Nitrous oxide is a potent greenhouse gas, and it destroys stratospheric ozone 1 . Seventeen per cent of agricultural nitrous oxide emissions come from the production of nitrous oxide in streams, rivers and estuaries 2 , in turn a result of inorganic nitrogen input through leaching, runoff and sewage. The Intergovernmental Panel on Climate Change and global nitrous oxide budgets assume that riverine nitrous oxide emissions increase linearly with dissolved inorganic nitrogen loads, but data are sparse and conflicting 2,3 . Here we report measurements over two years of nitrous oxide emissions in the Grand River, Canada, a seventh-order temperate river that is affected by agricultural runoff and outflow from a waste-water treatment plant. Emissions were disproportionately high in urban areas and during nocturnal summer periods. Moreover, annual emission estimates that are based on dissolved inorganic nitrogen loads overestimated the measured emissions in a wet year and underestimated them in a dry year. We found no correlations of nitrous oxide emissions with nitrate or dissolved inorganic nitrogen, but detected negative correlations with dissolved oxygen, suggesting that nitrate concentrations did not limit emissions. We conclude that future increases in nitrate export to rivers will not necessarily lead to higher nitrous oxide emissions, but more widespread hypoxia most likely will.Nitrous oxide (N 2 O) is responsible for about 9% of global climate forcing 4 but sources have been poorly quantified. Anthropogenic N 2 O is largely produced by microbial metabolism of reactive N from agricultural fertilizers and/or human waste 4 . The dominant pathways are nitrification of ammonium (NH 4 + ) to nitrate (NO 3 − ) and denitrification of NO 3 − to N 2 O and finally N 2 4 . The global N 2 O budget assumes a linear relationship between dissolved inorganic nitrogen (DIN) loads to rivers and riverine N 2 O emissions, estimating global riverine N 2 O emissions at 0.9 Tg y −1 , or about 17% of anthropogenic agricultural N 2 O emissions 2 . However, the global budget indicates an N 2 O increase of 5.4 Tg y −1 (range: −7.5 to 18.7), whereas atmospheric measurements indicate the value is 3.9 Tg y −1 (range: 3.1-4.7; ref. 4). Uncertainty may be caused by paucity and poor quantification of spatial and temporal N 2 O emission data from rivers (Supplementary Table S1). Only one previous estimate of N 2 O emissions includes diel data 5 . To better capture spatial and temporal trends in riverine N 2 O emissons and to examine the DIN/N 2 O relationship, we measured dissolved N 2 O concentrations and calculated emissions from the 300 km length of the seventh-order, highly nutrient-impacted Grand River.The Grand River is the largest Canadian river draining into Lake Erie (watershed area: 6,800 km 2 ; river discharge at mouth: 56 m 3 s −1 ; ref. 6 and Supplementary Fig. S1). Eighty per cent of the catchment is agricultural land 7 and 29 (waste-water treatment plants) WWTPs discharge DIN to the river and its tributaries

Diel (24-h) cycling of dissolved O2 (DO) in rivers is well documented, but evidence for coupled diel changes in DO and nitrogen cycling has only been demonstrated in hypereutrophic systems where DO approaches zero at night. Here, we show diel changes in N2O and DO concentration at several sites across a trophic gradient. Nitrous oxide concentration increased at night at all but one site in spring and summer, even when gas exchange was rapid and minimum water column DO was well above hypoxic conditions. Diel N2O curves were not mirror images of DO curves and were not symmetrical about the mean. Although inter- and intrasite variation was high, N2O peaked around the time of lowest DO at most of the sites. These results suggest that N2O must be measured several times per diel period to characterize curve shape and timing. Nitrous oxide concentration was not significantly correlated with NO3- concentration, contrary to studies in agricultural streams and to the current United Nations Intergovernmental Panel for Climate Change protocols for N2O emission estimation. The strong negative correlation between N2O concentration and daily minimum DO concentration suggested that N2O production was limited by DO. This is consistent with N2O produced by nitrite reduction. The ubiquity of diel N2O cycling suggests that most DO and N2O sampling strategies used in rivers are insufficient to capture natural variability. Ecosystem-level effects of microbial processes, such as denitrification, are sensitive to small changes in redox conditions in the water column even in low-nutrient oxic rivers, suggesting diel cycling of redox-sensitive compounds may exist in many aquatic systems.

One-quarter of anthropogenically produced nitrous oxide (N2O) comes from rivers and estuaries. Countries reporting N2O fluxes from aquatic surfaces under the United Nations Framework Convention on Climate Change typically estimate anthropogenic inorganic nitrogen loading and assume a fraction becomes N2O. However, several studies have not confirmed a linear relationship between dissolved nitrate (NO3-) and river N2O fluxes. We apply recursive partitioning analysis to examine the relationships between N2O flux and NO3-, dissolved oxygen (DO), temperature, land use and surficial geology in the Grand River, Canada, a seventh-order river in an agricultural catchment with substantial urban population. Results suggest that N2O flux is high when hypoxia exists. Temperature, not NO3-, was the primary correlate of N2O flux when hypoxia does not occur suggesting NO3- is not limiting N2O production and further increases in NO3- may not lead to comparable increases in N2O flux. This work indicates that a linear relationship between NO3- and N2O is unlikely to exist in most agricultural and urban impacted river systems. Most N2O is produced during hypoxia so quantifying the extent of hypoxia is a necessary first step to quantifying N2O fluxes in lotic systems. Predicted increases in riverine hypoxia via eutrophication and increased temperature due to climate change may drive nonlinear increases in N2O production.

Many environmental studies require the characterization of a large geographical region using a range of representative sites amenable to intensive study. A systematic approach to selecting study areas can help ensure that an adequate range of the variables of interest is captured. We present a novel method of selecting study sites representing a larger region, in which the region is divided into subregions, which are characterized with relevant independent variables, and displayed in mathematical variable space. Potential study sites are also displayed this way, and selected to cover the range in variables present in the region. The coverage of sites is assessed with the Quality Index, which compares the range and standard deviation of variables among the sites to that of the larger region, and prioritizes sites that are well-distributed (i.e. not clumped) in variable space. We illustrate the method with a case study examining relationships between agricultural land use, physiography and stream phosphorus (P) export, in which we selected several variables representing agricultural P inputs and landscape susceptibility to P loss. A geographic area of 110,000km was represented with 11 study sites with good coverage of four variables representing agricultural P inputs and transport mechanisms taken from commonly-available geospatial datasets. We use a genetic algorithm to select 11 sites with the highest possible QI and compare these, post-hoc, to our sites. This approach reduces subjectivity in site selection, considers practical constraints and easily allows for site reselection if necessary. This site selection approach can easily be adapted to different landscapes and study goals, as we provide an algorithm and computer code to reproduce our approach elsewhere.

External nitrogen (N) inputs originating from human activities act as essential nutrients accumulation in aquatic ecosystems or it is exported elsewhere, where the assimilation capacity is surpassed. This research presents a multi-annual case study of the dissolved inorganic nitrogen (DIN) in an urban river in Ontario (Canada), assessed changes in N downstream of the largest wastewater treatment plant (WTP) in the watershed. Changes in the DIN effluent discharge, in-river concentrations and loads were observed comparing the intra- and inter-annual variability (2010–2013) before, during and after WTP upgrades. These upgrades reduced the ammonium concentration in the river from 0.44 to 0.11 mg N-NH4+/L (year average), but the N load in the effluent increased. In the river, nitrate and ammonium concentrations responded to seasonal variability, being higher during the low temperature (>10 °C) and high flow seasons (spring and spring melt). Among years, changes in the DIN concentration are likely controlled by the effluent to river dilution ratio, which variability resides on the differences in river discharge between years. This suggest that the increasing trend in the DIN concentration and loads are the result of agricultural and urban additions, together with reduced N assimilation, in addition to N loads responding to variable river discharge. Finally, we propose monitoring both concentrations and loads, as they provide answers to different questions for regulatory agencies and water managers, allowing tailored strategies for different purposes, objectives and users.