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

Friday 16 January 2015

If ever there was a case for OSL dating, this is it......

This is a section of the big BRITICE map which shows the location of the key glacial features of the British Isles.  It's not very accurate, because inevitably there are many generalisations in the cartographic process and because it's bound to be out of date -- new information is always coming forward in the specialist literature.

In order to confirm the basic hypothesis of two major glacial episodes -- the Anglian and the Devensian -- in West Wales, we really need some OSL dating.  To explain:

One way of dating glacial landforms is optically stimulated luminescence dating (OSL). OSL is used on glacial landforms that contain sand, such as sandur or sediments in glacial streams. The OSL signal is reset by exposure to sunlight, so the signal is reset to zero while the sand is being transported (such as in a glacial meltwater stream). Once the sand grain has been buried and it is no longer exposed to sunlight, the OSL signal starts to accumulate.

As we can see from the map, there are three main groups  of fluvioglacial deposits in Pembrokshire -- shown brown on the map.  In the area hypothetically affected by Anglian ice only (shown yellow) the westernmost group is in the valley of the Western Cleddau river, mostly between Trefgarn Gorge and Haverfordwest.  The eastern group is in the valley of the Eastern Cleddau, around Gelli and Clynderwen.  And the third group is in the far north, in the Moylgrove-Monington area, in the area affected by the Devensian ice advance of the Irish Sea Glacier.

One of these groups -- the westernmost one -- should not be treated as a high priority for OSL dating because a great deal of meltwater related to the Devensian ice wastage phase was channelled through the Trefgarn Gorge and flowed southwards towards Haverfordwest and Milford Haven.  So the gravels in the river terraces and in patches on interfluves might well be made up of a mixture of very ancient gravels and newer ones -- that would give a confused OSL dating scenario, unless great care could be taken on a sampling programme.

But if we could get dates from sands in the other two groups of fluvioglacial materials, that would be fantastic.  I suggest initially -- as a test run -- that samples could be analysed from two locations:

(1) Llangolman gravel pit (described in a recent post here: and characterised by bedded sands and gravels that LOOK very old indeed:

(2) Trefigin gravel pit near Monington -- that's the northern one of the two pits shown on this satellite image.  Here the sands and gravels look much fresher and cleaner, as befits deposits claimed to have been laid down at the peak of the Devensian glaciation, around 20,000 years BP.

The two lower photos are from Charlie Bendall's Report on the Trefigin Quarry (2013)

My guess would be that the sand from the Llangolman Quarry will give an OSL date of around 450,000 years ago, and the sand from Trefigin Quarry about 20,000 years BP.

A Trefigin date would be quite interesting, because it might be accurate enough to tell us whether the sands and gravels on the higher ground on either side of the Teifi Valley (occupied to Glacial Lake Teifi) were laid down during the ice advance phase, or the ice retreat phase.

Has anybody out there got the expertise, the money, and the inclination to undertake this vital piece of research?


Trefigin Quarry Sedimentology Report 2013
Dr Charlie Bendall, DGES, Aberystwyth University, 50pp

Etienne JL, Jansson KN, Glasser NF, Hambrey MJ, Davies JR, Waters RA, Maltman AJ, Wilby PR. 2006. Palaeoenvironmental interpretation of an ice-contact glacial lake succession: an example from the late Devensian of southwest Wales, UK. Quaternary Science Reviews, 25: 739-762.


Constantinos Ragazas said...


Can OSL dating be used for local glaciation where there is no movement of sand? From your post I am inclined to say no. Do you know?


BRIAN JOHN said...

I really know very little about this, but from what I can gather the best materials to use are fluvioglacial deposits that have been exposed to direct sunlight and then buried. So material on braided outwash plains (sandar) is likely to be OK, as are kame terrace materials. But material in an esker, for example, might not be any good, since it might never have seen the light of day. And till is not any good either, for the same reason.