Machine Learning for Precipitation Nowcasting from Radar Images

TL;DR

This paper applies deep learning, specifically a U-Net convolutional neural network, to high-resolution precipitation nowcasting from radar images, demonstrating improved short-term weather prediction accuracy over traditional models.

Contribution

It introduces a novel application of deep learning for high-resolution precipitation nowcasting, framing it as an image-to-image translation problem using U-Net architecture.

Findings

U-Net outperforms optical flow, persistence, and NOAA HRRR models.

Deep learning enhances short-term precipitation prediction accuracy.

The approach enables high-resolution, one-hour precipitation forecasts.

Abstract

High-resolution nowcasting is an essential tool needed for effective adaptation to climate change, particularly for extreme weather. As Deep Learning (DL) techniques have shown dramatic promise in many domains, including the geosciences, we present an application of DL to the problem of precipitation nowcasting, i.e., high-resolution (1 km x 1 km) short-term (1 hour) predictions of precipitation. We treat forecasting as an image-to-image translation problem and leverage the power of the ubiquitous UNET convolutional neural network. We find this performs favorably when compared to three commonly used models: optical flow, persistence and NOAA's numerical one-hour HRRR nowcasting prediction.