Influences of stream ecosystem respiration on stream network denitrification: Results from a simulation modeling experiment

Denitrification can remove substantial amounts of NO3− from stream networks, but controls on the magnitude and distribution of network-scale denitrification remain poorly understood. Prior research using an empirical relationship between stream [NO3−] and denitrification efficiency to construct a stream network denitrification model suggested that smaller, lower order streams contributed disproportionately to whole-network NO3− removal when NO3− loads were low. We expanded this model by incorporating an empirical relationship between whole-stream aerobic respiration rate and denitrification efficiency, which decreased the simulated sensitivity of NO3− removal to [NO3−] and displayed additional heterogeneity in NO3− removal associated with variation in respiration rates. We explored the sensitivity of the expanded model across a set of scenarios representing variation in stream respiration and NO3−-loading rates. Stream respiration rates used in these scenarios were determined from a theoretical relationship between aerobic respiration rate and stream temperature and were calculated for model scenarios representing warm (high respiration rate) and cool (low respiration rate) conditions. Our results indicated that reach- and network-scale denitrification is apt to be strongly influenced by respiration rates when NO3− supplies are high relative to removal rates. Therefore, the distribution of respiration rates across stream networks likely plays a more important role in determining spatial patterns of denitrification rates than previously described, highlighting a mechanism by which larger (higher order) streams may contribute substantially to whole-network denitrification.