AQUAH: Automatic Quantification and Unified Agent in Hydrology
This work addresses the challenge of streamlining complex environmental modeling for hydrologists and decision makers, representing a novel application of AI agents in a domain-specific context.
The authors tackled the problem of automating hydrologic modeling by developing AQUAH, an end-to-end language-based agent that autonomously retrieves data, configures models, runs simulations, and generates reports from natural-language prompts, with initial experiments showing it can complete cold-start simulations and produce analyst-ready documentation without manual intervention.
We introduce AQUAH, the first end-to-end language-based agent designed specifically for hydrologic modeling. Starting from a simple natural-language prompt (e.g., 'simulate floods for the Little Bighorn basin from 2020 to 2022'), AQUAH autonomously retrieves the required terrain, forcing, and gauge data; configures a hydrologic model; runs the simulation; and generates a self-contained PDF report. The workflow is driven by vision-enabled large language models, which interpret maps and rasters on the fly and steer key decisions such as outlet selection, parameter initialization, and uncertainty commentary. Initial experiments across a range of U.S. basins show that AQUAH can complete cold-start simulations and produce analyst-ready documentation without manual intervention. The results are judged by hydrologists as clear, transparent, and physically plausible. While further calibration and validation are still needed for operational deployment, these early outcomes highlight the promise of LLM-centered, vision-grounded agents to streamline complex environmental modeling and lower the barrier between Earth observation data, physics-based tools, and decision makers.