Improving dynamical-statistical subseasonal precipitation forecasts using deep learning: A case study in Southwest China

Abstract

Subseasonal precipitation forecasting is challenging but critical for water management, energy supply, and disaster prevention. To improve regional subseasonal precipitation prediction, previous studies have proposed a dynamical-statistical projection model (DSPM). In this study, we develop a new method that combines the DSPM and deep learning (DL), called the DL-DSPM. The DSPM is developed using the observed relationships between large-scale atmospheric circulations and regional precipitation, and the dynamical forecasted atmospheric circulations from the European Centre for Medium-Range Weather Forecasts (ECMWF) model. The DL-DSPM improves upon the DSPM by correcting biases in atmospheric circulation forecasts from the ECMWF model using two DL models, namely, residual network and U-Net models. In the case of Southwest China (SWC), DL models can improve atmospheric circulation forecasts at lead times beyond 5 pentads, including large-scale drivers of SWC precipitation variability. The DL-DSPM outperforms the ECMWF model and DSPM forecasts in predicting precipitation anomalies beyond 4 and 5 pentads over most SWC regions, respectively. In addition, the DL-DSPM is more skillful than the ECMWF model and DSPM in predicting extreme precipitation events more than 4 pentads in advance. The successful combination of DL and the DSPM provides a new possible direction for DL applications in subseasonal precipitation forecasting.

Deep learning based sub-seasonal precipitation and streamflow forecasting over the source region of the Yangtze River

“…Despite general improvements of forecasts, they tend to smooth the extreme precipitation at sub-seasonal scales (Baño-Medina et al, 2021;Kim et al, 2022), likely due to insufficient heavy precipitation samples (Chen et al, 2022). Many studies have since introduced more recent variants of CNNs including U-Net (Ni et al, 2023) and SmaAt-UNet (Li et al, 2024), or coupled standard CNNs with different structures, such as Auto-Encoder (Ling et al., 2022), Transformer (Ling et al, 2024), and in particular ResNet, which shows the potential of mitigating the vanishing gradient issue by introducing the residual paths (Nie et al, 2024). Others have attempted to introduce specialized loss functions to balance heavy and light rains, such as the exponentially weighted mean squared error (Ebert-Uphoff et al, 2020) and Dice loss (You et al, 2022).…”