This month's Geophysical Research Letters brings more ice sheet melting data to be concerned about.  A paper by Eric Rignot and colleagues at JPL, Caltech, UC Irvine, and Utrecht University demonstrates the "Acceleration of the contribution of the Greenland and Antarctic ice sheets to sea level rise".  Here is a nice summary at ScienceDaily that goes into some broader implications for sea level rise.  In addition to putting a bunch of nice data and analysis on the table, Rignot et al will contribute substantially to a broader understanding of the overall ice/water system near the poles.

The paper describes work integrating a variety of methods to build up a two decade-long picture of ice mass loss in Antarctica and Greenland.  The numbers by themselves are pretty impressive: the ice sheet loss rate was ~478 Gigatons/year in 2006, with an acceleration of ~36 Gigatons/year^2.  Note that this means the acceleration is 7.5% of the rate -- in other words, ice sheet mass loss is speeding up at a remarkable clip for a process that is ongoing at continental length-scales.  Notably, the authors report a very small uncertainty in the acceleration (about 5%), which means that we can be quite certain there is a large non-linear contribution that is reducing ice sheet mass (one that is proportional to time^2).

Here are a few tidbits that are not in the paper or associated press stories.  I wrote to Dr. Rignot to satisfy my curiosity about a couple of points, and he graciously responded and gave me permission to quote the emails here.

First, after staring at the ice loss rate and acceleration data for a little while, I got to wondering why the authors extracted a linear change in the rate of ice loss, which results in a constant acceleration.  Given the data, you might wonder whether the acceleration was actually increasing rather than being a constant.  In our brief email exchange, Dr. Rignot said that while the linear fit was the simplest fit, it appears that, in fact, the acceleration is increasing in Antarctica (no word from Dr. Rignot about Greenland).  The team is going to wait for a few more years worth of data -- to increase their certainty and better constrain the statistical significance -- before they talk more about it.

This is pretty important.  We are talking about adding a highly non-linear term to models of the total ice sheet mass, one that is proportional to (time^3).  Depending on the size of the change in acceleration, this could radically change estimates of sea level rise from melting.

The present paper already demonstrates that ice sheet loss will account for substantially more sea level rise than is included in the IPCC models.  In addition, the authors observe that increased mass loss is likely to lead to a substantial increase in the speed at which glaciers deliver ice to nearby water, something that is not adequately addressed (or is simply not included, if you are following that story) in IPCC forecasts.

Perhaps it is time to revisit investing in water wings.