A Probabilistic Framework for Assessment and Interpretation of Quantitative Precipitation Forecasts from Storm-Scale Models

点击翻译按钮获取中文摘要

9714387 Foufoula The U.S. Weather Research Program (USWRP) is an interagency activity designed to perform and implement the research necessary to improve the delivery of weather services to the nation. Under this Program, the National Science Foundation, the National Oceanic and Atmospheric Administration (NOAA), the National Aeronautics and Space Administration and the Office of Naval Research are jointly evaluating and supporting research of high priority to the USWRP. Over the past decade, a significant efforts have been directed toward research and data collection initiatives that focus on improving numerical weather prediction. These investments have been somewhat successful and state-of-the-art in numerical weather prediction has truly advanced through a deeper understanding and improved parameterizations of the physical processes involved, new and better data assimilation techniques, and greatly improved numerical schemes. Despite this progress, however, precipitation still remains one of the most difficult variables to predict quantitatively in atmospheric models. Also, the inherent uncertainty associated with numerical predictions often makes the interpretation of forecasts difficult and limits their practical utility. The Principal Investigators have proposed an interdisciplinary approach to the problem of predicting precipitation on the storm-scale and interpreting the numerical forecasts. The approach will involve utilization of an advanced mesoscale meteorological model coupled with advanced statistical techniques. The Investigators will address the following three objectives, with emphasis on the use of storm-scale models, initialized using radar and other fine-scale observations, to generate quantitative precipitation forecasts and associated ensembles to be used in applications such as hydrologic modeling and flood prediction. Formulate and study a suite of multiscale statistical measures of forecast performance which can help to detect and correct deficiencies in microphysical parameterizations and obtain forecast assessments at application-appropriate scales, e.g., basin-average rainfall. Develop a probabilistic framework under which forecast improvements and the limits of predictability of current numerical weather prediction (NWP) models can be assessed while explicitly acknowledging forecast uncertainty and dependence on scale. Develop a probabilistic framework for assessing the practical utility of deterministic forecasts for risk-based decision making. Successful completion of this research will potentially open the doors to improved quantitative precipitation forecasting and better utilization/interpretation of numerical forecasts. ***