Ice Update: Not So Fast Antarctica

by Bill Chameides | October 22nd, 2009
posted by Erica Rowell (Editor)

Permalink | 1 comment

Post-glacial rebound — the very slow movement of the Earth’s crust still responding to the last ice age millennia ago — presents some challenges to our high-tech tools for measuring the Earth’s huge ice sheets. (Photo: NASA/GRACE team/DLR/Ben Holt Sr.)

Slower moving rocks under the Antarctic spell slower rates of Antarctic ice sheet melting.

In a recent post I discussed a paper by Isabella Velicogna of the University of California, Irvine, that reported two items of note:

  • both the Greenland and Antarctic ice sheets are melting and
  • the rate of loss has been increasing at an impressive clip.

The Velicogna findings were based on data gathered by the satellite-borne Gravity Recovery and Climate Experiment (GRACE) instrument.

How Do GRACE Satellites Measure Ice Mass on Earth?

GRACE is an amazingly elegant instrument: as it sweeps over the Earth’s surface, it uses highly precise measurements of the small variations in the Earth’s gravitational pull on the satellite to infer the mass of objects and material at the Earth’s surface. (More on GRACE here and here.)

In the case of large ice sheets like those of Greenland and Antarctica, repeated measurements over time make it possible to determine if the mass of the ice sheet is increasing or decreasing and at what rate.

Velicogna’s analysis revealed that between 2002 and 2006, the Antarctica ice sheet lost mass at a rate of 104 gigatons (Gt) of water per year, but between 2006 and 2009 the rate of loss increased to 246 Gt of water per year. (By way of reference, a gigaton of water is one billion tons — or roughly 250 billion gallons — of water, enough water to fill about 400,000 Olympic sized pools.)

Now, like most measurements that require high levels of precision, there is a potential for artifacts in the data that can lead to systematic errors. In the case of GRACE measurements of ice sheet mass, one of the largest potential sources of error is “post-glacial rebound” (PGR).

In order to infer the mass of the ice sheet at any one point, the nearby mass of the underlying rock must be subtracted out. Any error in the estimate of the rock mass will get propagated as an error in the estimate of the ice mass. Unfortunately, determining the mass of the rock is complicated because the Earth’s surface is very slowly rising as part of its long recovery from being buried under ice during the last ice age (which peaked about 20,000 years ago); we call this the post-glacial rebound or PGR. To accurately know the mass of rock, one must know the rate of this rebound.

Analyses like Velicogna’s use PGR models to correct for the rock mass rebound. Recognized as a poorly constrained area in ice mass estimates, these models typically have a large margin of error. This is reflected in Velicogna’s paper; she estimated that the PGR correction has an uncertainty of about 76 Gt per year for the Antarctic.

A Different View from Ground-Based Instruments Hooked to GPS

Against this backdrop comes last week’s paper in Geochemistry, Geophysics, Geosystems, in which Michael Bevis of Ohio State University and co-authors present data from a different source. Bevis’s team studied the output from a new network of field stations that use Global Positioning System (GPS) transmitters mounted into the bedrock across West Antarctica to determine the amount of Antarctica’s PGR.

What’s significant about the GPS network is that it provides much needed ground truth for the models used to correct for PGR and it will enable researchers to improve estimates of ice mass changes from GRACE in the future. The early result from the initial work by Bevis et. al. is that PGR models overestimate the amount of PGR and therefore also overestimate the rate of ice loss from the Antarctic ice sheet.

By how much? Other work will be needed to answer this question definitively. In the interim, Bevis et al. estimate the potential magnitude of the GRACE-inferred overestimation of ice loss may be on the order of about 33 Gt per year — implying a ~30 percent decrease in Velicogna’s rate of ice loss over the 2002 to 2006 time period and ~12 percent decrease over the 2006 to 2009 time period.

However, keep in mind that PGR occurs very slowly and so rapid changes in ice mass detected by GRACE over the span of a few years reflect ice loss, not PGR. And so while the work of Bevis et al. suggests that the rates of ice loss may need to be dialed down somewhat, it does not undermine Velicogna’s major findings that Antarctic ice loss is occurring and is accelerating. We’ll have to wait a while to see if this acceleration is a short-term phenomenon or part of a long-term trend.

filed under: climate change, faculty, global warming
and: , , , , , , ,

1 Comment

All comments are moderated and limited to 275 words. Your e-mail address is never displayed. Read our Comment Guidelines for more details.

  1. MattN
    Oct 22, 2009

    30% is alot to be off. In my line of work (Engineering) being that much off all the time would get me fired….

©2015 Nicholas School of the Environment at Duke University | Box 90328 | Durham, NC 27708
how to contact us > | login to the site > | site disclaimers >

footer nav stuff