Modeling denitrification in an agricultural catchment in Central New York

Denitrification, the microbially mediated reduction of NO₃⁻ to N₂, NO, or N₂O gas, provides an important ecosystem service by reducing N loads to downstream waters. To incorporate denitrification services into management planning, it is important to quantify the benefit; however, it is difficult to quantify denitrification rates and even harder to extrapolate them spatially and temporally to generate landscape-scale estimates of denitrification. We developed a coupled hydrologic-denitrification model that predicts daily denitrification rates across an agricultural watershed and calibrated it using in situ denitrification measurements and two types of hydrologic observations (streamflow and upland soil moisture). The model fits well with the observed denitrification (R² = 0.77, RMSE = 272 kg N ha⁻¹ yr⁻¹, NRMSE = 0.77), stream discharge (NSE = 0.66) and soil moisture (NSE = 0.77), and quantifies the denitrification ecosystem service provided by the watershed as a whole, as well as by the various land classes. Over the seven year model run, mean annual denitrification rates were 21 kg N ha⁻¹ yr⁻¹ watershed-wide, 47 kg N ha⁻¹ yr⁻¹ in the wetland, 52 kg N ha⁻¹ yr⁻¹ in cropped areas, and 4 kg N ha⁻¹ yr⁻¹ in pastures and forests. Quantification of ecosystem services is an essential prerequisite to accounting for those benefits in management decisions. This study is one step towards addressing the lack of field validated ecosystem service quantification studies. In the future, the model will allow us to further examine the spatial and temporal patterns of denitrification and to explore the implications of changes in management practices or climate.