FusionCast: Enhancing Precipitation Nowcasting with Asymmetric Cross-Modal Fusion and Future Radar Priors

TL;DR

FusionCast introduces an innovative multimodal fusion framework for precipitation nowcasting, leveraging asymmetric cross-modal fusion and future radar priors to significantly enhance prediction accuracy.

Contribution

The paper presents a novel framework that effectively fuses diverse data modalities using a gate mechanism and incorporates future radar priors for improved nowcasting.

Findings

FusionCast outperforms existing models in accuracy.

The gate mechanism effectively combines multimodal features.

Incorporating future radar priors enhances prediction quality.

Abstract

Deep learning has significantly improved the accuracy of precipitation nowcasting. However, most existing multimodal models typically use simple channel concatenation or interpolation methods for data fusion, which often overlook the feature differences between different modalities. This paper therefore proposes a novel precipitation nowcasting optimisation framework called FusionCast. This framework incorporates three types of data: historical precipitable water vapour (PWV) data derived from global navigation satellite system (GNSS) inversions, historical radar based quantitative precipitation estimation (QPE), and forecasted radar QPE serving as a future prior. The FusionCast model comprises two core modules: the future prior radar QPE processing Module, which forecasts future radar data; and the Radar PWV Fusion (RPF) module, which uses a gate mechanism to efficiently combine features from various sources. Experimental results show that FusionCast significantly improves nowcasting performance.