Self-Supervised Pre-Training for Precipitation Post-Processor
This work addresses the challenge of accurately predicting severe precipitation events, which is crucial for preventing hazardous weather impacts, but it appears incremental as it builds on existing deep learning and self-supervised techniques for a specific domain.
The paper tackles the problem of improving precipitation forecasts from numerical weather prediction models by proposing a deep learning-based post-processor that uses self-supervised pre-training on atmospheric data and transfer learning for segmentation tasks, achieving superior performance in experiments on regional NWP.
Obtaining a sufficient forecast lead time for local precipitation is essential in preventing hazardous weather events. Global warming-induced climate change increases the challenge of accurately predicting severe precipitation events, such as heavy rainfall. In this paper, we propose a deep learning-based precipitation post-processor for numerical weather prediction (NWP) models. The precipitation post-processor consists of (i) employing self-supervised pre-training, where the parameters of the encoder are pre-trained on the reconstruction of the masked variables of the atmospheric physics domain; and (ii) conducting transfer learning on precipitation segmentation tasks (the target domain) from the pre-trained encoder. In addition, we introduced a heuristic labeling approach to effectively train class-imbalanced datasets. Our experiments on precipitation correction for regional NWP show that the proposed method outperforms other approaches.