Improved Quantitative Precipitation Forecasts by MHS Radiance Data Assimilation with a Newly Added Cloud Detection Algorithm

Abstract

Abstract Satellite microwave humidity sounding data are assimilated through the gridpoint statistical interpolation (GSI) analysis system into the Advanced Research core of the Weather Research and Forecasting (WRF) model (ARW) for a coastal precipitation event. A detailed analysis shows that uses of Microwave Humidity Sounder (MHS) data from both NOAA-18 and MetOp-A results in GSI degraded precipitation threat scores in a 24-h model forecast. The root cause for this degradation is related to the MHS quality control algorithm, which is supposed to remove cloudy radiances. Currently, the GSI cloud detection is based on the brightness temperature differences between observations and the model background state at two MHS window channels. It is found that the GSI quality control algorithm fails to identify some MHS cloudy radiances in cloud edges where the ARW model has no cloud and the water vapor amount is low. A new MHS cloud detection algorithm is developed based on a statistical relationship between three MHS channels and the Geostationary Operational Environmental Satellite (GOES) imager channel at 10.7 μm. The 24-h quantitative precipitation forecast is improved rather than degraded by MHS radiance data assimilation when the new cloud detection algorithm is added to the GSI MHS quality control process. The temporal evolution of 3-h accumulative rainfall distributions compared favorably with that of multisensor NCEP observations and GOES-12 imager observations. The precipitation threat scores are increased by more than 50% after 3–6 h of model forecasts for 3-h rainfall thresholds exceeding 1.0 mm.