STEPCLIM: Severe Thunderstorm Evaluation and Predictability in Climate Models

While thunderstorms are increasingly recognized as an important hazard to life and property in Europe, still relatively little is known about the effects of climate change on the frequency and intensity of these localized and short-lived hazards. The European Severe Storms Laboratory (ESSL) addresses this issue in the project STEPCLIM (“Severe Thunderstorm Evaluation and Predictability in Climate Models”), whose main objective is to provide a suite of physical metrics to assess the frequency and intensity of severe thunderstorm hazards from climate model data. These metrics are developed using reanalysis data and hindcasts of a developmental decadal climate modeling system MiKlip, and quality-controlled severe storm reports from the European Severe Weather Database (ESWD).

In order to characterize the local state of the atmosphere, a set of parameters is defined which have a physical meaning in the dynamics of convective storms. These include instability, vertical wind shear, and measures for lifting support or possible trigger mechanisms, as well as other quantities which describe the vertical profiles of temperature, moisture and wind. A first raw version of these metrics is developed by comparing ECMWF reanalyses with ESWD storm reports in order to distinguish between situations with and without an occurrence of local severe weather phenomena. Subsequently, the set of metrics is refined by the use of regionalized reanalyses and hindcasts of the MiKlip climate model.

The method extends earlier work conducted in the United States by performing a re-calibration of the parameter space with respect to the thunderstorm and severe weather climate in Europe, where severe weather frequently occurs in environments that can be considered relatively benign compared to North American standards. Further novel aspects of the present study comprise a stratification of the developed metrics with respect to different types of severe weather (tornadoes, severe wind gusts, hail, and excessive precipitation), the development of a proxy for convective initiation, the investigation of the influence of different resolutions of the underlying climate model, and the application of the developed metrics on future regionalized climate projections.

First preliminary results of the present study are presented, including an outline of typical environments in which severe weather in Europe occurs, and selected aspects of the raw version of the developed metrics for the characterization of severe weather environments.