Sources of seasonal sea-ice bias for CMIP6 models in the Hudson Bay Complex

• Published in: Annals of Glaciology
• Main Authors: Crawford, Alex D., Rosenblum, Erica, Lukovich, Jennifer V., Stroeve, Julienne C.
• Other Authors: Canada Research Chairs, Natural Sciences and Engineering Research Council of Canada
• Format: Article in Journal/Newspaper
• Language: English
• Published: Cambridge University Press (CUP) 2023
• Subjects: Annals of Glaciology; Hudson Bay; Sea ice
• Online Access: http://dx.doi.org/10.1017/aog.2023.42; https://www.cambridge.org/core/services/aop-cambridge-core/content/view/S0260305523000423

Abstract The seasonal ice-free period in the Hudson Bay Complex (HBC) has grown longer in recent decades in response to warming, both from progressively earlier sea-ice retreat in summer and later sea-ice advance in fall. Such changes disrupt the HBC ecosystem and ice-based human activities. In this study, we compare 102 simulations from 37 models participating in phase 6 of the Coupled Model Intercomparison Project to the satellite passive microwave record and atmospheric reanalyses. We show that, throughout the HBC, models simulate an ice-free period that averages 30 d longer than in satellite observations. This occurs because seasonal sea-ice advance is unrealistically late and seasonal sea-ice retreat is unrealistically early. We find that much of the ice-season bias can be linked to a warm bias in the atmosphere that is associated with a southerly wind bias, especially in summer. Many models also exhibit an easterly wind bias during winter and spring, which reduces sea-ice convergence on the east side of Hudson Bay and impacts the spatial patterns of summer sea-ice retreat. These results suggest that, for many models, more realistic simulation of atmospheric circulation would improve their simulation of HBC sea ice.