Breakdown of hydrologic patterns upon model coarsening at hillslope scales and implications for experimental design

Spatial sensitivity studies that assess the reliability of simulated patterns are important aspects of the modeling process. To date, limited knowledge exists on the impacts of model coarsening at the hillslope scale or on the threshold resolution below which the fidelity of simulated patterns are no longer reliable. In this study, we utilize the Triangulated Irregular Network (TIN)-based Real-time Integrated Basin Simulator (tRIBS) to investigate how model aggregation leads to the breakdown of spatial hydrologic patterns in a ponderosa pine hillslope parameterized at fine-resolution (∼0.3 m). Results indicate that spatial patterns in soil moisture are controlled by small-scale curvature features at fine resolutions and by larger-scale vegetation patches at coarser resolutions. Model aggregation quickly eradicates curvature features and its spatial control on hydrologic patterns, while the level of coarsening possible in the hillslope still preserves vegetation patchiness. A threshold resolution of ∼10% of the original topographic field is identified through analyses of homogeneity indices, correlation coefficients and spatial errors. Below this resolution, model aggregation leads to unrealistic patterns in soil moisture and a transition from curvature-controlled lateral fluxes to vegetation-mediated vertical fluxes. Based on spatial error analyses, we evaluate the use of the distributed hydrologic model to identify sampling sites that represent the hillslope behavior and minimize the sensitivity to model resolution. Our findings demonstrate that spatial sensitivity occurs within hillslope domains depending on the characteristics of the spatial features that control the hydrologic response.