Large Model Driven Solar Activity AI Forecaster: A Scalable Dual Data-Model Framework
Jingjing Wang, Pengyu Liang, Tingyu Wang, Ming Li, Yanmei Cui, Siwei Liu, Xin Huang, Xiang Li, Minghui Zhang, Yunshi Zeng, Zhu Cao, Jiekang Feng, Qinghua Hu, Bingxian Luo, Bing Cao
arXiv:2508.06892·astro-ph.SR·Published 2025-08-09
Solar activity drives space weather, affecting Earth's magnetosphere and technological infrastructure, which makes accurate solar flare forecasting critical. Current space weather models under-utilize multi-modal solar data, lack iterative enhancement via expert knowledge, and rely heavily on human forecasters under the Observation-Orientation-Decision-Action (OODA) paradigm. Here we present the "Solar Activity AI Forecaster", a scalable dual data-model driven framework built on foundational models, integrating expert knowledge to autonomously replicate human forecasting tasks with quantifiable outputs. It is implemented in the OODA paradigm and comprises three modules: a Situational Perception Module that generates daily solar situation awareness maps by integrating multi-modal observations; In-Depth Analysis Tools that characterize key solar features (active regions, coronal holes, filaments); and a Flare Prediction Module that forecasts strong flares for the full solar disk and active regions. Executed within a few minutes, the model outperforms or matches human forecasters in generalization across multi-source data, forecast accuracy, and operational efficiency. This work establishes a new paradigm for AI-based space weather forecasting, demonstrating AI's potential to enhance forecast accuracy and efficiency, and paving the way for autonomous operational forecasting systems.
TopicsAstrophysics & Cosmology
Tagsweather-forecasting
arXiv categoriesastro-ph.SR, physics.space-ph
arXiv abstract pagePDF