Lake-effect Snow: Understanding Predictability and Dynamics through Ensemble-Based Convective-Permitting Data Assimilation, Modeling, and Sensitivity Analysis

湖效应雪(LES)是冷气团平流到相对温暖的水面上的结果，是美国东半部一些最大的降雪堆积的原因。这些风暴可以产生每小时几英寸(5-10厘米)的强烈降雪率，导致一天中积雪超过一米，并可能伴随着接近零的能见度、强风、严寒甚至雷鸣。该项目将研究美国东北部五大湖区的湖泊效应降水：这是一种将强烈的天气和中尺度强迫与精细对流尺度结构相结合的多尺度现象，给预测带来了挑战。这项研究将有助于多个学生的毕业论文和学位论文，并为数值天气预报和集合数据的有效利用的课堂练习提供材料。NSF赞助的安大略省冬季湖泊效应系统(OWLES)野外活动为评估数据同化技术提供了一个极好的测试环境，因为它拥有丰富的观测数据集，包括探空系统、地面测量、飞机传感器和移动雷达，以及一大群研究基础科学的合作者的丰富、不同的科学兴趣。本项目将实施和比较最先进的四维集合和混合数据同化系统，并评估它们在分析和预报湖泊效应雪方面的相对优势和劣势。冬季天气的雷达产品将被同化，并确定它们对可预测性时间尺度的影响。结果将量化每个观测系统对预报质量的影响，建立湖效应雪的内在和实际可预测性，并评估湖面边界、模式误差、天气和中尺度初始条件及其基本动力的贡献。另外一个更广泛的影响是开发和评估最佳数据同化技术。随着国家迈向国家对流尺度集合，这有可能为行动提供指导。下一代区域业务预报系统将需要千米尺度的对流模式分辨率和快速更新，利用最有效的四维集合和/或混合数据同化技术吸收所有可用的观测数据。改善湖泊效应事件的预测提前时间和准确性将对湖泊效应易发地区的居民产生积极的社会影响。这项研究产生的再分析场是合作者分析湖泊效应雪带的结构和演变以及上游湖泊-大气相互作用作用的重要组成部分。

Lake-effect snow (LES), the result of a cold air mass being advected over relatively warm water, is responsible for some of the heaviest snowfall accumulations in the eastern half of the United States. These storms can generate intense snowfall rates of several inches (5-10 cm) per hour, leading to accumulations of more than a meter of snow in the course of a day, and can be accompanied by near zero visibility, strong winds, bitter cold, and even thunders. This project will investigate lake-effect precipitation over the Great Lakes region of the Northeastern U.S.: a multi-scale phenomena combining strong synoptic and mesoscale forcing with fine convective-scale structures that present a prediction challenge. This research will contribute to the graduate theses and dissertations of multiple students, as well as provide material for classroom exercises on numerical weather prediction and effective use of ensemble data. The NSF-sponsored Ontario Winter Lake-effect Systems (OWLeS) field campaign provides an excellent test environment for the evaluation of data assimilation techniques due to the rich observation dataset, including sounding systems, ground measurements, aircraft sensors, and mobile radars, as well as the rich, diverse scientific interests of a large group of collaborators investigating fundamental science. This project will implement and compare the most advanced four-dimensional ensemble and hybrid data assimilation systems, and evaluate their relative strengths and weaknesses for analysis and prediction of lake-effect snow. Radar products for winter weather will be assimilated, and their impact on timescales of predictability determined. Results will quantify the impacts of each observing system on forecast quality, establish the intrinsic and practical predictability of lake-effect snow, and assess the contributions of the lake surface boundary, model errors, and synoptic and mesoscale initial conditions and their underlying dynamics.An additional broader impact is the development and evaluation of the best data assimilation techniques. This has the potential to provide guidance to operations as the nation moves toward a national convective scale ensemble. The next-generation regional operational prediction systems will require kilometer-scale convective-permitting model resolution and rapid updates ingesting all available observations using the most effective four-dimensional ensemble and/or hybrid data assimilation techniques. Improved forecast lead time and accuracy for lake-effect events will have positive societal impacts on residents of lake-effect prone regions. Reanalysis fields produced by this research are a vital component to the analyses of collaborators investigating the structure and evolution of lake-effect snow bands and the role of upstream lake-atmosphere interactions.