A New Vertical Grid Nesting Capability for Multiscale Simulations Using the WRF Model

Mesoscale atmospheric models are increasingly used for high resolution (< 3 km) and multiscale simulations. This is accomplished using grid nesting, a procedure where multiple computational domains of increasing resolution are integrated either concurrently or in series. A constraint in the concurrent nesting framework offered by the Weather Research and Forecasting (WRF) model, is that mesh refinement is restricted to the horizontal dimensions. This limitation prevents control of the grid on individual domains, leading to grids with excessive aspect ratios or skewness, and prevents optimization of the vertical grid for each domain. A procedure extending mesh refinement to the vertical direction for concurrent simulation has been developed, and is planned for inclusion in the public WRF v3.8.0 release. Implementation of this new feature is validated through idealized cases, and the benefits of vertical nesting are demonstrated using both mesoscale and large-eddy simulations (LES). Mesoscale simulations of the Terrain-Induced Rotor Experiment (T-REX) show that vertical grid nesting can alleviate numerical errors due to large aspect ratios on coarse grids, while allowing for higher vertical resolution on fine grids. Furthermore, the coarsening of the parent domain does not result in a significant loss of accuracy on the nested domain. LES of neutral boundary-layer flow shows that, by permitting optimal grid aspect ratios on both parent and nested domains, use of vertical nesting yields improved agreement with the theoretical logarithmic velocity profile on both domains. Vertical grid nesting in WRF opens the path forward for multiscale simulations, allowing more accurate simulations spanning a wider range of scales than previously possible.