Ground truth and modeling verification of the hail quadrature parameter

Abstract

The purpose of this study is to determine the usefulness of a new hail detection and size classification algorithm known as the Hail Quadrature Parameter (HQP). The parameter is explored from a theoretical, modeling, and ground truth perspective. HQP uses scatterer size and intensity information contained in horizontal reflectivity (Zh) and shape information contained in differential reflectivity (Zdr) (combined in the differential reflectivity hail signal, Hdr). It also adds the particle canting and surface irregularity information detectable by the linear depolarization ratio (LDR). The purpose of this study is to determine whether this additional information from LDR assists in hail detection and size classification. A modeling sensitivity study was performed using the T-matrix model to determine how HQP would react to changes in certain hailstone characteristics such as size, water-coat thickness, axis ratio, canting angle, ice density and percent water volume. Of greatest interest was whether small hailstones with thicker water coats would appear the same to the radar as larger hailstones with thinner water coats, causing potential false alarms in HQP. It was found that increasing water-coat thickness did not produce consistently larger HQP values. Rather, the water-coat thickness simply determined at which diameter hailstones entered the Mie regime and began displaying unpredictable and widely varying values. A ground-truth verification study was also conducted to determine how well HQP detected the presence and size of hail at the surf ace. Post-storm surveys were conducted in northern and eastern Colorado. The polarimetric radar parameters of interest were calculated at each survey point and relationships between these relevant parameters and hail size and damage were explored. Through both statistics and raw data plots it was determined that HQP does not perform better than for hail detection or size classification. It was also determined that it is unlikely that LDR adds as much information about hail size as was previously thought.