Representation and annual to decadal predictability of Euro-Atlantic weather regimes in the CMIP6 version of the EC-Earth Coupled Climate Model

• Published in: Journal of Geophysical Research: Atmospheres
• Main Authors: Delgado Torres, C., Verfaillie, D., Mohino Harris, Elsa, Donat, M.G.
• Format: Article in Journal/Newspaper
• Language: English
• Published: American Geophysical Union 2022
• Subjects: Física atmosférica; North Atlantic; North Atlantic oscillation
• Online Access: https://eprints.ucm.es/id/eprint/75640/; https://eprints.ucm.es/id/eprint/75640/1/Moh%C3%ADnoHarris%2012%20libre%28EMBARGO%20HASTA%2015-1-2023%29.pdf; https://doi.org/10.1029/2022JD036673

Weather regimes are large-scale atmospheric circulation states that frequently occur in the climate system with persistence and recurrence, and are associated with the occurrence of specific local weather conditions. This study evaluates the representation of the four Euro-Atlantic weather regimes in uninitialized historical forcing simulations and initialized decadal predictions performed with the EC-Earth3 coupled climate model. The four weather regimes are the positive and negative phases of the North Atlantic Oscillation (NAO+ and NAO−, respectively), Blocking, and Atlantic Ridge in winter; and the NAO−, Blocking, Atlantic Ridge, and Atlantic Low in summer. We also analyze the impact that the model initialization toward the observed state of the climate system has on the ability to predict the variability of the weather regimes' seasonal frequency of occurrence. We find that the EC-Earth3 model correctly reproduces the spatial patterns and climatological occurrence frequencies of the four weather regimes. By contrast, the skill in predicting the inter-annual to decadal variations of the weather regimes' seasonal frequencies is generally low, and the initialization does not significantly improve such skill. The observed teleconnections between the weather regimes and the North Atlantic sea surface temperatures are generally not reproduced by the model, which could be a reason for the low skill in predicting the temporal variations of the weather regime frequencies.