Multi-method quantification of the contribution of circulation changes to summer temperature trends in the northern hemispheric mid-latitudes

Observed summer temperature trends differ strongly around the northern hemispheric latitudes. Besides changes in aerosol emissions, changes in atmospheric circulation patterns - whether forced or not - are expected to contribute to considerable variation in summer temperature trends. Different statistical and machine learning methods have been developed and applied to quantify this contribution of circulation changes to summer temperatures. Here we test the accuracy of multiple methods by applying them to historical climate simulations and comparing the circulation contribution obtained by different methods to trends found in nudged circulation simulations with wind fields of the historical simulations but pre-industrial control forcing. After validating the methods we apply them to ERA5 and over the entire northern hemispheric mid-latitudes (over land). Our results consistently suggest that especially over Europe circulation changes have contributed to an increase in summer temperatures. In parts of central Asia and eastern North America, circulation changes have contributed to a cooling in summer temperatures.

While providing a systematic overview of circulation contributions to local temperature trends in the northern hemispheric mid-latitudes we also show how nudging experiments can help to validate and consolidate methods. We argue that such method evaluation studies become increasingly important with the ongoing expansion of applications of statistical and machine learning analyses on the observational record.