SIMULATION OF A DISTRIBUTED FLOOD CONTROL SYSTEM USING A PARALLEL ASYNCHRONOUS SOLVER FOR SYSTEMS OF ODES

摘要

A recently developed parallel asynchronous solver for systems of ordinary differential equations (ODEs) is used to simulate flows along the channels in a river network. In our model, precipitation is applied over the hillslopes adjacent to the river network links and water movement from hill-sope to link and along the river network is represented as a system of ODEs. The numerical solver is based on dense output Runge-Kutta methods that allow for asynchronous integration. A static partition method is used to distribute the workload among different processes, enabling a parallel implementation that capitalizes on a distributed memory system. Communication between processes is performed asynchronously. We illustrate the solver capabilities by integrating flow transport equations for a ~32,000 km~2 river basin subdivided into 574,000 sub-watersheds that are interconnected by the river network. We show that the runtime for an eight month-long simulation forced by 1-km resolution NEXRAD rainfall is completed in under 4 minutes using 64 computing nodes. In addition, we include equations to simulate small reservoirs spread throughout the river network and estimate changes in hydrographs at multiple locations. Our results provide a firm theoretical basis for the concept of distributed flood control systems.