Relative sea level data -- western Ireland and western Wales

This model for ice thicknesses in the British and Irish Ice Sheet (BIIS) was run as part of an experiment to recreate what happened in the Devensian.  The actual situation 21,000 years ago was rather different -- but the model does seem to represent rather well what happened in the Anglian -- and it matches other "maximum glaciation" models published by Hubbard et al and Patton et al.  Somerset, and most of Devon and Cornwall, are shown with an ice cover; under a scenario similar to this, an ice edge could well have been located on Salisbury Plain.

Edwards et al. Resolving discrepancies between field and modelled relative sea-level data: lessons from western Ireland. Journal of Quaternary Science 32(7) 957-975.

DOI: 10.1002/jqs.2969


One of the big problems in Quaternary science just now is this: how do models of ice sheet behaviour for the Devensian actually relate to field data? Modelling versus ground truthing -- actually this is not a conflict, but an attempt on all sides to make modelling more accurate by building into the models more accurate field data.  Inevitably, this makes the models more and more complex -- but, we hope, more reliable.

One of the interesting features in this paper by Robin Edwards et al (which is really a study of relative sea levels) is a model for ice thicknesses during the deglaciation phase of the BIIS or Celtic Ice Sheet. One the maximum model ice is shown as having a thickness of c 1000m over Pembrokeshire 21,000 years ago. That doesn't match up with reality as far as the Devensian is concerned, but interestingly, it does seem to represent rather closely what we now know of the Anglian -- which appears to have been Britain's biggest glaciation. This seems to be the best fit for the data on the ground.

So in theory c 300m of isostatic depression would have been possible during the Anglian Glacial episode— but (if the Devensian was anything to go by) the ice expansion episode may have been so short-lived that isostatic adjustment never reached equilibrium. The situation in Pembs was fairly similar to that of W Ireland with respect to ice thickness, loading and the position of the current coast with respect to the ice edge…...


ABSTRACT

Accurate reconstruction of late glacial and Holocene relative sea-level (RSL) histories is complicated where mismatches exist between geological data and RSL curves generated by models of glacio-isostatic adjustment (GIA). In Ireland, such discrepancies have profound implications for interpreting the glacial history of the British Isles and for the use of glacial rebound models to predict future sea-level changes. To address this issue we present new RSL data from four sites along the western coast of Ireland, including seventeen data points from the critical time period before 5000 BP for which very few data are available. We generate new RSL simulations from an existing GIA model, incorporating a thickened Irish Ice sheet component. Simulated curves from Co. Mayo and Co. Donegal accommodate the higher than present late glacial RSL inferred from glaciomarine muds whilst still meeting the requirement for below present RSL indicated by the new terrestrial limiting data points. Relaxation of trimline constraints on maximum ice sheet thickness provides considerable scope for improved GIA performance. These results demonstrate inferences about RSL drawn from GIA modelling and glacio-sedimentary data are not mutually exclusive, and represent a significant step toward resolving a long-standing debate between the field-based and modelling communities.