Joshua Gebauer- Mar 13

Name:     Joshua Gebauer

Title:    Impact of a Vertical Vorticity Constraint on Dual-Doppler Retrieved Vertical Velocity

Location: NWC 5600

Date:     2020/03/13

Time:     3:00 PM

Series:   Convective Meteorology (Mesoscale Dynamics)

Abstract: Previous observation simulation systems experiments (OSSEs) have found that a variational dual-Doppler retrieval with both a mass conservation and vertical vorticity constraint can improve retrieved vertical velocity, although with the limitation that rapid radar volume scans are needed in order to accurately calculate the vorticity tendency. These studies, however, did not replicate the errors that are typical in radar datasets and these errors might affect the performance of the retrievals. Therefore, the OSSE results need to be confirmed using real radar data. With the advancements in research weather radar, real-data tests are now possible. This study uses radar observations of a convective storm collected by the AIR and RaXPol on 4 Sept 2018 to test the impact of using a vertical vorticity constraint on vertical velocities retrieved from dual-Doppler analysis (DDA). The retrieved vertical velocity from the DDA was verified using vertically pointing radar observations from SMART-R 3.

DDA was conducted for a 37-minute period several times with varying grid spacing, time between volume scans, and methods to calculate the vorticity tendency. Additionally, a traditional DDA that obtains vertical velocity by vertically integrating the analastic mass-conservation equation and a variational DDA with no vorticity constraint were performed. When the time between volume scans is 30 s, the DDA with a vorticity constraint outperforms both the traditional and no-vorticity DDA, even when the vorticity tendency is calculated using a simple centered difference. However, for longer times between volume scans, using a simple centered difference reduces the accuracy of the retrieved vertical velocity. When the time between volume scans is longer, the retrieved vertical velocity can be improved by using advection correction to shorten the timestep when calculating the vorticity tendency. The improvements from the DDA with a vertical vorticity constraint are largest when the analysis is conducted on a 125 m grid, but the loss in accuracy with increasing time between volume scans is greatest at this grid spacing. The increase in skill of the DDAs with a vorticity constraint are mostly due to improvements in the structure of the vertical vorticity features. Additionally, the vorticity constraint appears to compensate for errors in the radar observations. Trajectory analysis using the DDA retrievals indicates that the differences between the DDAs with no vorticity constraint and DDAs with a vorticity constraint are significant, as the final heights of the trajectories in some regions of the storm can vary by as much as 3 km. Overall, these results suggest that future observational studies using dual-Doppler retrievals should strive to implement rapid-scan radars and a DDA procedure that includes a vertical vorticity constraint.