Overview
Large hydrodynamic modeling domains are often split into smaller, geographically adjacent models. This study presents a dynamic-coupling tool designed to link several such models built with the Environmental Fluid Dynamics Code (EFDC), dynamically transferring time-series data among them at their common boundaries.
Tool Development
The dynamic link was implemented using the Message Passing Interface (MPI) together with a custom coupling code. Development required overcoming several challenges, including porting the code to a Linux environment, implementing MPI, and selecting compiler flags for optimum performance.
Results
The tool was applied to a test case in which three EFDC hydrodynamic models were run together with the dynamic-coupling tool on a Linux cluster. Run times were compared against a sequential run of the three models in a Windows environment. By exploring and finding an optimal combination of Intel Fortran compiler flags, the coupled run achieved a speedup of 8.53.