The Impact of Assimilation of GPM Clear Sky Radiance on HWRF Hurricane Track and Intensity Forecasts

The impact of GPM microwave imager (GMI) clear sky radiances on hurricane forecasting is examined by ingesting GMI level 1C recalibrated brightness temperature into the NCEP Gridpoint Statistical Interpolation (GSI)- based ensemble-variational hybrid data assimilation system for the operational Hurricane Weather Research and Forecast (HWRF) system. The GMI clear sky radiances are compared with the Community Radiative Transfer Model (CRTM) simulated radiances to closely study the quality of the radiance observations.  The quality check result indicates the presence of bias in various channels. A static bias correction scheme, in which the appropriate bias correction coefficients for GMI data is evaluated by applying regression method on a sufficiently large sample of data representative to the observational bias in the regions of concern, is used to correct the observational bias in GMI clear sky radiances.

Forecast results with and without assimilation of GMI radiance are compared using various hurricane cases from recent hurricane seasons. Diagnoses of data assimilation statistics during the cycling experiments show that the bias correction coefficients obtained from the regression method can correct the inherent biases in GMI radiance data, significantly reducing observational residuals. The removal of biases also allows more data to pass GSI quality control and hence to be assimilated into the model. Forecast results demonstrate that the quality of analysis from the data assimilation is sensitive to the bias correction, with positive impacts on the hurricane track and intensity forecasts when systematic biases are removed from the radiance data. Details will be presented at the symposium.