Deep learning‐based precipitation bias correction approach for Yin–He global spectral model.

In this paper, a data‐driven bias correction approach based on deep learning is proposed, which is appropriate for the Yin–He global spectral model (YHGSM) re‐forecasting. The proposed architecture involves four U‐Net‐based networks estimating the proper bias correction models for YHGSM re‐forecasting that consider as correction factors the geopotential, specific humidity, and vertical velocity on three pressure levels from the YHGSM model. The proposed models are then evaluated for their bias correction capability on the 3‐h cumulative precipitation over the region of China between 15°–54.5° N, and 63°–122.5° E. The results revealed that U‐Net‐based models could reduce the root mean squared error (RMSE) and improve the threat scores (TSs), especially for heavy precipitation and rainstorms.The architecture of U‐Net based model. When the red arrow indicates double convolution, the original U‐Net and Att‐UNet are indicated, respectively, according to whether the attention module is used or not. When the red arrow indicates the residual, it indicates Res‐UNet, and if using the attention module then it is RA‐UNet. [ABSTRACT FROM AUTHOR]