Data-driven Precipitation Nowcasting Using Satellite Imagery

TL;DR

This paper introduces the Neural Precipitation Model (NPM), a satellite imagery-based deep learning approach for real-time, high-resolution precipitation nowcasting up to six hours ahead, especially useful for regions lacking ground radar.

Contribution

The paper presents a novel neural network model that leverages geostationary satellite data with positional encoders for improved precipitation prediction, addressing limitations of traditional ground-based systems.

Findings

NPM predicts rainfall with 2 km resolution in real-time.

The model effectively captures seasonal and temporal precipitation patterns.

Experimental results show improved accuracy over existing models.

Abstract

Accurate precipitation forecasting is crucial for early warnings of disasters, such as floods and landslides. Traditional forecasts rely on ground-based radar systems, which are space-constrained and have high maintenance costs. Consequently, most developing countries depend on a global numerical model with low resolution, instead of operating their own radar systems. To mitigate this gap, we propose the Neural Precipitation Model (NPM), which uses global-scale geostationary satellite imagery. NPM predicts precipitation for up to six hours, with an update every hour. We take three key channels to discriminate rain clouds as input: infrared radiation (at a wavelength of 10.5 $\mu m$), upper- (6.3 $\mu m$), and lower- (7.3 $\mu m$) level water vapor channels. Additionally, NPM introduces positional encoders to capture seasonal and temporal patterns, accounting for variations in precipitation. Our experimental results demonstrate that NPM can predict rainfall in real-time with a resolution of 2 km. The code and dataset are available at https://github.com/seominseok0429/Data-driven-Precipitation-Nowcasting-Using-Satellite-Imagery.