WMOD00513 - Laserfiche WebLink

Home Browse Search

Reprinted from JOURNAL OF ApPLIED METEOROLOGY, Vol. 19, No.3, March 1980 American Meteorological Sodety Printed in U. S. A. Predictor Variables of the Maximum Radar Echo Activity on Convective Days GERARD E. KLAZURA AND ROBERT G. PRITCHARD Office of Atmospheric Resources Management, Division of Research, Bureau of Reclamation, Denver, CO 80225 1 March 1979 and 10 December 1979 ABSTRACT Digital radar data and atmospheric sounding information were analyzed with the intention of be- ginning a search for atmospheric parameters which are easily attainable, are independent of whether or not clouds are seeded, and either individually or in concert with others can be used to predict the po- tential size, intensity and coverage of convective precipitation as estimated by radar. Stability indexes and upper level wind speeds seemed to be the dominant predktor variables. 1. Introduction The objective of this investigation has been to begin a search for atmospheric parameters which are easily attainable, are independent of whether or not clouds are seeded, and either individually or in concert with others can be used to predict the potential size, intensity and coverage of convective precipitation as estimated by radar. The approach used was to statistically compare the temperature, moisture and stability factors that were computed from a nearby sounding with the maximum size and intensity of radar echo characteristics for each con- vective cloud system. This was a first attempt to identify potential predictor variables for use in the Bureau of Reclamation's High Plains Cooperative Program (HIPLEX). 2. Data base and methodology Digital radar data were collected May-July 1976 at Miles City, Montana in support of HIPLEX. The radar used in this study is a sensitive, narrow-beam, 5 cm wavelength system which records echo data on computer-compatible magnetic tape. The antenna scanned continuously in a volume mode of 3600 in azimuth and 120 in elevation. The time interval for a complete volume scan was - 5 min. Most of the pre- cipitation echo data were from natural cloud sys- tems; but a few small, isolated clouds that were seeded were also included. The radar data process- ing and analysis products. used in this study are described by Schroeder and Klazura (1978). Echo characteristics have been compiled for each precipitation system over an area defined by range rings from 25 to 150 km from the radar. These ~~J parameters (dependent variables) were all computed from a 5 min volume scan recorded at or near peak convective periods of the day. The time of peak activity was determined through the analysis of composite B-scans, a range-azimuth plot which dis- plays radar reflectivity, and echo top information (Schroeder and Klazura, 1978). All of the composite B-scans for each day were visually scanned and the one which contained the most echo area coverage and highest reflectivities was chosen. The soundings taken at Miles City were used to search for predictor variables of these echo characteristics. The soundings were taken from 0 to 12 h before the peak convective period. Relationships between sounding characteristics (predictor variables) and radar reflectivity param- eters (dependent variables) were studied using simple and multiple linear correlation and regres- sion analyses. The radar echo characteristics and sounding variables chosen for this study are defined in Table 1. As mentioned earlier, the radar echo char- acteristics were computed over the scanning area between the 25 and 150 km range rings. The maxi- mum echo height, maximum reflectivity factor and maximum rain were obtained from individual radar bins. Maximum rain rate was obtained by convert- ing the maximum reflectivity observed in the 10 ele- vation sweep (30-150 km range) or the 20 elevation sweep (25-29.9 km range) by using the Marshall and Palmer (1948) Z-R relationship. The maximum rain volume was computed by multiplying rainfall rates by their associated bin areas and accumulating them over the entire scanning area. The maximum total volumes were computed by accumulating all ",.,4uj~;;" -"'''""''''<k ~--,,,,