AI Agent Zephyrus Aims to Revolutionize Weather and Climate Analysis

Researchers develop natural language AI agent to interact with weather forecasting models and data

Published on Mar. 11, 2026

Got story updates? Submit your updates here. ›

Computer scientists and weather researchers at the University of California San Diego have created an AI agent named Zephyrus that can analyze weather and climate data, and answer questions about it in natural language. The goal is to make it easier for students, scientists, and the public to access and understand critical weather and climate information.

Why it matters

Advances in AI-driven weather forecasting have improved predictions, but analyzing the resulting data remains difficult. Zephyrus aims to bridge the gap between complex weather models and language-based interactions, democratizing access to this crucial scientific data with implications for agriculture, disaster preparedness, transportation, and energy management.

The details

The Zephyrus AI agent can handle language-based queries, translate them into code to interact with weather models and data, and then translate the code-generated answers back into plain language. In initial tests, Zephyrus performed well on simple tasks like finding locations with specific weather conditions or forecasts, but struggled with more complex queries about extreme weather events. Researchers plan to expand Zephyrus' capabilities by using larger training datasets and fine-tuning open-source language models for climate-focused tasks.

  • The research team will present Zephyrus at the 14th International Conference on Learning Representations (ICLR) in Rio de Janeiro from April 23-27, 2026.
  • Researchers plan to use larger training datasets and fine-tune open-source models for climate-focused tasks in the next iteration of the Zephyrus AI agent.

The players

University of California San Diego

The research team that developed the Zephyrus AI agent is from the UC San Diego Jacobs School of Engineering and Scripps Institution of Oceanography.

Duncan Watson-Parris

A study co-author and faculty member at the UC San Diego Scripps Institution of Oceanography.

Rose Yu

A study co-author and faculty member in the UC San Diego Department of Computer Science and Engineering.

Got photos? Submit your photos here. ›

What they’re saying

“Our goal is to increase access to critical data and predictions by lowering the barrier to entry to analyzing these data. We want to increase the speed with which we can reason about multimodal data and learn about the Earth by making it easier for students and young scientists to interact with different datasets.”

— Duncan Watson-Parris, Faculty member, UC San Diego Scripps Institution of Oceanography (Mirage News)

“Our vision is to democratize earth science. Zephyrus is a crucial step toward creating AI co-scientists that dramatically lower the barrier to entry, allowing students and researchers everywhere to access and reason about critical weather and climate data at unprecedented speeds.”

— Rose Yu, Faculty member, UC San Diego Department of Computer Science and Engineering (Mirage News)

What’s next

Researchers plan to use larger training datasets and fine-tune open-source models for climate-focused tasks in the next iteration of the Zephyrus AI agent.

The takeaway

The development of the Zephyrus AI agent represents a significant step towards democratizing access to critical weather and climate data, empowering students, scientists, and the public to better understand and respond to the challenges posed by our changing environment.