Forecasting Weather Extremes using Atmospheric Analogs and Stochastic Weather Generators

Forecasting extreme weather events remains challenging, particularly at local scales and when multiple hazards occur together. In this study, we explore the potential of atmospheric analogs and stochastic weather generators (SWGs) as complementary approaches to conventional numerical weather prediction. Two frameworks are presented, combining large-scale atmospheric circulation information with stochastic simulations to forecast extreme precipitation and wind across Europe. Their performance is evaluated against ECMWF ensemble forecasts, highlighting improvements in probabilistic forecast skill and the ability to reproduce simultaneous and sequential compound extremes. The findings demonstrate the potential of SWGs to enhance the prediction of weather extremes and support early warning systems.

Meriem Krouma is a researcher at Researcher at the Department of Earth Sciences; Program for Air, Water and Landscape Sciences; Meteorology at Uppsala University.

Moderation: Sebastian Scher