Item

Abstract

Meteorological extremes on the seasonal time scale have received increased attention due to their relevance for society and economy. A recently developed approach is applied here to ERA5 reanalyses from 1950-2020 to identify hot and cold, wet and dry, and stormy and calm extreme seasons globally. The approach consists of (i) fitting a statistical model to seasonal mean values at each grid point, (ii) selecting a local return period threshold above which seasonal mean values are deemed extreme, and (iii) forming spatially coherent extreme season objects. The open-access ERA5 extreme season explorer, https://intexseas-explorer.ethz.ch, enables researchers to visualise, download, and investigate the dynamics of extreme season objects of any of the six types in their region of interest. We discuss examples of top 10 extreme seasonsglobally, show how they relate to anomalies in cyclone frequency and/or intensity, and provide insight into the substructure of extreme seasons. For instance, extreme hot seasons can emerge due to a strong warming of the warmest or coldest days only. Such a substructure analysis shows geographic variability, which is related, e.g., to the onset of monsoons, physical boundaries like the sea ice edge, or the frequency of occurrence of Rossby wave breaking. This study presents a first catalogue of objectively identified extreme seasons in the last decades, shows exemplarily how large-scale dynamics can lead to such seasons and their particular substructure, and with the help of the explorer supports the community in accelerating research in this important area at the interface of weather and climate dynamics.