Geology - The Shape of the Landscape

Note added 29/04/2009: Linked Figures and Maps are not yet available for these pages.

The Senni valley has been formed by the gradual erosion of the rocks which buried the Senni Beds after the Devonian period. Some of this erosion follows weaknesses caused by fault lines – stand a little south of the triangulation point on Fan Frynych ●2 and you can look down Cwm Du and up the head of the Senni valley along the line of a fault which continues south-west into the head of the Swansea valley and north-east along the north side of the Brecon Beacons.

The most spectacular erosion took place during the Quaternary – about two million years of repeated ice-ages and warmer periods such as the one in which we now live (known to geologists as the Holocene interglacial period). There is no reason to think that the Pleistocene ice-ages are at an end and that the ice will not one day return to Senni (fig 1.1. Senni under the ice).

The most recent glaciation – the Devensian glaciation - ended about 10,000 years ago. It went out with a bang because, after the earth had been gradually warming up from the very coldest period (about 20,000 years ago), it again, briefly, got very cold indeed before finally warming again at the beginning of the Holocene.

There are several kinds of evidence for this glaciation written into the Senni landscape:

1                    The shape of the valley

The Senni valley has a broad floor and relatively steep sides – a classic U-shaped glacial valley. You can see valleys like this, with the glacier still in them, in many colder parts of the world today – such as the Alps and Greenland.

2                    The crags of Cwm Du, Fan Nedd and the Senni Quarry

Cwm Du and Fan Nedd have cliffs formed by glacial erosion, with great spreads of ice-plucked debris at the base (fig 1.2. the cliffs of Cwm Du), while the bowl beneath the Senni Quarry (fig 1.3. The cirque of Senni quarry) is the remains of a small glacial cirque which probably formed during the very cold period at the end of the last ice age, about 11,000-10,000 years ago. Other such bowls around the Beacons contain deposits no earlier than 10,000 years or so whereas deposits as old as 12,000 years are found on lower ground nearby (such as in the Traeth Mawr bog near the Mountain Centre on Mynydd Illtud). We think this is because the deposits in these upland glacial cirques were removed by the small glaciers which formed in them during this late, very cold glacial period while those on lower ground survived: see a photo of a double moraine below Fan Bwlch Chwyth.

3                    Till on the valley sides and the till ridge through Heol Senni

Most of the Senni valley sides are covered in a mixture of sand, silt and clay containing cobbles and boulders plucked from the local rock. This mixed material is called a Boulder Clay or a Till (fig 1.4. a till exposure) and is made up of the remains of rocks, eroded by the glacier which filled the Senni valley as it ground over its bed flowing downhill towards the Usk valley. The hamlet of Heol Senni, from near Maes Walter to the bridge, is built on a ridge which may well be formed from this Till ●3 and could represent a terminal moraine – a point in the valley where the glacier stopped and where it deposited a larger quantity of the rock debris it was carrying.

It would be interesting to know if this is really the case – something we could best find out by coring down through the ridge to a depth of 10 metres or more using a mechanical corer and, perhaps, carrying out some geophysical studies using instruments which can, in effect, look through the ground to see what it is made of. If it is a till ridge then we can imagine a time, during the last ice-age, when the Senni valley floor, upstream of the bridge, was filled with one large glacier but when the hills around and the area downstream – the lands of Pant-y-Ffordd, Bailea and all the way down to Sennybridge, was bare glacial debris with, perhaps, a few hardy lichens and mosses surviving in the bitterly cold climate (fig 1.5. A modern glacier (Cook) and pro-glacial lake)

4                    Lake deposits in Cwm Du 

A little downstream of the bridge which carries the Sarn Helen track over the Cwm Du stream is a small cliff being undercut by the stream ●4. Towards the top of this cliff you can find thin, alternating layers of silt and clay which look very much like lake sediments laid down in alternating summer and winter conditions (fig 1.6. ).  

But when could there have been a lake here? The cliff stands at the mouth of the Cwm where the river starts to descend rapidly into the valley – so wouldn’t the water have just drained away? 

Not if it was being blocked by a glacier or a moraine. The Cwm Du sediments may well have been laid down in the bed of a small lake which built up here because the water which flowed down Cwm Du was blocked by the moraine left by the small glacier which filled the Cwm at the end of the last ice age. There are the remains of just such a moraine just downhill and this might well have once continued right across the valley.  

5                    Land-slips and Soliflucted deposits 

The valley below Craig Cerrig-gleisiad ●5 is famous, amongst British geologists at least, because it contains the remains of what may have been a vast land-slip which probably took away much of the western cliffs above the Cwm.  

Looking down from the cliffs to the west you can see the great tongue of mixed-up rock in the centre of the valley, dividing the two steams that flow out of it towards the Afon Tarell. Close by where you are standing on the cliff top are a sequence of steps - the tops of slabs of rock which did not fall away entirely – and depressions running parallel to the cliff-edge – the cracks at the back of each slab (fig 1.7. ). Once you realise what you are standing on it all feels a bit more precarious! But don’t worry, it hasn’t moved for a long time. 

This vast slip probably happened towards the end of the last glaciation when the melting ice within the rock and soil lubricated the movement of the ground and caused such slips to take place.  

The Senni Beds are susceptible to this kind of ground slippage because they contain weak beds through which water tends to drain and, once you know what to look for, you can find the scars of many more such slips around the area. 

There is a particularly spectacular slip on the south-east slopes of Fan Dringarth, above the Ystradfellte reservoir ●6 where the upper slope has fallen away leaving a great series of slipped-rock ridges on the upper slope and a mass of debris, like a rucked-up carpet, down in the valley below. This too may have happened during the latter part of the last glaciation because the water provided by the melting ice meant that it was a period of particular slope instability.  

Senni still has some unstable slopes, such as that crossed by the road on the zig-zags climbing over to Ystradfellte. Look at the lower slope of the hillside below the first bend going up the zig-zags ●7 and you can see a change both in the angle of the slope and in the vegetation – dry-soil grasses above, wet-loving rushes below. This too could be the remains of a large land slip which took away the lower slopes of Fan Nedd (a very long time ago).  

There is evidence of less spectacular but more widespread slope-movement all around the valley. Stand at the cattle-grid by the forestry on the Ysradfellte road ●9  and look up the Llia valley. The Llia stream is confined on a gravel bed in a trench several metres deep, between a series of bluffs (fig 1.8. ). You can see from a few bits of exposed soil, where the stream has cut into the bluffs, that they are made of glacial till. 

We might guess that the trench has formed because the stream has cut down through the till but it probably has quite a different explanation. The bluffs lie at the bottom of slopes covered by glacial till and this has slid gently down into the valley, thus confining the stream.  

This kind of slow slope movement happened over Britain in the latter part of the last glaciation and was caused by the repeated freezing and then thawing of the ground which caused it to partly liquefy and thus to flow downhill. This process is called solifluction and played a large part in shaping the British landscape – but it hasn’t taken place in Senni for at least 10000 years.

Lake Senni?

Another reason why we might think that the ridge through Heol Senni is a terminal moraine is found nearby. At the bridge over the Senni ●10, just downhill from Ty Isaf, the river flows over the bedrock and then drops a little more steeply before reaching more level meadows to the north of Felin Senni (there are nice cut-off meanders here too which show how the Senni river has moved across the valley and left its former branches cut off to form curved depressions).

Upstream of the bridge the river meanders a little across a slightly broader valley bottom where it is gradually reworking its banks, cutting into the outside of each bend and depositing new shingle on the opposite bank.

It is possible that the bedrock, just upstream of the bridge, was carved out to a slightly greater depth by the glacier which may have ended at the moraine and that the bridge lies roughly at the end of what was once a small lake which filled the depression left when the ice melted. The lake would have filled rapidly with sediments and is likely to have disappeared entirely in the first part of the current interglacial period – perhaps by 8000 years ago. But it may be that lake sediments here will contain evidence of the earliest plants to colonise the valley once the ice had melted and the land had warmed – another target for coring and geophysical survey to investigate.


There are a number of small bogs and depressions around the Senni valley ●11 and ●12 which contain peat with pollen and the preserved remains of plants many thousands of years old – perhaps reaching back more than 10000 years to the last glaciation.

One of these bogs ●12 is particularly interesting because it is close to the small prehistoric field system ●13 (it could be a very small, primitive farm or just an ancient stock enclosure) at the head of the Llia valley – as well as to Maen Llia and a number of other archaeological remains of about the same period.

It is therefore possible that the bog may preserve pollen from the plants which might have been grown as crops by our prehistoric ancestors who lived at the Llia valley site. We would not think of growing crops here now but it was easier for them because the climate was warmer and the upland soils were often more fertile than now.

This bog, and others around the valley are yet more places to survey and core.


The hillsides around the Senni valley are often slightly stepped (fig 1.1. ). These are “structural benches” – areas of gentler and steeper hillside which reflect the harder and softer beds of rock beneath. Where the rock is softer it has been eroded into gentler slopes, where harder the slope steepens. A ridge of harder rock also forms a beautiful waterfall at Gorslŵyd farm.

The harder beds are those laid down on the floodplain to each side of the Devonian rivers – finer deposits which have become better cemented. The softer beds are those laid down in the river channel itself. Thus the steps in our hillsides describe the gradual meandering of the ancient river across its desert floodplain down to the sea, a little to the south.

The Soils

Take the road from Senni over to Crai and walk up the track, closed by a heavy gate, which heads south over the Cnewr estate towards Fan Gyhirych ●14 (with thanks to the generous access policy of our neighbours at Cnewr). Look closely at the ditches on both sides of the track as you walk up the hill.

For the first 500 yards the soils exposed in the ditch sections are a dull, slightly bluish-grey and the vegetation contains quite a lot of plants that like plenty of water, such as the common rush Juncus (fig 1.1. ). When the first track leads off to the left and into the forestry, however, the soils change abruptly ●15 (fig 1.2. ) and the vegetation changes to acid-tolerant grasses such as the moorland mat grass nardus stricta. Soon after the forest gate you can see an excellent section, exposed by the ditch to the south of the track, where the soil is divided into three bands – dark at the top, light in the middle and brown below. Continue along the track and, after a small quarry, and a couple of bends, a rather larger bank of soil and rock has been exposed ●16 where it has been cut into by the track. Here you can see the pattern of black, light and brown overlain by repeated bands of light and dark (fig 1.3. ).

Now head back towards Senni and look at the soils exposed in the road ditch just east of Pen-waen-dwr ●17. This soil is largely uniform red-brown right from the till at the bottom to the root-mat at the top (fig 1.4. ).

Finally, have a look at any soil you see exposed in the valley itself – for example in the sections cut by the river by the Heol Senni bridge ●18 (you can climb over the stile here on the right-of-way across the field belonging to the Nuadd farm but please keep to the river bank). These soils too are largely a uniform reddish-brown.

What does it all mean?

Soil forms over hundreds and thousands of years. Rain falls on the bare rock, till or river gravel (or whatever the parent material is at a particular place), plants grow and the finer particles and soluble minerals near the ground surface get washed downwards. Some end up in rivers and others are left behind in the lower parts of the soil. Meanwhile the remains of the plants growing – and dying – at the surface become mixed in at the top by the abundant creatures, such as earthworms, which live in the soil. Thus the top 10cm or so become more organic and we end up with a sequence of horizons from the organic surface A horizon, through the B horizon where material has accumulated from above and finally into the weathered surface of the parent material – the C horizon.

Down in the Senni valley the till and river sediments have been gradually changing in this way since they were first laid down. For the glacial till this was under the ice more than 12000 years ago while the river sediments are being actively redeposited now and may be no more than a few years or centuries old. Thus the valley soils show this simple progression from A to C horizon, although the strong red colour of the rock (and thus the parent materials formed from it) makes it hard to see the organic matter accumulated near the surface or the fine material redeposited in the B horizon, which are much clearer in soils elsewhere.

On the hills, however, the process has been a bit more complex and the result a little different. Site ●14 lies on soliflucted till, which contains few pores and through which rainwater flows only very slowly. Thus there has been less movement of fine matter downwards and the B horizon is less well-defined. The wetness of the soil, and the lack of pores, has also meant that it contains very little air and the few microscopic plants and animals which can live in such harsh conditions have removed such oxygen as was originally present and thus changed the minerals in the soil. The most visible effect of this has been to change the red iron oxide (the oxide of a form called Iron III) into a blue-grey form (Iron II). At the surface the organic matter, from dead plants and animals, have not decayed as much as they would in a dry soil and the result is a slightly darker, peaty A horizon.

Further up the hill at ●15, however, we cross from the till, which slid down the hillside by solifluction during the last glaciation, to the exposed rock it left behind. Here the ground is much better – though not perfectly – drained. The sequence of dark, light and brown bands are the result of two competing processes, waterlogging-weathering at the bottom and downward acid-solution above.

Look at the till exposed at ●16 and in sections further along the track nearby. You can see the remains of stones which have been so weathered that they are now soft enough to cut with a penknife. This weathering has happened because the water held in the soil has been gradually dissolving these stones for more than 10,000 years – a process accelerated here by the lack of oxygen in the soil when it is at its wettest. Here, and at ●14, you can see brown patches and mottles where a little oxygen, in water descending root pores, has allowed the iron to change back to its original Iron III form.

The black surface horizon at ●15 is an accumulation of organic matter, mixed into the surface soil minerals and preserved not so much by the wetness of the soil but by acidity. The living and decaying plants release weak acids, as does the rain, and this makes the surface soil inhospitable to earthworms and other plants and animals which would otherwise break the organic matter down. The lighter band below has formed because these same acids have dissolved the minerals which colour the soil – iron especially – and left behind only uncoloured, resistant minerals like quartz.

This kind of soil is called a Podzol (which means Ash Soil in Russian, since the bleached, light horizon looks a bit like wood-ash), the wet soil at ●14 (and in the wetter bits of the Senni valley bottom) is a Gley and the normal soils of the valley are called Brown Earths.

The multiple bands of colour seen at ●16 are the result of something rather different. Look at the hillside immediately above and you can see that the slope here has an unusual, stepped, disturbed appearance. These disturbances are the remains of drove-ways. The bands in the soil profile below are made up of soil which has been eroded from the slope above when the drove-ways became cut-up by the hooves of the animals. You can see that this happened in episodes of erosion and stability by the bands of eroded mineral soil and the more organic topsoil which has formed on top of them.

Returning to the valley itself, two recent bits of building work have exposed the soil profile and allowed us to compare soils formed in different materials. The construction of a new house close to the village hall ●19 created a 2 metre deep section through the soil. We found largely unweathered till at 1m from the surface – and we can tell it was unweathered because the stones within it had small cappings of fine, compacted silt which formed when the soil was frozen, perhaps 12-18,000 years ago, and which have been preserved ever since.

The section at Gorslwyd farm ●20 nearby looks similar but exposed the broken bedrock, rather than till, showing that the Senni glacier of the last glaciation had been grinding away at the rock but had not deposited much till here. The lack of bedrock at a depth of 2m at ●16 and nearby, but not at Gorslwyd, is further evidence that the village is built on a thicker mass of till, which might be a moraine.