Abstract Extreme daily temperatures have important societal impacts which can in part be mitigated with advance warning. Understanding the relationship between the seasonal mean and daily extreme is of considerable interest, not least because a sound understanding of this relationship within reanalysis may lead to scope for improving seasonal forecasts of extremes of temperature. We show that for many regions of the Northern Hemisphere winter, seasonal mean temperature and coldest day are strongly correlated, and that the strength of the correlation varies from one location to another. We investigate this spatial variability by examining the way known dynamical drivers act on both the mean and extremes of the daily mean temperature distribution. We show that regions of high correlation between mean and extreme daily temperature overlap with regions where these dynamical drivers have a strong impact. In summer, the mean–extreme relationship is typically weaker than in winter and much less driven by large scale dynamical drivers.
Maidens et al. (Wed,) studied this question.