Uncertainty within Euro-Atlantic Weather Regime Definitions

Weather regimes are valuable for energy sector applications as some are associated with higher risk of energy shortages, e.g. blocking regimes in winter. However, there are many uncertainties linked to the implementation of weather regimes, with connected risks for the energy sector. Especially relevant is the impact of climate change on the energy sector and the question whether weather regimes are suitable to use for that.

The weather regime uncertainties are investigated by studying three methodologically different Euro-Atlantic weather regime definitions. These definitions consist of 4, 6 or 7 weather regimes and are determined using different pre- and post-processing steps. We systematically compare the three methods and analyse their methodological and spatio-temporal sensitivity using ERA-5 reanalysis data. We find that the occurrence and persistence of the identified weather regimes can be sensitive to small changes in the methodology, with a significant impact of the training period. When considering 4 or 7 weather regimes, this results in unstable regime patterns.

Next, we investigate the representation of weather regimes in CMIP6 models. The regimes that are identified for the studied CMIP6 models are often different to those identified for ERA-5. The definition with 4 weather regimes is best represented in the CMIP6 models, where the definitions using 6 or 7 regimes are unstable with model-dependent regime patterns. This raises the question whether CMIP6 models are suitable to study the effects of climate change when using weather regimes to get to energy sector impacts. As there is substantial instability and uncertainty associated with the different weather regime definitions, a cautious approach is recommended.