Witches Cauldron exposures confirm "two glaciations" scenario

Oblique and vertical photos of the coastal footpath in the vicinity of the Witches Cauldron.  The  best exposures are in the gully downstream of the footbridge.

I've been back to look at the Witches Cauldron, to the west of Ceibwr on the north Pembrokeshire coast.  It's turning out to be a rather important site -- not very easy to examine, except at low tide, and even then, if there is a sea running, you are likely to get soaked by  plumes of spray shooting out of gullies and holes in the wave-cut platform.  The complex network of tunnels, blowholes, caves and stream channels is quite something.......

This is the Quaternary sequence just along the coast at Ceibwr:

Modern soil -- c 20 cms. Uncemented
Sandloess and colluvium -- c 50 cms. Uncemented. Holocene?
Clay-rich Irish Sea Till -- up to 2 m thick. Uncemented. Devensian?
Brecciated slope deposits -- up to 50 cms thick. Uncemented
Clay-rich colluvium -- c 20 cms thick. Uncemented but stained / gleyed. Ipswichian interglacial?
Glaciofluvial gravels -- c 1.5m thick. Stained and cemented. Anglian?
Stony till -- up to 1m thick. Stained and cemented. Anglian?
Brecciated slope deposits -- up to 1 m thick

At the Witches Cauldron there are partial exposures all over the place, but the most complete and interesting sequence of deposits can be seen in the channel downstream of the footbridge, where the stream meets the sea.  There are several footpaths giving access to the foreshore and the spectacular overhangs -- but care is needed when the rocks are wet and slippery.  If you are not careful, you might just slip down a rather deep hole and never be seen again........  risky business, geomorphology........

The stream plunging into a tunnel over a bedrock ledge.  To the left and right there are exposures of cemented slope breccia, till and glaciofluvial gravels -- all stained with iron oxide and manganese oxide.

Projecting mass of cemented and stained Quaternary deposits.  At high tide wave action is eroding away the shale bedrock beneath these sediments, which are more resistant to erosion -- thus creating the spectacular overhang.

Two exposures of stained and cemented till -- so solid that it is difficult to examine.  Erratics of many different shapes and sizes, some rounded and sub-rounded, in a sandy and gravelly matrix.  

The till complex here is up to 2m thick, and it is clearly seen to overlie c 50 cm of slope breccia full of angular bedrock fragments.  Some of these blocks are up to 30 cm long -- with very irregular (frost-shattered?) shapes.  The till also incorporates horizons of bedded sands and gravels -- up to 50 cm thick.  This is important, indicating oscillations in the depositional environment and suggesting that the till is probably not a lodgment till.  Much slope breccia has been incorporated into the lower layer of the till.  On the stream bed there are many large stained boulders that have been released from the till.

In this exposure we see cemented slope breccia at the base, resting on shale bedrock, then a layer (c 40 cm thick) of cemented till, and then above the level of the notebook stratified cemented gravels.

Upstream of the footbridge we can see, in the river banks, various exposures of the cemented till, and there are abundant "conglomerate boulders" -- some over 1m across -- made from the cemented till and glaciofluvial deposits which clearly exist within the steep-sided valley.   There are also very many other rounded and sub-rounded erratic boulders.  These deposits appear to be extensive and thick. 

In many locations adjacent to the footpaths we also see much more recent uncemented Quaternary deposits.  Here are there disturbed slope breccias (usually less than 50 cm thick) are seen beneath fresh till which is up to 2m thick.  The till is very complex, and in many places it is seen to incorporate lumps and patches of cemented (ancient) till.  This is, after all, what we might expect if Irish Sea Ice has come in from the coast and has overridden pre-existing deposits.  There are no fresh glaciofluvial deposits above the fresh till, but here and there we see pseudo-stratified slope deposits around 30 cm thick.  Most of the fragments are small -- up to 10 cm diameter -- and they are made of local soft shales and mudstones which have simply moved downslope.  Above that, we see a thin layer (c 20 cm) of colluvium, sandy loam and modern soil.  

Exposure adjacent to the footpath at the top of the slope above the "cauldron".  This is a sandy till, but in the lower part of the exposure (around the notebook) there is a high percentage of broken bedrock fragments derived from overridden slope breccia horizons. 

Modern soil -- Uncemented
Sandloess and colluvium -- Uncemented. Holocene
Thin brecciated slope deposits ("Upper head")
Sandy and gravelly till -- up to 2 m thick. Uncemented, but inclusions of old cemented deposits. Late Devensian
Brecciated slope deposits -- up to 50 cms thick. Uncemented.  Early Devensian?

Hiatus?  No visible raised beach or other interglacial deposits

Stony till -- up to 50 cm thick. Stained and cemented. Anglian? Glaciofluvial gravels -- c 1.5m thick. Stained and cemented stratified sands and gravels. Anglian?
Stony till -- up to 50 cm thick. Stained and cemented. Anglian?
Brecciated slope deposits -- up to 50 cm thick.  Cemented.  Anglian? 
Black and grey mudstones and shales -- bedrock

I am now convinced that two glacial cycles are represented here at the Witches Cauldron.  Work is ongoing -- watch this space.......