Examining the Impact of Geostationary Microwave Observations in a Regional Hurricane OSSE

Weather forecast observing system simulation experiments (OSSEs) are commonly used to evaluate the information contained in a future observing system. Regional OSSEs, which target a specific type of weather event, are used when the goal is evaluation of a dataset with high spatial and/or temporal resolution. This presentation will describe regional weather forecast OSSEs that make use of a real case, hurricane Harvey (2017). The experiments are aimed at determining the impact of frequent in time observations from a new generation of geostationary microwave instruments, the geostationary synthetic thinned aperture radiometer (GeoSTAR). A nature run is produced for hurricane Harvey using a newer and higher resolution version of the forecast model run with different initial and boundary conditions and with different parameterization of physical processes. The control forecast and OSSE are performed using a 60-member EnKF DA system that accommodates rapid (15 minute - 1 hour) updates, and makes use of the high time frequency geostationary observations. This EnKF system has been proven to produce highly realistic forecasts of past tropical cyclone events, and has been validated for use in regional OSSEs. We describe results for the following experiments: A control simulation that assimilates conventional data only. An experiment that assimilates simulated radiances from the current generation GOES imagers. A future-platform experiment that tests the impact of assimilating data from a geostationary microwave sounder, which is being considered as a candidate for use in NOAA's future geostationary observing system We discuss the relative impacts of each source of data on the hurricane forecast, and also the lessons learned for the design and implementation of a regional OSSE system.