E-WP7 - STEPCLIM: Severe thunderstorm evaluation and predictability in climate models

In Europe, thunderstorms are increasingly recognized as an important hazard to life and property. According to the Munich Re Group, a yearly total damage of about € 5 billion (Munich Re Group, 2006) is inflicted by thunderstorm-related severe weather events (hail, flash floods, straight-line winds, tornadoes, and lightning) in Europe. Yet, relatively little is known about the predictability of such events on a seasonal to decadal timescale. STEPCLIM investigates predictability, as well as the processes that create and limit it.

The spatial and temporal resolution of global as well as regional climate models is insufficient to simulate thunderstorms and their attendant hazardous phenomena. However, these models do contain information related to the occurrence of these events. Within STEPCLIM Phase 1, a method has been developed that allows climate models to be used for the prediction of the frequency and intensity of hail, tornadoes and severe wind gusts in the future. Within STEPCLIM Phase 2, the method is improved and applied to both global and regional decadal climate predictions.

The statistical model relates atmospheric variables to probabilities of severe weather occurrence by means of an additive logistic regression in conjunction with the BMBF project ARCS (Analysis of Risk of Convective Storms in Europe). The development the logistic model requires that quality-controlled datasets of severe weather occurrence be available. Therefore the European Severe Weather Database (ESWD, managed by ESSL), which contains reports of hail, tornadoes, severe wind gusts and extreme rainfall events, is optimized for the analyses within the project.

Furthermore, the ability of the MiKlip system to represent atmospheric stability in the troposphere is investigated. During STEPCLIM Phase 1, several biases in the MiKlip datasets have been identified. An identification and corrections for any biases in the MiKlip datasets is done in conjunction with WPE-1 MOSQUITO.

Finally, the statistical model is applied to the MiKlip datasets to build a climatological dataset of a number of thunderstorm hazards. After bias-correction, performed using reanalysis datasets, the model becomes a component of the user-oriented MiKlip Common Evaluation System (CES).


  • To improve a statistical model that relates atmospheric variables to probabilities of severe weather occurrence.
  • To expand the ESWD dataset to mitigate inhomogeneity.
  • To identify and correct for biases in atmospheric stability in the MiKlip datasets
  • To apply the statistical model to the MiKlip system so that a dataset of thunderstorm hazards can be created for the current MiKlip system.
  • To develop a software tool for the Common Evaluation System that enables a user to build thunderstorm hazard risks of future MiKlip datasets.

STEPCLIM is conducted by the European Severe Storms Laboratory seated in Weßling, Germany.


European Severe Storms Laboratory
Lars Tijssen

Pieter Groenemeijer

Identification of favorable environments for thunderstorms in reanalysis data

2016 - Meteorologische Zeitschrift

Westermayer A.T. | P. Groenemeijer, G. Pistotnik, R. Sausen and E. Faust

Validation of Convective Parameters in MPI-ESM Decadal Hindcasts (1971–2012) against ERA-Interim Reanalyses

2016 - Meteorologische Zeitschrift, Vol. 25 No. 6, pp. 753-766

Pistotnik, G | P. Groenemeijer, and R. Sausen

A Climatology of Tornadoes in Europe: Results from the European Severe Weather Database

2014 - Monthly Weather Review, Vol. 142 (12), pp. 4775-4790

Groenemeijer, P. | T. Kühne