FlowCast : advancing precipitation nowcasting with conditional flow matching

Perrone Ribeiro, Bernardo (avtor), Faganeli Pucer, Jana (avtor)

Radar-based precipitation nowcasting, the task of forecasting short-term precipitation fields from previous radar images, is a critical problem for flood risk management and decision-making. While deep learning has substantially advanced this field, two challenges remain fundamental: the uncertainty of atmospheric dynamics and the efficient modeling of high-dimensional data. Diffusion models have shown strong promise by producing sharp, reliable forecasts, but their iterative sampling process is computationally prohibitive for time-critical applications. We introduce FlowCast, the first end-to-end probabilistic model leveraging Conditional Flow Matching (CFM) as a direct noise-to-data generative framework for precipitation nowcasting. Unlike hybrid approaches, FlowCast learns a direct noise-to-data mapping in a compressed latent space, enabling rapid, high-fidelity sample generation. Our experiments demonstrate that FlowCast establishes a new state-of-the-art in probabilistic performance while also exceeding deterministic baselines in predictive accuracy. A direct comparison further reveals the CFM objective is both more accurate and significantly more efficient than a diffusion objective on the same architecture, maintaining high performance with significantly fewer sampling steps. This work positions CFM as a powerful and practical alternative for high-dimensional spatiotemporal forecasting.

• Jezik: Angleški jezik
• Ključne besede: conditional flow matching, precipitation nowcasting, generative models, spatiotemporal forecasting, machine learning, machine learning in environmental science
• Vrsta gradiva: Drugo
• Tipologija: 1.08 - Objavljeni znanstveni prispevek na konferenci
• Organizacija: FRI - Fakulteta za računalništvo in informatiko
• Status publikacije: Objavljeno
• Različica publikacije: Objavljena publikacija
• Leto izida: 2026
• Št. strani: 21 str.
• PID: 20.500.12556/RUL-183935
• UDK: 004.85:551.578.1
• COBISS.SI-ID: 275895299
• Datum objave v RUL: 22.06.2026

Gradivo je del monografije

• Naslov: The Fourteenth International Conference on Learning Representations : ICLR 2026, Rio de Janeiro, Brazil, Apr. 23, 2026
• Kraj izida: [S. l.
• Založnik: s. n.]
• Leto izida: 2026
• COBISS.SI-ID: 275890435

Licence

• Licenca: CC BY 4.0, Creative Commons Priznanje avtorstva 4.0 Mednarodna
• Povezava: http://creativecommons.org/licenses/by/4.0/deed.sl
• Opis: To je standardna licenca Creative Commons, ki daje uporabnikom največ možnosti za nadaljnjo uporabo dela, pri čemer morajo navesti avtorja.

Sekundarni jezik

• Jezik: Slovenski jezik
• Ključne besede: pogojno ujemanje tokov, zelo kratkoročno napovedovanje padavin, prostorsko-časovno napovedovanje, strojno učenje in znanosti o okolju

Projekti

• Financer: ARIS - Javna agencija za znanstvenoraziskovalno in inovacijsko dejavnost Republike Slovenije
• Številka projekta: P2-0209-2022
• Naslov: Umetna inteligenca in inteligentni sistemi