Improving the Short-Range Precipitation Forecast of Numerical Weather Prediction through a Deep Learning-Based Mask Approach

Abstract

Due to various technical issues, existing numerical weather prediction (NWP) models often perform poorly at forecasting rainfall in the first several hours. To correct the bias of an NWP model and improve the accuracy of short-range precipitation forecasting, we propose a deep learning-based approach called UNetMask, which combines NWP forecasts with the output of a convolutional neural network called UNet. The UNetMask involves training the UNet on historical data from the NWP model and gridded rainfall observations for 6-hour precipitation forecasting. The overlap of the UNet output and the NWP forecasts at the same rainfall threshold yields a mask. The UNetMask blends the UNet output and the NWP forecasts by taking the maximum between them and passing through the mask, which provides the corrected 6-hour rainfall forecasts. We evaluated UNetMask on a test set and in real-time verification. The results showed that UNetMask outperforms the NWP model in 6-hour precipitation prediction by reducing the FAR and improving CSI scores. Sensitivity tests also showed that different small rainfall thresholds applied to the UNet and the NWP model have different effects on UNetMask’s forecast performance. This study shows that UNetMask is a promising approach for improving rainfall forecasting of NWP models.