PREEVENTS Track 2: Collaborative Research: Developing a Framework for Seamless Prediction of Sub-Seasonal to Seasonal Extreme Precipitation Events in the United States

第二轨道：合作研究：开发一个框架，无缝预测的亚季节到季节极端降水事件在美国极端降水是一种自然灾害，对生命，社会和经济构成风险。其影响包括水流湍急造成的死亡率和发病率、受污染的供水、水传播疾病以及水坝倒塌、电力和交通中断、严重侵蚀以及对自然和农业生态系统的破坏。这些影响涉及多个部门，包括水资源管理、能源、基础设施、交通、健康和安全以及农业。然而，在许多决策者规划、准备和建立连续性所需的时间框架上，“亚季节到季节（S2 S; 14到90天）”的预测技能很差，因此不存在足够的预测工具。此外，如果研究人员、预报员和地方或区域决策者之间没有建立双向沟通渠道，就无法提高社会对这些事件的复原力。因此，该项目的目标是加强对S2 S极端降水事件的科学理解，改善其预测，并增加研究和利益相关者社区之间关于此类事件的沟通。总体结果将是预测模型的发展，有可能减少死亡率，发病率和S2 S极端降水事件造成的损害，并扩大参与科学包括联邦，部落和地方利益相关者。三个目标用户群体;水资源管理人员、应急管理人员和部落环境专业人员将通过讲习班参与整个项目期间的工作。知识的共同生产将引导科学专注于对使用和依赖预测的人最重要的有用特征，从而有助于知识共享和提高预测有意义的能力。该项目将加强对美国S2 S极端降水事件的大尺度动力学和强迫的基本理解，并提高模拟和预测此类事件的能力。该项目召集了一个由科学家和利益相关者组成的专家团队，通过回答四个科学和社会相关的研究问题来缩小S2 S极端降水事件的预测差距：1）美国邻近地区S2 S极端降水事件的天气模式和特征是什么？2)大尺度气候变率模式是否调节了这些事件？如果是，如何做到？3)跨时间尺度的S2 S极端降水事件的可预测性如何？以及4）我们如何创建一个信息丰富的预测S2 S极端降水事件优化决策和规划？为了回答这些问题，该项目将首次将联合收割机观测与新型机器学习技术、高分辨率雷达复合材料、动态气候模型（国家多模型Ensemble和耦合模型相互比较项目第5阶段）以及让利益攸关方参与共同生产知识的研讨会相结合。该项目将确定与美国各地S2 S极端降水事件相关的基本天气和气候过程，从小到单个风暴到大到海洋盆地。S2 S极端降水事件的预测技能将通过增加对历史事件的机械理解和对模拟这些事件及其特征模式的模型性能的定量评估来提高。本项目中开发的统计和共同制作框架将具有灵活性，可在全球其他区域的极端气象事件和时间尺度上应用，并与未来的气候模型模拟以及其他利益攸关方社区一起应用，以减少极端气象事件的影响并提高对极端气象事件的复原力。

PREEVENTS Track 2: Collaborative Research: Developing a Framework for Seamless Prediction of Sub-Seasonal to Seasonal Extreme Precipitation Events in the United StatesExtreme precipitation is a natural hazard that poses risks to life, society, and the economy. Impacts include mortality and morbidity from fast-moving water, contaminated water supplies, and waterborne diseases as well as dam failures, power and transportation disruption, severe erosion, and damage to both natural and agro-ecosystems. These impacts span several sectors including water resource management, energy, infrastructure, transportation, health and safety, and agriculture. However, on the timeframe required by many decision makers for planning, preparing, and resilience-building "subseasonal to seasonal (S2S; 14 to 90 days)" forecasts have poor skill and thus adequate tools for prediction do not exist. Additionally, societal resilience to these events cannot be increased without established, two-way communication pathways between researchers, forecasters, and local or regional decision makers. Therefore, the goal of this project is to enhance scientific understanding of S2S extreme precipitation events, improve their prediction, and increase communication between research and stakeholder communities with regard to such events. The overarching results will be the development of predictive models that have the potential to reduce mortality, morbidity, and damages caused by S2S extreme precipitation events and broadening participation in science by including federal, tribal, and local stakeholders. Three targeted user communities; water resource managers, emergency managers, and tribal environmental professionals will engage throughout the project duration via workshops. The co-production of knowledge will steer the science to focus on useful characteristics that matter most to the people who use and rely on predictions, thus contributing to knowledge-sharing and improving the capability to predict what is meaningful.This project will enhance fundamental understanding of the large-scale dynamics and forcing of S2S extreme precipitation events in the U.S. and improve capability to model and predict such events. This project assembles an expert team of scientists and stakeholders to narrow the prediction gap of S2S extreme precipitation events by answering four scientifically and societally relevant research questions: 1) What are the synoptic patterns associated with, and characteristics of, S2S extreme precipitation events in the contiguous U.S.? 2) Do large-scale modes of climate variability modulate these events? If so, how? 3) How predictable are S2S extreme precipitation events across temporal scales? and 4) How do we create an informative prediction of S2S extreme precipitation events optimized for policymaking and planning? To answer these questions, this project will for the first time, combine observations with novel machine-learning techniques, high-resolution radar composites, dynamical climate models (the National Multi-Model Ensemble and the Coupled Model Intercomparison Project phase 5), and workshops that engage stakeholders in the co-production of knowledge. This project will identify the fundamental weather and climate processes that are tied to S2S extreme precipitation events across the U.S. from scales as small as individual storms to those as large as ocean basins. The prediction skill for S2S extreme precipitation events will be improved through an increased mechanistic understanding of historical events and a quantitative evaluation of model performance for simulating these events and their characteristic patterns. The statistical and co-production frameworks developed in this project will have the flexibility to be applied across meteorological extremes and timescales, in other global regions, with future climate model simulations, and with other stakeholder communities to reduce the impact of and increase resilience to extreme meteorological events.