ausblenden:
Schlagwörter:
-
Zusammenfassung:
The Pan-Arctic has experienced accelerated warming over past decades resulting in enhanced loss of sea-ice. Previously ice-covered surface waters are now directly subject to atmospheric conditions (e. g., heating and wind mixing). We use two mesoscale-resolving atmospheric reanalysis forced ocean/sea-ice simulations to examine ocean/sea-ice feedbacks resulting from these changes. One model uses the Parallel Ocean Program 2 (POP2) coupled to CICE5 on an ultra-high resolution global grid (8 km at the equator to 2 km at the poles - UH8to2) while in the other, the Hybrid Coordinate Ocean Model (HYCOM) is coupled to CICE5 on a regional Arctic 1/25º domain (04HYCICE). Both are forced with Japanese 55-year Reanalysis-driving ocean (JRA55-do) and were initialized from a data assimilative global 1/25º ocean/sea-ice state.
Near-present (2017-2020) distributions of seasonal sea-ice thickness and concentration are realistically reproduced by both models but with low biases in thickness in the central and eastern Arctic in the fall. In the eastern basin, UH8to2 Atlantic Water (AW) is shallower, warmer, and saltier than World Ocean Atlas 2018 climatology (2005-2017). These biases, combined with a weaker than observed stratification in the upper 100 m of the UH8to2, contribute to the seasonal biases in UH8to2 sea-ice thickness. AW analysis will be extended to the 04HYCICE to understand how differences in model stratification, among others, impact sea-ice distribution. A possible lagged response of eastern Arctic Ocean temperature to a 2016/2017 upper ocean warming event in the eastern Subpolar North Atlantic, simulated by the UH8to2 in agreement with observational results, will be discussed.