These conclusions are drawn from the synthesis of 36-min resolution ground-based microwave radiometric and Doppler wind profiler data, in combination with a 1-km nested grid Numerical Weather Prediction (NWP) model initialized with WSR-88D radar rdata by way of a multi-scale variational assimilation method. Superior radiometric retrievals are achieved by coupling the radiometric data with background fields from the NWP model using a 1-D variational technique, as compared to the traditional pure statistical inversion method. The model successfully portrayed the rapid development of a supercell matching the general characteristics of the Windsor storm.
It is shown that the extremely rapid nature of the changes in the local storm environment were associated with development of a narrow low-level jetlet that advected moist, unstable air northwestward toward Windsor just ahead of a pronounced dryline. These details and the success in capturing the supercell were only evident in model analyses that utilized the WSR-88D reflectivity and radial velocity data in the initialization procedure.