Investigating transformer‐based models for spatial downscaling and correcting biases of near‐surface temperature and wind‐speed forecasts

Abstract

High‐resolution and accurate prediction of near‐surface weather parameters based on numerical weather prediction (NWP) models are essential for many downstream and real‐world applications. Traditional dynamical or statistical downscaling methods are insufficient to derive high‐resolution data from operational NWP forecasts, making it essential to devise new approaches. In recent years, an increasing number of researchers explored the implementations of deep learning (DL) based models for spatial downscaling, motivated by the similarity between the super‐resolution (SR) problem in computer vision (CV) and downscaling. Furthermore, while transformer‐based models have become state‐of‐the‐art models for many SR tasks, they are yet rarely applied for downscaling of weather forecasts or climate projections. This study adapted the transformer‐based models such as SwinIR and Uformer to downscale the temperature at 2 meters (T2m) and wind speed at 10 meters (WS10m) over Eastern Inner Mongolia, encompassing the area from 39.6°N to 46°N latitude and 111.6°E to 118°E longitude. We used the high‐resolution forecast (HRES) data from the European Centre for Medium‐range Weather Forecast (ECMWF) with a spatial resolution of 0.1° as the input and the gridded observation data, China Meteorological Administration (CMA) Land Data Assimilation System (CLDAS), at a spatial resolution of 0.01° as the target. Given that the models use observation data rather than a coarse‐grained version of forecast data as the target, the models accomplish both bias correction and spatial downscaling. The results demonstrate that the performance of the SwinIR and Uformer are superior to that of two convolutional neural networks (CNNs) based models (UNet and RCAN). Additionally, we introduced a novel module to extract features of varying resolutions from the high‐resolution topography data and applied a multi‐scale feature fusion module to merge features of different scales, contributing to further enhancement of the Uformer's performance.This article is protected by copyright. All rights reserved.