34th Conference on Radar Meteorology

P4.8

Performance of a new velocity dealiasing algorithm for the WSR-88D

Arthur Witt, NOAA/NSSL, Norman, OK; and R. A. Brown and Z. Jing

Improving the quality of the base velocity data in the WSR-88D is an ongoing effort. Although the current velocity dealiasing algorithm in the WSR-88D Radar Product Generator (RPG) generally performs well, this is not always the case. In particular, when the WSR-88D is operating in Volume Coverage Pattern (VCP) 31 (the low-Nyquist clear-air mode), and wind speeds or vertical wind shear are high, there are often numerous velocity dealiasing errors. This causes many WSR-88D sites to instead primarily operate in VCP-32 (the high-Nyquist clear-air mode), despite the greater sensitivity of VCP-31 at detecting clear-air return.

Given the limited computer capacity that existed when the WSR-88D was initially fielded, its algorithms needed to be kept relatively simple (i.e., they could not use large amounts of memory and/or CPU). Since then, substantial upgrades to the WSR-88D's computer systems allow for the implementation of more sophisticated algorithms. This includes a new velocity dealiasing algorithm, initially targeted for use exclusively when the WSR-88D is operating in VCP-31.

The performance of the new dealiasing algorithm was compared to the current WSR-88D dealiasing algorithm, using a quantitative scoring procedure, on several recently collected VCP-31 cases, where the current algorithm performed very poorly (i.e., it generated numerous/extensive errors). The performance comparison was also done for several severe-storm and hurricane events, where the WSR-88D was operating in a variety of the precipitation-based VCPs. Test results indicate, for the VCP-31 cases, major improvements for the new algorithm versus the current algorithm, with more modest improvements for the severe-storm and hurricane cases.

extended abstract  Extended Abstract (680K)

Poster Session 4, Wind Profilers / Operational Needs
Monday, 5 October 2009, 1:30 PM-3:30 PM, President's Ballroom

Previous paper  Next paper

Browse or search entire meeting

AMS Home Page