EWE: An Agentic Framework for Extreme Weather Analysis

2025-11-26

AI Summary

Frequent extreme weather events exacerbate global climate risks, yet conventional expert-driven diagnostic paradigms suffer from low efficiency and hinder scientific progress. To address this, we propose the first AI agent framework specifically designed for extreme weather analysis, integrating knowledge-guided task planning, closed-loop reasoning, and a customized meteorological toolkit to enable end-to-end automated diagnosis—from heterogeneous multi-source data ingestion to multimodal visualization and scientifically grounded interpretation. We introduce the field’s first benchmark dataset and a stepwise evaluation metric suite. Experiments across 103 high-impact events demonstrate substantial improvements in diagnostic efficiency and reproducibility. Our framework delivers a low-cost, scalable meteorological intelligence capability—particularly beneficial for developing countries—and advances weather diagnostics toward automated scientific discovery.

Technology Category

• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Humans and AI: Brain-Sensing and Analysis
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Extreme weather events pose escalating risks to global society, underscoring the urgent need to unravel their underlying physical mechanisms. Yet the prevailing expert-driven, labor-intensive diagnostic paradigm has created a critical analytical bottleneck, stalling scientific progress. While AI for Earth Science has achieved notable advances in prediction, the equally essential challenge of automated diagnostic reasoning remains largely unexplored. We present the Extreme Weather Expert (EWE), the first intelligent agent framework dedicated to this task. EWE emulates expert workflows through knowledge-guided planning, closed-loop reasoning, and a domain-tailored meteorological toolkit. It autonomously produces and interprets multimodal visualizations from raw meteorological data, enabling comprehensive diagnostic analyses. To catalyze progress, we introduce the first benchmark for this emerging field, comprising a curated dataset of 103 high-impact events and a novel step-wise evaluation metric. EWE marks a step toward automated scientific discovery and offers the potential to democratize expertise and intellectual resources, particularly for developing nations vulnerable to extreme weather.

Problem

• Automating diagnostic reasoning for extreme weather mechanisms
• Overcoming labor-intensive expert-driven analysis bottlenecks
• Developing intelligent agent framework for meteorological data interpretation

Innovation

• Agentic framework emulating expert weather analysis workflows
• Knowledge-guided planning with closed-loop reasoning system
• Autonomous multimodal visualization from raw meteorological data