How Antarctic ice shelves melt

A new theory explaining the two different ways in which ice shelves melt is presented in this week’s Proceedings of the Royal Society A. This theory may help to explain satellite measurements of inland ice thinning near the coast of Antarctica.

Flowing ice can respond in two entirely different ways to the encroachment of warm ocean water. Scientists at the British Antarctic Survey have discovered that the way in which ice streams melt depends on whether the force acting on them changes suddenly, or gradually.

Warm ocean water can thin the ice shelves, hundreds of kilometres across and hundreds of metres thick, which currently fringe the Antarctic. This force lessens the ice shelves’ ability to push back against ice streams draining from the inland Antarctic ice sheet and this accelerates the flow of ice into the ocean which in turn raises the sea level.

C. Rosie Williams and her team examined how inland ice streams respond to different magnitudes and frequencies of stress, from the thinning ice shelf front. The results, if extrapolated backwards, might help to explain the thinning of inland ice in Antarctica, via the nature of past ocean-ice interactions.

The scientists discovered that how quickly a stress moves upstream, and effects inland ice thickness, depends on the frequency of the force. This was observed along two distinct behavioural branches.

At low frequencies (a force once every hundred or thousand years), slope, thickness and shallow ice approximations are sufficient to explain the stress travelling upstream; it does not depend on the stress on the ice membranes.

At high frequencies (a force inland every 10 years or less), the force travels upstream via direct transmission through membranes and thus penetrates tens of kilometres inland. The speed of the flowing ice adjusts rapidly to such forcing; however, the thickness varies little.

Although they found that at high frequencies, the forces travelling upstream did not significantly effect ice thickness, the team only investigated regular periodic forcing. In the real world changes can be rapid, non-periodic and non-linear, and hence further investigation is required to assess real-world implications.

It is also important to note that the properties found vary significantly between the 29 Antarctic ice streams considered. This further demonstrates the dynamic response to sudden ice shelf change. The results imply that each ice stream requires a separate analysis; within a decade a good forcing and thinning record will have been obtained for most ice streams.