Abstract Best management practices (BMPs) are popular approaches used to improve hydrology and water quality. Uncertainties in BMP effectiveness over time may result in overestimating long-term efficiency in watershed planning strategies. To represent varying long-term BMP effectiveness in hydrologic/water quality models, a high level and forward-looking modeling framework was developed. The components in the framework consist of establishment period efficiency, starting efficiency, efficiency for each storm event, efficiency between maintenance, and efficiency over the life cycle. Combined, they represent long-term efficiency for a specific type of practice and specific environmental concern (runoff/pollutant). An approach for possible implementation of the framework was discussed. The long-term impacts of grass buffer strips (agricultural BMP) and bioretention systems (urban BMP) in reducing total phosphorus were simulated to demonstrate the framework. Data gaps were captured in estimating the long-term performance of the BMPs. A Bayesian method was used to match the simulated distribution of long-term BMP efficiencies with the observed distribution with the assumption that the observed data represented long-term BMP efficiencies. The simulated distribution matched the observed distribution well with only small total predictive uncertainties. With additional data, the same method can be used to further improve the simulation results. The modeling framework and results of this study, which can be adopted in hydrologic/water quality models to better represent long-term BMP effectiveness, can help improve decision support systems for creating long-term stormwater management strategies for watershed management projects.

中文翻译：

---

代表最佳管理实践在解决水文和水质问题方面的长期有效性的建模框架：使用贝叶斯方法开发和演示框架

摘要 最佳管理实践 (BMP) 是用于改善水文和水质的流行方法。随着时间的推移，BMP 有效性的不确定性可能导致高估流域规划策略的长期效率。为了在水文/水质模型中表示不同的长期 BMP 有效性，开发了一个高水平和前瞻性的建模框架。框架中的组成部分包括建立期效率、启动效率、每次风暴事件的效率、维护之间的效率和生命周期内的效率。结合起来，它们代表了特定类型实践和特定环境问题（径流/污染物）的长期效率。讨论了可能实施该框架的方法。模拟草地缓冲带（农业 BMP）和生物滞留系统（城市 BMP）在减少总磷方面的长期影响以证明该框架。在估计 BMP 的长期性能时捕获了数据差距。使用贝叶斯方法将长期 BMP 效率的模拟分布与观察到的分布相匹配，假设观察到的数据代表长期 BMP 效率。模拟分布与观察到的分布很好地匹配，只有很小的总预测不确定性。通过附加数据，可以使用相同的方法进一步改进模拟结果。本研究的建模框架和结果可用于水文/水质模型，以更好地代表 BMP 的长期有效性，