Simultaneous Assimilation of Dual-Polarization Radar and All-Sky Satellite Observations to Improve Convection Forecasts

Accurate forecasts of the development and evolution of deep, moist convection in convection-allowing models (CAMs) remain a challenge in part owing to the difficulties inherent in modeling the microphysical and internal structures of convection, which can affect storm mode, intensity, and longevity. We hypothesize that underused Weather Surveillance Radar-1988 Doppler (WSR-88D) and GOES-16 observations can improve forecasts of deep convection in CAM ensembles. Since the upgrade to the national network of WSR-88D radars was completed in 2013, polarimetric radar data offer a wealth of information about the shape, size, and type of hydrometeors present in precipitation. Several distinct polarimetric signatures within deep convection have been identified, such as the differential reflectivity (Z_DR) column, that can aid significantly in characterizing internal storm structures and improve CAM representation of convection. In addition, GOES-16 infrared all-sky brightness temperatures (BTs) provide complimentary information on cloud structures and cover that radars cannot directly measure. The CAM ensemble in this study uses the Advanced Research version of the Weather Research and Forecasting Model with the High-Resolution Rapid Refresh configuration at 3-km horizontal grid spacing and with 40 ensemble members. Observations are assimilated using an Ensemble Kalman Filter, where the radar and satellite observations are assimilated jointly and separately, and results are compared in four proof-of-concept experiments. Results indicate that the assimilation of BT observations improves forecasts of a severe convective event, which are further improved with the assimilation of Z_DR observations. While BT assimilation alone improves the convective forecast, Z_DR observations provide additional improvements to updraft helicity tracks, precipitation, and hail forecasts.