Uncertainty-aware segmentation for rainfall prediction post processing

TL;DR

This paper develops and compares uncertainty-aware deep learning models, including a novel SDE U-Net, for post-processing precipitation forecasts to improve accuracy and reliability in weather prediction.

Contribution

It introduces a tailored SDE U-Net model for rainfall segmentation and demonstrates its superior performance over existing models in uncertainty quantification.

Findings

Deep learning models outperform baseline NWP forecasts.

SDE U-Net achieves best accuracy-reliability trade-off.

Uncertainty-aware models enhance operational weather forecasting.

Abstract

Accurate precipitation forecasts are crucial for applications such as flood management, agricultural planning, water resource allocation, and weather warnings. Despite advances in numerical weather prediction (NWP) models, they still exhibit significant biases and uncertainties, especially at high spatial and temporal resolutions. To address these limitations, we explore uncertainty-aware deep learning models for post-processing daily cumulative quantitative precipitation forecasts to obtain forecast uncertainties that lead to a better trade-off between accuracy and reliability. Our study compares different state-of-the-art models, and we propose a variant of the well-known SDE-Net, called SDE U-Net, tailored to segmentation problems like ours. We evaluate its performance for both typical and intense precipitation events. Our results show that all deep learning models significantly outperform the average baseline NWP solution, with our implementation of the SDE U-Net showing the best trade-off between accuracy and reliability. Integrating these models, which account for uncertainty, into operational forecasting systems can improve decision-making and preparedness for weather-related events.