Modeling hydrologic controls on denitrification: sensitivity to parameter uncertainty and landscape representation

This paper presents a general discussion of the interplay between model structure and hydrologic parameters in the context of denitrification estimation using coupled hydro-ecosystem models at a watershed scale. Given the key role played by hydrology in denitrification models, sensitivity analysis of hydrologic parameters is needed to determine both uncertainty in denitrification estimates and to suggest how measured data, such as streamflow, can be effectively used to reduce this uncertainty. This paper contributes to the broad goal of sensitivity analysis by examining the linkage between landscape tessellation, calibration, and the ability of models to capture hot-spot contributions to watershed scale denitrification across a range of N-loading. For a small mid-Atlantic forested watershed, denitrification estimates using RHESSys (regional hydro-ecologic simulation system) are compared across different strategies for calibration and landscape tessellation. Results demonstrate the utility of several potential approaches to account for hydrologically mediated hot-spots within landscapes.