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

Saturday 29 February 2020

Devensian till at Amroth -- an Eureka moment

In this photo we can see a stony blue-grey matrix-supported diamicton beneath a broken cap of peaty material.  The capping material belongs to the submerged forest,  and wave action has already removed most of it from the vicinity.  Normally it would all be buried beneath a beach sand layer.

In this exposure the peaty / silty layer is thicker and more continuous, but beneath it the diamicton is clearly from the same horizon as that shown in the top photo.  Gary's description:  "top layer of woody peat lays on top of a blue organic clay in the lower section of the clay are pieces of limestone which match the limestone of the Ludchurch area".

Here we see several isolated fragments of grey clay layer (belonging to the submerged forest sequence?) with signs of stratification beneath.  There is one quite broad gravelly layer.  Is this the top of the till layer? The sediments are resting on a broken bedrock platform with a litter of broken stones and boulders of local origin.   With a few more storms these patches of clay will be entirely obliterated........

It's always a pleasure to report on observations / discoveries made by others, and today there is some big news from Amroth, courtesy of Gary Davies.  In the context of a discussion about the fresh exposures of the submerged forest around the Pembrokeshire coast, Gary posted some 2014 photos on the Pembrokeshire Geology Facebook page, and I was duly astounded.  Anyway, Gary has kindly given me permission to reproduce them here...... in his original post he says: ........"there's a layer of blue limestone pebbles under the peat and clay layers. Probably brought down from Luchurch as the ice melted ....... prior to the vegetation starting to establish itself."

In the top caption I refer to the clay-rich blue-grey deposit as a "diamicton" -- but there is really not much doubt as to what it is.  If I was to see this in Greenland or Iceland it would be accepted instantly as a clay-rich till laid down at the base of a glacier where lodgement processes were at work.  It is certainly not a river or shoreline deposit, and it is not a brecciated slope deposit or a turbidite.  It has all of the characteristics of a true till -- highly varied stone shapes and sizes, many different lithologies, faceted faces, and no sign of layering or other arrangement.  Are there striations on any of the pebbles?  That could be a clincher......... (There may of course be signs of layering at the top of this bed, where the material becomes richer in organic materials and grades up into the peaty layers of the submerged forest.)

At the moment I see no reason why this deposit should not be interpreted as Irish Sea Till, laid down by the Irish Sea Glacier as it pressed eastwards into Carmarthen Bay.  There are clear similarities with the deposits exposed in Freshwater West and West Angle.......  It is uncemented and fresh-looking, and so my provisional interpretation is that it is of Late Devensian age, like all of the other unconsolidated tills of West Wales.

Grey-blue clay-rich Irish Sea till at West Angle -- interpreted as having been derived from pre-existing sea-floor deposits.

A section from the beach at Freshwater West, interpreted as showing clay-rich Irish Sea till overlain by organic deposits from the submerged forest.  This is remarkably similar to the exposures at Amroth.

Gary thinks that the limestone fragments in the Amroth exposures look as if they might have come from Ludchurch, a few miles to the north.  I'm not very convinced about that, as they could equally well have come from the limestone outcrops of the South Pembrokeshire coast.  The local rocks in Amroth all belong to the Coal Measures -- sandstones, shales and mudstones, and coal bands as well.  That having been said, erratics can be moved about in different directions in different glaciations, and it is known that many erratics rom North and mid Pembrokeshire were indeed transported southwards and eastwards during the Anglian glaciation.  Erratics from Ludchurch could have been introduced to the Amroth area 450,000 years ago and then redistributed during the Late Devensian glaciation.

What does all this mean?

This is all rather provisional, until the "Devensian till" and other deposits have been properly examined, but all of our assumptions about the extent of Devensian ice now have to be subjected to intense scrutiny.  Readers of this blog will know that I have wrestled endlessly with the concept of the "South Pembrokeshire enclave" -- hemmed in by ice from the east, north and west.  But now we are seeing increasingly convincing evidence of ice coming from the south as well, to the point where it becomes ludicrous (ice does not like flowing in a direction opposite to that from whence it came.......).

My latest reconstruction of the Late Devensian maximum ice extent in West Wales.  The large "enclave" shown is problematic, and needs revisiting........  Amroth is located close to the Pembs / Carms border, north of the word "Unaffected" on the map. 

Readers of this blog will also know that I am in a jolly dispute with certain other senior glacial geomorphologists about the nature and age of "diamictons" (which I interpret as Devensian till exposures) all the way along the South Pembrokeshire coast, culminating on the east coast of Caldey Island.  That's the basis for the line shown on the above map.  I have done many posts describing these exposures, which occur mostly on clifftops.

But now are we needing to extend this line right up into Saundersfoot Bay and suggest that the ice of the Irish Sea Glacier travelled NORTHWARDS to reach the coast at Amroth?  What other evidence is there for glacial action on this stretch of coast?  I have to admit that it is the part of Pembrokeshire that I know least well.

But the idea of an enclave pretty well encircled by active glacier ice looks increasingly untenable, especially in a lowland location like this.  This is no upland nunatak.  So did the ice of the Devensian Irish Sea Glacier affect the WHOLE of Pembrokeshire?

Watch this space ..........

This is a nice image (courtesy Pembs Coastal Photography) showing just how extensive the Amroth submerged forest is, when exposed:

And here are some more, gleaned from the web:

In this surface exposure the peat is very thin, and we can see that it grades down into 
a sticky blue clay

It's also worth remembering that the Amroth foreshore is one of the most important Mesolithic sites in Wales:

A paper entitled "Flint-working sites on the submerged forest bordering the Pembrokeshire coast, by Mr. A.L. Leach, F.G.S., will be found in the Proceedings of the Geologists' Association for 1918 (vol. 29, part 2), from which the following remarks are taken.

"Amroth, Site B2. -- Below the western end of this village evidence of flint-working abounds on a site first noted in August, 1912, and examined each summer and winter since. The sea washes away the soft blue silt, leaving the flakes projecting more or less noticeably. On each occasion I removed all visible flints and by the time of the next visit a fresh crop had become exposed. In August, 1917, for the first time in my experience, the whole site lay buried under several inches of sand. Objects in flint and chert collected inclued: one hollow scraper, one long flake, ridge-backed and serrated (saw); two shorter flint saws, two conical cores, one core trimmed to yield small flakes, three contiguous flakes, three long cores of cherty flint, two cores of black glossy flint, ten flint pebbles partly chipped into cores, fourteen small blades, twelve large flakes, two calcined flints, some scores of roughly chipped and broken fragments.”

See also Royal Commission Inventory for Pembs 1925.

Leach, 1912:

One locality near Amroth, in Carmarthen Bay, yielded cores and flakes in abundance; the circumstances indicate the existence of a chipping-floor or implement-factory on this part of the submerged land-surface, which now, during spring tides, is covered by not less than 20 ft. of water. In the patch of submerged forest recently exposed at Fresh-water West, in southern Pembrokeshire (see NATURE, March 28, 1912), a few small flint implements were also found.
Ref: Antiquity of Neolithic Man. A. L. LEACH — Nature volume 90, page134 (1912)

This is from the 1925 Royal Commission inventory:

Friday 28 February 2020

Abermawr -- the base of the Irish Sea Till

A fresh exposure of Quaternary deposits at the northern end of Abermawr bay.  Above the broken bedrock we can see a coarse rockfall breccia grading upwards into finer brecciated slope deposits.  Then above a coarser layer of broken bedrock slabs is the base of the Irish Sea Till -- weathered to a foxy red colour.  Above that is the unmodified blue-grey clay-rich Irish Sea till with its weathered upper surface -- and then the "rubble drift" ow interpreted as a modified and maybe remobilised meltout till capped by some blown sand and modern soil.  The till wedges out from right to left.

The severe storms during the month of February 2020 have not come from the north, but nonetheless there has been severe damage to the northern drift cliff at Abermawr, with cliff retreat in places, and a complete rearrangement of the storm beach at the base of the Quaternary exposures.  Pebbles have been piled up high against the face, and a number of fresh features have been exposed.

At the moment the basal contact between the Irish Sea Till and the brecciated slope deposits is particularly easy to examine.  The contact is very sharp, indicative of a sudden arrival of the ice front on a surface frozen solid by permafrost.  There are signs that this is an erosional contact, meaning that the overriding ice has removed some of the brecciated slope materials;  there is some evidence of shearing, and of the incorporation of rock debris into the glacier base.  Others have studied the glacier tectonics here in the past.  As we see elsewhere, the lowest 10 cm or so of the Irish Sea till have been weathered by water percolating upwards from within the slope deposits.  But this is pure clay-rich Irish Sea till right down to the base of the glacial layer.

The contact between the brecciated slope deposits (below) and the Irish Sea Till.  The massive clay-rich till with scattered erratics is weathered and stained to a foxy red colour in the 
lowest 15 cm - 20 cm -- from the trowel handle to the tip of the blade.

Evidence of glacitectonic features on the glacier bed

Newgale -- 2020 exposures of the submerged forest

Extensive peat beds running beneath the storm beach ridge, which had been flattened and reshaped during the storms.  On the right, the residual "boulder bed" which is now a prominent feature of the area normally occupied by the sandy beach.  The beach has probably been lowered by around 2m.  

Gullied peat bed, with alternating rills and ridges.  This feature was much more obvious in 2014 -- created by the downwash from breaking waves.

Remnant of a fallen tree trunk resting on peat.  There are many such features currently exposed at low water.

Broken tree trunk with remnants of the root system

A wonderful photo from Jamie Crofts, with acknowledgement!  It shows the residual boulders interspersed with slight sandy embankments and exposures of the peat beds.  I suspect that many of these boulders are residuals from a till layer that is not very far beneath the surface.

An erratic boulder with sharp edges exposed far out in the bay -- very different in appearance from the rounded beach pebbles.  There are many such boulders interspersed among more rounded boulders -- again suggestive of the presence of a till layer which is intermittently exposed.

The submerged forest at Newgale is currently very well exposed, following two severe storms during the month of February.  The photos show that the exposures are very extensive across a wide area littered with concentrations of well-rounded beach pebbles and also boulders up to 1 m in diameter.  In truth, the boulder litter is more spectacular than the peat beds and remnants of trees, branches and roots -- and there are no clear vertical sections in which one can see stratigraphic relationships.

In 2014 the relations between the various deposits were easier to see and interpret.  This is what I said at the time:

In trying to work out what has happened here, my best bet is as follows. There are some very old deposits here -- maybe dating from the last interglacial. These materials must have been overridden during the Devensian ice advance of the Irish Sea Glacier, but we see no signs of till or periglacial materials out here in the open bay. Peat beds and woodland then started to form after the ice retreat, when the coastline was far out to the west. So some peat beds might be more than 7,000 years old. I think these peat beds and the forest might have survived for at least 5,000 years -- this could be established by pollen analysis and examination of the tree species represented. The sea rose inexorably, and as it did so it drove a storm beach ridge eastwards, covering the old peaty / woodland area bit by bit, leaving it submerged beneath pebbles and sand. Within the last 2,000 years or so the eastwards advance of the storm beach ridge has slowed down as sea level has stabilised, but we know that it has continued because historical records show that there have been at least two other inns at Newgale, each of them in the area now submerged beneath the beach. The current Duke of Edinburgh Inn, alongside the road that has been blocked by pebbles several times this winter, is equally vulnerable -- and one wonders how long it will survive!

The interbedding of storm beach pebbles and peat beds suggests to me individual storm events in which waves have overtopped the ridge, flinging pebbles onto the peaty boggy area on the landward side -- then followed by more peat formation, when "normal" waterlogged or lagoonal conditions returned.

So it is quite possible that the peat beds are of all sorts of ages, with some of them shown in the photos above no older than a few centuries........

A 2014 photo showing thin peat resting on top of beach gravels.  I am inclined to think that this peat might date from historic time and may be less that 500 yrs old, developed in a boggy area on the landward side of the storm ridge as it moved inexorably eastwards.  During extreme events, pebbles are thrown over the crest of the storm beach ridge, settling in the boggy area beyond it.  Peat then forms above such pebbly layers and accumulates until the next big storm event occurs -- so there may be intercalations of several pebble beds and peat beds. They may not be laterally continuous.   Palynological and radiocarbon research will soon sort this problem out........

In Storm Barney in 2014 the storm beach ridge was breached in at least two places and huge quantities of pebbles were flung across the road and into the flooded area on the landward side of the ridge.  Events such as this must have occurred during the life of the ridge as it advanced eastwards from the outer part of St Bride's Bay over a period of 5,000 years or more.  Pebble layers derived from such events would then have been interbedded with peaty layers in the Newgale lagoon.

I see no reason to revise any of that.   But it is still difficult here to establish any sort of stratigraphy, and because storm damage is so extensive there are detached "rafts"of peat all over the place (sometimes slabs of peat rest on top of modern beach sand), and the remnants of tree trunks, branches and root systems which we can see are not guaranteed to be in their original positions.  This is a very dynamic environment!

On our visit to the beach, Ruth Crofts drew my attention to two subangular boulders which were stuck to pillars or pedestals of sticky grey clay.  We know that there are clay-rich layers in the peat beds and that some of them are clearly defined -- but the clay here is different, and this may be the most important observation of the afternoon!

Two subangular boulders exposed at the edge of the storm beach ridge.  Note that both of them rest on pedestals of sticky grey clay with small stone inclusions.  I suggest that we are looking here at the remnants of a layer of sticky Irish Sea till.

Nearby, the grey clay is exposed in a small pool and is seen to contain a wide variety of broken erratic fragments of many different shapes and sizes.  This appears to be till, directly overlain by a thin bed of peat.  This relationship is duplicated in Newport Bay.

In addition to the detached peaty remnants, there are several detached remnants of the concreted gravels which I mentioned in 2013:

I'm puzzled by these, because they appear to be beach gravels solidly cemented with manganese oxide cement.  Are they fragments of a very ancient ( interglacial) raised beach which may be present beneath the Newgale till layer, or are they simply fragments derived from the interior of the storm beach ridge where hydrological conditions have been suitable for the precipitation of manganese and iron oxides?  There is a project there for somebody.........

There are still many questions to be asked and answered at Newgale, but my current best guess at a stratigraphy is as follows:

5.  Modern sandy beach capping oll of the other deposits.
4.  Storm beach pebbles contained within the migrating ridge and left behind on its seaward flank.
3.  Holocene peat beds broadly assigned to the "submerged forest" -- formed during sea level rise.  They may range in age from 8,000 yrs BP to just a few centuries BP.  Formed on the landward side of  the advancing storm beach ridge.
2.  Layer of clay-rich till containing many foreign erratics (Late Devensian).  Many of the large boulders scattered across the beach are derived from this till layer.
1.  Cemented interglacial (?) beach gravels -- detached fragments are now scattered across the beach.

Complications occur because of the breaking and repositioning of peat rafts etc during storms and intermittent stripping away of the beach sand, and because of the intercalation of pebble beds and peat beds on the landward side of the migrating storm beach ridge.

Thursday 27 February 2020

Abermawr raised beach -- more exposures after the storms

The main raised beach exposure, including some boulders over 50 cm in diameter.  The exposure is up to 2 m thick and 4 m from edge to edge, and is capped by brecciated slope deposits and then the famous Abermawr "rubble drift"  -- now interpreted as a rearranged or mobilised meltout till madee of far-travelled materials mixed with local slope deposits.

Following the two severe storms within the last fortnight, I think some spray from the highest waves has been reaching the raised beach exposures in the northern cliff section at Abermawr -- about 5m above HWM.  I have not ben able to fix the altitude properly, because the exposures are inaccessible at present.  Anyway, there are now two exposures separated by a gully -- and in all, the beach remnants stretch about 10 m laterally.

The second exposure of the raised beach, which extends for c 6m and rests on an inaccessible  bench or platform on top of a rough bedrock cliff (which is steep and slippery).  The brecciated deposits resting on the beach are of Early and Middle Devensian age, accumulated over a period of maybe 50,000 years.

On this remnant of the raised beach platform we can see clear traces of a smoothed wave-eroded surface below the rock overhang.  It's steeply sloping, and when -- in due course -- it becomes possible to examine the raised beach and its platform at close quarters, it will be interesting to see how this smoothed rock slab relates to everything else......

As indicated in earlier posts, the assumption is that this is a Last Interglacial raised beach capped by a suite of Devensian periglacial and glacial deposits.

Wednesday 26 February 2020

How stable are horns and knife-edged ridges?

The Matterhorn in Switzerland -- the most famous example in the world of a "horn" or residual peak with three or more intersecting faces or rock walls.  Peaks such as these are often too steep for snow and ice to accumulate, although small impermanent ice masses and snow patches can present grave dangers to climbers since they are likely to break off without warning and go crashing down at high velocity.

Here is another splendid image of the Matterhorn -- the most photographed mountain in the world?

Further to this recent post:

there has been some discussion in the literature about the stability of steep pyramidal peaks (horns) and knife-edged ridges or arētes.  They occur in particular in glaciated alpine landscapes, but they are common too around the edges of ice caps and ice sheets where remnants of old interfluves stand well clear of the surfaces of outlet glaciers and where local cirque glaciation comes into play.

By the way, I'd like to see the word "tind" used instead of "horn" because it has a much longer association with the landscape.  According to Wikipedia:  From Middle English tind, tynd, from Old English tind (“tine, prong, tooth”), from Proto-Germanic tinduz, tindaz (“prong, pinnacle”), from Proto-Indo-European (e)dont- (“tooth, projection”). Cognate with Dutch tinne (“battlement”), German Zinne (“pinnacle, battlement”), Danish tinde (“pinnacle, battlement”), Dutch tinne (“tooth of a rake”), Icelandic tindur (“spike, tooth of a rake or harrow, pinnacle, peak, battlement”). Cf. the related tine. Also more distantly related to Dutch tand (“tooth, tine”), English tooth.  In Norwegian the word tind is also used specifically for a high and spectacular mountain pinnacle.

Ama Dablam, Nepal, Himalayas -- a mountain horn whose summit is at 6,812 m above sea level

Ulvetanna Peak, Queen Maud Land, Antarctica -- altitude 2,930 m.   Here we can see the remnants of ridges running out from the peak.  That in the foreground has almost been removed by coalescing cirque glaciers, and that in the middle distance is being whittled away, leaving very spectacular pinnacles and arete remnants behind.

Machapuchare, Himalayas, altitude 6,993m.  A sacred mountain which may or may not ever have been climbed.  Climber are asked not to set foot on the summit.  Really there are two summits connected by a vicious knife-edged ridge.

Fitz Roy in the Argentinian / Chilean Andes -- summit altitude 3,405 m.  This is the highest pinnacle in an area of jagged peaks and ridges affected by multiple small glaciers with many different orientations.

Mount Assiniboine in the Canadian Rockies.  Altitude 3,618 m.  Known as "the Canadian Matterhorn".

Stetind, Nordland, Norway, summit at 1392 m.  Unofficially Norway's "national mountain"..........  On the summit there is a remnant of an ancient plateau surface.

Aguille du Dru, in the French Alps.  Summit altitude 3754 m.  This very spectacular summit is the highest point on a long ridge of which much remains.

Tre Cime di Lavaredo -- three distinctive and famous peaks in the Italian Dolomite.  The highest summit lies at 2,999 m.  These are really the last remnants of a mountain ridge -- currently subject to intense frost weathering -- note the steep banks of scree, which do not accumulate in situations where flanking glaciers are still present.

I have not included Ketil (Greenland), Half Dome (United States), Mount Asgard (Baffin Island), and Mount Thor (Baffin Island) because although they are all located on ridges they are also flanked by glacier outlet troughs which have cut away, in each case, one particularly spectacular and vertical rock face.  So they are asymmetrical features, somewhat different in origin from the horns created at the intersections of mountain ridges affected by cirque and valley glacier erosional processes.

Back in the 1960's Professor David Linton had a theory that horns and knife-edged ridges are very stable features which can survive for many thousands of years in a state of equilibrium.  He never did explain this theory very well, and I doubt that it is very reliable.  He thought that horns formed at the intersections of ridges affected by cirque glaciers.  As cirques expand, their headwalls are pushed back further and further,  eating into the ridge until the ridge crest is breached, at which point glacial processes are reduced in effectiveness.  As noted above, many horns have three or four faces.  Linton thought that "dilatation of the rock parallel to each face" was an important mechanism of erosion, and he thought that this process would become less effective as the horn gets smaller, eventually reaching an equilibrium state.  In this state rock expansion and spalling will take place all over the horn rather than being concentrated on the flattish horn faces.  I don't buy into this hypothesis -- and it must surely be the case that exposed pyramidal peaks must be affected most of all by frost shattering on the edges or corners between the flattish faces -- and this process would increase as a horn gets smaller, rather than reducing.  If the flanking cirque glaciers are rendered ineffective as a result of ridges being breached or destroyed, it is possible that the mechanism for the glacial removal of rockfall debris might be reduced, and this might result in a buildup of scree, as we see in the case of the Dolomites.

I suspect, however, that theories of evolution and equilibrium states are all rendered redundant by the complex and unique histories of glacier / climate oscillations experienced by  all of the sites at which horns and knife-edged ridges are found.   Geology and structure play a part too.  No two sites have experienced exactly the same history -- so horns never have perfect asymmetry, and if you look at all of the examples illustrated above, the similarities are obvious -- but so are the deviations from the ideal.

Tuesday 25 February 2020

Sixty years ago, in Iceland.......

I was rummaging through some box files today, and came across this scruffy document.  It's the original plane table map which I made with David Sugden (now Emeritus Professor at Edinburgh University) in the summer of 1960.  Sixty years ago...........  and the memories of our Oxford Icelandic Expedition are as fresh as ever.  Well, some of them are.....

I am reminded how things have changed.  In 1960 we walked everywhere in and around the valley of Kaldalon -- no helicopters, Land Rovers or skidoos.   I spent hours every working day at the plane table, taking sightings, building up the triangulation, and putting notes all over the map and into my field notebook.    No aerial photos or satellite images, no GPS signals, and not even a theodolite.  Dave and I worked very well together, discussing all our observations endlessly, and learning our trade as geomorphologists on the hoof -- after just three terms as students in Oxford University.  It was really rather presumptuous of us, to assume that we could discover new things and say something useful, having spent no time at all studying glacial geomorphology with specialists.  But while we were in the valley, living under canvas for six weeks or so, we did learn rather a lot, and afterwards we wrote our first scientific paper, called "The morphology of Kaldalon, a glaciated valley in Iceland."  It wasn't really very good, but to his eternal credit Professor Gunnar Hoppe of Stockholm University accepted it for publication in "Geografiska Annaler", probably because he thought that we were young men who should be encouraged in our efforts to undertake serious and meaningful field research.    And so, after another expedition in East Greenland in 1962,  and after a spell with BAS in the Antarctic, David and I embarked upon our academic careers.

Happy days..........

Some shots of the valley as it looks today.  The glacier was quite healthy in 1960, but now it is on the way out -- more and more of the trough head / rockwall at the end of the trough is exposed every year, and soon the supply of ice from Drangajökull will be cut off.

Sandur in Kaldalon Valley, with the pitted outwash area in the foreground. (Drone image)

Eskers and pitted outwash in the area referred to as "The Trout Pools".  Our base camp was located near here.

See also: