A note on ‘Collapse’

There is a lot in the media at the moment about the ‘collapse’ of the West Antarctic Ice Sheet. See my previous blog post for more information. But when we talk about ‘ice sheet collapse’, what do we actually mean? When we talk of people ‘collapsing’, they fall down right in front of us in the street. Buildings collapse. Bridges collapse. It’s a very bad thing. Right?

Collapse: rapid, irreversible recession

When scientists talk about the ‘collapse’ of an ice sheet, they mean irreversible, rapidly increased rates of recession. The rates at which the grounding lines of the ice streams recede will increase. It’s a positive feedback cycle, a viscious loop that means that future recession of the ice stream is inevitable (see Marine Ice Sheet Instability and previous post). Given that Pine Island Glacier has a sea-level equivalent of 1.5 m and the West Antarctic Ice Sheet a sea level equivalent of 3.3 m1, this is something to be concerned about. The entire recession of the ice stream will still take centuries, but this is very difficult to model and there are huge uncertainty ranges with many unknown parameters.

Floating ice shelves have also been discussed using terms like ‘collapse’. Here, the ice shelves are thinned for decades from warm water from below, leaving them vulnerable during warm summers when melt water ponds on their surface. This can lead to rapid iceberg production and fragmentation and distintegration of the floating ice shelf over the course of weeks. This can have long-term consequences, as the tributary glaciers accelerate, thin and recede in response to their changed boundary conditions. Ice shelves ‘hold back’ or buttress these tributary glaciers, and once they’re gone, the glacier is destabilised and must find a new equilibrium. This adjustment can take decades, as we’ve observed on the Antarctic Peninsula.

Ice-stream collapse in West Antarctica

Mouginot and colleagues found that ice streams in West Antarctica are accelerating, with a sustained increase in ice discharge2. Ian Joughin and colleagues also suggested that Thwaites Glacier may be in the early stages of irreversible recession, but we are still some way from passing a threshold that resulted in dramatically higher rates of recession. They suggested that we are heading inexorably towards this threshold, which could be passed anytime in the next 250-900 years3 (we’re heading towards 250 years at the moment). Future modelling efforts will reduce the timescale range, and the rates of ice-sheet recession that will occur once this threshold has been passed. This work is not unprecedented, and follows decades of work and several recent papers discussing marine ice sheet instability in West Antarctica (e.g., references 4, 5, 6).

Ice streams of Antarctica with Pine Island Glacier and Thwaites glacier highlighted.
Ice streams of Antarctica with Pine Island Glacier and Thwaites glacier highlighted.

‘Collapse’ is slow

But the West Antarctic Ice Sheet will be with us for some time yet. The West Antarctic Ice Sheet is not going to melt away tonight, this decade, this century or even next century. But this work is certainly alarming; the IPCC did not include ice dynamical changes such as these in their future sea level predictions. This research indicates that rates of sea level rise over the next few centuries could be very rapid. London is unlikely to flood anytime soon, but small changes in sea level exacerbate storm surges, increase coastal erosion, and flood low-lying areas in places without barrages like London. Future generations will face significant challenges associated with sea level rise. We should be concerned about this, and we should try and consider ways in which we can reduce the melting of these ice streams. That means combating climate change.

A better word for collapse?

This post brings to mind the work of Somerville and Hassol7, who suggested a series of words that could be better used for science communication. I include their table below. Perhaps we should find a better word for ‘collapse’. ‘Irreversible melt’? ‘Irreversible decline’? ‘Rapid, irreversible recession’? It just doesn’t have the same ring. Any other suggestions? Answers in the comments.

Table from Somerville and Hassol, 2011. What scientists say, and what they mean.
Table from Somerville and Hassol, 2011. What scientists say, and what they mean.

Further reading


1.            Bamber JL, Riva REM, Vermeersen BLA, Le Brocq AM. Reassessment of the potential sea-level rise from a collapse of the West Antarctic Ice Sheet. Science 2009, 324(5929): 901-903.


2.            Mouginot J, Rignot E, Scheuchl B. Sustained increase in ice discharge from the Amundsen Sea Embayment, West Antarctica, from 1973 to 2013. Geophysical Research Letters 2014: 2013GL059069.


3.            Joughin I, Smith BE, Medley B. Marine Ice Sheet Collapse Potentially Underway for the Thwaites Glacier Basin, West Antarctica. Science 2014.


4.            Overpeck JT, Otto-Bliesner B, Miller GH, Muhs DR, Alley RB, Kiehl JT. Palaeoclimatic evidence for future ice sheet instability and rapid sea level rise. Science 2006, 311(no. 5768): 1747-1750.


5.            Favier L, Durand G, Cornford SL, Gudmundsson GH, Gagliardini O, Gillet-Chaulet F, et al. Retreat of Pine Island Glacier controlled by marine ice-sheet instability. Nature Climate Change 2014, 4: 117-121.


6.            Hughes T. Abrupt climatic change related to unstable ice-sheet dynamics: toward a new paradigm. Palaeogeography, Palaeoclimatology, Palaeoecology 1992, 97(3): 203-234.


7.            Somerville RCJ, Hassol SJ. Communicating the science of climate change. Physics Today 2011, October: 48-63.

11 thoughts on “A note on ‘Collapse’”

  1. The Amundsen embayment ice streams are toast, although sharp action might slow things down a bit. The remaining ice sheets are a different matter, although with emissions trending the way they are…

    “The entire recession of the ice stream will still take centuries(.)”

    Hmm, no conditional on that statement. Rignot says that at *present* rates the studied ice streams will be gone in 200 years. That is (barely) “centuries,” but given that the process will inevitably accelerate as the ice moves back into deeper water we already know it’s going to be rather less than that. He places no speed limit on the process, so “decades” is not out of the question.

    This is the first major tipping point, officially passed. Be careful, be scientific, but please don’t be reticent when it come to such things.

  2. Just to add, we speak freely of the Roman Empire having collapsed and that took a couple hundred years. And as one of the NASA scientists said in the Monday press call, perhaps “collapse” really is apt given that it’s ice sheets we’re talking about. Are people really too stupid to be asked to think on such time scales?

  3. It just occurred to me that I seem to recall “collapse” being used extensively, in and out of the scientific literature, with regard to deglacial ice sheet retreats, and at a quick glance that seems to be the case. So is your hesitancy just because it’s about the future?

  4. Aha, I just now saw that Revkin posted this response from Joughin:

    ‘In defense of collapse, if I say, “The collapse of the Roman Empire,” you all pretty much understand what I mean and am talking about about a timescale similar to that for the WAIS. And I think most members of the public would too. If collapse can’t refer to extended periods such as decades to centuries, someone should contact Jared Diamond and ask him to change the title of his bestseller.

    ‘And someone should also talk to the astronomers: “The collapse to a white dwarf takes place over tens of thousands of years, (Wikipage on gravitational collapse)”. I think if we are clear about the time scales, the public is smart enough to figure out what we mean (that doesn’t mean we shouldn’t be cautious with how we word things).’

  5. Mike Bromley

    Regardless of the semantics of collapse, the amount of conjecture in the whole ‘finding’ reduces its meaning to a mere guess. It could take centuries. It could not. It might be a problem. It might not. But regardless of the conjectural bent in the recently-reported articles, the media has spun it into a contorted feeding frenzy, the out come of which is a fear reaction among some, and a healthy yawn about wolf-crying among others. And THAT is no longer scientific reporting, but fear-mongering. Not informative, but speculative and devious.

    1. Bethan Davies

      I hope that by being clear in my description of these findings I have been informative. I have been clear on uncertainties and always seek to provide original information and data in my website!

      Numerical models are a simplification of reality. There are many assumptions and uncertainties. Many parameters are unconstrained or not validated. This is the case in any model. However, models are still useful; they help us understand complex physical processes and feedback loops, and they provide insights into past, present and future glacier behaviour. That makes them important, and more than just mere conjecture and speculation.

  6. It should be so obvious to all that manmade global warming takes place and accelerates. Still quite a few people prefer to continue to turn the blind eye to realities. Short term profit and limitless greed are of course factors that transform some “scientists” into prostitutes, and non-scientists to insist on confusing as many as possible in order to paralyse decision makers. It is time to make this year the climate year – and many years to follow http://ufbutv.com/2014/01/11/make-2014-the-climate-year/

  7. I think its worth pointing out that it is the energy from the sun that melts ice, in the case of Antarctica mostly via seawater warmth that gets under the ice shelf. The sun did not start shining with the industrial revolution.

    If there is a clear anthropogenic effect then by all means discuss it, but there is a widespread tendency to blame man for any melting ice, which is clearly not the case. Since it is local climate (ocean currents) that matters for melting it is not even clear that a warming global climate will warm the particular ocean current that is relevant here.

    1. Joseph Sobry

      Overlooking a couple of things are we?
      Clearly the sun has for eons warmed ocean waters. The long wave infrared back radiation due to greenhouse gases has done the same for eons.
      It is the increase in greenhouse gases that has caused the increase in the heating of the ocean’s waters (or if you prefer the decrease in cooling). We think, correctly, that the increase in greenhouse gases is mainly caused by human beings either directly or indirectly.
      We do not blame man for “any” melting ice we blame man for the increase in melting which clearly is the case.
      You go to extremes to avoid accepting the truth.
      The local climate is not the same as an ocean current. In any event infrared warming happens everywhere around the planet and 24 hours per day. The whole ocean is being warmed no matter where you think your local current may be hiding to somehow avoid being warmed.
      There is a clear planet wide anthropogenic effect.

Leave a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

This site uses cookies. Find out more about this site’s cookies.