The Paviland Moraine: its bearing on the bluestone debate

The Paviland Moraine (pecked line) runs for almost 10 km inland of the south coast.  It is best seen  just inland of the coast at Port Eynon Bay (after Bowen, 1985). Port Eynon Bay is the southernmost bay on Gower.  The moraine has an arcuate form and may be up to 40m thick.  The north-facing flank is steeper than the south-facing flank, according to the 2015 QRA Field Guide.  Bowen's presumed Devensian ice limit is shown by the dashed line.

The Paviland Moraine in western Gower is widely cited as one of the oldest glacial constructional features in the UK -- and for many years Prof DQ Bowen has promoted the idea that it is of Anglian age, on the basis that it LOOKS very old and that it contains till and clasts of many sizes that appear to be heavily weathered.  Also, it lies beyond or outside the Late Devendsian ice margin reconstructions made by Bowen and many other workers over the years. It also rests on a ridge crest not far from the famous Paviland Cave, within which a human skeleton has been found and which has been dated to the Middle or Late Devensian.  The thinking has been, for many years, that if people were inhabiting the Gower caves at this time, the ice margin could not have been located in the vicinity.  I have done various earlier posts on this topic:

http://brian-mountainman.blogspot.co.uk/2012/02/paviland-and-ice-edge.html

http://brian-mountainman.blogspot.co.uk/2011/08/monster-from-daylight-rock-1.html

http://brian-mountainman.blogspot.co.uk/2011/08/monster-from-daylight-rock-2.html

http://brian-mountainman.blogspot.co.uk/2012/02/arthurs-stone-gower.html

On the QRA / GLWG field trip the other day, we visited the Paviland Moraine itself, and while it is singularly unspectacular (essentially a low capping of till and other sediments on a limestone rock ridge more or less parallel with the coast) it does have an irregular crest line and a nice "micro morphology" of low ridges, hollows, valleys, hummocks and steps which is rather convincing.  In short, the moraine is no more "degraded" than any of the Devensian constructional features in South Wales (like the hummocks in the Moylgrove-Cardigan area) which can be dated with a fair degree of certainty.  As soon as I saw it, I thought: "If somebody wants to tell me that this is a Devensian feature, I would be inclined to believe it."  Also, John Hiemstra and his colleagues showed us a very convincing sections from an 11m core through the moraine, showing that the contained clasts are quite fresh, and free of the weathering crusts that might be expected if there had been almost half a million years of weathering since deposition.  The other very convincing argument for a Late Devensian age for this feature is the presence of "plugs" of chaotic fluvioglacial debris at the mouths of Western Slade and Eastern Slade valleys, suggestive of pulses of very turbulent and even catastrophic meltwater flow from an ice margin in close proximity.  The debris exposed in the cliffs is very fresh, and cannot have been emplaced here if a Devensian ice margin had been located  several kilometres to the north, on the other side of the limestone ridge.

 A one-metre section of the 11m core, laid on its side.  The sediment here is reddish brown in colour, with large clasts set in a matrix of gravel, sand and silt.  There is less clay in the matrix than there is in the lower part of the core.  This is clearly a till.  Note that the clasts (sliced through by the drill) are clean, with no weathering crusts.

So I'm convinced that the Paviland Moraine is Late Devensian in age.  Whether this represents the MAXIMUM extent of Devensian ice on the Gower is another matter -- I reckon the jury is still out on that one.  But there is no reason in principle why there should not have been intermittent human habitation of the Gower caves round about the time that the Devensian ice coming out of the South Wales Valleys was at its maximum. 

 Now we come to the relevance of all this for the bluestone transport debate.  Here is a 1994 quote:

"New evidence confirms the suggestion that the Llanddewi Formation of Gower (Bowen, 1969b) and its Paviland Moraine should be correlated with the Anglian (Bowen et al., in preparation). This, however, represents glaciation from due north, and does not support the hypothesis of an Anglian ice-sheet (~450 ka) that transported 'bluestones' from Preseli to Stonehenge (Kellaway, 1971; Thorpe et al., 1991). Cogent arguments have been assembled against this hypothesis (Kidson and Bowen, 1976; Green, 1973;(in press); Darrah, 1993), and are supported by a 36Cl age determination on an igneous rock from the Stonehenge collection in the Salisbury Museum. This shows that it was still buried at its source outcrop during the Anglian (400 ka), and did not become exposed by denudation to the atmosphere until the Late Devensian (Bowen et al., in preparation), after which it was presumably quarried by prehistoric people and taken to Stonehenge."

Bowen, D.Q. (1994) Late Cenozoic Wales and South-west England.  Proc Ussher Society 8, pp 209-213.  (Scott Simpson lecture 1994)

Leaving the highly controversial 36Cl dating issue to one side, what Bowen is arguing here is that if the Anglian ice from the South Wales valleys was able freely to press across Gower as far as the position of the Paviland Moraine,  there could not have been Irish Sea ice in the vicinity -- and if there was no Irish Sea Ice here in the Anglian, it cannot possibly have transported bluestones from Pembrokeshire all the way to Stonehenge.  (The gradient of the Irish Sea Glacier must have been continuous from Preseli, across the tip of Gower and all the way to Somerset.)

That is all perfectly logical IF THE PAVILAND MORAINE IS REALLY OF ANGLIAN AGE.

If, however, that dating is faulty, as it now appears to be, and if the moraine is just another Late Devensian feature, Gower could well have been overridden by Irish sea Ice during the Anglian and the bluestones could well have been transported from Preseli to Stonehenge.  There must have been Irish Sea erratics lying around on Gower, or contained in ancient deposits, when the Late Devensian ice from the valleys arrived and flowed across the peninsula from north towards south.  This would tie in with the observations of George, Bowen and many other researchers that are scattered Pembrokeshire erratics in raised beach deposits and in the head which in places underlies the Devensian till and fluvioglacial materials.  The situation is very similar in Pembrokeshire, and in Devon and Cornwall, where ancient erratics (and sometimes very large boulders) sit on the raised beach platform and are incorporated into slope and other deposits that pre-date the LGM.

It all slots together.......

 This is the BRITICE map for West Wales, including Pembrokeshire and the Gower.  The cream-coloured area was supposedly ice free during the LGM.  Note that most of Gower is shown as ice-free during the Devensian (the dotted line shows the supposed limit) -- but the mapping here is now shown to have been based upon the flimsiest of evidence, with the old dating of the Paviland Moraine responsible for a multitude of sins.

This map from Thorpe et al (which assumes that the Anglian Irish Sea ice reached Salisbury Plain) shows that the big ice stream must have crossed much of Pembrokeshire and Gower en route.  This explains the presence of Pembrokeshire erratics on Gower, at Pencoed and on the Vale of Glamorgan.   The ice was much more limited in extent during the Late Devensian, when ice from the valleys of the Nedd, Tawe, Tywi and Taf flowed broadly southwards, but with much variation induced by topography.  Sometime near the peak of this very recent glaciation, the Paviland Moraine appears to have been  formed.