Impact of mid-latitude/polar winds on Antarctic sea ice and tropical climate

Abstract:

Antarctic sea ice extent has slightly increased over the satellite observational period (1979 to present) despite global warming. Several mechanisms have been invoked to explain this trend, such as changes in winds, precipitation, or ocean stratification, yet there is no widespread consensus. Additionally, fully coupled earth system models run under historic and anthropogenic forcing generally fail to simulate positive sea ice extent trends over this time period.

In this work, we quantify the role of atmospheric winds on sea ice trends and variability with an earth system model run under historic and anthropogenic forcing and with winds poleward of 45 degrees nudged to observed winds taken from the ERA-Interim reanalysis from 1979 to 2018. Simulated Antarctic sea ice captures the observed interannual variability in sea ice extent over 1979-2018 and the observed long-term trends from the early 1990s onwards, yet for the longer 1979-2018 period it does not simulate a positive sea ice trend, in part due to faster-than-observed warming at the global and hemispheric scale in the model. We also find negligible impact of the sea ice initial conditions in 1979 on long term trends. At the regional scale, simulated sea ice shows higher skill compared to the pan-Antarctic scale both in capturing trends and interannual variability. Additionally, we find that the model captures a significant component of observed tropical Pacific sea surface temperature trends and interannual variability, indicating a forcing of tropical climate from the mid-latitudes and polar regions.