Development of a Radar-based Hail-Detection-Product

At the KNMI we are developing a tool for the detection and display of severe weather phenomena related to convective systems, like heavy rainfall, thunderstorms, windgusts, and summer hail. This ``Severe Weather Product'' is to be used for nowcasting and short-term forecasting of these weather phenomena. Radar data, both reflectivity and velocity, will be the primary source of information, and it will be completed with other observations and data from numerical weather prediction (NWP) models. Currently, the KNMI operates two C-band Doppler Radars which are performing low-elevation volume-scans every 5 minutes and extensive volume-scans every 15 minutes. The first sub-product under consideration is a tool for the detection and display of summer hail using non-Doppler Radar data and temperature information from NWP models.

From literature five methods for the detection of summer hail using Radar reflectivity have been selected. First of all, the method based on the bare Radar reflectivity with a threshold and that on the Vertically Integrated Liquid (VIL) with a threshold are selected. The method developed recently in New Zealand by Auer, in which Radar reflectivity is combined with cloud-top temperatures, is selected as well. Finally, the two methods which are currently in use within the framework of NEXRAD, i.e., the method of Waldvogel and the Severe Hail Index (SHI), are chosen. These five different methods have been tested on severe weather events in the Netherlands during the summer of 1999. As ground truth, observations by the 321 volunteers of the (rainfall) observer network of the KNMI as well as detailed hail damage reports from agricultural insurance companies have been taken. Evidently, some hail events will not have been observed due to their small spatial extent and/or their short time duration. Both the effect of this incomplete ground truth dataset as the effect of possible spatial mismatches between the high-altitude Radar observations and the ground truth observations on the scoring parameters, like the Critical Success Index (CSI), have been investigated. The results of the comparison between these five methods for detection of hail will be presented in detail.