PowerPoint

[31] Assessing the whole suite of data from all streams, we find a few exceptions to the proposed 2.7 μM cut-off. If we assess all 29 cases where NO3 concentrations were below 2.7 μM, N2O supersaturation was observed in three cases (112–181% saturation), all of which were at the downstream site of Jackson Creek. If we instead assess all 23 cases where saturation was <85% (this threshold is set to help restrict the effect of diel temperature variability), we find four cases with low saturation at NO2+3 concentrations above 2.7 μM. This occurs once in Stouffville Creek (upstream site; NO2+3 = 35 μM, 79% saturation), once in Layton Creek (NO2+3 = 8.2 μM, 60% saturation), and twice in Jackson Creek (NO2+3 = 2.9 μM, 80% saturation; NO2+3 = 35 μM; 80% saturation). In addition to variables identified in the regression tree analysis, we note the highest degree of under-saturation was also associated with high temperature, low SO42− and low NH4+ (Figure 8).

image

Figure 8. Environmental conditions associated with low N2O % saturation. NO2+3 concentrations are shown on two plots: (a) all data with N2O < 120% saturation and (b) all data where NO2+3 < 10 μM. (c–f) Plots for other parameters show all data where saturation was less than 120%. No Mariposa Brook data fall on the plots. The dashed line represents equilibrium N2O (100% saturation).

Download figure to PowerPoint

4. Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

4.1. Annual N2O Fluxes

[32] With the addition of these five study streams, annual N2O flux estimates now exist for twenty low order streams and drainage canals worldwide [Beaulieu et al., 2008; Harrison 2002; Harrison and Matson, 2003; Hasegawa et al., 2000]. Shorter-term or seasonal studies exist for a much broader range of streams and rivers systems. While our data set represents the longest-term study, others have focused more upon spatial gradients [Stow et al., 2005], or among-stream variation [Harrison and Matson, 2003; Beaulieu et al., 2008]. Emissions from our study streams were considerably lower than those observed from eutrophic subtropical streams [Harrison and Matson, 2003]; but were similar to emissions from streams in the midwestern United States [Beaulieu et al., 2008].

[33] We saw relatively strong relationships between mean annual NO2+3 concentrations and stream N2O fluxes, and these relationships were supported by inclusion of annual flux estimates from the broader literature. The relationship between NO2+3 and N2O can be explained by either higher rates of denitrification at higher NO3 concentrations [Beaulieu et al., 2010a], higher N2O yields of denitrification at increased NO3 concentrations, or both factors acting in concert [Silvennoinen et al., 2008b]. Although elevated nitrification rates could also contribute to higher concentrations of NO3 and increased N2O production, we performed isotopic analyses of N2O in four of these streams, and results were indicative of N2O produced via denitrification [Baulch, 2009].

[34] If we compare N2O fluxes to fluxes from agricultural lands growing the dominant crops in our study catchments (corn, soy, wheat and alfalfa) (http://www.statcan.gc.ca/pub/95-629-x/2007000/4123856-eng.htm) we find that areal emissions from agricultural lands are similar in magnitude, or greater than emissions from our study streams [Drury et al., 2008; Wagner-Riddle et al., 1997]. When the small area covered by streams and much larger area in crops is considered, N2O fluxes from our study streams are a small component of catchment N2O budgets [Baulch, 2009], as has been shown for streams and rivers in New Zealand and the United States [Cole and Caraco, 2001; Wilcock and Sorrell, 2008].

[35] Nations that are signatories to the UN Framework Convention on Climate Change are responsible for estimating their greenhouse gas emissions, including the emissions of N2O from streams resulting from agricultural activity [IPCC, 2006]. The method to tabulate emissions developed by the IPCC estimates emissions from rivers based on simple assumptions regarding nitrogen cycling and N2O yields [IPCC, 1996; Seitzinger and Kroeze, 1998]. However, these assumptions may not hold across all rivers, which vary markedly in important factors such as length [Clough et al., 2006]. Emissions from surface drainages (i.e., aquatic sources influenced by groundwater inputs) are estimated based on an empirical relationship between N2O-N and NO3-N [IPCC, 1996, 2006]. This method also has significant limitations. Perhaps most critically, it does not account for gas exchange (therefore can vary over very short distances [Reay et al., 2003]) and it does not correct for saturation N2O [Beaulieu et al., 2008]. We suggest the relationship between mean annual NO3 concentrations and N2O emissions presented here (Table 4; based on streams in Mexico, the midwestern United States, southern Canada and Japan) may provide the foundation for development of an alternative IPCC method. The observed N2O-NO3 relationship could be incorporated into riverine nutrient models [e.g., Whitehead et al., 1998], allowing large-scale estimates of N2O fluxes. These fluxes could then be attributed to different nations based on their contributions to riverine nitrogen loading. More research is necessary to fill in gaps among different stream orders and stream chemical conditions, and expand the geographical range used in tabulating this relationship. Perhaps most importantly, more data are needed from streams which function as hot spots of emissions. Although the log-log relationship we report has significant uncertainty and would require an estimate of stream and river surface area, it may ultimately be an improvement upon the IPCC method. The current IPCC method for aquatic ecosystems has a 50-fold range of uncertainty (based on N2O emissions per kg leached N) [IPCC, 2006], combined with conceptual problems [Reay et al., 2003; Clough et al., 2006; Beaulieu et al., 2008].

4.2. Spatial and Temporal Variation in Stream N2O Fluxes

[36] N2O fluxes were highly variable in space and time, as reported in previous studies [Harrison and Matson, 2003; Reay et al., 2003; Beaulieu et al., 2008]. Even in streams where environmental conditions were quite similar between the sites, fluxes differed markedly (e.g., Layton Creek). Mariposa Brook showed the greatest degree of spatial variability in fluxes, possibly due to the influence of tile drains.

[37] Fluxes from Stouffville Creek were much lower at sites where gas dynamics were not influenced by the upstream reservoir. The degree of temporal variability is greatest in Stouffville Creek, with high fluxes associated either with high flow following heavy rainfall or occurring during winter (at the site that was not ice covered). In each of these streams, a bias toward summer sampling would have yielded quite different flux estimates. For example, sampling during June–September would have overestimated annual fluxes from Layton Creek by 38–77% (based on different annual time windows), by 18–75% from Black River, and by 6–83% from Mariposa Brook. These findings highlight the importance of long-term sampling, and the potential for bias associated with summer-only flux estimation. In Stouffville Creek, the direction of bias resulting from summer sampling varied and could yield either overestimates or underestimates of annual flux. In Jackson Creek, N2O uptake was consistently highest in summertime.

[38] While on an annual scale NO3 is a reasonable predictor of N2O dynamics, shorter-term patterns are clearly more complicated. Although NO2+3 was frequently correlated with daily N2O fluxes, it was not a significant correlate in all cases, and in many cases explained only a small proportion of variability in daily N2O fluxes. This contrasts with past results from two river systems, where NO3 explained 52–68% of variation in N2O emissions over time [McMahon and Dennehy, 1999; Stow et al., 2005]. However, the poorer relationships we report are commonly observed, particularly in areas affected by groundwater N2O inputs [Clough et al., 2006]. Other variables including ammonium and nitrite show significant correlations in some of our systems. Although nitrite is only ever present in low concentrations, it explains a significant amount of the variation in daily N2O fluxes in Stouffville Creek (Table 5). Nitrite addition experiments have been shown to yield higher N2O production ratios than the addition of NO3, and strong correlations between N2O fluxes and nitrite concentrations have been shown within estuaries [Dong et al., 2002]. Ammonium was also a significant correlate in one of our study streams, but was the best single predictor of N2O fluxes in a group of small eutrophic Mexican streams [Harrison and Matson, 2003]. In our study, the range of ammonium concentrations is relatively small. As a result, if ammonium is an important predictor in streams, this might not be evident in our data. The reason that annual fluxes are more predictable than daily fluxes in these data is not clear, but is likely explained, at least in part, by the aggregation of large amounts of data in annual flux estimates which may minimize the influence of error in daily flux measurements and the effect of factors such as diel variation and spatial variation in N2O and N chemistry.

4.3. Under Ice N2O

[39] Periods of ice cover allowed us to assess gas dynamics in the absence of significant atmospheric gas exchange. As in the annual data, NO2+3 emerged as an important predictor of N2O under ice. While to our knowledge under-ice N2O concentrations have not been measured in streams, these findings are consistent with studies showing strong relationships between N2O and NO3 in groundwater [e.g., LaMontagne et al., 2003], which is thought to be the major source of streamflow under-ice during mid-winter [Cunjak et al., 1998]. In addition to groundwater N2O inputs, N2O production may still occur within streams during periods of ice cover. Although rates of microbial N processing are expected to be low, N2O yields may increase at low temperatures (Table 1).

4.4. N2O Consumption

[40] Transient periods of N2O undersaturation and uptake have been observed [Hemond and Duran, 1989; Stow et al., 2005; Beaulieu et al., 2008], even in streams and rivers with high annual loads of dissolved inorganic N [McMahon and Dennehy, 1999; Rajkumar et al., 2008]. However, Jackson Creek is the first system to be shown function as a net N2O sink on an annual basis.

[41] N2O undersaturation in streams often reflects the ability of denitrifiers to use free N2O as an electron acceptor. Periods of net N2O consumption have been associated with low ambient NO3 concentrations, low O2 concentrations, low flow, or some combination of these factors [Richey et al., 1988; Hemond and Duran, 1989; LaMontagne et al., 2003; Stow et al., 2005; Rajkumar et al., 2008]. In soils, where N2O consumption is more widely documented, factors including temperatures greater than 5°C and low O2 (or saturation and porosity that drive low O2) have been associated with net N2O consumption [Blackmer et al., 1980; Chapuis-Lardy et al., 2007]. In reservoirs, low O2, low NO3 and high organic carbon availability have been cited as factors associated with net N2O consumption [Hendzel et al., 2005]. This is consistent with the fact that N2O-reductase activity can be inhibited by the presence of O2 [Cavigelli and Robertson, 2001]. However, the clearest link to N2O consumption, as in our study, is low concentrations of inorganic N [Chapuis-Lardy et al., 2007].

[42] Our data are indicative of a threshold NO2+3 concentration of approximately 2.7 μM, below which net N2O consumption is often observed. We suggest that anthropogenic N enrichment may have reduced the frequency with which streams act as N2O sinks. Background nitrate concentrations in headwaters streams of nearby regions are estimated to be <11 μM [Smith et al., 2003]. If natural NO3 concentrations were at the low range of historic [Livingstone, 1962] and background NO3 estimates [Meybeck, 1982; Smith et al., 2003], our results suggest N enrichment of streams may have changed some streams in this region from net N2O sinks to net N2O sources. Although there are few estimates of N2O flux from low NO3 streams, N2O undersaturation was observed in another low NO2+3 (<0.5 μM) stream [Baulch, 2009]. However, surprisingly high annual emissions of N2O were reported for a low NO3 stream in the midwestern United States (2.1 μM) [Beaulieu et al., 2008], suggesting the threshold we report may not be applicable across all landscapes.

[43] The importance of soil N2O uptake is the focus of considerable attention, with recent research suggesting a large area of the earth's surface is covered by soils that are highly prone to net N2O uptake [Kroeze et al., 2007]. Even if stream N2O uptake is a common occurrence, the importance of this phenomenon to global budgets is likely to be small. Observed areal N2O uptake rates from our study streams are low in comparison to soil uptake rates [Chapuis-Lardy et al., 2007], and streams cover a much smaller area of the landscape than soils. Stream N2O uptake rates are also lower than mean emissions from other study streams.

4.5. Sources of Error

[44] There are several sources of uncertainty in our analyses. Daytime sampling underestimates time-integrated daily fluxes in these streams by approximately 9% [Baulch, 2009]. Error in measurement and modeling of gas transfer is also significant. Gas tracer experiments are typically estimated to be accurate within 6–10% [Hibbs et al., 1998; Moog and Jirka, 1998]. Our results suggest a mean error of ∼6%, but as high as 28%. Error associated with O2-based estimates of gas transfer is similar to tracer-based measurements, or higher [Rosamond et al., 2011]. Where we had multiple measurements of gas transfer at different stream discharges, we found relatively good agreement between measured and modeled gas transfer coefficient in the Black River (1–3% difference between measured and modeled) and Stouffville Creek (1–22%), but poorer results in Jackson Creek (13–44%), Mariposa Brook (30–41%), and Layton Creek (38–42%). However, the overall accuracy of this model-based extrapolation remains uncertain [Moog and Jirka, 1998] because it depends upon the accuracy of the underlying model structure which can be tested only using a large number of measurements of gas transfer velocity. In addition to error associated with estimation of gas transfer velocity, and the diel sampling bias we note, spatial variability in emissions is also large (Figures 3 and 4). As a result, there is a relatively high degree of uncertainty in these and other N2O flux measurements. More frequent, and more spatially intensive sampling could help improve accuracy of flux estimates, as would work to better parameterize gas transfer velocity.

5. Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

[45] Our findings indicate there is a strong link between mean annual NO3 concentrations and N2O fluxes. Although more research is required, this relationship could serve as a foundation for a revised IPCC method to estimate stream N2O emissions, given the relationship is supported across the four geographic regions and 20 streams studied thus far.

[46] Reducing N2O emissions is important to help mitigate climate change, and to help lessen the role of N2O in stratospheric ozone depletion [Ravishankara et al., 2009]. Although reduced NO3 loading to streams would lead to improved water quality, our results indicate that the benefits in terms of reducing stream N2O emissions may be relatively small, due to low emissions from these systems. Instead, efforts to improve N management are likely to lead to the greatest emissions reductions directly from agricultural soils – which at the scale of our study catchments have much larger emissions. Important mitigation gains might also be attained by remediating systems that constitute emissions hot spots, such as those influenced by sewage or tile drains [Reay et al., 2003; Beaulieu et al., 2010b].

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

[47] We would like to thank Jason Venkiteswaran, Richard Elgood, Heather Broadbent, and others for assistance with modeling, field work, and lab analyses. Funding support was provided by an NSERC Discovery grant to P.J.D. and scholarship support of H.M.B.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
gbc1819-sup-0001-t01.txtplain text document2KTab-delimited Table 1.
gbc1819-sup-0002-t02.txtplain text document2KTab-delimited Table 2.
gbc1819-sup-0003-t03.txtplain text document0KTab-delimited Table 3.
gbc1819-sup-0004-t04.txtplain text document1KTab-delimited Table 4.
gbc1819-sup-0005-t05.txtplain text document1KTab-delimited Table 5.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.