A U-LSGAN Network for Downscaling Mesoscale Outputs to LES Simulation

Sub-kilometer-grid numerical simulations are highly desired for many weather-sensitive applications. A U-LSGAN model that integrates Generative Adversarial Networks with Least-Squared loss in a U-Net framework is developed to downscale 1km × 1km mesoscale model outputs to 200m × 200m large-eddy-simulation (LES) grids for Lushan and surrounding regions in Eastern China, an area characterized by steep mountains and complex land cover. Three key meteorological variables - 2-meter temperature (T2), 2-meter relative humidity (RH2), and 10-meter wind speed (WSPD10) - are downscaled, and the impacts of auxiliary inputs including terrain height, land uses, accumulated precipitation, and surface pressure are evaluated. The model uses MAE (mean absolute error), smooth L1 (smooth MAE), and reduced-weight adversarial losses to optimize the downscaling process between coarse-resolution inputs and fine-scale targets, enabling accurate reproduction of sub-kilometer weather patterns. Results based on 2522 test samples show that U-LSGAN can effectively capture the magnitude and fine-scale structures of all three variables with power spectra closely matching the LES model outputs. In comparison to bilinear interpolation, U-LSGAN achieves dramatic improvements, with MAEs of RH2, T2, and WSPD10 reduced by 36.05%, 43.16%, and 52.02%, respectively.