A “one stop shop” runoff simulation model blending hydrological data boundaries

While literature offers a plethora of hydrological models, their practical implementation is still hampered by the amount of information needed, either to support calibration or to provide the essential inputs for a physically-based description of underlying pro-cesses. To leverage the diverse data sources that may be available across a river basin at multiple levels of detail, a hybrid simulation scheme is introduced. Its architecture synthesizes the three characteristic schematization levels, i.e., distributed, semi-distributed and lumped. The first involves rainfall-runoff transformations and soil moisture balance processes, which are represented through a modified NRCS-CN scheme on a cell basis. The semi-distributed level involves surface runoff routing over the soil cells and along the river network, via a novel velocity method that accounts for the concept of event-varying time of concentration. Finally, a lumped approach is ap-plied for soil and groundwater processes, by means of conceptual layers receiving the distributed vertical fluxes from the overlying cells to produce interflow and baseflow. The model runs in event-based mode and can easily be extended for continuous simu-lations, to support associated engineering studies. Its overall advantage is the flexibility of acquiring its inputs from easily retrievable spatial information, thus ensuring balance between data requirements, parsimony with respect to parameter setting, and accuracy in the description of process heterogeneity.