Improving Short-Term QPFs with a WRF-LETKF Radar Data Assimilation System: OSSEs on Typhoon Morakot (2009)

This study develops a Doppler radar data assimilation system which couples the local ensemble transform Kalman filter (LETKF) with the Weather Research and Forecasting (WRF) model. The benefits of this system to short-term quantitative precipitation forecasts (QPFs) are evaluated with observing system simulation experiments (OSSEs) on Typhoon Morakot (2009), which brought record-breaking rainfall to Taiwan when interacting with Taiwan's topography. The results indicate that the assimilation of radial velocity and reflectivity observations improves three-dimensional winds and rain mixing ratio most significantly because of the direct relations in the observation operator. The patterns of spiral rainbands become more consistent between different ensemble members after the radar data assimilation. The rainfall intensity and distribution during the 6-hour deterministic forecast are also improved, especially for the first 3 hours; however, both the forecasts with and without the radar data assimilation have similar trends driven by larger-scale (synoptic-scale) conditions. Moreover, a series of sensitivity experiments are carried out to develop proper assimilation strategies, and a new mixed-localization method which applies different covariance localizations to different analysis variables is proposed and found to give further QPF improvement in this tropical cyclone case.