Impact of Improved Initialization on Warm-Season Mesoscale Rainfall Prediction in a Nonhydrostatic Eta Model

这项研究是由美国国家科学基金会/美国国家海洋和大气管理局开展的工作的继续。天气研究计划支持使用NCEP Eta模式的非流体静力版本，具有精细（~ 10公里）水平分辨率，研究改进初始化对美国中部暖季降雨预测的影响。新的模拟将在中西部北部、爱荷华州和明尼苏达州进行，那里有一个密集的地面观测点中网，可以为验证和核实提供广泛的数据集。主要重点将是改进与中尺度强迫相关的风暴初始化和模拟，包括冷池/流出边界、低层大气湿舌和弱锋情况。利用密集的地表中网资料和卫星云资料同化可以改善边界层温度和湿度的初始化。模拟中将包括对陆地-大气相互作用的详细处理。将考虑在推导的rneso-beta初始特征的不确定性范围内的扰动进行综合模拟，以评估模型预测的可信度。改进初始化的一些技术是现成的，而另一些技术将在研究的第二阶段开发或修改。这项工作的独特之处在于，它将利用非常规观测输入，探索精细网格初始化对当地中尺度降水预测的影响。这项研究将通过确定现有的高分辨率初始数据的最有益用途，促进业务预报的未来发展。此外，这项研究将提高我们对中西部上游地区强烈局部风暴发展的理解。

This research is a continuation of work begun with NSF/NOAA-U.S. Weather Research Program support using a nonhydrostatic version of the NCEP Eta model with fine (~ 10 km) horizontal resolution to study the effect of improved initialization on predictions of warm-season rainfall in the central United States. New simulations will be run for cases in the upper Midwest, in Iowa and Minnesota, where there is a dense mesonetwork of surface observing sites exists that can provide extensive data sets for validation and verification. The primary emphasis will be improved initialization and simulation of storms associated with meso-beta forcing, including cold pool/outflow boundaries, lower atmosphere moist tongues and weak front situations.The use of data from the dense surface mesonetwork as well as satellite cloud data assimilation will improve initialization of the temperature and moisture in the boundary layer. A detailed treatment of land-atmosphere interaction will be included in the simulations. Ensembles of simulations will be performed considering perturbations within the range of the uncertainties in the derived rneso-beta initial characteristics to assess model prediction credibility. Some of the techniques for improved initialization are readily available, whereas others will be developed or modified during this second phase of the research. The uniqueness of the effort is that it will explore refined grid initialization impacts on local mesoscale precipitation prediction, using non-routine observational input. This research will facilitate future advances in operational forecasting by establishing the most beneficial uses of available higher resolution initial data. Also, the study will improve our understanding of development of intense local storms in the upper Midwest.