Catchment-scale multi-process modeling with local time stepping

摘要: Numerical modeling is an important tool for prediction, analysis and understanding of the dynamics land surface processes. To increase usage impact such tools, it crucial to decrease runtime by increasing computational efficiency. Dynamic processes as water flow are typically described higher-order differential equations. Solving these accurately requires numerical integration over time, where errors depend on time steps taken. Typically, simulation use smallest required in a model’s domain simulate flow. In this paper, we analyze local stepping, catchment-scale flow, infiltration, slope stability landslide runout. scheme, temporal cell specific, allowing higher The implemented scheme works with fully free that synchronized only visualization. We implement method monotonic upwind conservation laws (MUSCL). investigate influence resulting changes computation accuracy hydrology-coupled, flood simulation. Results show stepping can be total variation diminishing (TVD) second-order spatially accurate. Simulation results both 1D dam-break scenarios flash insignificant result, while reduces 50%. Finally, successfully multi-process lands model hydrology, flooding, failure, implementation dynamic could widely increased efficiency without significant loss accuracy.