Impact of Assimilating GOES-R Infrared Brightness Temperatures on the Forecast of Tropical Cyclones

An ensemble-based data assimilation system is used to assess the impact of assimilating satellite infrared radiance data in both clear and cloudy skies on the forecast of tropical cyclones. The current study builds on recent improvements in tropical cyclone forecasts via assimilation of high-resolution ground-based and airborne Doppler radar observations that are available around coastlines or for select Atlantic basin storms within reach of reconnaissance aircraft. The new generation geostationary satellite infrared radiance data, including those from the Advanced Himawari-8 Imager (AHI) on the Geostationary Meteorological Satellite that launched in October 2014 and the Advanced Baseline Imager (ABI) on the Geosynchronous Operational Environmental Satellite (GOES) that will be launched in 2016, have or will have near global coverage at all times with high spatial and temporal resolution. GOES-ABI and GMS-AHI will provide two times higher spatial and temporal resolution radiance data than the satellites currently in orbit. Both contain 10 infrared channels with 2 km x 2 km spatial resolution with images produced every 15 minutes. The assimilation of such high-resolution satellite observations in both clear and cloudy skies is challenging given their strong non-linear relationships to the underlying model fields, the general lack of effective quality control on them, the need to apply bias corrections to them, and the necessity of data synthesizing and thinning for their application to regional scale numerical weather prediction. These difficulties are especially relevant to the cloudy radiances. For the current study we couple the Community Radiative Transfer Model (CRTM) to the ensemble Kalman filter (EnKF) data assimilation system developed at Penn State University (PSU) and built around the Weather Research and Forecasting model (WRF). This new framework, together with our assimilation strategies that include superobbing and effective data quality control, enables us to directly assimilate multiple channel brightness temperatures with high temporal and spatial resolution into the EnKF. The impact of assimilating brightness temperatures from these new advanced imagers is assessed through both examining the dynamical covariance between the satellite radiances and the state variables estimated from an ensemble and performing extensive observing system simulation experiments (OSSEs) for Hurricane Karl (2010). Preliminary experiments show promising improvements in hurricane initialization and forecasting via assimilation of these next generation brightness temperatures. Ongoing research also explores the effectiveness of using the successive covariance localization (SCL) technique of Zhang et al. (2009 MWR), the impact of assimilation window length and the observation density, observational error covariance. We will be also exploring the impact of assimilating the satellite radiance in comparison to (and in addition to) other in-situ and remotely sensed observations.