TianJi:An autonomous AI meteorologist for discovering physical mechanisms in atmospheric science

arXiv cs.AI / 3/31/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper introduces TianJi, an “AI meteorologist” system designed to autonomously uncover physical causal mechanisms in atmospheric science rather than only performing statistical weather prediction.
TianJi uses a large language model–driven multi-agent architecture to conduct literature research, propose scientific hypotheses, and plan verification experiments by driving complex numerical weather models.
The system separates research into a cognitive planning phase (meta-planner creating experimental roadmaps) and an engineering execution phase (specialized worker agents handling data prep, model configuration, and multi-dimensional analysis).
In two atmospheric dynamics test cases (squall-line cold pools and typhoon track deflections), TianJi achieves expert-level end-to-end experiment execution with no human intervention and compresses the research cycle to a few hours.
The authors position TianJi as a shift for AI in Earth system science—from a “black-box predictor” to an “interpretable scientific collaborator” that can assess and explain hypothesis validity from its outputs.

Abstract

Artificial intelligence (AI) has achieved breakthroughs comparable to traditional numerical models in data-driven weather forecasting, yet it remains essentially statistical fitting and struggles to uncover the physical causal mechanisms of the atmosphere. Physics-oriented mechanism research still heavily relies on domain knowledge and cumbersome engineering operations of human scientists, becoming a bottleneck restricting the efficiency of Earth system science exploration. Here, we propose TianJi - the first "AI meteorologist" system capable of autonomously driving complex numerical models to verify physical mechanisms. Powered by a large language model-driven multi-agent architecture, TianJi can autonomously conduct literature research and generate scientific hypotheses. We further decouple scientific research into cognitive planning and engineering execution: the meta-planner interprets hypotheses and devises experimental roadmaps, while a cohort of specialized worker agents collaboratively complete data preparation, model configuration, and multi-dimensional result analysis. In two classic atmospheric dynamic scenarios (squall-line cold pools and typhoon track deflections), TianJi accomplishes expert-level end-to-end experimental operations with zero human intervention, compressing the research cycle to a few hours. It also delivers detailed result analyses and autonomously judges and explains the validity of the hypotheses from outputs. TianJi reveals that the role of AI in Earth system science is transitioning from a "black-box predictor" to an "interpretable scientific collaborator", offering a new paradigm for high-throughput exploration of scientific mechanisms.