Noah-Agent: Multi-Expert AI Agent Framework for Fortran Climate Models

Noah-Agent is a multi-expert AI agent framework for automated parameterization and validation of large-scale Fortran climate models.

Project Overview

Large Fortran climate model codebases are hard to parameterize and validate by hand. Noah-Agent coordinates multiple specialized agents to automate those steps.

Status

Preprint, version 0.1, released on Zenodo, in preparation.

Approach

• Multiple specialized agents handle parameterization and validation of large Fortran model codebases.
• The framework automates parameter selection and validation against reference behavior.

Relation to Other Work

Noah-Agent connects to the broader Noah-MP land surface modeling effort and to ESM-bench, which evaluates whether AI agents understand Earth System Model physics and code.

Links

• Preprint (Zenodo): https://zenodo.org/records/17862049
• Publication entry: Noah-Agent
• Blog post: Noah-Agent: a multi-expert agent framework for Fortran climate models