Downscaling from Mesoscale Weather to Neighborhood Environments Using Large-Eddy Simulation

Understanding climate impacts on neighborhoods or even smaller scales is essential for investigating public health in highly heterogeneous urban environments. Given synoptic-scale forcing predicted by community climate models, mesoscale weather models can provide information at horizontal resolutions of a few kilometers. Such resolutions, however, are still too coarse to represent environmental conditions on neighborhood resolutions of tens of meters or smaller experienced by individual residents. The objective of this work is to design a downscaling framework that bridges from mesoscale weather to neighborhood environments. Using mesoscale Weather Research and Forecasting (WRF) with reanalyses prepared for the Baltimore Social-Environmental Collaborative (BSEC), large-eddy simulation (LES) runs are performed in both nested and idealized approaches. The nested WRF-WRF-LES approach provides direct downscaling of horizontal grid spacing from tens of kilometers to tens of meters. The idealized WRF-LES approach enhances the flexibility of testing configuration options, including but not limited to lateral boundary conditions, domain size, grid spacing, and urban canopy parameterization. This framework will enable direct evaluation of modeling results against near-surface weather, flux, and boundary layer profiling observations to be collected by BSEC, and provide high-quality input data to urban microclimate and indoor air quality models.