Community HydroAgent: Formalizing Forecaster Expertise into Skill-Orchestrated Flood Forecasting Workflows
Abstract
Operational flood forecasting depends on tacit forecaster expertise that is difficult to formalize, audit, and transfer. Although artificial intelligence methods have advanced flood prediction and model-error correction, few studies explicitly represent the tacit expert rules, review checkpoints, and workflow constraints that connect model outputs to operational warning decisions. Here, we propose Community HydroAgent for Flood Forecasting (CHA-Flood), a skill-orchestrated agent framework that embeds Large Language Models (LLMs) into a model-driven flood forecasting workflow, where each skill encodes explicit rules to bound LLM reasoning. We validated its effectiveness using five state-of-the-art LLMs in the South Yamhill River basin (USA). Prior judgment captures observed peak flow and flood volume within 5% tolerance in 10 and 11 out of 14 events, and 5-fold cross-validation over 129 events yields Pearson correlations of 0.62 and 0.84. Building on a high-baseline scheme library (average KGE 0.890), the guided scheme selection further improves KGE by 0.023-0.154. All five tested LLMs successfully execute the CHA-Flood workflow with comparable judgment accuracy (each spanning 40%-80% across cross-validation folds) but with substantial cost differences. Removing only the language-model narrowing lowers median KGE from 0.723-0.753 to 0.662 for all five LLMs, indicating that the gain comes from LLM reasoning rather than from the choice of LLM. CHA-Flood does not aim to replace human forecasters; instead, it translates their tacit expertise into an auditable and reproducible workflow. This skill-orchestrated paradigm demonstrates how explicit rule boundaries can guide LLM reasoning to complement physically based simulation in next-generation flood forecasting.
Related
- “When the Emerald wood dove calls, the earth prepares to drink”: Community elders lived experiences of indigenous weather forecasting in eastern Mount Kenya region
- Graph-based probabilistic earthquake clustering and forecasting in Sumatra
- Cost-aware evaluation of time-series foundation models for urban air-quality and temperature forecasting
- Physically constrained Mamba forecasting supports multiobjective reservoir water quality decisions
- XAI-driven analysis of influencing factors in streamflow-based flood forecasting via fine tuned MTV19Anet
BibTeX
@misc{yang2026community,
title = {{Community HydroAgent: Formalizing Forecaster Expertise into Skill-Orchestrated Flood Forecasting Workflows}},
author = {Qingyi Yang and Siqian Qiu and Bing Li and Xu Shan and Jia Feng and Shunan Zhou and Xudong Zhou and Tiantian Xing and Jiale Guo and Xiaoyi Dong and Gaoyu Liu and Xiaohuan Liu and Haiqing Pu and Qingwen Deng and Xun Zhang and Zhongrun Xiang and Haiyang Qian and Ying Yan and Yongkang Xu and Nuo Lei and Tianlong Jia and Baoying Shan and Carlo De Michele},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2607.23983},
eprint = {2607.23983},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.23983},
}
RIS
TY - UNPB
TI - Community HydroAgent: Formalizing Forecaster Expertise into Skill-Orchestrated Flood Forecasting Workflows
AU - Qingyi Yang
AU - Siqian Qiu
AU - Bing Li
AU - Xu Shan
AU - Jia Feng
AU - Shunan Zhou
AU - Xudong Zhou
AU - Tiantian Xing
AU - Jiale Guo
AU - Xiaoyi Dong
AU - Gaoyu Liu
AU - Xiaohuan Liu
AU - Haiqing Pu
AU - Qingwen Deng
AU - Xun Zhang
AU - Zhongrun Xiang
AU - Haiyang Qian
AU - Ying Yan
AU - Yongkang Xu
AU - Nuo Lei
AU - Tianlong Jia
AU - Baoying Shan
AU - Carlo De Michele
PY - 2026
DA - 2026/11/01
PB - arXiv
DO - 10.48550/arxiv.2607.23983
UR - https://arxiv.org/abs/2607.23983
AB - Operational flood forecasting depends on tacit forecaster expertise that is difficult to formalize, audit, and transfer. Although artificial intelligence methods have advanced flood prediction and model-error correction, few studies explicitly represent the tacit expert rules, review checkpoints, and workflow constraints that connect model outputs to operational warning decisions. Here, we propose Community HydroAgent for Flood Forecasting (CHA-Flood), a skill-orchestrated agent framework that embeds Large Language Models (LLMs) into a model-driven flood forecasting workflow, where each skill encodes explicit rules to bound LLM reasoning. We validated its effectiveness using five state-of-the-art LLMs in the South Yamhill River basin (USA). Prior judgment captures observed peak flow and flood volume within 5% tolerance in 10 and 11 out of 14 events, and 5-fold cross-validation over 129 events yields Pearson correlations of 0.62 and 0.84. Building on a high-baseline scheme library (average KGE 0.890), the guided scheme selection further improves KGE by 0.023-0.154. All five tested LLMs successfully execute the CHA-Flood workflow with comparable judgment accuracy (each spanning 40%-80% across cross-validation folds) but with substantial cost differences. Removing only the language-model narrowing lowers median KGE from 0.723-0.753 to 0.662 for all five LLMs, indicating that the gain comes from LLM reasoning rather than from the choice of LLM. CHA-Flood does not aim to replace human forecasters; instead, it translates their tacit expertise into an auditable and reproducible workflow. This skill-orchestrated paradigm demonstrates how explicit rule boundaries can guide LLM reasoning to complement physically based simulation in next-generation flood forecasting.
ER -