ASSIMILATION OF RADAR DATA FOR THE PREDICTION OF A SMALL HURRICANE, OGNI, NEAR INDIA COAST USING THE ARPS ASSIMILATION SYSTEM

The Center for Analysis and Prediction of Storms (CAPS) has developed in recent years data assimilation systems suitable for assimilating and other sources of observations. The system consists of several principal components: 1) a program that quality-controls and data preprocessing, 2) an objective analysis system based on the Bratseth successive correction scheme (ADAS), and an alternative three-dimensional variational analysis method (3DVAR), 4) a cloud and hydrometer analysis package for cloud, moisture and adiabactic fields, and 5) a non-hydrostatic forecast model known as the ARPS that completes the data assimilation system. In this study, the system is applied to a small typhoon named OGNI, which formed over Bay of Bengal, India during last week of October 2006. Three experiments are carried out to test the impact of the radar data from Chennai, India. These experiments use, for the ARPS forecasts, respectively, initial and boundary conditions, (1) interpolated from NCEP operational GFS analyses, (2) produced by ADAS and the cloud analysis, and (3) the 3DVAR and cloud analysis. The comparison of the experiments with and without the radar data reveals that typhoon OGNI is reasonably well simulated when radar data are analyzed using either ADAS or 3DVAR. Judging from the 850 hPa vorticity and wind fields, the forecast is improved in terms of the magnitude and direction of wind speed, time and location of typhoon center by the radar data assimilation with 3DVAR and ADAS. The genesis and northward movement of the typhoon is also well simulated with the use of radar data. The timing of the typhoon landfall is, however, too early.