Impact Studies of GOES-R Instruments for the TC Core Region Data Assimilation: Assimilating MSG SEVIRI Radiances as GOES-R ABI Proxy Data

In the past, satellite radiances were not directly assimilated in TC vortex scale in either global or regional system. The difficulty of such assimilation is caused by the necessity to capture the strong spatial variability of flow-dependent background error covariance structure near storm center, as well as by nonlinearity of cloud-affected effects at high resolution. Variational-ensemble hybrid methods for data assimilation can address both issues. At CIRA/Colorado State University, the Maximum Likelihood Ensemble Filter (MLEF) is applied to NOAA operational Hurricane WRF (HWRF with NMM core) system to directly assimilate cloud-affected satellite radiances in TC core region. The system components also include the forward components of the Gridpoint Statistical Interpolation (GSI) system and the Community Radiative Transfer Model (CRTM).

To illustrate the promise of the approach, a priority in this study is given to assimilation of high resolution MSG SEVIRI channel 9 (IR 10.8μm) radiances, into the core region of Hurricane Fred (2009). We will first examine the performance of MLEF-HWRF assimilating MSG SEVIRI channel 9 radiances, and then we focus on quantifying information measures with respect to different experimental configurations, and combination with other spectral bands. This study represents the first time that MSG SEVIRI as GOES-R proxy data is successfully assimilated into the TC core region and highlights the importance of Infrared imager radiance assimilation in TC vortex scale for improved TC forecasts and analyses. The results demonstrate that given the high degree of accuracy afforded by modern and future satellite instruments such as MSG SEVIRI, GOES-R Advanced Baseline Imager (ABI), and GOES-R Geostationary Lightning Mapper (GLM), the reliability of direct satellite radiance assimilation no longer depends on instrument calibration and noise as much as on the choice of assimilation method, observation forward models and spectral bands. The hybrid data assimilation method will prepare the NWP user community for merging current satellite data with the next generation satellite missions (i.e., NPP, JPSS and GOES-R) that have critical importance for weather forecasting, natural hazards warning, and many other applications.