Data-driven Precipitation Nowcasting Using Satellite Imagery

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 globalscale 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 µm), upper- (6.3 µm), and lower- (7.3 µm) 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 realtime with a resolution of 2 km. The code and dataset are available at https://github.com/seominseok0429/Datadriven-Precipitation-Nowcasting-Using-Satellite-Imagery.