Greenland's Glacial Melt May Decelerate

Glaciology: From the front

Stephen Price1

1. Stephen Price is with the COSIM project, Fluid Dynamics and Solid Mechanics Group, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, USA.
e-mail: sprice@lanl.gov

Abstract

The causes of recent dynamic thinning of Greenland's outlet glaciers have been debated. Realistic simulations suggest that changes at the marine fronts of these glaciers are to blame, implying that dynamic thinning will cease once the glaciers retreat to higher ground.
Introduction

For the past decade, many outlet glaciers in Greenland that terminate in the ocean have accelerated, thinned and retreated. To explain these dynamic changes, two hypotheses have been discussed. Atmospheric warming has increased surface melting and may have also increased the amount of meltwater reaching the glacier bed, increasing lubrication at the base and hence the rate of glacier sliding1. Alternatively, a change in the delicate balance of forces where the glacier fronts meet the ocean could trigger the changes2, 3, 4. On page 110 of this issue, Faezeh Nick and colleagues5 present ice-sheet modelling experiments that mimic the observations on Helheim glacier, East Greenland, indicating that the dynamic behaviour of outlet glaciers follows from perturbations at their marine fronts.

Greenland's ice sheet loses mass partly through surface melting and partly through fast-flowing outlet glaciers that connect the vast plateau of inland ice with the ocean. As the outlet glaciers flow into the sea, icebergs calve from their fronts. As highlighted in the fourth assessment report of the Intergovernmental Panel on Climate Change6, earlier ice-sheet models have failed to reproduce the dynamic variability shown by ice sheets over time. It has therefore not been possible to distinguish with confidence between basal lubrication from surface meltwater and changes at the glaciers' marine fronts as causes for the observed changes on Greenland's outlet glaciers.

The distinction bears directly on sea-level rise — the motivation for much of modern-day glaciology. If the recent dynamic mass loss from Greenland's outlet glaciers is linked to changing atmospheric temperatures, it may persist for as long as temperatures continue to increase. However, if the source of the dynamic mass loss is a perturbation at the ice–ocean boundary, these glaciers will lose contact with that perturbation after a finite amount of thinning and retreat. Therefore, the first hypothesis predicts continued retreat of outlet glaciers into the foreseeable future, whereas the second does not — provided the bedrock topography prohibits a connection between the retreating glacier and the ocean.

Nick and colleagues5 tested the physical mechanisms of each hypothesis in an innovative ice-flow model, and used that model to match a time series of observations from Helheim glacier, one of Greenland's three largest outlet glaciers (Fig. 1). They found that a reduction in resistance at the glacier front — which might result, for example, from the loss of a floating ice-tongue or a change in calving rate — triggers glacier behaviour in the model that is in broad agreement with observations7 from 2001–2006. Importantly, the model captures the observed pattern of a relatively minor initial acceleration, followed by more rapid acceleration and thinning as the glacier terminus retreats into deeper water across a bedrock low, and subsequent deceleration and stabilization as the glacier retreats into shallower water. In contrast, for experiments where basal lubrication was altered to simulate increased sliding from meltwater input, the modelled velocity and geometry show little similarity to the observations.

Along with many observations2, 3, 4, 7, 8, 9, 10, Nick and colleagues' simulations strongly support the contention that the recent retreat of Greenland's outlet glaciers is the result of changes at their marine fronts. Furthermore, the simulations confirm the earlier hypothesis7 that bedrock topography largely controlled Helheim glacier's rapid acceleration and retreat in 2004 and 2005, and its deceleration and stabilization in 2006. Finally, the current work indicates that, if requirements of observational data (high-resolution bedrock topography) and computational resources (fine computational grid resolution) can be met, improved predictive capability for ice-sheet models is attainable. With respect to the concerns raised by the Intergovernmental Panel on Climate Change, this study signals progress.

OTOH, Antarctica may continue to accelerate for a looong time.