Collaborative Research: What Drives the Most Extreme Rainstorms in the Contiguous United States (US)?

极端降水及其经常引起的山洪暴发每年造成无数人死亡和重大损失。对极端降雨的准确预报仍然难以捉摸，这在一定程度上是由于人们对哪些大气因素能够区分最极端的降水事件与那些产生较强但不那么极端的降水事件的认识不完全。该项目将使用观测、机器学习和数值模型模拟来识别支持这些事件的环境条件的关键差异。该项目的结果将有助于天气预报员更好地识别导致极端降雨的具体条件，从而提高对山洪暴发事件的预警能力。此外，该项目还有一个重要的教育和推广部分，目的是提高公众的科学素养和参与度。极端降水，这里定义为平均复发间隔（ARI）为1000年的降雨事件，在美国几乎所有情况下都会造成重大的社会影响。该项目将侧重于提高对美国大陆最极端降雨事件的理解，以及这些事件与严重但不那么极端的降水的区别。该项目有三个主要目标。研究小组将首先从多个降水数据集中建立一个最极端降雨事件的数据库，重点关注6、24、48和72小时积累期的10年、100年和1000年ARIs的超过情况。对于目标1中确定的事件，研究人员将从大气再分析数据集中收集有关环境条件的数据，并使用自组织地图（一种机器学习技术）来确定哪些大气参数与更多或更少的极端降雨事件相关。在项目的最后一步，将使用云模式1 （CM1）进行理想数值模拟，以测试大气湿度、中尺度上升、垂直风切变和其他由早期分析确定的成分的作用，以确定对中尺度对流系统的影响，以及在美国中部频繁出现的大型雷暴复合体。该项目由物理与动态气象学计划和促进竞争研究的既定计划（EPSCoR）共同资助。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。

Extreme precipitation, and the flash flooding it often causes, is responsible for numerous fatalities and significant damage every year. Accurate forecasts of extreme rainfall remain elusive, and this stems in part from incomplete understanding of what atmospheric factors distinguish the most extreme precipitation events from those that produce heavy, but not as extreme, rainfall. This project will use observations, machine learning, and numerical model simulations to identify key differences in environmental conditions that support these events. The project results should help weather forecasters better identify the specific conditions that lead to extreme rainfall, thereby improving the advanced warning for flash flooding events. The project additionally has a significant education and outreach component with the goal of increased public scientific literacy and engagement.Extreme precipitation, defined here as rainfall events with an average recurrence interval (ARI) of 1000 years, causes significant societal impacts in almost all cases in the United States. This project will focus on improving understanding of the most extreme rainfall events in the continental United States, and what distinguishes these events from heavy, but less extreme precipitation. The project has three main objectives. The research team will initially develop a database of the most extreme rainfall events from multiple precipitation datasets, focusing on the exceedance of the 10-, 100-, and 1000-year ARIs for 6-, 24-, 48-, and 72-hour accumulation periods. For the events identified in Objective 1, the researchers will collect data about the environmental conditions from atmospheric reanalysis datasets and use Self Organizing Maps, a machine learning technique, to identify which atmospheric parameters are associated with more and less extreme rainfall events. In the final step of the project, idealized numerical modeling will be conducted using Cloud Model 1 (CM1) to test the roles of atmospheric moisture, mesoscale ascent, vertical wind shear, and other components identified by the earlier analyses to determine the impacts on mesoscale convective systems, the large thunderstorm complexes that are frequent in the central US. This project is jointly funded by the Physical and Dynamic Meteorology program and the Established Program to Stimulate Competitive Research (EPSCoR).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.