WRF: Collaborative Research: Extended-range forecasts of atmospheric rivers for adaptive management of flood risk, water supply, and environmental flows in California

这项研究将开发方法，以提高人类和环境的水供应稳健性，在加州和类似地区的高度变化的气候。将设计明确考虑极端风暴事件预报不确定性的自适应控制政策。这项工作将明确地为区域风暴路径模式和这些风暴类型的相关预报误差量身定制水系统操作。预期结果包括三个主要的科学进展：1)在不同地点和提前期的冷季降水、温度和洪水的扩展范围预测中，对时空不确定性的表征和建模；2)这种对预测误差结构的改进知识将与水资源控制政策设计的计算方法相结合，以制定对预测不确定性具有鲁棒性的适应性政策；3)确定应如何设计基于预测的控制政策，以应对以十年尺度的干旱和洪水为代表的长期气候不确定性。通过这些成果，这项工作将支持向将最先进的气候信息与加州水务部门的决策相结合的转变，并且这项工作的发现将可转移到具有类似洪水状况的其他半干旱地区。这项工作将在整个项目期间与加州水管理机构的主要利益相关者进行广泛的互动和技术转让，以促进研究成果和方法的转化。为这个项目执行的所有软件和数据分析都将是开源的，并托管在GitHub上，使世界各地的研究人员能够复制和扩展项目的发现。这些软件工具将被设计用于支持台式计算机和高性能计算集群的分析，包括NSF XSEDE资源，以支持一系列决策过程。该团队将推进数据科学和软件设计关键领域的研究生和本科生教育，这些领域在最近的美国国家科学院报告中被确定为科学可重复性和劳动力准备的关键问题。这些教育扩展包括康奈尔大学的pi统计学课程和加州大学戴维斯分校的水资源工程课程，结合了该项目简化的数据分析和建模任务。教育发展将通过ASCE教育任务委员会（ECSTATIC）广泛分享。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。

This research will develop ways to improve the robustness of human and environmental water supplies in California and similar regions with highly variable climates. Adaptive control policies that explicitly account for uncertainty in forecasts of extreme storm events will be designed. This work will explicitly tailor water system operations for regional storm track patterns and associated forecast errors for those storm types. Expected outcomes include three major scientific advancements: 1) characterization and modeling of spatial and temporal uncertainty in extended-range forecasts of cold-season precipitation, temperature, and floods at different locations and lead times 2) this improved knowledge of forecast error structure will be coupled with computational approaches for water resources control policy design to develop adaptive policies that are robust to forecast uncertainty; 3) determination of how forecast-informed control policies should be designed for long-term climate uncertainty represented by decadal-scale droughts and floods. Through these outcomes, this work will support a shift toward integrating state-of-the-art climate information with decision-making in the water sector of California, and findings from this work will be transferable to other semi-arid regions with similar flood regimes. This work will enable extensive interactions and technology transfer with key stakeholders in water management agencies in California throughout the project to promote the translation of research findings and methods into practice. All software and data analysis performed for this project will be open source and hosted on GitHub, enabling researchers around the world to reproduce and extend the project's findings. These software tools will be designed to support analyses on desktop computers as well as high-performance computing clusters, including NSF XSEDE resources, to support a range of decision-making processes. The team will advance graduate and undergraduate education in the key areas of data science and software design, identified in recent National Academy reports as critical issues for scientific reproducibility and workforce preparedness. These education expansions include of the PIs' statistics courses at Cornell and water resources engineering courses at UC Davis, incorporating simplified data analysis and modeling tasks from this project. Educational developments will be shared broadly through the ASCE Task Committee on Education (ECSTATIC).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.