A framework for WRF to WRF-IBM grid nesting to enable multiscale simulations

With advances in computational power, mesoscale models, such as the Weather Research and Forecasting (WRF) model, are often pushed to higher resolutions. As the model’s horizontal resolution is refined, the maximum resolved terrain slope will increase. Because WRF uses a terrain-following coordinate, this increase in resolved terrain slopes introduces additional grid skewness. At high resolutions and over complex terrain, this grid skewness can introduce large numerical errors that require methods, such as the immersed boundary method, to keep the model accurate and stable. Our implementation of the immersed boundary method in the WRF model, WRF-IBM, has proven effective at microscale simulations over complex terrain. WRF-IBM uses a non-conforming grid that extends beneath the model’s terrain. Boundary conditions at the immersed boundary, the terrain, are enforced by introducing a body force term to the governing equations at points directly beneath the immersed boundary. Nesting between a WRF parent grid and a WRF-IBM child grid requires a new framework for initialization and forcing of the child WRF-IBM grid. This framework will enable concurrent multi-scale simulations within the WRF model, improving the accuracy of high-resolution simulations and enabling simulations across a wide range of scales.