I have been looking again at the QRA "Quaternary of Gower" Field Guide edited by Rick Shakesby and John Hiemstra, while checking on the likelihood of Devensian ice filling Carmarthen Bay. That's an interesting and rather significant matter, given that I am now convinced that Late Devensian ice from the west affected all of the clifflines of South Pembrokeshire, at least as far to the east as the eastern coast of Caldey Island. I have shown this on many of my maps on this blog.
But what happened during the late Devensian in the area just off the SE corner of the above map? As of 2011 this was the accepted situation, as portrayed by the BGS:
This shows an ice edge looping across Gower, leaving the southern and western coasts ice-free. It shows a prominent glacier lobe filling Swansea Bay, fed by the Neath and Tawe Glaciers, but not much of a lobe in Carmarthen Bay. The ice front is not very convincing, given that the Tywi Glacier must have been big and powerful, and given that there must also have been supplements to discharge from the Black Mountains and from the Cynin Glacier.
In the old days, Prof Neville George and others accepted that idea that Gower had been affected by an extensive "Older Drift Glaciation" and a more limited "Newer Drift Glaciation". These can be assumed to be equivalent to the Anglian and Late Devensian glaciations elsewhere. The field evidence for these was tied in closely to their stratigraphic relationships with the raised beach -- and there were some rather simplistic assumptions too about the "Older Drift" ice coming from the west and the "Newer Drift" ice coming from the north. Then everything was screwed up even further when Prof DQ Bowen started re-classifying everything and placing deposits into a scheme of "aminostratigraphy" based upon the technique of amino acid dating which was anything but mature -- and which has now been shown to be wildly unreliable without very careful calibration and date correction. For more than twenty years, confusion reigned...........
Now aminostratigraphy has been dumped, and there is a chance that new field investigations and stratigraphic studies by Hiemstra, Shakesby, and McCarroll will lead to some real progress.
Back to the 2015 QRA Guide. From the detailed research reported, the following now seems to be quite well established:
1. The whole of the Gower was ice-covered during the Late Devensian by Welsh ice moving from the north.
2. The Paviland Moraine, assumed by some to be an ancient feature, is most appropriately interpreted as a Late Devenisian moraine, possibly marking a retreat stage.
3. The deposits ar Rotherslade do not mark a "maximum" Devensian ice edge position.
4. The ice lobe in Swansea Bay was much more powerful and extensive than previously thought.
5. Since there are apparently fresh glacigenic deposits on the western edge of Gower in Broughton Bay, Rhossili Bay and on Worms Head, there must have been a substantial southwards-flowing ice lobe in Carmarthen Bay which incorporated old sea-floor sediments before pressing across the present coast.
This all makes sense, and ties in with the sea-bed research which we have described earlier on this blog:
1. For example, there is some very convoluted reasoning relating to the glacigenic and related deposits on the south Gower coast, involving ice from the north and a steeply sloping ice front; the authors have not considered at all the possibility that Irish Sea ice from the west might have approached or even crossed the present coastline, creating a dynamic and confusing contact zone between two ice masses in which a multitude of different sedimentological situations may have occurred. Maybe they have too easily accepted the old assumptions of the Irish Sea Glacier ice edge being positioned far to the west, off the west coast of Pembrokeshire? As I have explained many times on this blog, those assumptions should have been dropped long since -- and if Irish Sea ice at the LGM reached the eastern side of Caldey Island it is perfectly feasible that it also reaches western and southern Gower.
2. While the authors of the Field Guide admit to the evidence of an older west - east glacial advance across Gower, leaving evidence in the erratic suite of the raised beach and in other scattered erratics also in western Gower in particular, they seem to go to some pains on page 61 to "explain it away". The fact that there was an "Older Drift" Glaciation is incontrovertible, and if Irish Sea ice at some stage transported Pembrokeshire erratics to Pencoed and beyond, the same ice stream is very likely indeed to have flowed across the Gower. I do not understand why that should be perceived to be a problem -- unless the Field Guide authors have been wondering why they cannot find any "Older Drift" deposits that are still coherent and exposed.
3. On that latter point, I am intrigued by the cemented till which is described in the Field Guide from Watch-house Bay and Foxhole Cove, near Southgate. A number of exposures are described, but the authors do not publish a stratigraphic sequence, so it is difficult to see how the deposits at the site (including cemented raised beach and limestone breccia) relate to one another. The authors (Danny McCarroll and his geologist daughter Bethan McCarroll) assume throughout their description that the cemented till is Late Devensian in age and is related to all the other till deposits described in the Field Guide. But nowhere is that assumption supported by evidence, and as far as I can see there is no exposure that shows that the cemented till is stratigraphically younger than the cemented raised beach. I wonder, therefore, whether the cemented till is OLDER than the raised beach, and is related to the Anglian (or Wolstonian) Glaciation of the Gower? The fact that there are also uncemented (and presumably Late Devensian) tills in the vicinity is interesting but irrelevant as far as dating is concerned. My reasoning here is related to Lydstep. If we pop across the water and take a look at the Pembrokeshire coast, only about 50 km away, we find TWO tills at Lydstep, one cemented and capped by cemented roackfall and slope breccia, and the other fresh and unconsolidated. Lydstep, Watch-house Bay and Foxhole Cove are all limestone coasts with typical coastal karst scenery and morphological features.