Predicting sea level rise: Understanding how icebergs form could lead to better forecasts

November 23, 2010
A helicopter view of a crack in the Amery Ice Shelf in East Antarctica that has been propagating at about 3 meters per day for close to two decades. Eventually, this crack system is expected to yield an iceberg. Photo by Jim Behrens

( -- In an effort to understand how fast sea level could rise as the climate warms, a University of Michigan researcher has developed a new theory to describe how icebergs detach from ice sheets and glaciers.

This process of "iceberg calving" isn't well understood. While scientists believe it currently accounts for roughly half of the mass lost in shrinking ice sheets, current models don't take changes in iceberg calving into account in their predictions, says Jeremy Bassis, an assistant professor in the Department of Atmospheric, Oceanic and Space Sciences.

Bassis is the author of a paper on the new theory published online today in the . It will appear in print in the January edition.

"Our models cannot predict about half of the mass balance. We don't know how much of an effect this will have, but we've seen several prominent examples where calving is connected with speed-up of the ice-retreat process," Bassis said.

The Larsen B ice shelf in Antarctica—a 2,000-square-mile, 700-foot-thick slab that had been stable for thousands of years—disintegrated in about six weeks between January and March of 2002. Scientists believe rising temperatures and ice fracturing primarily caused the disintegration. But they don't have a handle on exactly how it happened.

When ice breaks off of ice shelves, it doesn't directly or immediately cause melting and sea level rise. But scientists believe it can contribute to and hasten those processes. The icebergs can float into warmer parts of the ocean and melt. And ice calving can perhaps lead to more fracturing.

"What we've been lacking is a way to quantify whether calving will increase or decrease in the future, and do it in a way that doesn't involve information from a particular glacier. There are thousands and thousands of glaciers that might break up because of change. We'd like to have a theory that explains how this happens for any regime and this is a step in that direction," Bassis said.

Bassis' theory is a statistical methodology that allows him to predict the average iceberg calving rate and how much that can be expected to fluctuate for a given region based on the amount of stress within the ice and its thickness. He was able to construct a more general model than current ones, he says, because he assumed that cracking happens somewhat randomly. He didn't try to predict where every fracture would occur.

"You don't need to understand what every glacier is doing if you know what are doing on the whole," Bassis said.

Current predictions of sea level rise range from about four inches to 2.5 feet by 2100. This is a wide range that Bassis' theory could help to narrow.

"From a societal perspective, a significant portion of the Earth's population lives very close to sea level. We'd like to be able to better predict how much sea level will rise so we can build defenses or, if necessary, plan for migrations," Bassis said.

Explore further: Footloose Glaciers Crack Up: What Happens When Glaciers Float On Ocean Surface

More information: The paper is called "The statistical physics of iceberg calving and the emergence of universal calving laws."

Related Stories

Footloose glaciers crack up

July 14, 2010

Glaciers that lose their footing on the seafloor and begin floating behave very erratically, according to a new study led by a Scripps Institution of Oceanography, UC San Diego researcher.

Scientists expect increased melting of mountain glaciers

January 20, 2006

Sea level rise due to increased melting of mountain glaciers and polar ice caps will be much lower in the 21st Century than previously estimated. However, decay of mountain glaciers in due to global warming will be much more ...

Sea level rise of 1 meter within 100 years

January 8, 2009

New research indicates that the ocean could rise in the next 100 years to a meter higher than the current sea level - which is three times higher than predictions from the UN's Intergovernmental Panel on Climate Change, IPCC. ...

Greenland and Antarctic ice sheet melting, rate unknown

February 16, 2009

The Greenland and Antarctica ice sheets are melting, but the amounts that will melt and the time it will take are still unknown, according to Richard Alley, Evan Pugh professor of geosciences, Penn State.

Recommended for you

Climate change made Harvey rainfall 15 percent more intense

December 14, 2017

A team of scientists from World Weather Attribution, including researchers from Rice University and other institutions in the United States and Europe, have found that human-caused climate change made the record rainfall ...

East Antarctic Ice Sheet has history of instability

December 13, 2017

The East Antarctic Ice Sheet locks away enough water to raise sea level an estimated 53 meters (174 feet), more than any other ice sheet on the planet. It's also thought to be among the most stable, not gaining or losing ...

Hydraulic fracturing negatively impacts infant health

December 13, 2017

From North Dakota to Ohio to Pennsylvania, hydraulic fracturing, also known as fracking, has transformed small towns into energy powerhouses. While some see the new energy boom as benefiting the local economy and decreasing ...


Adjust slider to filter visible comments by rank

Display comments: newest first

2 / 5 (4) Nov 23, 2010
The sea level isn't rising any more than it has for thousands of years.
3.7 / 5 (3) Nov 24, 2010
"The sea level isn't rising any more than it has for thousands of years."

Thanks all my fears are assuaged, praise NotParker for his deep knowledge and well expounded critique of the current paradigm

2.3 / 5 (3) Nov 24, 2010
Take a look at the grpah of the last 8000 years.


And guess what ... the sea level will gradually rise until it stops and then it will stop rising as the end of this interglacial approaches and we plunge into the rest of this ice age.

It will get about 10C cooler, pretty much ending agriculture in large parts of the earth.

Take a look at the bottom graph:


3 / 5 (4) Nov 24, 2010
NotParker, you are a subversive, resorting to facts. Shame on you.
2.3 / 5 (3) Nov 24, 2010
Quote from an Aussie journalist who worked for Murdock is not a source of science, might as well qoute Palin or any other schill on Fox News :)

2.3 / 5 (3) Nov 24, 2010

"Not long ago, I read the Skeptics Handbook which displays some fundamental misunderstandings of how our climate works. I wondered whether a rebuttal of this document would be worthwhile but when I floated the idea to a few people, the general response was "been there, done that". ...

"The 'Skeptics Handbook' begins by asking "what evidence is there that more CO2 forces temperatures up further?" It then lays out 4 arguments: the greenhouse signature is missing, CO2 lags temperature, it's not warming and the CO2 effect is saturated. The great irony of the 'Skeptics Handbook' is when you examine these 4 arguments and the full body of empirical evidence that goes with them, what you actually find is the evidence that more CO2 forces temperatures up further.

1 / 5 (1) Nov 26, 2010
This is a little dangerous in principle. In the title of the paper he uses the term "statistical physics", which is what he is proposing here since theoretical physics can't predict glacier processes very well (if I'm reading it correctly).

In a chaotic system like those involved in thermodynamics, it seems entirely appropriate to use statistical physics to be able to predict the behavior of a system. It's been demonstrated that the statistically derived laws of thermodynamics produce a good predictive model of many systems under many circumstances. I'm not sure you can do that with ice bergs with as much confidence. How do you verify the efficacy of the statistical model? Can you test it in an isolated way over a meaningful time span? Can/should the model be used before it is verified by testing?

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.