Evaluation of the benefits of assimilation of Meteosat-8 observations in an NWP system over the Indian Ocean region

This paper describes the impact of direct assimilation of Indian Ocean Data Coverage (IODC) Meteosat-8 SEVIRI Clear Sky Brightness Temperatures (CSBTs) and the Atmospheric Motion Vectors (AMVs) in a 4D-Var framework through a set of Observing System Experiments (OSEs) during May 2018. Meteosat-8 located at 3.5°E was relocated to 41.5°E and fully functional over the IODC region since February 2017. The impact of CSBTs from selected infra-red and water vapor channels of SEVIRI and AMVs derived from visible, infrared and water vapor channels is studied through identical OSEs using National centre for Medium Range Weather Forecast (NCMRWF) Unified Model (NCUM) assimilation and forecast system. The study discovered that the analysis increments of humidity and temperature are largely driven by AMVs rather than the CSBTs over the IODC region. The complementarity of AMV and CSBT assimilation is noticed in the wind analysis. Humidity forecasts from NCUM in the medium range show that the CSBT contribution is less than that of AMVs. CSBT assimilation showed near neutral impact in the temperature forecasts; however, it has contributed significantly to the wind forecast, at different pressure levels. Impact parameter estimated based on European Centre for Medium-Range Weather Forecasts (ECMWF) ERA5 global reanalysis shows that Meteosat-8 AMV assimilation produced noticeable improvement over the Indian Ocean region, particularly in the lower and upper tropospheric levels. Results from this study indicate the importance of wind assimilation over the Tropics, compared to the use of CSBTs. Though the AMV assimilation improved the humidity and wind fields, the statistical skill scores analyzed during the extreme precipitation event associated with two cyclones during the study period show that the precipitation over the IODC region is predicted well when the AMVs are denied from the assimilation.