MEASUREMENT OF CLEAR-AIR GRADIENTS AND TURBULENCE PROPERTIES WITH RADAR WIND PROFILERS

Citation
Ee. Gossard et al., MEASUREMENT OF CLEAR-AIR GRADIENTS AND TURBULENCE PROPERTIES WITH RADAR WIND PROFILERS, Journal of atmospheric and oceanic technology, 15(2), 1998, pp. 321-342
Citations number
54
Categorie Soggetti
Metereology & Atmospheric Sciences","Engineering, Marine
ISSN journal
07390572
Volume
15
Issue
2
Year of publication
1998
Pages
321 - 342
Database
ISI
SICI code
0739-0572(1998)15:2<321:MOCGAT>2.0.ZU;2-O
Abstract
An experiment comparing balloon and profiler observations was carried out to evaluate the capability of Doppler radar wind profilers to remo tely measure useful meteorological quantities other than wind. The sit e chosen was in Southern California during a time of year when it offe rs a natural laboratory for investigating extreme contrasts in tempera ture and humidity. To evaluate the new capabilities, it was found that new and additional treatment of the radar data was necessary. For exa mple, the adequacy of the usual radar wind observations obtained by ed iting the Doppler spectral moments was found to be very questionable f or shortterm observations, so the authors extended the editing to the raw spectra, and substantial improvement was found. The advantages of the redundancy in five-beam systems are investigated and are also foun d to be very necessary to obtain the accuracy needed. A technique for minimizing the variances of the differences of the four redundant pair s of radials is described. The resulting improved vertical velocity es timates substantially improve the agreement between radio acoustic sou nding system (RASS) temperature retrievals and balloon-measured temper atures. The ability of the profilers to measure turbulence intensity w as rested, and the accuracy of techniques using the spectral width to measure turbulent dissipation rate when complicated spectra are presen t is examined. Two different techniques for optimizing the calculation of spectral width are compared and the errors assessed. One technique integrates over the uncontaminated range of the chosen spectral peak and then extrapolates a Gaussian function to infinity. The other metho d uses the slope of the log least squares best fit of the uncontaminat ed points to a Gaussian function. Profiler-measured length scales of w ind and scalar quantities are measured and compared. Profiles of radar -measured gradients of refractive index are compared with gradients me asured by balloon. It is shown how gradients of humidity can be calcul ated to about the same accuracy as refractive-index gradients by combi ning the temperature gradients from RASS with the refractive-index gra dient observations from the radar.