Detecting and forecasting large hail in the Netherlands

Abstract

In this thesis, the Hail Detection Algorithm (HDA) developed by Witt et al. [1998] is used to detect large hail. This algorithm is based on single-polarization radar and NWP model data and gives an estimate of the Maximum Expected Hail Size (MEHS). Some difficulties like the occurrence of ground clutter and storm tilt were encountered while applying the HDA, but eventually methods have been developed to overcome these problems. Using the outcomes of the algorithm, a clima- tology has been made for the Netherlands for the time that the volume data of the two KNMI radars (De Bilt and Den Helder) has been archived (2008-2016). Also, the HDA is used in a case study of 23rd of June 2016, where a single hail storm caused over 500 million euros of damage to houses, cars and agricultural properties. To prevent severe damage in the future, it is of great importance that one can indicate whether or not hail will occur in a certain situation. Moreover, one ideally wants information about the hail size. It was found that the most unstable CAPE (MUCAPE) had the strongest relation with hail hail size. Moreover, under very weak deep layer shear conditions (< 5.14 m/s), there seems to be a linear relation between the maximum possible hail size and MUCAPE. Physically, this can be explained by the fact that the maximum possible hail size is achieved when the terminal velocity equals the maximum updraft velocity. According to the simple parcel theory, the maximum updraft velocity is directly related to the amount of MUCAPE.