Self-Supervised Learning with Probabilistic Density Labeling for Rainfall Probability Estimation

TL;DR

This paper introduces SSLPDL, a self-supervised learning approach with probabilistic density labeling to improve rainfall probability estimation from NWP models, especially for extreme weather events.

Contribution

It proposes a novel self-supervised post-processing method with probabilistic density labeling to enhance rainfall forecasts and address class imbalance in extreme weather prediction.

Findings

Outperforms existing precipitation forecasting models in regional post-processing.

Demonstrates competitive extension of forecast lead times.

Effectively handles class imbalance in heavy rain event prediction.

Abstract

Numerical weather prediction (NWP) models are fundamental in meteorology for simulating and forecasting the behavior of various atmospheric variables. The accuracy of precipitation forecasts and the acquisition of sufficient lead time are crucial for preventing hazardous weather events. However, the performance of NWP models is limited by the nonlinear and unpredictable patterns of extreme weather phenomena driven by temporal dynamics. In this regard, we propose a \textbf{S}elf-\textbf{S}upervised \textbf{L}earning with \textbf{P}robabilistic \textbf{D}ensity \textbf{L}abeling (SSLPDL) for estimating rainfall probability by post-processing NWP forecasts. Our post-processing method uses self-supervised learning (SSL) with masked modeling for reconstructing atmospheric physics variables, enabling the model to learn the dependency between variables. The pre-trained encoder is then utilized in transfer learning to a precipitation segmentation task. Furthermore, we introduce a straightforward labeling approach based on probability density to address the class imbalance in extreme weather phenomena like heavy rain events. Experimental results show that SSLPDL surpasses other precipitation forecasting models in regional precipitation post-processing and demonstrates competitive performance in extending forecast lead times. Our code is available at https://github.com/joonha425/SSLPDL