Tuesday, December 16, 2014

Searching for the dragons - with Richard Alley at AGU14

Professor Richard Alley gave one of the first lectures at AGU14. (Click here for how to view AGU14 videos.)

Richard talks in his usual lively manner, with an early reference to the possibility that some things may happen very quickly, much faster than economic discounting allows. He talks about how faster, less-expected changes would be very damaging to economies and ecosystems. He says that "in some sense, we are searching for the dragons that are out there" - in a reference to a fifteenth century map showing where "here be dragons".

He points out that when we discover a possible dragon part we examine it further - and often find that it's "not likely to eat us" - or we may find that it is indeed really dangerous.

Examples of "not likely to eat us", or not yet, are belching methane sediments in the sea floor and "The Day After Tomorrow" events. He spoke of how most things we'll be faced with are gradual climate change pushing systems across thresholds (of our making), such as occurred with Sandy and Katrina; and the Big Wet in Australia; regional droughts such as in California and eastern Australia (some of the examples are mine, not Richard's.)

Antarctica: the most interesting one that is hanging out there

Dr Alley says that the "most interesting one that is hanging out there" is the West Antarctic ice sheet. It's not the only one, he said, but it's very interesting. That's been featured here quite a bit, with a lot of research papers published on the subject and undoubtedly many more to come.

Richard talked about how there was "a general picture out there" that meant much higher sea levels. And then, he said, there's this one: the Heinrich events.

Heinrich events are defined off ice-rafted debris (IRD), which is a "crappy indicator" - said Richard. "You can make IRD peaks by rolling over an iceberg, you can make them by changing a drift path, you can make them by making the ocean colder so the icebergs survive longer. But if you see a lot of IRD, you know one thing with high confidence. There is no ice shelf."

The important point is that with very few exceptions, every ice shelf in Antarctica is melting on the bottom. An ice shelf is just a debris filter. It stops ice sheets from producing IRD. It holds the debris against the ice sheet until the debris is largely melted out. Only then does it make icebergs. So if there are IRD peaks, then there aren't ice shelves.

We're told there are two ways ice shelves melt out - both requiring heat ("making it hot where they care"):
  • Melting from the top - meltwater wedging from warm air and the ice breaks off, as with Larsen B
  • Melting from the bottom - warm water at grounding line depth (Jakobshavn).

Richard goes into some detail about how ice shelves melt. For example, he talks about Heinrich events coming out of a cold ocean, which he said was a real puzzle. Eventually Shaun Marcott, Peter Clark and co found that in a Heinrich event, a cooling of the surface leads to a delayed subsurface warming. I'll divert for a moment to quote from a ScienceDaily article on their paper:
"We don't know whether or not water will warm enough to cause this type of phenomenon," said Shaun Marcott, a postdoctoral researcher at Oregon State University and lead author of the report. "But it would be a serious concern if it did, and this demonstrates that melting of this type has occurred before."
If water were to warm by about 2 degrees under the ice shelves that are found along the edges of much of the West Antarctic Ice Sheet, Marcott said, it might greatly increase the rate of melting to more than 30 feet a year. This could cause many of the ice shelves to melt in less than a century, he said, and is probably the most likely mechanism that could create such rapid changes of the ice sheet.

Richard Alley mentions that the main source of melting at the grounding lines in Antarctica is circumpolar deep water, which is North Atlantic Deep Water (NADW). He finishes by saying that the most important aspects when it comes to the melting of ice sheets are:
  • for Greenland, air surface temperature - Greenland will most like melt from above
  • for Antarctica, circumpolar deep water and where it is (as well as winds).

And he doesn't sound all that confident that the dragons in Antarctica won't bite.

From AGU14

The dry version of Richard Alley's lecture is:
PP11ET Hose or Not to Hose: Mechanisms of Abrupt Climate Variability I
PP11E-02 Ice-Ocean Interactions and Heinrich Events
Heinrich events likely represent very strong feedbacks on millennial climate changes, as reviewed here. Ice shelves are debris filters, removing rocks from basal ice before iceberg formation, so ice-shelf loss is the easiest way to make an ice-rafted-debris event when marine-ending ice sheets exist. In turn, ice-shelf loss can be triggered by warmer ocean waters reaching deep grounding lines.
Recent work (Marcott et al., 2011, PNAS, etc.) shows subsurface warming in response to “shutdown” of the Atlantic Meridional Overturning Circulation (AMOC) as part of Dansgaard-Oeschger (DO) cycling, whether shutdown was caused by hosing, salt oscillations, or something else. A hybrid model may best fit the data, with a MacAyealian thermal oscillator in Hudson Strait allowing only some Nordic-Seas-triggered DO subsurface warmings to remove a Hudson Strait ice shelf and trigger a Heinrich event, which then extends the “shutdown” into the Labrador Sea to cause longer-lasting, larger far-field climate anomalies.
Notable uncertainties remain, however, and the final story is likely to be more involved than this. 

J. White, RB Alley, D Archer, AD Barnosky, J Foley, R Fu, M Holland, B Lozier, J Schmitt, LC Smith, G Sugihara, DWJ Thompson, A Weaver, S Wofsy. National Research Council. Abrupt Impacts of Climate Change: Anticipating Surprises. Washington, DC: The National Academies Press, 2013. (Free download from NAP with free registration.)

Alley, Richard B., Jochem Marotzke, W. D. Nordhaus, J. T. Overpeck, D. M. Peteet, R. A. Pielke, R. T. Pierrehumbert et al. "Abrupt climate change." science 299, no. 5615 (2003): 2005-2010. DOI: 10.1126/science.1081056 (avail here)

Marcott, Shaun A., Peter U. Clark, Laurie Padman, Gary P. Klinkhammer, Scott R. Springer, Zhengyu Liu, Bette L. Otto-Bliesner et al. "Ice-shelf collapse from subsurface warming as a trigger for Heinrich events." Proceedings of the National Academy of Sciences 108, no. 33 (2011): 13415-13419. doi: 10.1073/pnas.1104772108  (open access)


  1. I would be reluctant to second guess Richard Alley, but I'm not too sure about Greenland melting from the top down. There are three areas which have the potential to produce bottom-up melting to an extent varying from notable to catastrophic. The biggest imponderable is the insufficiently well-known topography of the underlying bedrock. This might be helpful as a general guide to Greenlands bedrock.

    In detail:
    In the northwest, Humboldt and Petermann Glaciers are two of the biggest. Both sit over large bays which would allow the grounding lines to retreat rapidly in the right conditions. Petermann is also one of the fastest flowing Greenland glaciers, reflecting that this problem is occurring. While the bedrock behind these glaciers is above sea level, it is only slightly so - there may be relatively small features that would allow sea water to penetrate under the whole Greenland icecap.

    In the northeast, Zahariae (sp?) and 79N are much the same. The bay they sit over is much larger, although the terrain behind them is higher and unlikely to lead to anything too disastrous.

    In the west is Jakobshavn (mentioned above as an archtypal bottom melt). Not only is the large and fast-moving glacier rapidly retreating, but as with the Petermann, there is the possibility that the bedrock behind the glacier is low enough to permit ingress of seawater into the depression underlying the whole icecap.

    Jacobshavn is situated where the northwards current along the coast brings a lot or relatively warm water into Baffin Bay. Petermann is more subject to cold water flowing out of the Arctic basin, but it would not take unfeasible changes in wind and water to accelerate its withdrawal.

    In the short term, bottom melt could lead to rapid loss in the Humboldt and the northeastern glaciers (and is already doing so for the other two), making a significant contribution to overall mass loss. In the longer term, Petermann and Jacobshavn could - and I stress could - be gateways to undermining the bulk of the Greenland icecap - the same mechanism invoked in talking about the rapid loss of the WAIS.

    I don't doubt that top-down loss has been the dominant process and will probably going on being so. But equally, I would think that bottom melt is non-trivial now, and in the right circumstances (insert uncertainty caveats here) could be a significant contributor in the future.

  2. Thanks for posting this up, Sou.


Instead of commenting as "Anonymous", please comment using "Name/URL" and your name, initials or pseudonym or whatever. You can leave the "URL" box blank. This isn't mandatory. You can also sign in using your Google ID, Wordpress ID etc as indicated. NOTE: Some Wordpress users are having trouble signing in. If that's you, try signing in using Name/URL. Details here.

Click here to read the HotWhopper comment policy.