Wind structure in a supercell thunderstorm as measured by a UHF wind profiler

Citation
Gs. Lehmiller et al., Wind structure in a supercell thunderstorm as measured by a UHF wind profiler, M WEATH REV, 129(8), 2001, pp. 1968-1986
Citations number
45
Categorie Soggetti
Earth Sciences
Journal title
MONTHLY WEATHER REVIEW
ISSN journal
00270644 → ACNP
Volume
129
Issue
8
Year of publication
2001
Pages
1968 - 1986
Database
ISI
SICI code
0027-0644(2001)129:8<1968:WSIAST>2.0.ZU;2-8
Abstract
A mesocyclone associated with a tornadic supercell passed almost directly o ver the Lamont, Oklahoma, 404-MHz wind profiler on 24 May 1998. The archive d spectral moment data, including 6-min resolution radial velocity measurem ents from the vertical and two oblique beams of the profiler, provided a ra re opportunity to construct a detailed time-height section of the three-dim ensional wind structure of a severe convective storm near and in its mesocy clone. Supplemental information used to create this cross section was provi ded by nearby operational Next Generation Weather Radar/Weather Surveillanc e Radar (NEXRAD/WSR-88D) radars, as well as visual and surface observations . The most difficult challenge in analyzing the profiler data was the removal of hydrometeor fall-speed contamination in order to determine the true air motions. Since the profiler beams saturate under heavy precipitation condi tions, the use of the returned signal power in the profiler beams to estima te precipitation fall speeds was not reliable. To overcome this difficulty, radar reflectivity data from four surrounding NEXRAD/WSR-88D radars were o bjectively analyzed for the region scanned by the profiler. The components of vertical velocity due to precipitation were then estimated by developing empirical statistical models between the objectively analyzed reflectivity factors and profiler-measured vertical velocities located well outside the storm updraft. This technique was successful in that it removed fully 80% of the explainable variance associated with the precipitation fall speeds; the corresponding estimated expected error, on the order of 1 m s(-1), was fully one magnitude smaller than the scale of the measured vertical velocit ies within the updraft. Although the recovery of the horizontal winds was less successful than the recovery of the vertical velocities, horizontal winds were recoverable in r egions where the horizontal vertical velocity gradient was reduced, and whe re corroborated by observations from nearby operational NEXRAD radars. The resulting time-height cross section of wind exhibited a region of strong up draft near and in the mesocyclone and downdraft flanking it. Weak downdraft s were also present at low levels surrounding the region of strongest updra ft. The updraft peak velocity of 50 m s(-1) represents one of the largest v ertical velocities ever measured in a convective storm. Evidence is present ed of significant modification of the horizontal wind structure away from t he updraft and of storm-generated buoyancy waves in the surrounding environ ment.