Assessing Errors in Variational Dual-Doppler Wind Syntheses of Supercell Thunderstorms Observed by Storm-Scale Mobile Radars

Abstract

Abstract Dual-Doppler wind retrieval is an invaluable tool in the study of convective storms. However, the nature of the errors in the retrieved three-dimensional wind estimates and subsequent dynamical analyses is not precisely known, making it difficult to assign confidence to inferred storm behavior. Using an Observing System Simulation Experiment (OSSE) framework, this study characterizes these errors for a supercell thunderstorm observed at close range by two Doppler radars. Synthetic radar observations generated from a high-resolution numerical supercell simulation are input to a three-dimensional variational data assimilation (3DVAR) dual-Doppler wind retrieval technique. The sensitivity of the analyzed kinematics and dynamics to the dual-Doppler retrieval settings, hydrometeor fall speed parameterization errors, and radar cross-beam angle and scanning strategy is examined. Imposing the commonly adopted assumptions of spatially constant storm motion and intrinsically steady flow produces large errors at higher altitudes. On the other hand, reasonably accurate analyses are obtained at lower and middle levels, even when the majority of the storm lies outside the 30° dual-Doppler lobe. Low-level parcel trajectories initiated around the main updraft and rear-flank downdraft are generally qualitatively accurate, as are time series of circulation computed around material circuits. Omitting upper-level radar observations to reduce volume scan times does not substantially degrade the lower- and middle-level analyses, which implies that shallower scanning strategies should enable an improved retrieval of supercell dynamics. The results suggest that inferences about supercell behavior based on qualitative features in 3DVAR dual-Doppler and subsequent dynamical retrievals may generally be reliable.