Storm-Scale Quantitative Precipitation Forecasting Using Advanced Data Assimilation Techniques: Methods, Impacts and Sensitivities

University Of Oklahoma Norman Campus, Norman OK

Investigators

Abstract

Convective storms and associated strong winds and heavy precipitation cause billions of dollars of damage and numerous deaths annually; at the same time, accurate forecasting of severe weather and precipitation amounts are among the most challenging tasks in meteorology. The Principal Investigator will develop and apply advanced techniques and tools for predicting localized precipitation, and will study the impact of assimilated data and the sensitivities associated with initial and boundary conditions and model physics. The project will include detailed analyses of high-resolution numerical simulations in order to understand fundamental physical processes that determine how, when and where convective storms are initiated. The knowledge gained from the process and sensitivity studies will be applied to the design and improvement of data assimilation systems. The research will exploit observations collected in prior field campaigns. Intellectual Merit: Fundamental advances will include a significant improvement in the ability to optimally utilize the huge volume of real-time data from the national Doppler radar (WSR-88D) network for the initialization of high-resolution NWP models and for accurate short-term prediction of severe, high-impact weather. A cost-effective 3D variational analysis system coupled with a complex cloud analysis procedure will be refined and tested. The project is expected to produce, for the first time, analyses of convective systems, together with their environment, using the ensemble Kalman filter method from real observations, that would allow accurate predictions of individual storms for up to several hours. Significant progress will also be made in the fundamental understanding of convective initiation processes. The understanding of forecast sensitivity to initial conditions as well as the associated error growth will provide important guidance for the optimal design and deployment of high-resolution observational networks and can lead to a better understanding of the predictability of convective systems. The work with the slant-path GPS (Global Positioning System) data using an advanced 3DVAR technique is expected to lend further support for the deployment of a national high-density GPS surface receiver network. Broader Impacts: The research will directly address one of the three key research themes of the US Weather Research Program, namely, the improvement of forecasting heavy precipitation and flooding through optimal use of observational data and improved numerical precipitation guidance. This project will provide much needed education and training for graduate students and post-doctoral scientists in the increasingly important areas of advanced data assimilation, numerical weather prediction and ensemble forecasting. The research findings will have a direct path to operations through the group's involvement in the Weather Research and Forecast (WRF) model development and testing, and their work with the version of WRF 3DVAR targeted for operational use at the National Centers for Environmental Prediction of National Weather Service. The research therefore will directly contribute to the understanding of precipitating weather systems and to the improvement of NWP.