The observed increase in Antarctic sea ice is thought to be mostly driven by surface winds. These winds drive the motion of sea ice and shift warm air southwards to melt ice in some regions.

However, global climate models do not reproduce observed sea ice trends. Ocean-sea ice models that are driven by observed wind changes do reproduce the pattern of observed sea ice trends, but the trends are weaker than observed.

To understand what was happening, CLEX researchers analysed the sea ice volume budget in 10 climate models to see how much the simulated sea-ice change was driven by dynamic processes, such as wind‐driven sea ice motion, and how much was due to thermodynamic freezing and melting processes.

They found that the models generally agreed on changes to average yearly cycle of freeze and melt, with dynamic processes dominating the sea ice edge and thermodynamic processes dominating the interior of the sea ice pack.

However, the models disagreed about the trends of sea ice volume.

In the sea-ice interior, each process dominates around half of the models. Thermodynamic processes tend to dominate in models with low‐magnitude averages of sea ice motion.

These results provide insight into the estimates and patterns of Antarctic sea ice in global climate models, helping to understand the disparity between simulated and observed sea ice trends.

Based on these results it seems justified to re-examine whether observed trends are indeed dominated by sea ice motion or whether the dynamical effects of wind stress play a smaller role than warm- or cold-air advection in driving Antarctic sea ice change than is currently presumed.

The disparity between wind and sea ice trends in some of the models also suggests that ocean-sea ice processes may contribute strongly to driving changes to Antarctic sea ice and require further examination in global climate models. A major problem is that it is extremely difficult to accurately measure Antarctic sea ice thickness by satellite, so previous studies have just considered changes in sea ice cover, rather than the total amount of sea ice. How well processes affecting sea ice cover relate to changes in total sea ice volume is not well known.


  • Paper: Schroeter, S., Hobbs, W., Bindoff, N. L., Massom, R., & Matear, R. (2018). Drivers of Antarctic sea ice volume change in CMIP5 models. Journal of Geophysical Research: Oceans, 123.