A human-centered modeling approach to simulate best management practices and behaviors under uncertainty to meet water quality guidelines

当人们与环境相互作用时，往往会对水造成危害。总最大日负荷（TMDL）是处理水质问题的重要方法。它设定了水体（如河流）每天可以处理多少污染的限制。水质最佳管理规范（BMPs）是防止污染物进入水体和实现TMDL设定目标的常用方法。这一行动通常需要不同专家之间的合作、数据共享和利益相关者的参与。这个NSF项目旨在提高我们对如何在不同空间尺度下更好地管理水质的理解。关于人与环境相互作用的三个理论将为我们的分析奠定基础。本项目将建模、数据分析和调查相结合，研究农民在不同空间尺度下的BMP实施决策。该项目围绕三个研究任务（RTs）展开。RT1通过定期会议召集项目顾问委员会的利益相关者和专家。研究小组对当地农民进行了访谈和调查，以评估影响研究地区实施bmp的因素。RT2涉及开发一种基于双向耦合代理的水质模型，其中农民是代理。该模型采用基于社会心理学理论的贝叶斯推理方法模拟agent的决策和BMP的实现。RT3利用数值实验来评估未来气候和社会经济情景以及人类行为的不确定性对TMDL目标的影响。这些设想是与项目咨询委员会共同制定的，并在项目总结讲习班上传播。这个与切萨皮克湾实际从业人员的合作项目产生并分享结果，以通知切萨皮克湾计划的科学和技术咨询委员会。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

When people interact with their environment, they often induce harm to the water. Total Maximum Daily Load (TMDL) is a vital method for undertaking water quality issues. It sets limits on how much pollution a water body, like a river, can handle each day. Water quality best management practices (BMPs) are a popular method to prevent pollutants from entering the water bodies and to achieve the goals set by TMDL. This action usually requires cooperation among different experts, data sharing, and engaged stakeholder involvement. This NSF project aims to improve our understanding of how to manage water quality better at different spatial scales. Three theories regarding human-environment interactions will lay the groundwork for our analyses. This project combines modeling, data analysis, and surveys to study farmers’ BMP implementation decisions at different spatial scales.The project is structured around three research tasks (RTs). RT1 engages stakeholders and experts in the project advisory board via regular meetings. The research team interviews and surveys local farmers to evaluate factors influencing the implementation of BMPs in the study area. RT2 involves the development of a two-way coupled agent-based water quality model, in which farmers are agents. The model simulates the agents’ decisions and implementation of BMP using the Bayesian inference method based on socio-psychological theories. RT3 uses numerical experiments to evaluate the effect of future climate and socioeconomic scenarios and human behavior uncertainty on TMDL targets. These scenarios are co-developed with the project advisory board and disseminated in the project wrap-up workshop. This collaborative project with real-world practitioners in the Chesapeake Bay produces and shares results to inform the Scientific and Technical Advisory Committee of the Chesapeake Bay Program.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.