Why is Summertime Arctic Sea Ice Drift Speed Projected to Decrease?
Abstract. Alongside declining Arctic sea ice cover during the satellite era, there have also been positive trends in sea ice Arctic-average drift speed (AADS) during both winter and summer. This increasing sea ice motion is an important consideration for marine transportation as well as a potential feedback on the rate of sea ice area decline. Earlier studies have shown that nearly all modern global climate models (GCMs) produce positive March (winter) AADS trends for both the historical period and future warming scenarios. However, most GCMs do not produce positive September (summer) AADS trends during the historical period, and nearly all GCMs project decreases in September AADS with future warming. This study seeks to understand the mechanisms driving these projected summertime AADS decreases using output from 17 models participating in the Coupled Model Intercomparison Project phase 6 (CMIP6) along with 10 runs of the Community Earth System Model version 2 Large Ensemble (CESM2-LE). The CESM2-LE analysis reveals that the projected summertime AADS decreases are due to changes in sea surface height (SSH) which act to reduce sea ice motion in the Beaufort Gyre and Transpolar Drift. During March, changes in internal stress and wind stress counteract these tilt force changes and produce positive drift speed trends. The simulated wintertime mechanisms are supported by earlier observational studies, which gives confidence that the mechanisms driving summertime projections are likely also at work in the real world. However, additional research is needed to assess whether the simulated summertime internal stresses are too weak compared to the tilt forces. The projected summertime SSH changes are primarily due to freshening of the Arctic Ocean (i.e. halosteric expansion), with thermal expansion acting as a secondary contribution. The associated ocean circulation changes lead to additional piling up of water in the Russian shelf regions, which further reinforces the SSH increase. CMIP6 models also show evidence of SSH-driven projected decreases in summertime Arctic sea ice motion, but the models show a wide range of regional SSH trend patterns. Due to small ensemble sizes and the unavailability of required daily output, we were not able to further examine mechanisms in the CMIP6 models. Altogether, our results motivate additional studies to understand the role of SSH in driving changes of Arctic sea ice motion.
Viewed (geographical distribution)