Skip to main content
Surfaces and interfaces

Surfaces and interfaces

Fast-moving glaciers slide more easily

05 Jan 2015
Photograph of Neal Iverson and the Iowa State University Sliding Simulator
Slippery slope Neal Iverson pictured next to the Iowa State University Sliding Simulator, which tests how glaciers slide across their beds. (Courtesy: Bob Elbert/Iowa State University)

As glaciers move faster, they experience less friction between the ice and the ground below. This is the conclusion of Lucas Zoet and Neal Iverson of Iowa State University in the US, who used a new experimental tool to simulate glacial sliding and demonstrate the importance of understanding how ice deforms to create cavities as it flows across large obstacles.

Given their potential for contributing to sea-level rise, understanding how glaciers move is vital to predicting their response to changing climates. However, gaining insights into how the underside of a huge piece of ice travels across the rough surface of the Earth is an extremely challenging problem. When modelling the flow of ice sheets, an increase in a glacier’s sliding speed was assumed to result in a corresponding increase in drag. In the late 1950s, however, the French glaciologist Louis Lliboutry proposed a more complicated sliding law – one in which increasing speed could ultimately result in a decrease in drag, once a threshold sliding speed has been exceeded. At the heart of this alternative theory is the impact of cavities that form in ice in the wake of obstacles on the Earth’s surface.

Bumps and pockets

“As ice slides forward, it has to viscously deform to get around a bump, much like water has to viscously deform to get around a stone at the bottom of a stream,” explains Zoet. Unlike with water, however, the ice is slow to fill in behind the obstacles that it passes around. “This leaves a pocket behind the bump of a size that is dependent on how fast the ice is sliding and how much pressure is acting on the ice to close the pocket,” Zoet adds.

As the glacier slides faster, the sizes of the cavities formed increase, extending out further behind the topographic obstacles that created them. According to Lliboutry’s theory, on an idealized glacier bed comprising a series of sinusoidal bumps, drag can be decreased when the cavities become large enough to extend beyond the inflection point of the next bump in the series. While this “double-value theory” has been the subject of debate, the difficulties of studying sub-glacial processes in the field – via boreholes, for example – have prevented it from being empirically tested.

To better explore these processes in a controlled setting, Zoet and Iverson created a new experimental device for simulating glacial sliding in the laboratory. Their simulator consists of a ring of ice 90 cm in diameter and 21 cm thick that is rotated above a rigid, sinusoidal bed. A hydraulic ram applies a constant downward force on the ice, simulating the weight of an overlying glacier while also allowing for the growth or contraction of cavities at the bed interface. The simulator operates within a cold room, with an additional fluid cooling system maintaining the ice ring at its pressure-melting temperature of 0.01 °C. Windows in the wall of the simulator allow the internal deformation of the ice to be observed, through the displacement of plastic marker beads embedded into the ice.

More realistic sliding rules

The researchers conducted a number of experiments to determine the relationship between the drag exerted on the ice and its sliding speed. “Lliboutry’s predictions matched our results well,” Zoet told physicsworld.com, adding that the results “will give theorists firm ground to stand on, moving forward, so that more complicated and realistic sliding rules can be developed”.

Ian Willis – a glaciologist at the University of Cambridge who was not involved in the study – calls the work “exceedingly valuable”, commenting that “it provides ammunition to the idea that existing large-scale numerical glacier and ice-sheet models – the sort of models that are used to project ice-mass responses to future warming – should incorporate such ‘double-valued’ sliding laws”.

Commending the design of the researchers’ experiment, glaciologist Martin Sharp of the University of Alberta agrees, adding that the double-valued relationship “could help to explain the episodic occurrence of ice avalanches, and the sudden changes in rates of glacier flow that are increasingly observed in the era of satellite monitoring of glacier velocities”.

The study is described in the Journal of Glaciology.

Copyright © 2024 by IOP Publishing Ltd and individual contributors