As ice shelves are sandwiched between the ocean and the atmosphere, they are sensitive elements of the climate system. A warming atmosphere and a warming ocean will directly affect them. There is no such thing as a standard ice shelf – they come in many different types. Some are captured in a bay, others are connected with the interior only on one side, some have very fast ice streams draining into them, some have marine ice freezing underneath onto the ice shelf and thickening it, others lose a lot of mass by melting from their underside, some are subject to surface melt and others not. There is a long list of manifestations. But although we can classify the structures into groups, the dynamics of an individual ice shelf are determined by factors that depend on its specific location. That said, there's one question we want to have answered about all ice shelves: are they at risk of becoming unstable?
We have observed ice shelves using field surveys since early last century and from space since the 1960s; a relatively short time on the typical 500-year timescale of ice shelf dynamics. Marine sediment records collected by the ANDRILL project have shown that the Ross Ice Shelf, which today is supposed to be stable, has retreated and advanced about 60 times in the past 10 million years. And in the past 30 years or so, seven ice shelves have completely disintegrated.
As curious human beings, we not only want to know how the cryospheric system works, but also what will happen to ice shelves in our lifetimes and how it will affect our daily lives. When ice shelves disintegrate, the inland ice masses next to them experience less back-pressure and their draining glaciers start to accelerate. This has already happened along the Antarctic Peninsula, leading to more mass transport of ice to the sea and an increase in sea-level.
All the ice shelves that have disintegrated since 1980 were located along the Antarctic Peninsula, an area which has experienced extraordinary warming. For example, air temperature has increased by up to 2.5 K in the last 50 years at Faraday/Vernadsky station on the Antarctic Peninsula. Although it’s not yet clear whether this is regional warming or linked to global warming, it begs the question whether it’s the cause for the ice shelf disintegration.
Going with the flow
The temperature of the atmosphere, as well as the temperature of the ocean, affects the temperature of the ice itself. Ice dynamics happen on the microscale, although the dimensions of the ice masses themselves are quite large. When ice is flowing at speeds of hundreds of metres per year, the grains of this polycrystalline material deform: they slide on their crystal planes, recrystallize, grow or nucleate. There are many factors that influence the deformation, for example stress acting on the ice, impurities like small rock particles, crystal size and temperature. Warmer ice is softer than colder ice, and the faster it flows, the softer it gets. So we have to deal with many contributions which all need to be quantified in some way.
Before we can determine whether the temperature increase of the past few decades is causing the disintegration, we need much more information about these factors. Unfortunately, this requires field measurements that are both time- and cost-intensive. To obtain a vertical temperature profile of an ice shelf requires drilling through some hundred metres of ice, freezing a thermistor string in and re-visiting the site over the next few years, once temperatures have stabilised after the drilling. Then that gives us the temperature of one location in an ice plate that’s some thousand square kilometres in size. To know more about the physical and chemical composition of the ice, we need to obtain an ice core and analyse it, which increases the effort exponentially. But these “physical cores” are exactly what we need to expand our knowledge of the material parameters of a single ice shelf.
Seeing the changes
As we saw during the break-up events in February, May, and June/July of 2008 on the Wilkins Ice Shelf, satellite remote sensing is a very useful tool for observing changes in an ice shelf – once the changes are manifested in a way that makes them “visible” to the sensors of the satellites. This means that there are either changes on the surface of the shelf, such as fracturing of the ice or increased summer surface melt, or variations at its margins like iceberg formation.
But remote sensing can do much more: once satellites are retrieving images repeatedly under specific settings, we are able to determine the deformation pattern of an ice shelf, which tells us its stress regime. Other sensors provide us with information on ice shelf elevation and how it changes over time, giving us an indication of ice shelf thickness. As a result, the multitude of current and upcoming satellite systems increase our knowledge, but they need to acquire continuous time series and they must be combined and validated with ground measurements.
Theoretical models of ice shelves describe the deformation of an ice shelf in a continuum mechanical way, assuming the ice to be an incompressible fluid. They use balance equations, such as mass balance and energy balance, and constitutive equations like flow law in which strain is related to stress. With the benefit of some simplifications, for example that ice shelves are very flat and that there is no shear stress at the ice/ocean interface, we end up with a set of equations we can solve numerically. Of course, this requires input parameters, including ice thickness, the inflow speed of the glaciers draining into the ice shelf, the melt rate at the ice/ocean interface and many more. Once we know these parameters precisely enough, we can test the theory and see if our current theoretical description reflects reality well. At that stage we can perform sensitivity studies to see the influence of warming of the atmosphere, changes in snow accumulation or melting at the ice/ocean interface.
On the next level we can go even further and use damage mechanics to calculate where and under which conditions fractures will occur. Under this theory, the ice acts in many ways like a fluid and also partly like a solid. Cracks start on a crystal level with micro-cracks and move up to the macro-scale, where fractures open when the stress is above a critical threshold – yet another parameter. This already increases the complexity of the theoretical description tremendously and we are still missing one aspect: the different mechanisms of ice shelf break-up, some of which await discovery.
But although we might never be able to predict the date and the exact location of an ice shelf break-up, we are able to get a sufficient understanding of ice shelf dynamics to allow us to answer the question of their stability – provided that we increase our efforts to obtain the required parameters.
For those ice shelves that have already completely disintegrated, most of these parameters will not be accessible. For the Wilkins Ice Shelf and all others that still remain intact, we still have a chance to get enough data to be able to study the impact of warming at the required level of precision and to assess the ice shelves’ risk of becoming unstable.
References:
M. Braun, A. Humbert, A. Moll, “Changes of Wilkins Ice Shelf over the past 15 years and inferences on its stability”
The Cryosphere Discuss., Vol. 2, pp. 341-382, 2008.
Fox, D., “ANTARCTICA: Freeze-Dried Findings Support a Tale of Two Ancient Climates”
Science, Vol. 320., No. 5880, pp. 1152 – 1154, doi: 10.1126/science.320.5880.1152
H. D. Pritchard, D. G. Vaughan, “Widespread acceleration of tidewater glaciers on the Antarctic Peninsula” J. Geophys. Res., Vol. 112, F03S29, pp. 1-10, doi:10.1029/2006JF000597, 2007.
A. Humbert, R. Greve, and K. Hutter. "Parameter sensitivity studies for the ice flow of the Ross Ice Shelf, Antarctica". Journal of Geophysical Research, 110(F4)(F04022):doi:10.1029/2004JF000170, 2005.
1 comment
Comments on this article are now closed.
Date and the location of ice shelf break-up
Edited by abm33 on Oct 8, 2008 10:22 AM.
Comments on this article are now closed.