A comparison of the atmospheric response to the Weddell Sea Polynya in atmospheric general circulation models (AGCMs) of varying resolutions

• Published in: Weather and Climate Dynamics
• Main Authors: Ayres, Holly C., Ferreira, David, Park, Wonsun, Kjellsson, Joakim, Ödalen, Malin
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2024
• Subjects: Weddell; Weddell Sea; Sea ice
• Online Access: http://nora.nerc.ac.uk/id/eprint/537647/; https://nora.nerc.ac.uk/id/eprint/537647/1/wcd-5-805-2024.pdf; https://doi.org/10.5194/wcd-5-805-2024

The Weddell Sea Polynya (WSP) is a large opening within the sea ice cover of the Weddell Sea sector. It has been a rare event in the satellite period, appearing between 1973 and 1976 and again in 2016/2017. Coupled modelling studies have suggested that there may be a large-scale atmospheric response to the WSP. Here, the direct atmospheric response to the WSP is estimated from atmosphere-only numerical experiments. Three different models, the HadGEM3 UK Met Office model, the ECHAM5 Max Planck Institute model, and the OpenIFS ECMWF model, each at two different resolutions, are used to test the robustness of our results. The use of large ensembles reduces the weather variability and isolates the atmospheric response. Results show a large (∼100–200 W m−2) turbulent air–sea flux anomaly above the polynya. The response to the WSP is local and of short duration (barely outlasting the WSP) with a similar magnitude and spatial pattern of lower-tropospheric warming and increase in precipitation in all six configurations. All models show a weak decrease in surface pressure over the WSP, but this response is small (∼2 hPa) in comparison to internal variability. The dynamic response is inconsistent between models and resolutions above the boundary layer, suggesting a weak or null response that is covered by internal variability aloft. The higher resolution does not alter the pattern of the response but increases its magnitude by up ∼50 % in two of the three models. The response is influenced by natural variability in the westerly jet. The models perform well against ERA5 reanalysis data for the 1974 WSP in spatial response and magnitude, showing a turbulent heat flux of approximately 150 W m−2.