Station2Radar: query conditioned gaussian splatting for precipitation field

TL;DR

This paper introduces Query-Conditioned Gaussian Splatting (QCGS), a novel framework that fuses weather station and satellite data to generate accurate, real-time precipitation fields with improved efficiency and resolution flexibility.

Contribution

QCGS is the first framework to selectively render precipitation regions by combining a rainfall location proposal network with an implicit neural representation, enhancing precipitation field generation.

Findings

Over 50% RMSE improvement over traditional methods

Efficient real-time precipitation field generation

High performance across multiple scales

Abstract

Precipitation forecasting relies on heterogeneous data. Weather radar is accurate, but coverage is geographically limited and costly to maintain. Weather stations provide accurate but sparse point measurements, while satellites offer dense, high-resolution coverage without direct rainfall retrieval. To overcome these limitations, we propose Query-Conditioned Gaussian Splatting (QCGS), the first framework to fuse automatic weather station (AWS) observations with satellite imagery for generating precipitation fields. Unlike conventional 2D Gaussian splatting, which renders the entire image plane, QCGS selectively renders only queried precipitation regions, avoiding unnecessary computation in non-precipitating areas while preserving sharp precipitation structures. The framework combines a radar point proposal network that identifies rainfall-support locations with an implicit neural representation (INR) network that predicts Gaussian parameters for each point. QCGS enables efficient, resolution-flexible precipitation field generation in real time. Through extensive evaluation with benchmark precipitation products, QCGS demonstrates over 50\% improvement in RMSE compared to conventional gridded precipitation products, and consistently maintains high performance across multiple spatiotemporal scales.