An assessment of the CMIP6 performance in simulating Arctic sea ice volume flux via Fram Strait

Numerical models serve as an essential tool to investigate the causes and effects of Arctic sea ice changes.Evaluating the simulation capabilities of the most recent CMIP6 models in sea ice volume flux provides references for model applications and improvements.Meanwhile,reliable long-term simulation results of the ice volume fux contribute to a deeper understanding of the sea ice response to global climate change.In this study,the sea ice volume flux through six Arctic gateways over the past four decades(1979-2014)were estimated in combination of satellite observations of sea ice concentration(SIC)and sea ice motion(SIM)as well as the Pan-Arctic Ice-Ocean Modeling and Assimilation System(PIOMAS)reanalysis sea ice thickness(SIT)data.The simulation capability of 17 CMIP6 historical models for the volume flux through Fram Strait were quantitatively assessed.Sea ice volume flux simulated from the ensemble mean of 17 CMIP6 models demonstrates better performance than that from the individual model,yet IPSL-CM6A-LR and EC-Earth3-Veg-LR outperform the ensemble mean in the annual volume flux,with Taylor scores of 0.86 and 0.50,respectively.CMIP6 models display relatively robust capability in simulating the seasonal variations of volume flux.Among them,CESM2-WACCM performs the best,with a correlation coefficient of 0.96 and a Taylor score of 0.88.Conversely,NESM3 demonstrates the largest devi-ation from the observation/reanalysis data,with the lowest Taylor score of 0.16.The variability of sea ice volume flux is primarily influenced by SIM and SIT,followed by SIC.The extreme large sea ice export through Fram Strait is linked to the occurrence of anomalously low air temperatures,which in turn promote increased SIC and SIT in the corresponding region.Moreover,the intensified activity of Arctic cyclones and Arctic dipole anomaly could boost the southward sea ice velocity through Fram Strait,which further enhance the sea ice outflow.