Mapping model structure to catchment structure: Complex observational data synthesis for model assessment

How do we reconcile and fully utilize complex field observations and catchment structure in model specification and assessment? We pursued this grand challenge and developed a new hydrologic model based on the distribution of landscape elements along the stream network to simulate landscape-scale hydrologic connectivity and catchment runoff. We synthesized complex field observations at an experimental catchment to independently assess model consistency with observed catchment processes. The wealth of data available allowed us to assess the model's process consistency and diagnose model structure discrepancies. The selection of the catchment was particularly motivated by the availability of detailed spatial and temporal hydrologic connectivity data (the hydrologic connectivity between hillslopes, riparian areas and the stream network). These detailed observations provided a unique opportunity to assess internal dynamics of the model simulations and directly assess model performance. Our results suggest that model structure informed by catchment morphology and topology and assessed with high frequency observations distilled to landscape connectivity signatures provide strong catchment simulations and opportunity for model structure diagnosis.