Development and testing of a terrain-based hydrologic model for spatial Hortonian Infiltration and Runoff/On

Efficient numerical simulation of process interactions between infiltration and Hortonian runoff is needed to evaluate patterns of internal state variables and fluxes within a watershed. A fully distributed rainfallerunoff model was developed for event-based studies of spaceetime watershed processes. A routing hierarchy was defined over the watershed using the D-infinity contributing area algorithm. Computation of ponding time was included to handle variable run-on and rainfall intensity. The GreeneAmpt model was adopted to calculate surface infiltration, and the kinematic wave model was used to route Hortonian runoff and channel flow. The model can handle input rainfall, soil parameters, and other properties that vary in space and time. The model was tested first against analytical solutions for idealized overland planes. After a sensitivity analysis to identify the most significant parameters, it was then calibrated and verified using rainfall and streamflow data from the USDA-ARS Walnut Gulch experimental watershed in Arizona, USA. The coefficients of efficiency for runoff volume, peak flow, and time to peak flow with respect to calibration/validation are 0.95/0.65, 0.85/0.09, and 0.69/0.88, respectively. Example applications of the model show its potential for simulating internal states and fluxes. The open-source model is provided for spaceetime simulation and scaling of event-based Hortonian runoff and infiltration. 2007 Elsevier Ltd. All rights reserved.