Real-time computer techniques in the detection and analysis of severe storms from digital radar data

Abstract

An improved computer method was developed by which multi-tilt digital radar data can be interpolated in three dimensions and reduced to a two-dimensional display of partially vertically-summed reflectivity (Z) maps (PVSZ) in near real time. The computer method was developed by using digital radar data collected with the 1 0 -cm radar at the National Severe Storms Laboratory in Norman, Oklahoma. Various combinations of interpolation schemes were used to develop the new computer method, and the resultant products were compared to determine whether or not significant features of a severe storm evident in constant altitude reflectivity (Z) maps (CAZM) are retained by the new reduction technique. In addition, the number of PVSZ layers was varied to determine the minimum needed for adequate depiction of the tilt of the storm core. Finally, severe storm data from New England were processed by using the new data-reduction technique to find out whether or not any of the severe-storm signatures observed in analyses of Oklahoma storms were evident in the New England digital radar data. The new computer method developed during this investigation resulted in significant savings of computer processing time and memory as compared to a previously used method and yet retains all of the significant features of the severe storm complex revealed by the other, more cumbersome method. Three PVSZ layers appear to be sufficient to depict adequately the tilt of the storm core. A tornadic storm in Maine presented many severe-storm signatures that have been observed in analyses of Oklahoma tornadoes.