Radar and lightning detection

You are here: Radar and lightning detection » Research & Development » Hail detection

Hail detection

Detection of hail using radar observations has been investigated in close collaboration with the Netherlands weather service (KNMI). This research resulted in a hail detection product operationally used at the weather office. 

The hail detection algorithm produces a probability of hail which is derived from volume reflectivity data following the method of Waldvogel. It is based on the difference between the height of the freezing level and the echotop-45 dBZ, i.e., the maximum height at which a reflectivity at least equal to 45 dBZ is observed. The product is generated every 15 minutes from the Wideumont radar and every 5 minutes from the Zaventem radar.


Example of hail detection product generated from the Wideumont radar observations.

The quality of radar echo top heights and the implications for hail detection has been investigated. The results of this study can be found in:


Delobbe, L. and I. Holleman, 2006. Uncertainties in radar echo top heights used for hail detection. Meteor. Appl., 13, 361-374.

Hail detection and MACCBET project

In the frame of the MACCBET project, RMI performed the reanalysis of the archived data of the Wideumont radar. Ten years of observational radar data were analysed by newly implemented enhanced hail detection algorithm (Witt et al., 1998). While previous hail detection algorithm indicate the pobability of hail, the new algorithm also indicates whether hail is severe (diameter ≥ 19 mm), and estimates the maximum expected hail size. The algorithm is based on the calculation of Severe Hail Index (SHI). Arthur Witt and coauthors developed this parameter as the primary predictor variable for severe-size hail. The SHI is a thermally weighted vertical integration of reflectivity profiles. Estimation of maximum expected size of hail (MESH) and probability of severe hail (POSH) depends on the values of the SHI parameter.


Example of maximum expected size of hail product generated from the Wideumont radar observations.

Example of probability of severe hail product generated from the Wideumont radar observations.

New products of the enhanced hail detection algorithm are now in the verification phase. After thorough verification, they will become operational. In the MACCBET project the statistical analysis of occurrence and severity of hail in Belgium will be used for the verification of regional climate model.


Detailed information on the enhanced hail detection algorithm can be found:

Witt, Arthur, Michael D. Eilts, Gregory J. Stumpf, J. T. Johnson, E. De Wayne Mitchell, Kevin W. Thomas, 1998: An Enhanced Hail Detection Algorithm for the WSR-88D. Wea. Forecasting, 13, 286–303.


Contact person for this topic: Maryna Lukach