THE BOOK
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
HERE

Thursday 27 January 2011

The strata do not lie


The two diagrams above come from a paper of mine which was published in 1973.  Click to enlarge.

The top diagram is a model showing how the "sediment record" varies at various points along the long profile of a big glacier such as the Irish Sea Glacier. Close to the ice shed area, ice will be present for a very long time -- almost from the outset of a glacial episode through to the final catastrophic wastage.  This means that there will be thick glacial deposits and not much in the way of periglacial deposits or head. About 250 km downstream, there will have been a long period of permafrost / periglacial conditions before the ice arrives, and so there may be thick periglacial head accumulations, overlain by an almost equal thickness of till and other glacial materials.  Closer to the eventual ice edge (at the outer limit of glaciation) there may be even thicker head and just a thin layer of till -- since ice is present for no more than a few centuries -- or even less.  Beyond the ice limit there will be no till -- just a thick series of head deposits of different sorts, representing fluctuations of periglacial climate -- sometimes wetter, sometimes drier, sometimes colder etc etc.

This is of course a model -- and it takes no account of bed conditions or glacier thermal regimes -- which can lead to accelerated erosion in some places at some times, and freezing-on and effective protection of the bed (and underlying sediments) at other times.  Also, fluvioglacial and other sediments can accumulate well beyond a glacier edge -- for example in valleys occupied by meltwater streams, and in the sea or in lakes.  So variable erosion and deposition rates and types have to be built into the model, which becomes more and more complex as we go on........

But if we look at the real sediment sequences we see that they do match the model reasonably well.  The two sequences -- from Abermawr in North Pembrokeshire and Poppit on the Teifi Estuary -- show a thick head sequence beneath the Irish Sea Till which marks the last glaciation of the area.  This till must have accumulated over a very long period of time while the glacier was "waxing" -- maybe 15,000 years or more.  The 4 facies of head at Abermawr show that there were climatic oscillations during this time -- and probably during the Early Devensian as well.

How long was the glacier ice present at these two sites?  Maybe a few centuries?  The ice was active enough to erode away some of the head layers -- in places we can see how there is an unconformity, with the junction between till and head clearly truncating or cutting across the "pseudo-bedding" of the head layers.  So the ice was flowing and eroding -- which means that the glacier was warm-based and sliding.  We must be talking about the period around 20,000 yrs BP.

After this glaciation, what do we see?  Well, first a wasting ice phase, as the glacier melted away -- with flow tills and abundant fluvio-glacial deposits here and elsewhere in West Wales.  Then, after a weathering episode, another cold spell, which must represent the Older Dryas - Younger Dryas period, around 12,000 - 10,000 years ago.  This was a short phase, and a thin layer of head accumulated, and incorporated much glacial material -- so other authors and I have referred to it as "rubble drift" to differentiate it from the clean "main head" below.  Then we have sandy loam, shown by analysis of sand grains to have been wind-blown at a time of bare ground surface conditions, strong dry winds, and abundant exposed glacial and fluvio-glacial materials on the land surface.  The modern soil is developed on and in this sandy loam deposit which caps the stratigraphic sequence.

The stratigraphy does not lie -- and it cannot be explained in any other way than that outlined above.

No comments: