Hindcast simulations reveal sea ice melting varies in the Arctic, suggesting thermodynamic and dynamic models are effective tools for prediction.
Satellite‐derived sea ice products and ice model simulations are analyzed in the Bering Sea during the winters of 2018–2019, featuring historically low sea ice coverage on the Eastern Bering Sea shelf, and 2019–2020, when ice conditions were close to average. Hindcast simulations are conducted using the Community sea Ice CodE (CICE) as a standalone regional ice model forced with ERA5 atmospheric fields and oceanic fields derived from realizations of the Regional Ocean Modeling System (ROMS) internally coupled with its own ice model component. CICE is able to predict the observed seasonal variability in ice concentration and ice thickness provided that ocean mixed layer temperature and salinity are strongly nudged toward an accurate ocean state. Ice mass balance analysis shows that sustained periods of southerly wind during the 2018–19 season led to increased melting and a larger ice export into the Arctic when compared to 2019–20. More ice was produced in 2018–19, the season with lower sea ice areal extent, offset by larger melting and transport out of the domain. The volume of ice exported to the Russian shelf and basin is comparable to the net transport through the Bering Strait. The consistency between the model and satellite products demonstrates that the representation of sea ice thermodynamic and dynamic models present in CICE are adequate for representing key physical processes driving sea ice variability in this region during both low and normal ice coverage years.
No takes yet. Share an insight, caveat, or question.
Smith et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: