The myth of "periglacial Dartmoor"

I have been having another look at the Evans et al paper on Dartmoor and pondering on the manner in which an "established wisdom" actually becomes established........

For many years I have been intrigued by the fact that professional geomorphologists, over several generations, have simply accepted the idea that the uplands of southern Britain (we'd better not call them mountains) have been affected by periglacial conditions but not by glacier ice.  Hardly anybody has been prepared to stand out from the crowd apart from Stephan Harrison, arising from his work on Exmoor -- maybe because the "periglacial paradigm" was argued so forcefully by famous establishment figures such as Prof David Linton.  But over and again we have seen "definitive" statements in the specialist literature to the effect that SW England has never been affected by glacier ice, and that everything in the landscape can be explained by reference to oscillating periglacial and more temperate conditions during the course of the Pleistocene.  Since geomorphologists have been saying this sort of thing, it is perhaps not surprising that archaeologists have seized upon their statements and have assumed that everything is sorted, and that nobody talks nowadays about glaciers affecting the counties of Somerset, Devon and Cornwall.

A number of things have contributed to the questioning and eventual destruction of the myth.  After Stephan's work on Exmoor suggested very strongly that glacial conditions were possible -- and had indeed existed on the uplands of the South-West, a number of researchers (including me, on this blog) pointed out that the Exmoor findings were perfectly in line with the evidence of glacial action in Somerset, the evidence of till at Fremington, the giant erratics on the coasts of Devon and Cornwall, and the occurrence of till on Lundy Island. (Use the search facility on this blog if you want more detail.)  Then along came glacial modelling, which allowed the reconstruction of ancient ice sheets and ice caps through the use of increasingly sophisticated data sets and calculation procedures.  This work, based at Aberystwyth University and elsewhere, showed that it was possible -- and indeed probable -- that the Irish Sea Ice Sheet had pushed across the coasts of SW England on more than one occasion, with the more extreme models showing glacier ice covering most of Wiltshire.  When James Scourse and others showed that Irish Sea ice had reached the Scilly Isles in the Devensian or Last Glacial Maximum,  only about 20,000 years ago, the computer models were shown to be essentially quite reliable. 

One would have thought that on the basis of all this new work, geomorphologists would have been happy to go on the record to state unequivocally that the Irish Sea Glacier HAD affected the counties of Somerset, Devon and Cornwall and that local ice caps must also have been present during the Pleistocene.  But when I asked senior geomorphologists to sign up to a simple letter to British Archaeology laying out the facts, only a couple of them agreed;  some of them refused, and most never bothered to respond.  I was disappointed and surprised by that, since it seemed to suggest that geomorphologists were reluctant to become involved in "the Stonehenge problem" even in a peripheral way, and that they did not want to be involved in anything that might upset their colleagues in Archaeology departments!  Whether these senior academics were afflicted by apathy or diplomacy, their silence simply served to encourage senior professors like MPP, GM and TD to believe that there was a consensus among earth scientists that no glacial processes had ever affected South-West England.  So, somewhat disgusted by the wimpish tendencies of some of my senior ex-colleagues, I decided to submit the letter under my own name -- and to his eternal credit, Mike Pitts published it in British Archaeology:
http://brian-mountainman.blogspot.se/2011/12/brit-arch-letter-published.html

So there we are then.  Has the myth of "periglacial Dartmoor" and the "ice-free South-West" finally been laid to rest? I doubt it -- too much academic capital has been invested in it for any instant change of attitude.  And don't let's forget that many archaeologists still accept that whatever Chris Clark, James Scourse and Chris Green may have written in rather influential articles in the past MUST be true.  They may not have noticed it, but the world has moved on -- and thanks to David Evans, Stephan Harrison and their colleagues we now have an influential article in the peer-reviewed literature that sits easily with all of the other field evidence which we have examined on this blog over the last couple of years.

Listen carefully, archaeologists.  The South-West of England has been affected by the ice of the Irish Sea Glacier, and by local ice caps, on several occasions during the Pleistocene.  This means that it is perfectly feasible for glacial erratics to have been carried from West Wales, Ireland and even Scotland on more than one occasion and dumped well inland of the Bristol Channel coast.   Got that?

Below I reproduce the Introduction to the recent article by Evans et al -- with key phrases highlighted in bold type.  It's a very interesting comment on how academic thinking (even in mainstream scientific disciplines) can be affected by fashions, conservatism and senior academics who have reputations to protect, to the point where perfectly sound observations presented by people who are not a part of the academic establishment can be ignored and -- dare I say it? -- even sneered at.

============================

The glaciation of Dartmoor: the southernmost independent Pleistocene ice cap

in the British Isles

David J.A. Evans, Stephan Harrison, Andreas Vieli, Ed Anderson


EXTRACT FROM THE INTRODUCTION

Although the granite uplands of Dartmoor (Fig. 1) have long
been considered to be relict permafrost and periglacial landscapes
that lay beyond the limits of Quaternary glaciations (Linton, 1949;
te Punga, 1956; Waters, 1964, 1965; Gerrard, 1988), the notion that
glaciers had developed in these areas was entertained by some
early researchers (Ormerod, 1869; Pillar, 1917). The evidence pre-
sented at that time was largely circumstantial even anecdotal. For
example, Ormerod (1869) reported that “he had not seen any
glacial markings on the Dartmoor granite, but that Professor Otto
Torrell, when visiting the Moor with him last autumn, gave an
unqualified opinion that many of the gravels were the remains of
moraines” (p. 99). In contrast, Somervail (1897) suggested that the
absence of small lakes on Dartmoor was incompatible with former
glaciations and wrote: “It is true that various attempts have from
time to time been made by various observers to refer certain
phenomena occurring on Dartmoor to local glaciation. None of
these, however, are, I think, the result of true glacial action, but
must be referred to the more common operations of running water.
During the cold of the Pleistocene period, and at its close, the floods
from melting snows would perfectly accomplish all the distribution
and arrangement of that deposit of angular rocky debris
surrounding Dartmoor so frequently referred to ice. The same cause
would also equally well explain these accumulations of scree
matter filling some of the valleys, which some have regarded as the
remains of ancient glacial moraines” (p. 388). Pillar (1917) later
argued that Dartmoor should have been glaciated given its prox-
imity to the Pleistocene ice sheets: “As the meteorological condi-
tions must have been the same in these contiguous areas, it seems
somewhat strange that land in such close proximity should be
considered as outside the range of Ice influence” (p. 179).

A more systematic and empirical approach was taken by Pickard
(1943), who argued for the former existence of extensive glaciers
and small ice caps on Dartmoor during the Quaternary based on
a large collection of varied features, some more convincing than
others. In particular he presents evidence of amphitheatre-like
valley heads or incipient cirques, glacially “worn boulders” and
grooved rocks, “moutonnée rocks”, perched boulders, moraines
comprising ridges of blocky debris, and potential glacifluvial
gravels. Although this evidence has never been directly refuted, the
predominant view since the 1950s has been that the Dartmoor
landscape of summit and valley-side tors is the product of peri-
glacial mechanical weathering and slope processes that have
exploited zones of rotten granite and exposed large coherent
bedrock residuals, which because of their dilatation joints or
pseudo-bedding resemble corestone stacks. The efficacy of these
cold climate processes, which must have operated over a large
proportion of the Quaternary, was enhanced by pre-existing granite
breakdown through deep weathering in the tropical climate of the
Tertiary (Linton, 1955) and/or pneumatolysis during much earlier
periods of deep thermal activity (Palmer and Neilson, 1962; Eden
and Green, 1971). Despite the greater substance of the large
volume of work undertaken since Pickard’s paper, it is not clear
how the present consensus regarding the absence of glacial ice on
Dartmoor came about. There are several possibilities, the first of
which is the assumption largely championed by Linton (1955) that
the development of tors required a lengthy period of ice free
conditions. Palmer and Neilson’s (1962) view that the Dartmoor
tors reflected prolonged periglacial action, rather than the opera-
tion of tropical deep chemical weathering processes as Linton had
argued, may also have served to further alienate notions of Dart-
moor glaciation and consolidate the periglacial paradigm. The
second possibility is that Pickard’s views may not have been taken
seriously. His 1943 paper was his Presidential address for the
Devonshire Association and was, presumably, not refereed. More-
over, although it appears he had a strong interest in natural history
and geology, he was an opthalmist by training and was likely
regarded as an enthusiastic amateur by later geomorphologists.

However, recent research on Exmoor (Fig. 1) by Harrison et al.
(1998, 2001) has demonstrated the existence of tills and associ-
ated glacial landforms in the vicinity of The Punchbowl, a north-
facing valley near the village of Winsford. The tills have been
deposited at altitudes down to 255 m asl probably by a glacier snout
that flowed into The Punchbowl from a small ice cap located on the
summit plateau of Winsford Hill (426 m asl). The presence of glacial
ice at relatively low altitudes on Exmoor enabled Harrison (2001) to
speculate on the likelihood that Dartmoor had been similarly
glaciated at times during the Pleistocene. The Exmoor glaciation
evidence is entirely predictable considering that areas of high
terrain, such as those located in SW England, are likely to have been
cold enough to host small ice caps and glacierets during full glacial
periods when British-Irish Ice Sheet limits extended as far south as
the Isles of Scilly (Scourse, 1991; Scourse et al., 1991; Scourse and
Furze, 2001; Hiemstra et al., 2006, Fig. 1). Indeed, numerical
modelling exercises invariably create ice masses over the SW
English uplands during the Last Glacial Maximum (LGM) (Hubbard
et al., 2009) merely because the environmental boundary condi-
tions for the model will develop glacier ice where the local equi-
librium line intersects the topography. Moreover, the plateau-style
of topography on Exmoor and Dartmoor is conducive to the accu-
mulation of snow on the broad summits in addition to its buildup in
the deeper valley heads due to snowblow (cf. Manley, 1959; Sissons
and Sutherland, 1976; Sissons, 1979; Sutherland, 1984; Mitchell,
1996; Rea and Evans, 2003, 2007; Coleman et al., 2009). There-
fore the style of glaciation will be similar to the plateau icefield
glacial landsystem, wherein predominantly thin, largely cold based
and protective ice on upland surfaces would drain into valley heads
radiating from the plateaux to form locally thick, warm-based
snouts capable of eroding the substrate (Rea et al., 1998; Rea and
Evans, 2003, 2007). Such erosion would have been responsible for
the production of The Punchbowl on Exmoor and potentially the
overdeepened valley segments around the high summits of
northern Dartmoor, identified by Pickard (1943) as evidence for
glacial modification.

The evidence for former glacier ice in marginally glacierized
terrains, such as those represented by Exmoor and Dartmoor, is
likely to be subtle for a number of reasons. First, the predominantly
thin ice located on low-angled slopes would only generate low
shear stresses and low flow rates, so the creation of well developed
bedrock erosional forms is unlikely. Additionally, the coarse crys-
talline nature of the Dartmoor granite is not suitable for the pres-
ervation of striae and other small scale glacial erosional features.
Second, the absence of high bedrock cliffs above the accumulation
zones of plateau icefields precludes the provision of extraglacial
rock debris, which together with the resistant nature of the granite
substrate would have resulted in small glacial debris loads and
hence weakly developed moraines and tills. Glaciers would have
instead only been able to incorporate periglacial slope deposits,
which after at least several hundred thousands of years of weath-
ering and gelifluction will have reached significant thicknesses in
valley bottoms (Waters, 1964). Moreover, such deposits would have
developed into large rock-fronted lobes and possibly rock glaciers
in some valley heads in the absence of, or between, glaciations,
similar to the stone runs of the Falkland Islands (Joyce, 1950;
Clapperton, 1975; Hansom et al., 2008; Wilson et al., 2008).
Harrison et al. (1996) have suggested such an origin for some
boulder lobes on Dartmoor. These rock-fronted, lobate forms, once
their ice content was removed, would resemble thin till sheets and
moraines in some settings. Finally, glacier ice that is frozen to its
bed effectively protects even delicate periglacial landforms and
sediments which therefore can survive one or more glaciations (cf.
Clapperton, 1970; Whalley et al., 1981; Dyke, 1993; Kleman, 1994;
Kleman and Borgström, 1994; Rea et al., 1996a, b). On Dartmoor, the
tors and clitter fields were initially regarded by Linton (1949, 1955)
as features that could not survive glaciation but he later modified
this view by arguing that they could be protected by thin and slow
moving ice. This principle of limited glacial erosion has since been
used to explain the preservation of tors in the Cairngorms (Sugden,
1968; Hall and Phillips, 2006; Phillips et al., 2006) and the presence
of preglacially weathered in situ bedrock and weathering pits in
glaciated east and northeast Scotland (Hall and Sugden, 1987; Hall
and Mellor, 1988; Hall and Glasser, 2003).

Although Dartmoor has been long established as an exemplar of
a mature periglacial landscape, no systematic assessment of
potential glacial landform evidence has ever been undertaken,
likely due to the overwhelmingly strong periglacial sediment and
landform signature. Given the recent advances in our under-
standing of plateau icefield landsystems and the preservation of
tors and associated deposits beneath cold based ice, the systematic
survey and mapping of the Dartmoor landscape for potential glacial
evidence is warranted. We now show evidence that northern
Dartmoor (Fig. 1) was glaciated and discuss the evidence for
glaciations in the context of alternative views of landscape inheri-
tance. As the resolution of these issues is critical to the recon-
struction of regional glaciation levels and palaeo-equilibrium line
altitudes in marginal glacierized terrains and hence the refinement
of boundary conditions for numerical ice sheet models, the
central aim of this paper is to assess the nature of the Dartmoor
landscape in the light of current understandings of plateau ice
landsystems.