Impact of Quikscat and AIRS Data Assimilation on Simulations of the Catarina Hurricane

The Catarina Hurricane hit the Southern Brazil coasts in March 2004 was the first recorded hurricane in the South Atlantic basin. A cyclone formed over the continent on 20/21 March. Initially the system moved east/southeastward as typically do extratropical cyclones that form in the rear of frontal systems over southeastern Brazil. On 24/25 March the cyclone slowed down and began the westward displacement, hitting the southern coast of Santa Catarina State in Southern Brazil on 28 March, where it caused much damage with strong winds and large accumulated rainfall amounts. The aim of this study is to evaluate the impact of satellite data assimilation in the Catarina event simulations with the BRAMS model. Different types of remote sensing data were used due to missing observed data available in the system trajectory area. The numerical simulations were performed during the period from 22-28 March 2004, which consisted in exploring the impact through assimilation of QuikScat 10-m winds and AIRS vertical profiles of temperature and relative humidity data. Assimilating the QuikScat estimated wind data the system presented a reduction in its northward deviation compared to the case without data assimilation (i.e., model initialized with the global analysis), but with slower displacement. Furthermore, there was an increase in the system intensity, mainly when satellite data was available in the region. The analysis of the mean apparent heat source (Q1) profile presented larger intensity at 900 hPa on the 22nd-23rd, which favored the surface pressure drop. The assimilation of retrieved AIRS profiles of temperature and humidity showed a system trajectory slightly different compared to the estimated satellite position and previous experiments. However, the model simulated the exact position of the system center by 1200 UTC on the 28th. In addition, the system became more intense, with lower surface pressure than the case without assimilation. The winds were also more intense in the first 72 hours of integration. The mean Q1 profile presented a reduction in the heating rate at various times, though between 00 and 09 UTC on the 23rd there was an increase in the heating rate mainly at 900 hPa, that contributed to intensify the pressure drop (and wind). The rain distribution analysis showed an increase in the contribution of the microphysics rain in the initial days of the system when the system showed greater intensification. The simulations with the remote sensing data assimilation showed significant improvement, primarily concerning the impact of the AIRS data. The vertical profiles of the apparent heat source and moisture sink indicated that the intensity and height of the maximum heating are crucial for the reproduction of the intensity of the Catarina event.