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

Sunday 31 May 2015

Kaldalon Sandur, NW Iceland

This is a photo from my 1960 collection -- the sandur or outwash plain in the Kaldalon Valley, NW Iceland.  The water is carried seawards from the snout of the Kaldalon Glacier (an outlet glacier from the Drangajokull ice cap).  This not really a "flood plain" because there is no flooding season -- the streams are so debris-rich that channels get choked with debris all the time, overflowing and constantly finding new channels.  A picture such as this would be different every day.

But river levels do rise and fall quite dramatically -- in the afternoons, during warm summer weather, the river will be high from snow and ice melting on the glacier, making crossings hazardous and often impossible.  The photo below is typical.  In the nights, when the temperatures are lower and melting rates are greatly reduced, crossings are possible with care, if you choose the right places......

Afternoon "high water" on the Kaldalon sandur.

Saturday 30 May 2015

Deeply symbolic......

One of my favourite images of Pentre Ifan.  To the right of the cromlech, two people having a jovial scientific discussion about the origins of megalithic monuments (or some such thing) and on the left a man with his head inside a cardboard box......

Friday 29 May 2015

Meltwater channels at Rhosyfelin

I thought I should record these for posterity, in a separate post.  On the image above we can see the edges of the Afon Brynberian Channel and (in the upper part of the image) its confluence with the gorge of the Afon Nyfer.  On the top edge of the photo you can see the edge of Castell Mawr -- another of the sites investigated by Prof MPP and his colleagues.  The gorge of the Afon Nyfer is quite complex -- it seems to have been cut in several phases, which are not well understood. Within it, at Felin y Gigfran, there are crags reminiscent of those at Rhosyfelin.

In the image above we see the "micro-morphology" associated with the crag of Craig Rhosyfelin.  There are undoubtedly two minor channels.  The western one has a "blind" intake.  The eastern one is actually a humped channel, and in this photo, with the shadows just right, we can see the intake channel which carried water uphill on the western side of the Brynberian Gorge, over a little col and then after swinging to the right, down along the NW flank of the spur.  At the NE tip of the spur, where we see much evidence of bedrock moulding and block abrasion by torrential sediment-rich meltwater, the flowing stream joined the much more substantial torrent flowing within the main valley.

These are classic features of sub-glacial meltwater erosion,  and although in their main outlines they may date from the Anglian Glaciation, I suspect that there has been considerable modification and "freshening-up" during the Devensian glaciation around 20,000 years ago.

Just a reminder:  these were the directions of meltwater flow:

Thursday 28 May 2015

The Rhosyfelin roadstone quarry?

I have been pondering at some length about the characteristics of the little "cwm" at the extreme SW extremity of the Rhosyfelin rock face.  This is the intake point of one of the subsidiary channels which carried glacial meltwater along on the western flank of the Rhosyfelin ridge.

Proposed Anglian (and possibly Devensian) meltwater flow in the vicinity of Craig Rhosyfelin. The long arrow indicated the flow direction of water in the main river gorge, and the two shorter and converging arrows indicate the flow directions which led to the creation of the two smaller channels.

The meltwater which flowed over the col at the base of the Rhosyfelin rhyolite ridge has had a dramatic effect on the exposed rock outcrops -- the moulded bedrock features are quite spectacular.  But the steep drop into the little "cwm" is unusual, and on examining it in detail I really wish that the archaeologists had not smashed everything to pieces when they moved in during Sept 2013 and made further inroads in Sept 2014.  There are piles of debris lying about here and there, and even bits of timber including an old telegraph pole; and I don't think all of this is is the responsibility of the archaeologists.

Is there an twentieth-century roadstone quarry here?  I have searched the records, and there is nothing on the 1888 six-inch map:

But we can see that in 1888 there was a trackway here, extending first in a SW direction from the steep bend on the modern road, and then running NW along the hedge line to join the road again.

I know from experience in other parts of Pembrokeshire that in the 1920's and 1930's there was a massive road building programme, and a huge demand for hard core or road metal, as tarmac roads were built to replace the old dirt tracks.  A lot of people who had access to roadstone made small fortunes from selling hardcore to the County Council.  In some places there were even crushing plants, which did their job for a few years and were then dismantled.  This was all on a "cottage industry" scale rather than an industrial one.  So did the local landowner at Rhosyfelin take advantage of this commercial opportunity and take a few hundred tonnes of rhyolite rubble from this location?  Worth thinking about.  It would have been very easy to transport the stone out to the roadway on the hillside.  Another possibility is that the local farmer has been using this site for taking stone for his own purposes, for improving farm tracks.  This is the location:

There has certainly been a lot of messing about at this location, and in relatively recent times.  So here are a couple of recent photos which show "the cwm" and the approx shape of the bounding slopes:

I suppose one could go back to Pembs CC or Nevern parish records to see if there has ever been a planning consent or a registered quarry here -- but I have my doubts.  In Pembrokeshire a lot of things have always gone on under the radar.

Anyway, that's a little theory, for what it's worth.......

Tuesday 26 May 2015

Where did the Rhosyfelin "proto-orthostat" come from?

We have had some discussion about the origin of that big rock slab Rhosyfelin.  There is some confusion out there.  Prof MPP and his colleagues think that it is now about 15m from its source.  That does not mean 15m from nearest part of the rock face, but the confusion is understandable, and the archaeologists should have made things clearer than they have.  Here is a pic of the relevant part of the face:

You can see the slab with the bucket on it.  MPP clearly does not think the stone has come crashing down from the high pinnacle at the top of the photo, but from one of the "fresher looking" exposures at the extreme right of the photo.  They would be about 15 m away.  He thinks it was quarried from there, and then manhandled or manoevred to its present position by being dragged along on these famous stone "rails"  -- on the subject of which we have spent enough time already......

Myris has informed us that only one rock sample has been taken from the slab itself, but maybe 20 or more have been taken from the rock face -- presumably with a view to establishing exactly where the slab came from.  We can also assume that the provenance has been established to the satisfaction of Drs Ixer and Bevins, who will be joint authors when the big paper is published in the autumn.  We'll see whether the petrographic evidence stands up -- there are certain aspects of it thus far that I have questioned on this blog!

Could the stone have got from one of the recesses on the right of the photo to its present position without human involvement?  I wouldn't see a problem with that -- the gradient is quite steep, and many stones in the neighbourfood have moved a long way from their places of origin, with the assistance of glacier ice, meltwater and periglacial processes.  There are some big slabs quite far out onto the valley floor, near the NE limit of the area excavated thus far.

Let's see the colour of the evidence......

Monday 25 May 2015

Other elongated pillars out on the valley floor at Rhosyfelin

I recall from various talks that the big flat-topped stone at Rhosyfelin is deemed to be unique because it is too far from the rock face to be in a 'natural' position, and because its alignment approx parallel with the rock face cannot be explained except as a result of human interference.  Somehow the word has got around that it is a very long way from the rock face; not at all, since one end is just 4 m from the rock face and the other end about 5.7 m away. 

As for uniqueness, the site is in a proper mess as a result of the excavations higher up the little meltwater channel; but we can still see three other elongated slabs that are approx parallel with the rock face and some distance away from it.  The one closest to the bend in the channel is a very rough and irregular elongated slab about 2.40 m long and about 3 m from the rock face. 

A second slab, of which we can only see about 1.70 m exposed, also seems to be aligned parallel with the rock face, and is 4.3 m away from it.

A third slab, about 5m further down the valley and also on the valley floor, is 2.20 m long at least.  It seems to be very heavily weathered, maybe because it has been exposed at the ground surface for many thousands of years while the 'proto-orthostat' in the main dig site has been buried beneath solifluction materials. Again it is aligned approx parallel with the rock face, and it is 4 m from the face. 

The archaeologists might argue that these three other elongated blocks were all aligned down-valley and parallel with the rock face because they were all carefully chosen as orthostats and were on their way down to a sort of assembly point or dressing area down on the main valley floor close to the tip of the spur.  I don't buy that at all -- these blocks are highly irregular in shape, and to me they show that some rocks -- just some, out of hundreds -- have slid or rolled into positions well out onto the valley floor either with the assistance of snowbanks or else because of the sheer velocity of their falls, from previous positions high up towards the ridge crest.

The crack of doom on the Rhosyfelin pseudo-proto-orthostat

I have had another look at that crack of doom which runs across the top of the famous stone at Rhosyfelin.  It isn't, after all, related to the two dipping fractures which are visible on the south face of the block, but when one looks carefully at the north face one sees that the fracture runs deep into the heart of the block, with a depth of 27 cms.  That's almost half the depth of the block.  It seems to terminate at another foliation plane which we can see as a little projecting ledge.  Yet again this is evidence that the fractures at Rhosyfelin are 'local' and discontinuous.  But this does mean that the strength and stability of this slab is seriously compromised, and I am more convinced than ever that it could never have been moved anywhere without breaking in half.

A place of beauty.....

Every now and then we need to remind ourselves that Rhosyfelin is a place of extraordinary beauty, tucked away in its wooded gorge, and with birdsong echoing through the woods.  This is the south/east side of the ridge this afternoon.  A nice contrast with the terrible archaeological mess  -- in more senses than one -- left by the archaeologists on the other side of the ridge.....

Rock mechanics at Rhosyfelin

Here again is Adam Stanford's fantastic Gigapan image of the Rhosyfelin rock face, scree bank and "proto-orthostat."  It helps to give the context for the previous post.

If we look at the big stone, the right (fresh) end was originally at the bottom of the standing pinnacle, resting on a jagged fracture or set of fractures,  and the  left (weathered) end was the one at the top, exposed to the atmosphere.  The bottom (buried) face of the stone would originally have been positioned approximately along the rock face as we see it today, facing NW.

By going to the source of the image you can zoom in very closely and examine things in detail.  Thoroughly recommended!

ADDED 25 May;  Here is that rather interesting diagram, courtesy Phil Morgan, showing the two mechanisms of rockfalling that we should be thinking about.  Although the rockfall debris has accumulated in an almighty jumle of blocks, slabs and fragments apparently at random, there are some signs of order.  One of the most interesting is that the weathered -- top -- ends of the pillars and pinnacles that have fallen tend to be furthest away from the rock face.  That means that the lateral fractures that run along the face, and which are at right angles to the foliation planes and the foliation surface fractures, have acted as ledges or hinges when the slabs have toppled over and outwards.

The top part of the rock face appears to have evolved as in Fig 4. and the lower part of the face might have evolved by sliding, as in Fig 5.  In reality the whole scenario has been far more messy, since accumulated banks of rockfall debris have provided a rough surface which, when snow-covered, could have acted to assist in the movement of blocks well away from the rock face.  Rolling and sideways sliding are also feasible.

Sunday 24 May 2015

The famous Rhosyfelin "proto-orthostat" -- the most useless bluestone ever?

The pseudo-proto-orthostat at Rhosyfelin, carefully protected from visiting tourists and marauding geomorphologists......

Largely as a result of media hype generated by Prof MPP and his colleagues, the large stone at Rhosyfelin which was first unearthed in 2011 has gained considerable notoriety.  It is known to thousands of faithful followers as "the bluestone that never made it to Stonehenge" or as "the bluestone that confirmed the presence of a Neolithic quarry at Rhosyfelin."

On p 286 of his 2012 book, Prof MPP refers to "an ancient ground surface", "hammer stones" and "a monolith left behind when quarrying ended" as showing that this is "the Pompeii of prehistoric stone quarries."  All a bit over the top, you might think.  But warming to his task, the good professor then says of the discovery of the stone:  "This was the smoking gun; the game was up for anyone still trying to argue that the bluestones were not quarried in Preseli during the Neolithic, and then taken to Wiltshire....."

In the midst of all this excitement, let's take a deep breath and take a look at the famous stone, leaving aside all the  speculation about fulcrums and rails, pillars and wedges and so forth.

The facts about the stone

As I have already indicated, it is made of local blue foliated rhyolite.  It is over 3m long and is probably twice as heavy as Prof MPP suggests -- at around 8 tonnes.  So it is much larger than any of the Stonehenge bluestones, and on that basis alone one might wonder how on earth our heroic Neolithic ancestors planned to get it out of the Brynberian Valley and all the way to Stonehenge, all in one piece. That would have been a Herculenean task, even if the rock was solid and robust, which it is not......

So let's take a look at it in detail.  We can of course only see five of the six sides, since its bottom face is hard to examine without risking life and limb.  The elongated, roughly rectilinear stone lies approximately E-W, so we can refer to the faces as "E,W, N, S and top" without any great risk of confusion.  The E and W faces are the "ends" of the rock and the N and S faces are the elongated "sides".

Here is a photographic record: 

The relatively fresh top and weathered N face of the slab.  The top surface is conformable with the foliations in the rock, and coincides with a foliation plane fracture.  The pitted surface, with foliation "ridges", is typical of the exposed surfaces on the crest of the rhyolite ridge today. One might speculate that the top was on the "inside" of the crag, protected from intensive weathering, when the slab was in its original position. If one looks carefully, one can see a fracture running across the top face of the rock, approx halfway along its length.

Close-up of the top face and the S face -- both relatively fresh and thus protected from weathering prior to the rockfall that brought the slab crashing down.  The tip of the pen shows the fracture that threatens to split the rock into two; it may be related to the dipping parallel fractures that we see on the S face.

The corner between the E face (end) at the right edge of the photo, and the elongated S face.  We see intense weathering features in the foreground, and in addition a weathered section of a substantial foliation plane fracture running along the S face.  Since the E face was the top of the rock when it was in its original position, the weathering (involving the rotting of at least some of the minerals) of this outcropping fracture plane is a consequence of concentrated rainwater running directly down the side of the slab, presumably in a fissure.

A close-up of the weathered E face of the rock, showing outcropping foliations, picked out by weathering over a very long period of time. This must originally have been the top of the pinnacle when the rock was in situ on the ridge.  

A close-up of part of the relatively fresh W (end) face of the rock, showing abundant parallel fractures unrelated to foliation planes.  Some of these fractures can also be traced on the top of the slab. The chunks knocked off have probably provided samples for the geologists to look at, and are nothing to do with Neolithic quarrymen.

[Note added on 25 May:  I suspect that the face that is now the "bottom face" of the recumbent stone was originally the one aligned approximately with the rock face as we see it today, ie coinciding with the sub-planar fracture surface which is so much discussed.  When it was in its original position, the faces were originally aligned as follows:
 E end was at the top of the pinnacle
W end was at the bottom
Top of the recumbent stone was facing SE (embedded in the crag)
Bottom of the recumbent stone was facing NW (probably exposed)
S face was facing NE (probably protected)
N face was facing SW (probably exposed)
The stone does not appear to have rolled into position.  Rather, it has fallen and slid, twisting in the process so that the bottom of the broken pinnacle now rests further from the rock face than the top end.  This is a three-dimensional jigsaw puzzle!  The archaeologists will of course say that the stone was manipulated into its current position by the famous Neolithic quarrymen, since it is too far from the face to be in a natural position.  I think it much more likely that snow, glacier ice or meltwater have played a role, given the context within which the stone is found.]

[Further Note added:  we had an interesting discussion in 2012 -- thanks to Phil Morgan -- on the rock mechanics of the site, reported here:
The diagrams are particularly helpful.  There were also lots of useful comments.]

In short, this old pinnacle is in a very fragile state, and it is a miracle that it has survived without breaking in two.  There are multiple fractures exposed on the flanks of the slab, unrelated to the foliation planes.  These render the slab completely unsuitable for long-distance transport or even for use in the immediate neighbourhood, as the slightest percussive impact might result in either breakage into two more or less equal portions, resulting in the loss of its "pillar" shape, or in the spalling off of large irregular chunks of rock such as those we see today beneath and on the flanks of the rock after the clearance work of the archaeologists.

In other words, this is a completely useless orthostat, since it has no integrity or inherent internal strength.  So it is a fantasy to pretend that it was in some way "valuable" or worth any expenditure of time and effort by our heroic ancestors.

Craig Rhosyfelin -- a seriously cracked-up crag

 Part of the rock face exposed during the archaeological dig.  If you get close enough to it and look along it, it certainly looks very flat, and one might think of it as "unnatural" in some way, and therefore a quarry face.  However...... things are not so simple.

I was intrigued the other day with Myris's comment that the "quarry face" at Rhosyfelin is  "a large planar surface that appears not to be structural in nature (absence of slickensides) and is difficult to explain if totally natural.  Not a fault plane nor master joint plane.........."

I commented on it at the time,  and have been back to take another look.  I am now more than ever impressed that the whole crag is so seriously riddled with multiple fractures running in all directions that it is a miracle that any of has survived the ravages of time and the Ice Age.  There are many posts touching on this subject, for example:

I published this rough illustration of the fractures in two dimensions.  The third dimension is the fracture plane(s) running along the face itself.

Things are actually a great deal more complicated still.  Let's concentrate for a moment on the "large planar surface" which might also be called a "sub-planar foliation surface" since it coincides with the foliations in the rock which have been given so much attention by the geologists.  It is not actually a single sub-planar surface at all, but a rock face made up of fragments of multiple surfaces.  Some parts of the rock face project at least 1.5 m out beyond other parts -- so we have recesses and projections when we examine the site carefully.  When we look at the tip of the spur, we see some of these surfaces end-on:

 Close-up of the tip of the spur.  Here we can see the steep dip of the "sub-planar foliation fracture surface" on the right, and then behind it, across the photo from right to left, a further 6 (at least) other fractures, more or less parallel.

 A close-up of another part of the spur tip.  There are multiple fracture surfaces coinciding with the foliation planes, some of them 10 cms apart and others less than 1 cm apart.

Looking in the other direction, down the slope from near the right edge of the photo with the blue and orange lines on, we get another cross-section of what is going on in the interior of the crag.  Again, we see multiple fractures related to the foliation planes in the rock.  There are at least ten pronounced fractures, and many more hairline cracks.

Now here's a question for the geologists, which was raised in a letter to me from Barry.  Why do these foliation plane fractures occur, in some places very closely spaced, and in others widely spaced?  Is it something to do with crystal alignment or arrangement, or something to do with cooling surfaces?  Rob and Richard, advice please?

So much for the fractures related to the foliation planes.  Then we get to all the other fractures.  I started to record them by compass directions, but gave up because there are so many of them, some traceable over 50m or more, undulating up and down along the rock face, and others restricted to just one semi-detached slab.  The message is that hundreds of these fractures are discontinuous and localised, developed in response to quite localised stresses.  So I would agree with Myris that there is no master fault plane or joint plane here, and certainly no bedding plane since we do not have any identifiable "beds" of rock as we do on sedimentary rock exposures. These are, after all, metamorphosed volcanic rocks, and the main processes of deformation shortly after their emplacement would have been related to cooling and contraction and maybe loading beneath later igneous and sedimentary accumulations.

So here are two more pics just to confuse things even further.

 Another exposure on the rock face, showing one long and undulating fracture running from left to right, and a series of steeply plunging fractures which are not parallel.  Just to the left of centre, there is a V-shaped arrangement of fractures.

This is the most complex arrangement of fractures on the whole rock face.  In a word, wholesale chaos in three dimensions......

Back to the original point.  This is certainly a highly complex rock face -- worth a structural geology field trip on its own account -- but I do not agree that it is difficult to explain if totally natural.  I see nothing whatsoever on the rock face to suggest human intervention, apart from the cosmetic work recently undertaken by the archaeologists. 

What intrigues me rather more is the role of geomorphology in all of this.  I mentioned the possible role of cooling, contraction and loading beneath later Ordovician and later rocks, but the more I think about it the more convinced I become that the rock wall is actually a meltwater channel wall, swept clean by turbulent meltwater flowing over the col at the base of the spur and then downslope in a subglacial channel.  I'm also quite attracted by the idea that there may have been actual fracturing of the crag as a result of ice loading during the Anglian Glaciation and -- to a lesser extent -- during the Devensian or Last Glaciation.  This process of compression / tension / pressure release might well have been a powerful factor in explaining the entrainment of blocks into the base of the over-riding glacier.  Another possibility is that we are looking at the results of unloading or pressure release linked to ice wastage.  We see evidence of this happening in many glacial environments today -- I have discussed this before in the blog.

Quarry face?  I suggest that we just forget the idea.

Friday 22 May 2015

Rhosyfelin -- how heavy is the big stone?

Can somebody please assist in some calculations?  Prof MPP says in his book that the Rhosyfelin "proto-orthostat" -- so greatly revered that some even take off their wellies to stand on it -- is about 13 feet long and weighs about 4 tonnes, i.e as much as the large bluestones at Stonehenge.

I measured it today, and found that it is c 3.80 m long, c 1.30 m wide and .60 m deep.  That gives a volume of 2.964 m3.  Multiply by 2.7 for rhyolite, and that gives a weight of just over 8 tonnes.

Has anybody done an "official" calculation?  If so, what is the "official weight"?  Myris, what do your geological friends say?

Many of us who have examined it are quite sure it is far too big and heavy to be a candidate for hauling off to Stonehenge -- quite apart from its lack of structural integrity, of which more anon......

In my estimation, if any attempt was to be made to move it away from this site, it would break in half in pretty short order.  "Ah yes", will say the archaeologists, "that's why the heroic Neolithic tribesmen  left it behind, following its rejection by the project engineer."

Thursday 21 May 2015

Rhosyfelin historic photos

I came across these wonderful B+W photos -- presumably taken by Dr Richard Bevins or one of his colleagues -- in the 2011 Archaeology in Wales publication by Ixer and Bevins called CRAIG RHOS-Y-FELIN, PONT SAESON IS THE DOMINANT SOURCE OF THE STONEHENGE RHYOLITIC ‘DEBITAGE’.

These images are reproduced from the on-line pdf version of the article.  They show the outer tip of the rhyolite spur, from slightly different angles.  The colour photo at the bottom shows what the site now looks like, after the stripping away of a vast amount of vegetation. To the right of the crag, the jungle has disappeared.

The B+W photos appear to have been taken in June 2010 when Richard was taking his eight samples from various rock outcrops on the ridge. No wonder he only took one sample from the "rock face" to the right of the ridge crest!

It is so long since the vegetation was stripped away that we tend to forget what the site looked like originally -- and we can all too easily forget what a huge amount of effort has been expended here by Prof MPP and his team.  Whether features have been "revealed" or "exposed" in the process, or "created" by the diggers, is a rather interesting question........  since the removal of the first tree and its roots in 2011, in the first of four seasons of digging, there have been multiple choices relating to which stones and which debris to take away, and which things to leave behind as evidence in support of the Neolithic Quarry hypothesis. 

Wednesday 20 May 2015

Garn Ffoi -- prehistoric stock management features?

Just for fun -- arising out of the suggestion that the Cursus at Stonehenge might have been a stock management structure rather than a Roman chariot racecourse........

This is a satellite image of Garn Ffoi near Newport, Pembs, showing the assumed Iron Age defended settlement site, with more modern enclosures to the SE and then towards the bottom of the image the wonderful stone walls (now ruinous) that can only have been used in association with the gathering and sorting of sheep and other stock normally left to run free on the common.  Could they be prehistoric?  Note the 90 degree bend in the southernmost corral. The other one has a funnel shape.  Not far away from here there is also a wonderful cattle pound, assumed to be medieval:

Saturday 16 May 2015

New video from Prof MPP -- nothing changes.......

From the Stonehenge News site -- info about a new AHRC short video in which Prof MPP talks about his assorted projects at -- and relating to -- Stonehenge.  A celebration of "remarkable discoveries" and media obsessions.  If AHRC can point to a certain acreage of press coverage for projects that it funds, all is presumably well with the world......... Nothing much changes.

"Bluehenge" doesn't feature much in the video, apart from being referred to as a "new stone circle", although it is clear from the press notice that MPP still thinks it held a circle of bluestones, in spite of zero evidence to support the idea.  No changes there then.

And why is Stonehenge where it is?  Well, it's all to do with the brilliant discovery of those periglacial stripes, which made Stonehenge a very special place in the natural world.  Same old pictures, same old argument.  No changes there then.

And the purpose of Stonehenge?  Well, again there has been a brilliant discovery, and it's all sorted.  It was built as a political statement and unifying feature, attracting people from all over the UK as a place where the ancestors could be revered.  No changes there then.

And the link with West Wales?  Ah yes, all those bluestones were brought from a very small area of West Wales,  as a confirmation and statement of political unity.  One thing is interesting -- namely that MPP still thinks the stones were used as well for a local stone circle, for which he and his team are still presumably searching........  No changes there then.

There is also mention of the fact that the "Feeding Stonehenge" project is now finished.  I was under the impression that the Rhosyfelin dig is a part of that particular project.  Does that mean there will be no more digging at the site in 2015?  If so, I hope somebody is going to come back and clear up all that mess.

Although the film is just 5 mins long, it suggests that MPP's core beliefs about Stonehenge and West Wales, as expressed in his book (following the 2011 dig at Rhosyfelin) are completely unwavering.  What was I saying about ruling hypotheses?

Stonehenge Riverside Project. A film from the AHRC
by stonehengenews
This film from the AHRC is the latest in our specially commissioned series which celebrates the AHRC’s 10th anniversary. This film looks back to the year 2007 and the Stonehenge Riverside Project.

Led by Professor Mike Parker Pearson, the Stonehenge Riverside Project brought together academics from around the globe in one of the largest field archaeology research studies of the 21st Century.

The project set out to further understanding of Stonehenge and neighbouring complex, Durrington Walls. The project looked at the sites in the context of the surrounding landscape, and in doing so several remarkable discoveries were made including a new stone circle dubbed, ‘Bluehenge.’ The discovery catapulted the project to the forefront of the world‘s media and made front page of newspapers all over the world.


The Stonehenge News Blog

Friday 15 May 2015

Rosemergy niche glacier, Cornwall

 The Rosemergy hollow or niche, about 10 km from Land's End.  From Harrison et al 2015.  The red line shows the catchment and the area within which the small glacier is thought to have been located.

Another interesting paper from Stephan Harrison and colleagues, providing evidence for a very small niche glacier tucked away at Rosemergy, in a NW-facing hollow near the outer tip of Cornwall.

This is so far to the south and west that most people until now have discounted the idea of Devensian glacier ice forming on the mainland and have assumed that the basic processes of landscape evolution during the Devensian (and earlier glacial episodes too) must have been periglacial rather than glacial.  That "traditional" view is not that sensible, given that periglacial environments in Devon and Cornwall during the Late Quaternary must have been both very cold and very snowy -- with extensive perennial snowpatches in favourable locations -- just as we see today, for example, in places like Arctic Canada and the South Shetland Islands in Antarctica.  Where there are perennial snowpatches there is a good chance that some of them will be thick enough for the formation of firn or glacier ice, and for them to be classified as "niche glaciers".  (Cirque glaciers tend to be larger, and need greater relief than we find in the South-west of England.)  So ice-related processes could -- in theory -- have operated in some places, and moving ice on the bed of some of these niche glacier features might even have resulted in the creation of striations on bedrock and on transported pebbles and boulders. The processes responsible for these subtle landscape alterations are referred to as "nivation" processes.

Landscapes in the South Shetland Islands, Antarctica.  These are extreme periglacial landscapes with permafrost and frost-related processes dominating.  In the spring we see a mosaic of snow-patches as the winter snow cover melts off.  Some of these snow-patches are seasonal, and others are perennial.  Some, in the most favourable locations, are more than 50m thick, and contain snow, firn and glacier ice which moves very slowly, at a rate of maybe a metre or two each year.  This is a reasonable "model" for the Devensian situation across much of Devon and Cornwall.

Another reason why we should tend towards acceptance of the evidence presented in this paper is that in the late Devensian the ice of the Irish Sea Glacier affected parts of the Scilly Islands, even further towards the SW; and at the same time there was a sizeable Dartmoor Ice cap which was thick enough and long-lasting enough to have left substantial features behind when it melted.  See here:

Then we have the evidence of Devensian glaciation on Lundy as well, which I have covered in a number of posts on this blog.  Here is one of them:

And as for Exmoor, there are yet more posts.  Find them with the search box.  Here is one:

If there were perennial snow-patches, niche glaciers and even local ice caps in the far South-West during the Devensian, there must have been a pretty extensive snow cover on Salisbury Plain as well.  In the Anglian Glaciation, which was by common consent much more prolonged and severe, the idea of glacier ice pressing far inland, maybe as far as Wiltshire, does not seem half as "extreme" as it might have done twenty years ago.


"The southernmost Quaternary niche glacier system in Great Britain"
 3 February 2015
JOURNAL OF QUATERNARY SCIENCE (2015)    ISSN 0267-8179. DOI: 10.1002/jqs.2772

Until recently, the scientific consensus has been that the uplands of south-west Britain remained unglaciated throughout the Quaternary, with glacial ice sheet limits lying to the north of the south-west peninsula. However, recent work has shown that small glaciers and ice caps existed in the uplands of Exmoor and Dartmoor during the late Quaternary, demonstrating that the consensus of an unglaciated south-west Britain requires considerable revision. Here we report geomorphological and sedimentary evidence supported by glacier- climate modelling for a Quaternary niche glacier from west Cornwall, south-west England. This niche glacier represents the southernmost such system from mainland Great Britain, and provides evidence for the presence of extra-glacial niche glaciers probably during the Last Glacial Maximum of the Devensian glaciation, and well outside the limits of the main British–Irish Ice Sheet.

Tuesday 12 May 2015

The Celtic Sea ice lobe -- again

There has been another paper on the Celtic Sea ice lobe -- ie the southernmost extension of the Irish Sea Glacier (otherwise referred to as the British and Irish Sea Ice Sheet or BIIS) during the Devensian maximum.  This one is by Mark Furze and others.  It contains detailed analysis of the sedimentary cores taken from the Celtic Sea Deep or Celtic Deep Basin, which we have discussed before in this blog. Neither the abstract nor the paper itself are easy to follow, but essentially the paper seeks to understand what happened when the glacier ice began to melt back towards the narrow constraints of St Georges's Channel at a time of rising sea level and isostatic recovery.  Was a temporary lake contained within the basin, or are the sediments best explained as glacio-marine, related to the gradual lifting of the ice margin as the sea flooded in from the south?  The evidence is equivocal, and to their credit the authors reserve judgment, and invite future researchers to come up with evidence that might prove conclusive.

What interests me rather more is the problem of ice limits.  The authors have chosen to stick with the lobe shape as proposed by Scourse and Furze in 2001 and as accepted by various authors since then.  I have never thought that the ice limit is very sensible, since the dashed red line on the map does not accord with the evidence for Devensian till on Lundy Island and Caldey Island and inside the mouth of Milford Haven, and neither does it fit well with what we know of the behaviour of large ice masses in unconstrained environments. 

That having been said, bit by bit the Devensian history of the Celtic Sea "theatre" becomes clearer.....


Furze, M. F. A., Scourse, J. D., Pien ́kowski, A. J., Marret, F., Hobbs, W. O., Carter, R. A. & Long, B. T. 2014 (January): Deglacial to postglacial palaeoenvironments of the Celtic Sea: lacustrine conditions versus a continuous marine sequence. Boreas, Vol. 43, pp. 149–174. 10.1111/bor.12028. ISSN 0300-9483.
Recent work on the last glaciation of the British Isles has led to an improved understanding of the nature and timing of the retreat of the British−Irish Ice Sheet (BIIS) from its southern maximum (Isles of Scilly), northwards into the Celtic and Irish seas. However, the nature of the deglacial environments across the Celtic Sea shelf, the extent of subaerial exposure and the existence (or otherwise) of a contiguous terrestrial linkage between Britain and Ireland following ice retreat remains ambiguous. Multiproxy research, based on analysis of 12 BGS vibrocores from the Celtic Deep Basin (CDB), seeks to address these issues. CDB cores exhibit a shell-rich upward fining sequence of Holocene marine sand above an erosional contact cut in laminated muds with infrequent lonestones. Molluscs, in situ Foraminifera and marine diatoms are absent from the basal muds, but rare damaged freshwater diatoms and foraminiferal linings occur. Dinoflagellate cysts and other non-pollen palynomorphs evidence diverse, environmentally incompatible floras with temperate, boreal and Arctic glaciomarine taxa co-occurring. Such multiproxy records can be interpreted as representing a retreating ice margin, with reworking of marine sediments into a lacustrine basin. Equally, the same record may be interpreted as recording similar conditions within a semi-enclosed marine embayment dominated by meltwater export and deposition of reworked microfossils. As assemblages from these cores contrast markedly with proven glaciomarine sequences from outside the CDB, a glaciolacustrine interpretation is favoured for the laminated sequence, truncated by a Late Weichselian transgressive sequence fining upwards into fully marine conditions. Reworked rare intertidal mol- luscs from immediately above the regional unconformity provide a minimum date c. 13.9 cal. ka BP for commencement of widespread marine erosion. Although suggestive of glaciolacustrine conditions, the exact nature and timing of laminated sediment deposition within the CDB, and the implications this has on (pen)insularity of Ireland following deglaciation, remain elusive.

Sunday 10 May 2015

Ancient huts and circles: Carn Goedog - Hafod Tydfil area

These are the hut circles, ring cairns and other enigmatic embanked enclosures that can be traced in the landscape around Carn Goedog and Hafod Tydfil on the northern flank of Mynydd Preseli.  This "group" of features is listed separately from the features clustered around Carn Alw, a kilometre or so to the east.  In reality there is no reason to think there are any cultural or tribal differences between the two areas -- this is a matter of convenience from the point of view of those doing the listing......
Thanks for Geo for reminding us about these listings. The full doc is easy to find on the web, and can be downloaded for free as a PDF.

PRN 11506
SUMMARY A group of hut circles situated at the base of the steep, north facing slope of Carn Goedog
LONG DESCRIPTION A group of hut circles situated at the base of the steep, north facing slope of Carn Goedog at 240m above sea level. There are at least 6 hut circles, in a linear spread roughly aligned E-W along the base of the slope. The hut circles are defined by sub circular stone and earth banks with hollowed interiors. The diameter of the hut circles varies from 3.0m to 5.5m and the banks have an average height of 0.5m. Two hut circles appear to be built on platforms terraced into the hillside. There are faint traces of linear earthen banks in close vicinity to the hut circle group but no clear pattern is discernable. Two of the hut circles are conjoined. FM & RR May 2009
Group of round huts and ancillary features on Preselis. RPS August 2001

PRN 11509
SUMMARY A roughly square enclosure situated on the northeast slopes of Mynydd Bach.
LONG DESCRIPTION A roughly square enclosure situated on the northeast slopes of Mynydd Bach at 250m above sea level. The enclosure measures c.25m square and is defined by a low earthen bank c.30m high and 1.0m wide. The bank is composed of large and medium stones protruding through the turf. The SW corner is oddly shaped with a 5m long indentation in the banks outline. Drewett's 1983 plan is a good representation of the enclosure on the ground. FM & RR May 2009
Rectilinear enclosure. RPS August 2001

PRN 9944
SUMMARY A large, well-preserved circular enclosure situated on the gentle north facing slopes of Carn Goedog. It lies within a complex of fields and other features (PRN 8403).
LONG DESCRIPTION A large, well-preserved circular enclosure situated on the gentle north facing slopes of Carn Goedog at 195m above sea level. It lies within a complex of fields and other features (PRN 8403).The site consists of a circular enclosure, c.35m in diameter, defined by an earth and stone bank c.4.0m wide and 0.5m high. The interior is somewhat sunken and is nearly covered with reed growth. In the SE quadrant of the interior is a spread of stone that appears to have some form - perhaps circular - possibly an indication of a former hut circle? Unfortunately the reed growth within the enclosure obscures any other evidence of internal features. A site visit durIng 2003 for the PFRS project concluded that this site was not a ring barrow as had previously been suggested and did not fall into any known category of prehistoric funerary and/or ritual monument. FM & RR May 2009

PRN 8403
SUMMARY A complex of features including rectilinear fields, hut circles and a well-preserved circular enclosure, situated on the gentle north facing slopes of Carn Goedog.
LONG DESCRIPTION A complex of features including rectilinear fields, hut circles and a well preserved circular enclosure (PRN 9944), situated on the gentle north facing slopes of Carn Goedog at 200m above sea level. It is difficult to assign a date to this group of features that show characteristics of the prehistoric through to the early medieval period. FM & RR May 2009

PRN 11511
SUMMARY A group of small circular earthworks spread along the banks of three closely spaced north flowing streams on the gentle north facing slopes below Carn Goedog.
LONG DESCRIPTION A group of numerous small circular earthworks spread along the banks of three closely spaced north flowing streams on the gentle north facing slopes below Carn Goedog at 170m above sea level. These small circular earthworks have an average diameter of 5.0m. The majority are low, sub circular, stony earthworks or mounds with hollowed interiors. The average height of any bank is 0.5m and width 0.7m. One enclosure at SN12193399 is very different in appearance- the drystone walls stand to c.1.0m high and are visible internally to several courses high. It is again c.5.0m in diameter with an entrance on the E, and there appears to be evidence of a small sub circular lobate enclosure on the S. The date and purpose of this group is unclear but must be related to their close proximity to the three streams. FM & RR June 2009 Group of between 15 and 20 sub-circular structures. RPS August 2001

PRN 13243
SUMMARY A roughly circular enclosure situated on the lower north facing slopes of Mynydd Bach.
LONG DESCRIPTION A roughly circular enclosure situated on the lower north facing slopes of Mynydd Bach at 240m above sea level. The enclosure measures c.18m in diameter and is defined by a spread, stony earthen bank c.2.0m wide. Internally there is evidence of possible hut platforms in the SW and NW quadrants. Possibly associated with the evidence for a field system in this area (PRN 13245). The enclosure is too large to be considered a hut circle. FM & RR June 2009
A roughly circular, grass covered bank circa 2m wide and 18m in diameter. It is much eroded and therefore not continuous. The interior is sunken and there is no ditch apparent, inside or out. There is evidence for some stone protruding, especially on the north west bank side. Despite being once recorded as a possible ring barrow, it is more likely that this site is some kind of enclosure or hut circle. RR 2004

PRN 13244
SUMMARY Originally identified from an aerial photograph this is a sub rectangular enclosure situated on the lower north facing slopes of Mynydd Bach. It is associated with long hut PRN 96867.
LONG DESCRIPTION Originally identified from an aerial photograph this is a sub rectangular enclosure situated on the lower north facing slopes of Mynydd Bach at 240m above sea level. In 2009 the site was discovered to be an enclosure measuring c.18m E-W by c.16m N-S, defined by the remnant of an earth and stone bank that is not continuous. The ground appears to have been built up on the north to form a level platform on which to construct the enclosure. There is no obvious entrance through the bank. Within the enclosure in the SW corner is evidence of small circular feature, diameter c.3.0m, visible as a number of medium stones protruding through the turf. Outside the enclosure on the east, a few metres away, is a rectangular 2-celled long hut (PRN 96867) and just outside of the enclosure on the south is a vague oval earthen mound. FM & RR June 2009

PRN 96867
SUMMARY A rectangular 2-celled long hut situated on the lower north facing slopes of Mynydd Bach. It lies close to enclosure PRN 13244.
LONG DESCRIPTION This rectangular 2-celled ‘long hut’ lies just to the east of enclosure PRN 13244 with which it is obviously associated. Both features are situated on the lower north facing slopes of Mynydd Bach at 240m above sea level. In 2009 the site was recorded as measuring approximately 8.0m E-W by 5.0m N-S. Dense reeds growing across the area heavily obscured the site but it appeared to be a 2-celled building of drystone construction. The walls are now visible as a low spread of stone protruding through the turf, but the internal partition wall was still clear. FM & RR June 2009

Saturday 9 May 2015

Devensian Till on Brynberian Moor

After my most recent trek across the trackless (well, almost) wastes of Brynberian Moor in the company of Chris Johnson, I'm more convinced than ever that there is no extensive spread of lake deposits.  So let's forget the idea of an extensive Lake Brynberian impounded against the face of Mynydd Preseli.  I have always been a little concerned about this theory, since if there had been a large lake there should be a deep spillway across the col somewhere near Tafarn-y-bwlch which carried water down towards the Gwaun Valley.  There isn't one, so although some maps from other geomorphologists have shown this lake with a surface altitude of c 285m, let's now forget it........

There do appear to be some thin lake clays, but they are at a much lower altitude, around 130m on the moor near the cottage of Glanyrafon Uchaf.  How deep was the water?  It's difficult to say -- but probably in excess of 10m?  And where was the ice edge at the time?  And where were the main inputs of debris-carrying meltwater?  And was the Brynberian river gorge -- running past the site of Craig Rhosyfelin -- the main meltwater escape route?  Questions, and more questions...... but my working hypothesis is that one or maybe several ephemeral lakes were present here on these lower slopes for maybe just a few decades or centuries during the period of ice wastage around 20,000 years ago.

As far as the moorland is concerned, there are surface exposures of clay-rich till all over the place.  The top photo shows such an exposure on the trackway leading across Waun Brwynant towards Hafod Tydfil.  The middle photo shows about 50 cms of churned-up clay-rich till overlain by about 20 cms of peat.  There is a widespread cover of peat, but I'm not sure that it is thick enough anywhere on this moorland to have been used in the past for peat-cutting "turbaries".  (In contrast, there are a number of sites high on the Presely ridge where the traces of past peat cutting can still be seen.)

The bottom photo also shows churned-up till about 60 cms thick on the moorland below Hafod Tydfil.  The churning suggests that there has been permafrost here, and that periglacial conditions existed at some time following the deposition of the till.  Although the best exposures are in stream cuttings, and are thus difficult to excavate and examine properly, they look to me like periglacial involutions and frost-heave features.  I do not think they are loading features such as we sometimes see in flow tills and other sediments that have behaved like very mobile fluids when additional loads have been dumped on top of them.  In true loading features, we would expect to see a layer (or several layers) of additional sediments on top of those displaying the structures -- and here there are no loads anywhere to be seen!  So let's live with the periglacial hypothesis for the time being.  It would make sense, since almost everywhere else where we find Devensian glacial and fluvio-glacial deposits in Pembrokeshire, we find within their uppermost layers signs of permafrost structures like fossil ice wedges and involutions, or else -- where there are substantial slopes and good supplies of slope materials -- a layer of "upper head."

So I'm going to stick with the idea of an extensive till plain here, with occasional patches where real morainic features project above the level of the wide stream basins.  There is more topography here than we might think when we look at the area from a distance, or look at satellite images.  Actual relief is more than 5m in many places -- and if we examine the mosaic of vegetation we can see areas of sphagnum, grasses and rushes interspersed with drier areas of bracken, with low rises or ridges of morainic material with boulders littered about and with vegetation dominated by gorse bushes.  There are some rock outcrops too.  And erratic boulders everywhere -- none of them very far-travelled, but all showing signs of wear and typical facets arising from breakage in transport.  Much more work -- and time -- is needed to sort out the glacial geomorphology of this vast wilderness which stretches for more than 6 km along the north face of the mountain.......

Friday 8 May 2015

Bedd yr Afanc again.....

Made another visit to Bedd yr Afanc in the company of Chris.  This is a very small megalithic monument, classified as a Neolithic passage or gallery grave of a rather unusual type. It's really rather insignificant in the landscape, and is difficult to find on a gently sloping surface of grassland and heath vegetation.  But here are the location details:

OS Ref (GB): SN108345 / Sheet: 145
Latitude: 51° 58' 34.38" N
Longitude: 4° 45' 18.18" W

For other posts relating to this monument, use the search box.

The stones only project above the surface for  40 cms or so, and they vary a lot in shape and dimensions -- and rock type too.  Some are made of dolerite, some are volcanic ashes, rhyolites and what appear to be gabbros.  I did not think it was a good idea to knock chunks off so as to examine the lithologies a bit more carefully.  The stones have almost certainly just been picked up in the immediate vicinity and built into the monument.  They are typical glacial erratics made of rocks from the Fishguard Volcanic Series -- and not one of them has travelled very far from its place of origin.  The facets and broken surfaces seem to be of several different ages, but none of the edges is particularly sharp.

Where are the capstones?  There are plenty of flattish stones embedded in the turf in the vicinity.  Some may actually be embedded in the turf in the "passage" itself.

As we can see in the photos, many of the stones lean inwards towards the axis of the monument.  The passage is not aligned towards anything significant.......

There are a number of boulders with rounded-off edges in the vicinity which seem to be made of a very similar bluish rhyolite as that which is exposed at Rhosyfelin.  The rock outcrops there are less than 2 km away, towards the NNE.   Without proper geological analysis, the origins of these boulders cannot be reliably guessed at -- and there are of course many other outcrops of bluish foliated rhyolite  in the Fishguard Volcanic Series to the N and NW.

One of the boulders made of bluish foliated rhyolite, close to Bedd yr Afanc