Evaluation and Projection of Arctic sea ice concentration under various emission scenarios using CMIP6 models

The rapid decline and warming of Arctic sea ice, particularly in September when the annual minimum occurs, confirms the manifestation of climate change in polar regions. September sea ice trend demonstrates the high likelihood of ice-free summers in Arctic in near future raising critical concerns for global climate systems. This study systematically evaluates the performance of 22 Coupled Model Intercomparison Project Phase 6 (CMIP6) models in simulating September Arctic Sea Ice Concentration (SIC) during the period 1959-2014. Using advanced skill metrics-root mean square error (RMSE), interannual variability skill score (IVS), Taylor skill score, and M-score-a comprehensive ranking identifies ten top-performing models, including EC-Earth3-Veg, ACCESS-CM2, and CESM2-WACCM. The analysis reveals that CMIP6 models effectively capture the September SIC and variability offering confidence in their predictive capabilities. Employing the best-performing models, probability density distribution functions of September SIC are generated under Shared Socio-economic Pathways (SSPs), projecting a significant reduction of up to 90% in Arctic SIC by the end of the century. Our findings highlight a critical acceleration in sea ice loss under high-emission scenarios, with profound implications for global climate feedbacks, polar ecosystems, and socio-economic systems. By addressing key gaps in model evaluation and future projections, our study contributes to advancing climate modelling efforts and improving our understanding of Arctic sea ice dynamics during this pivotal month.