Improving regional high resolution hurricane forecasts through data assimilation and diagnostic verification with satellite observations

A research version of high-resolution forecast and data assimilation system that is being developed in HRD at AOML/NOAA is used to study improvements to tropical cyclone (TC) numerical prediction, with satellite observations. The study approach is twofold consisting of : 1) diagnostic analysis of forecast errors by quantitative comparison of the forecast to satellite observations that are sensitive to the thermal and hydrologic state of the atmosphere, including clouds and precipitation and 2) assimilation of these observations into the model initial state to improve initial conditions on the vortex scale. The forecast and assimilation numerical system in this project is HWRFx-EnKF system (Experimental Hurricane Weather Research and Forecasting and Ensemble Kalman Filter system). The development and implementation of this system is partially supported by NOAA's Hurricane Forecast Improvement Project (HFIP). In the first phase of the present project, an observation operator model is developed for transformation of the forecast model data into infrared (IR) and microwave (MW) satellite observation space. This model consists of the Community Radiative Transfer Model (CRTM; supported by the Joint Center for Satellite Data Assimilation) and the Goddard Satellite Data Simulation Unit (SDSU, supported by NASA) and an interface between these models and the forecast model. The satellite observations that are used routinely in the operational observation-based TC analysis such as IR and passive and active MW measurements from geostationary and polar orbiting platforms are used in the study. Results of simulating the satellite observations for Hurricane Bill and Tropical Storm Danny of 2009 and comparisons to the corresponding satellite observations will be presented.