Large-ensemble assessment of the Arctic stratospheric polar vortex morphology and disruptions

• Published in: Weather and Climate Dynamics
• Main Authors: A. Kuchar, M. Öhlert, R. Eichinger, C. Jacobi
• Format: Article in Journal/Newspaper
• Language: English
• Published: Copernicus Publications 2024
• Subjects: Meteorology. Climatology; QC851-999; Arctic
• Online Access: https://doi.org/10.5194/wcd-5-895-2024; https://doaj.org/article/5a47cd62a5fe4d9f9fa9903e9577a6f6

The stratospheric polar vortex (SPV) comprises strong westerly winds during winter in each hemisphere. Despite ample knowledge on the SPV's high variability and its frequent disruptions by sudden stratospheric warmings (SSWs) in the Northern Hemisphere (NH), questions on how well current climate models can simulate these dynamics remain open. Specifically the accuracy in reproducing SPV morphology and the differentiation between split and displacement SSW events are crucial to assess the models in this regard. In this study, we evaluate the capability of climate models to simulate the NH SPV by comparing large ensembles of historical simulations to ERA5 reanalysis data. For this, we analyze geometric-based diagnostics at three pressure levels that describe SPV morphology. Our analysis reveals that no model exactly reproduces SPV morphology of ERA5 in all diagnostics at all altitudes. Concerning the SPV morphology as stretching (aspect ratio) and location (centroid latitude) parameters, most models are biased to some extent, but the strongest deviations can be found for the vortex-splitting parameter (excess kurtosis). Moreover, some models underestimate the variability of SPV strength. Assessing the reliability of the ensembles in distinguishing SSWs subdivided into SPV displacement and split events, we find large differences between the model ensembles. In general, SPV displacements are represented better than splits in the simulation ensembles, and high-top models and models with finer vertical resolution perform better. A good performance in representing the morphological diagnostics does not necessarily imply reliability and therefore a good performance in simulating displacements and splits. Assessing the model biases and their representation of SPV dynamics is needed to improve credibility of climate model projections, for example, by giving stronger weightings to better performing models.