Was there a Bristol Channel Ice Stream?

 

The revised model from Ely et al.  The white areas show thin ice and sluggish or non-existent ice movement. The purple areas show ice streams or zones of relatively rapid ice flow.  Note the big Irish Sea ice stream and the much smaller but associated Bristol Channel ice stream heading east and pushing onto the Somerset Levels.

Thanks to Bevins et al (2025) for pointing me towards this article:

J.C. Ely, C.D. Clark, S.L. Bradley, L. Gregoire, N. Gandy, E.Gasson, R.L.J. Veness, R. Archer
Early View. Behavioural tendencies of the last British-Irish Ice Sheet revealed by data-model comparison
Journal of Quaternary Science. (2024)

It's a comprehensive and very specialised study of the hundreds of computer models of the behaviour of the British and Irish Ice Sheet (which also goes by several other names)  -- most of them relating the the Devensian glaciation.  They range from maximal to minimalist models and from extreme ones to ultra cautious ones.  Many of the older models can be rejected because the parameters used in their creation are now inadequate -- but the work of the BRITICE project has stimulated a new generation of models with much greater and more careful ground truthing used to check on their viability and accuracy.

Anyway, in tying everything up at the end of an extensive analysis, the authors produced a map for the southern part of the Ice Sheet which showed that (a) the ice sheet must have crossed the coastal barrier of Devon and Cornwall and impinged upon the small local ice caps of ther SW Peninsula (Dartmoor and Exmoor in particular) and (b) that there might well have been an eastwards flowing ice stream in the Bristol Channel.

I must admit that the latter suggestion (it is no more than that) came as a surprise to me, since previoius models have almost always suggested that the Bristol Channel was largely ice free at the times of extensive glaciation.  Partly, as I have pointed out many times on this blog, this defect arose from the assumption of an ice-free enclave in South Pembrokeshire, which makes no sense from a glaciological perspective.  So now we have a modern and apparently reliable model that agrees substantially with the situation portrayed in these mapped reconstructions:

After Kellaway, Williams-Thorpe and others

After Gilbertson and Hawkins

My own suggestion

Note that the latest modelled reconstruction shows ice pressing across the whole of Cornwall, and that the model does not show the greatest extent of Devensian ice in the Celtic  Sea.  Here the modelled outer edge of the Celtic Sea piedmont glacier is shown around the position of the Isles  of Scilly.  But as Scourse and others have pointed out, the extreme position was much further to the south, at the shelf edge.  At that time the ice must have been thicker and more extensive in the Bristol Channel, with accelerated streaming and a progression well inland of the Somerset coast.

Gradually, it all comes together............