Some of the ideas discussed in this blog are published in my new book called "The Stonehenge Bluestones" -- available by post and through good bookshops everywhere. Bad bookshops might not have it....
To order, click

Wednesday, 28 December 2016

How did the Devensian BIIS collapse?

I never did like the look of this strange purge / surge lobe running from St Georges Channel all the way south into the Celtic Sea approaches, with no lateral spreading.  Does this new paper from Stokes et al question the "binge-purge" model?  It suggests that as deglaciation proceeds, ice streams are switched OFF and are less and less significant as vehicles for ice evacuation.

I'm a bit slow catching up on articles these days, but it was interesting to see this article and its findings.

Ice stream activity scaled to ice sheet volume during Laurentide Ice Sheet deglaciation
C. R. Stokes,  M. Margold,  C. D. Clark & L. Tarasov
Nature 530,322–326(18 February 2016)
Published online 17 February 2016

The contribution of the Greenland and West Antarctic ice sheets to sea level has increased in recent decades, largely owing to the thinning and retreat of outlet glaciers and ice streams1, 2, 3, 4. This dynamic loss is a serious concern, with some modelling studies suggesting that the collapse of a major ice sheet could be imminent5, 6 or potentially underway7 in West Antarctica, but others predicting a more limited response8. A major problem is that observations used to initialize and calibrate models typically span only a few decades, and, at the ice-sheet scale, it is unclear how the entire drainage network of ice streams evolves over longer timescales. This represents one of the largest sources of uncertainty when predicting the contributions of ice sheets to sea-level rise8, 9, 10. A key question is whether ice streams might increase and sustain rates of mass loss over centuries or millennia, beyond those expected for a given ocean–climate forcing5, 6, 7, 8, 9, 10. Here we reconstruct the activity of 117 ice streams that operated at various times during deglaciation of the Laurentide Ice Sheet (from about 22,000 to 7,000 years ago) and show that as they activated and deactivated in different locations, their overall number decreased, they occupied a progressively smaller percentage of the ice sheet perimeter and their total discharge decreased. The underlying geology and topography clearly influenced ice stream activity, but—at the ice-sheet scale—their drainage network adjusted and was linked to changes in ice sheet volume. It is unclear whether these findings can be directly translated to modern ice sheets. However, contrary to the view that sees ice streams as unstable entities that can accelerate ice-sheet deglaciation, we conclude that ice streams exerted progressively less influence on ice sheet mass balance during the retreat of the Laurentide Ice Sheet.


The authors admit that there are big differences between the Laurentide Ice Sheet and the modern ice sheets of Antarctica and Greenland -- both of which evacuate large quantities of ice via calving into the sea.  But there are similarities with the British and Irish Ice Sheet or BIIS (which Henry Patton is now calling the Celtic Ice Sheet) which was of course diminishing at the same time as the Laurentide Ice Sheet.  I'm particularly interested in the extent to which this new work calls into question the "binge-purge" model of ice sheet collapse as described by Hubbard et al in their seminal paper of 2009. 

This is from one of my earlier posts, in 2011 (seems a long time ago):

The British - Irish ice sheet almost melted away around 27,000 BP, and then (because there was a relatively prolonged period of increased accumulation without any major interruptions) the ice sheet grew and grew until it reached its maximum volume and extent around 20,000 BP.  There was a phenomenal rate of growth over 7,000 years, followed by a catastrophic collapse.  The authors refer, over the whole lifetime of the ice sheet, to "binge-purge cycles" of gradual accumulations followed by rapid ice evacuations.....................

Quaternary Science Reviews Volume 28, Issues 7-8, April 2009, Pages 758-776
Quaternary Glaciodynamics
Dynamic cycles, ice streams and their impact on the extent, chronology and deglaciation of the British–Irish ice sheet

Alun Hubbard, Tom Bradwell, Nicholas Golledge, Adrian Hall, Henry Patton, David Sugden, Rhys Cooper and Martyn Stoker


I have always been rather sceptical about the strange "surge lobe" described in a number of papers as occurring at the peak of the Devensian glaciation.  The assumption seems to be that this lobe represented the final "purge" which evacuated so much ice from the centre of the ice sheet that its demise was then inevitable.  Catastrophic ice sheet collapse then followed, with the whole ice sheet disappearing in just a few thousand years.  I have argued that this lobe, which supposedly brought ice from the NE all the way south the the Isles of Scilly, is glaciologically improbable, since according to the model it penetrated for 400 km southwards with hardly any lateral spreading, in a landscape devoid of any lateral constraints.  It was not flowing in a well defined trough, and there were no bounding hill masses or mountain walls such as those seen in Norwegian fjords or even on the flanks of current Antarctic and Greenland ice streams.

So does this new work question the "binge-purge" hypothesis at a fundamental level?  Could it be that the BIIS actually disintegrated through surface melting, with rapid ice stream evacuations switched OFF?  All comments welcome!

See also:


Alex Gee said...

There is an extensive Karst science literature that demonstrates that large cave systems under the LLangattock escarpment and the Blorenge above Blaenavon were largely formed by sub- glacial melt water.

The ice that formed the LLangattock caves entering the Severn vally at Abergavenny and the ice forming the Blaenavon caves entering the Severn valley at Pontypool!

Both cases beg two questions

First "Once free of its confining valleys why did the ice not continue straight ahead onto the north flank of the Mendip Hills?"

Second: with no confining landscape features, why does the accepted literature state that the ice then performed a 90 degree turn to flow South west down the existing Bristol Channel?

Surely the answer to the second question would be of great interest to Physicists and other silly people who believe that nature behaves according to certain immutable physical laws?

How did that happen?

BRIAN JOHN said...

Alex, I am somewhat confused! Abergavenny is not in the Severn Valley, and neither is Pontypool -- and which references are you quoting here with respect to ice movements etc? And I am not aware of anybody who wants glacier ice flowing south-westwards down the Bristol Channel....... more info please!

AG said...

Sorry brian that was complete gibberish!.

Alex Gee said...

Brian to clarify:

I meant that the Anglian age welsh ice that formed the caves at llangattock would have exited the mountains at Abergavenny with the ice forming the Blaenavon caves exiting the valleys at Pontypool!

The flatter land beyond is quite obviously not the Severn Valley! Apologies I was looking at a terrain model.

The Anglian ice from both valleys would than have flowed south over the flatter land before joining the Severn estuary somewhere between Newport and Chepstow.

Once the Anglian glaciation was at its peak thickness and extent, such valley flows would of course be somewhat of an irrelevance.

My point in writing was to highlight the fact that the existence of these Anglian age ice
streams is proven from published karst studies! They would have to had flowed from the Brecon area! and upon joining the Severn Valley would have had no reason to change direction until they hit the north flank of the Mendip Hills.

Perhaps the route of the Altar stone|??

BRIAN JOHN said...

Alex, what is the evidence for these large cave systems to have been largely formed by subglacial meltwater? Because, in Wales, ice cover has been relatively short-lived in each glacial episode, I would not have thought that subglacial conditions prevailed for long enough...... I can understand a situation in which large pre-existing cave systems might have been completely filled with meltwater under hydrostatic pressure, and maybe flushed out and remodelled........

There is a lot of work on the ice limits of SE Wales, from Colin Lewis, Geoff Thomas and others. The current assumption is that the Late Devensian ice did not project further south than Bridgnorth or Kiddermeinster, but that the Anglian ice cover was much more extensive, reaching Gloucester and the Cotswolds. The evidence is very subtle, tied in to drainage diversions and assorted phases of terrace formation. Clearly not properly sorted out yet.....

Altar Stone route? Not unless the Altar Stone lithology can be matched with one of the Senni Beds outcrops in SE Wales. My money is currently on Llansteffan or Laugharne, as suggested in another post!