Wednesday, May 14, 2014

Why Melting of the West Antarctic Ice Sheet Should Freak You Out



Two scientific studies on melting of the West Antarctic Ice Sheet were published this week, driving a firestorm of sensationalist headlines in the New York Times, National Geographic, CNN, and other websites on May 12th (unsurprisingly, Fox News has maintained radio silence on this topic at the time of this article’s publication). Dire warnings of unstoppable glacier melt and runaway sea level rise abound. Is this all media hype? If you read no further, the answer is no- this is real.

What's going on in West Antarctica?

Three major ice sheets remain on Earth today: Greenland, West Antarctica and East Antarctica (see my previous entry). Of these, the West Antarctic Ice Sheet, though smallest by volume, is considered the most vulnerable to future melting and collapse. This is because the outer margin of the West Antarctic Ice Sheet sits on bedrock that lies below sea level, unlike the other major ice sheets. Such a “marine-based” ice sheet is sensitive to melting both on the top of the ice, from above-freezing atmospheric temperatures, and below the ice, from incursion of ocean water. Even though the ocean around Antarctica is cold, its warmth and saltiness relative to the ice is sufficient to melt the ice sheets.

Today, the flow of warmer, salty ocean water beneath the West Antarctic Ice Sheet is constrained by the contact between the ice sheet and the ground beneath the ice. This is referred to as the “grounding line”. The position of the grounding line relative to the topography beneath the ice sheet is critical to determining the stability of marine-based ice sheets in response to warming1

Stable (a, left) versus unstable (b, right) configurations of marine-based ice sheets. From Hanna et al., 2013 (ref. 1)

Specifically, if the grounding line is located on a topographic upslope or ridge, the ice sheet is stable after initial retreat, because thinning and retreat of the ice sheet in response to warming is compensated by a shallower grounding line depth (panel a, above). If the grounding line is located behind a ridge or on a downslope, there is no topographic “catch” for the grounding line, so the ice sheet is unstable and will quickly melt and retreat (panel b, above).

Why does this matter? Glaciers from the Amundsen Sea in West Antarctica contain enough ice to raise sea level by 4 feet, and have rapidly melted over the past two decades2. The two scientific studies published this week provide complementary approaches to addressing the susceptibility of these glaciers to future melting (link to cool YouTube video from NASA/JPL). Rignot and colleagues3 provide an updated analysis of the topography of the bedrock underneath glaciers in the Amundsen Sea. The authors find that rapid retreat of these glaciers observed over the past 20 years is closely tied with elevation of the bedrock underneath the ice. Importantly, improved imaging of the bedrock underneath the current glaciers shows that, inward of the ridge upon which the grounding lines are presently located, there are no major topographic ridges or other features that may compensate for a thinning ice sheet. If these glaciers continue to melt as they are today, they will eventually enter a highly unstable regime where the entire glacier is susceptible to collapse.

Joughin and colleagues4 use a numerical model to forecast the timing over which such a collapse will occur. Their model correctly reproduces observed rates of melt and glacial retreat over the past 20 years, which gives confidence that the model accurately describes the timing and magnitude of melting in this glacier system. Projecting forward, the authors find that the Amundsen Sea glaciers reach a scenario of complete collapse within 200 to 500 years, and indeed are likely already experiencing an early stage of collapse today.

"Hysteresis"
Many phenomena in the Earth’s climate system are subject to hysteresis- the dependency of an outcome on not just the current state, but on previous states as well. Hysteresis is perhaps easiest explained visually, as a variable that does not respond in a linear fashion to changes in a causal variable. 


Hypothetical example illustrating the difference between linear responses and hysteretic responses. Note: this represents a major oversimplification of the climate system, and the exact relationships between CO2, temperature and sea-level rise shown here should not be accepted as scientific fact. 
In the hypothetical example shown above, the forcing agent (CO2, orange) causes a linear response in temperature (red). When CO2 rises or falls, temperature follows with a corresponding rise or fall. However, the bottom graph of sea-level rise (blue), which is assumed to be forced by temperature, shows a response that is very different from its forcing. The sea-level rise response has inertia from the previous state (increase in temperature), and so is unresponsive to the change in forcing (decrease in temperature). In this hypothetical example, sea-level rise exhibits hysteresis. Even if a future corrective action was applied to ameliorate CO2 levels and force a reduction temperature, the sea-level rise persists.

The example is illustrative of perhaps the scariest aspect of Amundsen Sea glacier melting: its hysteresis. Without shallower bedrock to “catch” the grounding line, there is no known mechanism by which collapse of a marine-based ice sheet can be prevented, or reversed. Although glaciology’s knowledge of ice sheet-bedrock interactions is imperfect, the Rignot and Joughin studies suggest complete melting of the Amundsen Sea glaciers could happen regardless of any future events.

What Happens once Glaciers Collapse?
Take Miami, Florida as an example. Miami-Dade County has a population of 2.56 million, and averages a mere six feet above sea level today. With a four-foot sea level rise from melting of the Amundsen Sea glaciers, 40% of urban Miami-Dade County will be underwater (blue shaded area, below). Moreover, under such a sea-level rise it would take only a two-foot storm surge to inundate a majority (60%) of modern-day Miami (blue plus red shaded area, below). A recent study estimated that the Miami region receives a two-foot or larger storm surge every ten to twenty years due to hurricane activity5, meaning the majority of modern-day Miami would be flooded on a semi-regular basis due to natural storm activity following collapse of the Amundsen Sea glaciers.


Hypsographic curve for Miami-Dade County, showing the percentage of land (horizontal axis) beneath a certain elevation above present sea-level (vertical axis) (adapted from ref. 6).
Sea-level effects from Amundsen Sea glacier collapse would extend far beyond South Florida. Globally, an estimated 634 million people, almost 10% of the global population, live within 30 feet of sea-level. Although more detailed demographic-elevation estimates are not yet available, a significant fraction of this population would likely be endangered from a four-foot sea-level rise, particularly in heavily populated, low-lying areas of Southeast Asia.

Can this be stopped?
With such risks from sea-level rise, halting the melting of the Amundsen Sea glaciers would appear to be of paramount importance. Any approach to engineer a cessation of melting, however, is quickly overwhelmed by the sheer scale of the problem. Melting of the Amundsen Sea glaciers today is driven by incursion of warmer ocean water; ceasing the melting would thus literally require cooling down the ocean.

Alternatively, the melting of the Amundsen Sea glaciers could be counteracted by an increase in ice gain from additional snowfall on the top of the glacier. However, the volume necessary is staggering: 18.6 billion Goodyear Blimps full of water would have to be frozen onto the Amundsen Sea glaciers each year, just to counteract the ~100 gigatons of ice lost to melt each year observed at present6. As the ice sheet continues to destabilize in the future, this volume is expected to rise significantly.

18.6 billion of these per year equals a lot of ice.
The (Tarnished) Silver Lining
If there is any good news, it is that the rate of collapse of the Amundsen Sea glaciers is estimated to be on the order of several centuries. In his blog, Andy Revkin points out the stark difference between geologic and societal meanings of “collapse”; for geology, a “sudden collapse” still requires multiple human generations. So we have plenty of time to prepare.

In addition, science is never settled. If the incursion of warm oceanic waters in the Amundsen Sea slows down, perhaps the glaciers will recover and avoid collapse. Of course, the opposite- increased warming and an even more rapid collapse- is equally plausible. We may yet learn of new mechanisms that lead to stabilization of marine-based glaciers. But with the state of scientific knowledge at this moment, there is little reason to suspect we can avoid, at a minimum, four feet of sea-level rise in the coming several centuries. Take your talents to South Beach soon.


References
1. Hanna, E. et al. (2013) Ice-sheet mass balance and climate change. Nature, 498, 51-59, doi:10.1038/nature12238.
2. Rignot, E. et al. (2013) Ice-shelf melting around Antarctica. Science, 341, 266-270, doi:10.1126/science.1235798.
3. Rignot, E. et al. (2014) Widespread, rapid grounding line retreat of Pine Island, Thwaites, Smith and Kohler glaciers, West Antarctica from 1992 to 2011. Geophysical Research Letters, doi:10.1002/2014GL060140.
4. Joughin, I. et al. (2014) Marine Ice Sheet Collapse Potentially Underway for the Thwaites Glacier Basin, West Antarctica. Science Express, doi:10.1126/science.1249055.
5. Klima, K. et al. (2012) Hurricane Modification and Adaptation in Miami-Dade County, Florida. Environmental Science and Technology 46, 636-642, doi:10.1021/es202640p.
6. Southeast Florida Regional Climate Change Compact Technical Ad hoc Work Group. April 2011. A Unified Sea Level Rise Projection for Southeast Florida. A document prepared for the Southeast FL Regional Climate Change Compact Steering Committee. 27 p.

Additional Reading