Patterns on the Irish Sea Floor


Ask a geologist to do a survey of the sea bed and they will find geology. Any mark or feature in the surface sediments will be interpreted in terms of the geomorphology that they expect to find. An archaeologist looking at the same buried sub-sea features might see something different, as with crop marks on land. But why would an archaeologist wish to investigate the sea bed other than to seek, perhaps, an ancient shipwreck?


The possibility of ancient human settlement on what is now the floor of the Irish Sea – or indeed anywhere around the British or Irish continental shelf – is raised by the stories in Celtic mythology, which variously suggest lost cities around the coast: Lyonesse off Land’s End, a submerged ‘Celtic Otherworld’, or a lost land called Annwn; and other myths from Mediterranean sources also pose questions. Welsh myths describe a plain “as large as the sea” off the coast of North Wales and Irish myths would portray a submerged “flowery plain” in the vicinity of the Isle of Man. Irish and Welsh legends speak of a golden tower in the sea explored by the earliest colonists. [1] To approach these myths logically either they are just that: ‘myths’ in the sense of ‘fiction’ or they must be based on an authentic root memory. If indeed they recall an ancient submergence then the next logical step is to ask, when?


We need not debate here the various mechanisms of sea-level change. On all conventional theories it must be remembered that the gradual eustatic rise of the sea since the ice age extended over thousands of years; a period of time longer than that which separates us from Dynastic Egypt. It is a failure of imagination to suggest that during these millennia there was no human settlement on the now-submerged regions of the continental shelf. The consensus is that present-day sea-level only stabilized around 5,000 years ago since which time there have been just minor fluctuations. In other parts of the world we find the opposite phenomenon at this same era: emergence and raised beaches rather than submerged features, together with evidence that people quickly colonised the newly exposed land (see: Raised Beaches and Submerged Forests).


For Britain and Ireland, submerged forest deposits are found all around the modern coasts (see: Submerged Forests around Britain and Ireland). These come from various ages, but most date from either of two periods; the first roughly 5-6000 BC around the time that the North Sea land bridge was flooded, while those around western coasts cluster around a later period closer to 3000 BC. So if we should seek any submerged archaeology then, logically, it should date from these eras. The earlier dates correspond to the Mesolithic period of supposed hunter-gatherers; the later dates correspond to the transition between the Middle and the Late Neolithic. Again, logically, any memories of real events in a legend would suggest that the former date is too early; and so we should be looking around 5,000 years ago to find underwater archaeology to match the legends.


We should expect that any submerged Middle Neolithic sites would look like those that we find on land from this era; such as the Court Cairns found all round the Irish Sea or the dolmens found further south; or perhaps settlements like those at Skara Brae; or field-walls such as those beneath Irish peat bogs. Cornish legends about the loss of a land called Lyonesse speak of some 140 ‘churches’ drowned in a sudden rise of the sea. [2] We would have to consider these as pagan religious sites and parishes. What might such submerged cairns and settlements look like now, ruined and buried beneath metres of silt on the sea floor? Imagination is required. And this is before we even consider looking for ‘lost cities’.


Older studies of sea level change around the British coast would speak freely of a ‘submerged forest period’ in the mid-Holocene (see: The Irish Sea Coast). The theory, before glacio-eustatic modelling of world sea-levels forced it out of favour, was that there was an initial period of eustatic sea level rise (i.e. melting of polar ice) which gradually drowned the coasts up to c.5500 BC. After this rise a regression exposed western coasts again, before a return of the sea to modern shores created the second wave of submerged forests around 3000 BC. Disagreement then arises as to how far out the sea may have receded during the warm mid-Holocene climate. Recent specialist studies would prefer a conservative ‘somewhere west’ of the present shores (as they have no data from the sea-floor and decline to speculate) or that the submerged forests were created by trapped freshwater lakes near the coasts.


The commonsense approach of archaeologists would be to seek submerged coastal sites such as we find at Westward Ho, Devon and at Skara Brae. However for the Irish Sea the legends of submerged plains, lost lands and towers are clearly pointing us further away from the shoreline than any archaeologist would consider; into depths that require expensive expeditions that so-far have been the preserve of geologists seeking oil and gas reserves or shallow sites for wind-farms.


Much of the geological mapping that we have for the central Irish Sea is acknowledged to be the work of just one man: the late Dr Robin Wingfield of the British Geological Survey (BGS) who, in the 1980s, identified ice-wedge polygons and circular features in sonar reflections taken north of Anglesey. [3] In the Strategic Environment Assessment (SEA) coastal resources survey of 2004 the Irish Sea region is defined as SEA6 and various reports are available via the BGS Geoindex database. Another survey investigated suitable sites for wind farms. This concentrated on an area closer to the Lancashire coast where the Morecambe Bay gas field and wind farms are situated and it did not encounter the patterned ground found in the older surveys; the sea-bed is instead described as mostly flat and featureless. The authors noted that the older sonar techniques were largely obsolete and that geophysical data was completely absent for large areas. [4] These results and subsequent studies are now embodied in the BGS Geoindex and in the updated report by Mellet-et-al of 2015, which summarises the state of knowledge as at that date.


The Geoindex shows the Holocene mud and sand deposits to vary between 5m and 40m thickness, with the deepest deposits lying off the Cumbrian coast. [5] Shipwrecks and modern pipelines are clearly identifiable from trailing sand patterns formed by the prevailing currents. Also noted are sub-surface features and ‘pock-marks’; once those defined as escaping methane (marsh gas) are excluded there still remain some interesting sites to explore. Of particular interest is a feature lying off Morecambe Bay described as a pingo (on land we might call these kettle-holes). This reveals a rounded-rectangular feature, which would be a football-field-sized; and were it on land it would be about the same size as a typical Neolithic henge. Some of the pock-marks nearby are described as unknown and ‘possibly man-made’. [6] This location would at some prehistoric era have been a river valley flowing into a lake that is now the Lune Deep. Another potential pingo is noted to the north of the Isle of Man.


To summarise the surface geology of the eastern Irish Sea as it is described would be as follows. The rectangular shelf slopes gently away from the English coastal dunes out to a depth of 50m between Anglesey and the Isle of Man. Close to the coast lies the Eastern Irish Sea Mud Belt overlaying the Central Irish Sea Gravel Belt that becomes exposed further west; this is mostly comprised of diamicton (a glacial till of mud and boulders). The smooth surface gives way to gravelly rough ground as the sea bed becomes deeper between Mann and Anglesey, where areas of bedrock are exposed. These rocks are a continuation of the onshore geology of Anglesey. Here the sea-bed has just a thin covering of Holocene mud and sand and it is in this region that the glacially patterned ground and polygons are described. This description may be investigated on the maps in the BGS database and its various options as linked here; together with the interpretive reports that have been published by the geologists. A summary map of the areas of ‘patterned ground’ and the other unexplained features is offered below as an introduction, since the BGS geological overlays give little indication of depth. Sea bed contours can be investigated via UK Admiralty chart 1826, or via the Admiralty online database.

Irish Sea Patterns.jpg


The public domain database of the British Geological Survey (BGS) can be found at: and click on ‘view the offshore Geoindex’.


The Admiralty wrecks and pipelines database plus a contour map (password required) is at:


Geologists would deduce all of the patterned-ground to be glacial scour features similar to those studied on land. In his 1987 paper, Wingfield had interpreted the polygons and near-circular features north-west of Anglesey as collapsed pingos, which typically occur in formerly-glaciated areas. The small undulations on the sea-bed are interpreted as drumlins and flutes. It is important to note that there has been little examination of these features other than by sonar reflection and whatever they are formed-of lies buried beneath the sand and gravel. Another discovery from the DTI surveys is that the only place where hard bed-rock penetrates the Holocene deposits lies to the north-west of Anglesey and it is interesting that the two highest peaks on the sea-bed at 32m and 35m depth correspond to areas where the bedrock is exposed. [7] The Holocene sand and mud varies in depth as shaped by the prevailing currents and it is likely that much has been scoured away in this region.


In 2002 I was given permission to reproduce some of the sonar interpretations in Atlantis of the West. I do recall in the 1990s speaking by telephone to a representative at the BGS (who may have been the late Dr Wingfield) who told me there was no other evidence then available. Later, in 2002, I spoke to his colleague Ceiri James who informed me that Dr Wingfield had died; and that he alone would have known on what evidence he described the patterned ground (as published in 1995 and cited in the later SEA6 reports). Recent enquiries of BGS could not identify from precisely which surveys the published evidence of ‘polygons’, in areas other than those near Anglesey, had actually originated. The public database shows numerous sonar tracks crossing these areas.


However, it is not really the relic features from the ice age that should interest us; rather we should ask: what lies hidden beneath the sand and mud layers? Another interesting feature from the DTI Environmental Assessment was a straight north-south ‘wall’ about 3m (10-feet) high where the mud has piled-up against it. [8] This lies close to the mapped boundary between the mud and diamicton. The survey describes the straight feature as unidentified and probably man-made – as it was not at that time shown on the Admiralty databases of wrecks or pipelines. However, it does correspond to where the sonar track crosses the modern pipeline between the Morecambe Bay gas platform and its outlying northern well! This shows us how quickly the modern currents can pile-up sediment over even a recent man-made structure; so who knows how much archaeology might lie buried deeper beneath the Holocene mud and gravel.


In this analysis I would seek only to open minds to what is possible in order to reconcile the incompatible cross-disciplinary picture that arises when the same evidence is approached from contradictory specialist viewpoints: geology, archaeology and mythology, each offering us a different perspective. Archaeologists are only now waking-up to the likelihood of Mesolithic human activity on the submerged ‘Doggerland’ of the North Sea (Sunday Times 8-Sept-2019), so perhaps the Irish Sea is worthy of equal consideration:


I was reminded by all this of the story recorded by the Byzantine historian Procopius of Caesarea. In the fifth century AD he recorded the legend of a long wall built by the people of ancient Brittia (northern Britain) to separate the land of the living from the land of the dead to the west. [9] To this same place, we are told, the souls of the dead were ferried! Were a man or any wild creature to venture beyond the wall they would surely be struck down dead within half an hour by the foul air. This would be quite a fair description of dense methane gas escaping from muddy pools. After all swamp gas, we are told, can also catch fire and create UFOs to confuse the credulous observer; so perhaps we should not be taken aback by a little bit of potentially explosive archaeology on the sea bed!





1) For a discussion of all these legends see Towers of Atlantis. The legends of a submerged tower in the Irish Sea are found in Nennius and in the Irish Book of Invasions.

2) As described by William of Worcester (fifteenth century) in his Itinerary of Cornwall.

3) See Wingfield 1987 and later citations.

4) See page 52 of the Kenyon & Cooper Sandwindfarms report.

5) Various overlay-maps are publicly available online via the BGS Geoindex offshore website, of which many are reproduced in the Mellet et al report of 2015.

6) See figure 16 on page 33 of the Holmes and Tappin 2005 report.

7) See the summary map ‘static bedforms’ as Figure 13 on page 28 of Holmes and Tappin 2005, which lists the sources from which it was compiled.

8) See Figure 17.5 on page 34 of the Holmes and Tappin 2005 report.

9) Procopius, History of the Wars, VIII, xx



References and Sources:


Wingfield, R T R. (1987) Giant sand waves and relict periglacial features on the sea bed west of Anglesey. Proceedings of the Geologists’ Association 98, 400–404.


James, J.W.C. and Wingfield, R.T.R. (1987) Aspects of the sea bed sediments in the southern Irish Sea. Proceedings of the Geologists’ Association 98, 404-406.


D.I., Jackson & R.T., Wingfield & D, Evans & R.P., Barnes & M, Arthur & M, Howells & R, Hughes & Petterson, Michael. (1995) The Geology of the Irish Sea, London: HMSO for the British Geological Survey.


2004 Strategic Environmental Assessment SEA6 SV Lia SEA6 Survey - seabed sampling survey with video and photography (Irish Sea)


Kenyon, N. and Cooper, W. (2005) Sand banks, sand transport and offshore wind farms. 10.13140/RG.2.1.1593.4807.       Download available via researchgate


Holmes, R. and Tappin, D. R. (2005) DTI Strategic Environmental Assessment Area 6, Irish Sea, seabed and surficial geology and processes, British Geological Survey Commissioned Report, CR/05/057.


Mellett, C. Long, D. Carter, G. Chiverell, R. and Van Landeghem, K. (2015) Geology of the seabed and shallow subsurface: The Irish Sea. British Geological Survey Commissioned Report, CR/15/057.



[Paul Dunbavin – September 2019]