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

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Henan, Xiong, Shengwu, Zhang, Yifang, Yin, Wenjie, Zhou, Chen, Zhang, Yuqiang, Duan, Pengfei
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917340465070080
author Wang, Henan
Xiong, Shengwu
Zhang, Yifang
Yin, Wenjie
Zhou, Chen
Zhang, Yuqiang
Duan, Pengfei
author_facet Wang, Henan
Xiong, Shengwu
Zhang, Yifang
Yin, Wenjie
Zhou, Chen
Zhang, Yuqiang
Duan, Pengfei
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2603_13298
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle FusionCast: Enhancing Precipitation Nowcasting with Asymmetric Cross-Modal Fusion and Future Radar Priors
Wang, Henan
Xiong, Shengwu
Zhang, Yifang
Yin, Wenjie
Zhou, Chen
Zhang, Yuqiang
Duan, Pengfei
Machine Learning
Artificial Intelligence
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.
title FusionCast: Enhancing Precipitation Nowcasting with Asymmetric Cross-Modal Fusion and Future Radar Priors
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2603.13298