Hull Geological Society

 Late Devensian glacigenic sediments exposed in coastal cliffs at Barmston (Holderness), Danes’ Dyke and South Landing (Flamborough Head) East Yorkshire: a field guide, by the Flamborough Quaternary Research Group of the Hull Geological Society: Colin Clark, Rodger Connell, Derek Gobbett, Dennis Haughey, Ian Heppenstall, Mike Horne, Stuart Jones, Brian Kneller, Chris Leach, Paul Richards and Rod Towse.

For the Sedimentology: Process and Product Conference at the University of Hull on Sunday 7^th October 2012.

Copyright Hull Geological Society 2012 and 2026.

1. Introduction.

The sites to be visited can be found on the following maps –

OS Explorer 1:25 000 sheet 301, Scarborough, Bridlington and Flamborough Head.

OS Explorer 1:25 000 sheet 295, Bridlington, Driffield and Hornsea

BGS 1985 1:50 000 series, England and Wales sheets 55 and 65, Flamborough and Bridlington, solid and drift provisional edition.

The coastal cliff exposures of Holderness and Flamborough Head, East Yorkshire, provide extensive exposures of variable quality, in glacial sediments deposited during the Dimlington Stadial of the Late Devensian, between approximately 22 000 and 15 000 calendar years before present. During this interval the North Sea Lobe of the last British-Irish Ice Sheet advanced down what is now the western North Sea, as far as north Norfolk, depositing a thick and varied succession of glacigenic sediments. See Catt (2007) for a regional review of the glaciation in the district and Bateman et al. (2011) for a re-evaluation of the chronology of the type site of the stadial at Dimlington, southern Holderness.

At Barmston, northern Holderness, we are able to investigate the sedimentology and structural geology of a range of glacigenic facies: diamictons of the Skipsea Till and laminated muds, sands and gravels deposited during the advance and retreat episodes of the North Sea Lobe ice sheet towards the close of the period. The site was studied by Evans and Thomson (2010), as part of a wider study in Holderness.

The sites at Flamborough Head, Danes’ Dyke and South Landing, provide evidence of palaeoenvironments before and during the advance of the North Sea Lobe, preserved within valleys cut into the Chalk bedrock of the headland. Ongoing research by the Flamborough Quaternary Research Group of the Hull Geological Society has recognised the potential of these sites, apparently forgotten since the late 19^th century (Lamplugh 1891), to provide significant new dating control on the initial advance of the North Sea Lobe into eastern England. Work at the sites is being prepared for publication (Heppenstall at al. 2010).

2. Barmston. Coastal cliff sections between TA172579 and TA 169600.

The coastal cliffs at Barmston (figure 3) are low and undulating, reaching a maximum height of 10m OD. Due to their ‘soft’ nature erosion rates can be rapid and average some 2m per year. The most recent published research on the glacial sediments here, and more widely in Holderness, is that of Evans and Thomson (2010). They have proposed that the sediments represent a series of minor readvances of the North Sea Lobe, each of which tectonised pre-existing deposits to produce a subglacial glacitectonite facies together with subaqueous outwash sediments, forming ridges at the ice margin. They identified three major litho-facies associations (LFA1, LFA2 and LFA3) within these depositional systems and these are usually visible in the Barmston Cliffs.

Figure 3. Air photograph of much of the Barmston section. Approximately 2km in length. Note the increased erosion south of Barmston Road End and Barmston Drain and locally increased erosion where gravel and sand bodies, loaded into Skipsea Till, have been preferentially washed out of the cliff. [We thank Brian Kneller of the FQRG for obtaining this photograph and Neil McLachlan of the East Riding of Yorkshire Council for giving permission for its use.]

Skipsea Till diamicton makes up the lower parts of nearly all the cliff exposures with from about 1m to 6m visible. It is typically a matrix supported massive diamicton of dark greyish brown colour (Catt 2007) though subtle colour variations are seen. Stratified and laminated facies are also recorded and both ductile and brittle deformation structures can be seen, indicating stress from the north or northeast mirroring the regional clast fabrics. Attenuated lenses and rafts of grey and reddish brown silty/sandy clay can also be seen in the section possibly derived from Mesozoic bedrock lithologies to the north.

Of particular interest is the undulating upper surface of Skipsea Till (LFA1). Whilst reaching over 6m above beach level in some locations it is also seen just above the beach in others (figures 7 and 16). This feature has been interpreted as either an irregular surface due to meltout and slumping as a surge lobe decayed together with possible squeeze ridges (Eyles et al 1994, Evans et al. 1995) or possibly as subtle proglacial deformation produced by re-advance, possibly that of the Withernsea Till ice (Evans & Thomson 2010).

Though not mentioned in any detail by Evans and Thomson (2010) Skipsea Till (LFA1) displays a very conspicuous sub-vertical joint system along the Barmston Cliff section, and indeed at other sites on the Holderness Coast. Penny and Catt (1967) first recorded these features and interpreted them as conjugate sheer joints formed by stress from the northeast during a small re-advance to form the “Barmston Push Moraine” (figure 7). Recent measurement of the sub-vertical joints along this section in September 2012 (figures 8 and 9) failed to replicate the orientations obtained by Penny and Catt. The previously visible joints, commonly present throughout the entire exposed thickness of the diamicton, do not show any vertical displacement of diamicton stratification and are commonly seen within apparently  diapirically intruded diamicton ‘columns’. This, together with the loaded bodies of gravel and sand which commonly overlie the diamicton (figures 7, 11 and 13) suggest its upper part was saturated at deposition. These features combined suggest the joints formed ‘late’ and are the product of dewatering and compaction of the diamicton.

Figure 7. Well jointed Skipsea Till (LFA1, Dmm) at TA1730 5792 (the spade for scale is 1m long). Barmston south. Note the irregular loaded contact with overlying gravels and sands (LFA3).

Figure 8. Sub-vertical joint orientations measured in the Skipsea Till at TA 1730 5791, close to Catt and Penny’s 1967 “push moraine” site. N = 60. The orientation of Catt and Penny’s conjugate shear joints are shown in red for comparison with the FQRG data from September 2012.

Figure 9. Sub-vertical joints measured along the Barmston section in the Skipsea Till from TA 1730 5791 to TA 1715 to 5881. N = 305.  The FQRG data is from September 2012. The orientation of Catt and Penny’s conjugate shear joints are shown for comparison.

Figure 11. Barmston central TA 1713 5883 (rucksack for scale). Deformation structures at the top of the Skipsea Till overlain by gravels and sands, with fine grained rhythmites high in the section.

Figure 13. Barmston central TA 1705 5916 (rucksack for scale). Gravel pendant structures. Saturated till loaded by overlying sandy gravels.

 At a few locations at Barmston south ‘inter-beds’ of gravelly sands can be seen between the Skipsea Till diamicton (LFA1) units (figure 10). The ‘upper till’ is typically weathered to a reddish brown colour. Evans and Thomson (2010) have interpreted such bodies as subglacial canals transporting meltwater and sediment to the ice front at times of high discharge ultimately to build ice front fan bodies. Whilst this seems a plausible explanation, Bateman et al.s’ (2011) investigation of a much larger body between Withernsea and Skipsea Tills at Dimlington suggest some possible proglacial sandar with overlying re-advance tills. Detailed investigation of each body is now warranted.

Figure 10. Barmston south, TA 1722 5820 (spade and geologist for scale). Well jointed Skipsea Till overlain by gravelly sand and a further diamicton unit. Is this subglacial channel fill or proglacial outwash overlain by a minor re-advance till?

In most of the available exposures LFA1 diamicton is overlain by small and large bodies of gravels and sands (LFA3 of Evans and Thomson 2010) (figures 7, 11 and 13). These range from large pods to smaller ‘pendant’ structures (figure 13). All appear to have been loaded into originally saturated, plastic, diamicton. LFA3 can be found both where the diamicton is low or high in the cliff and locally (as in figure 11) the deformation structures appear to be overturned in a consistent northward direction.

Along the Barmston central and northern sections LFA1 and LFA3 are typically overlain by a sequence which appears to ‘drape’ a pre-existing topography. The sequence coarsens upwards from sand/silt/clay rhythmites (LFA2b of Evans and Thomson 2010) (see figures 5, 12 and 16) into massive and de-watered sands, rippled sands (LFA2a) and finally into sandy and gravelly clinoforms (LFA2a and LFA3) marking progradation of subaqueous fan bodies within the glacilacustrine basins (figures 16, 17 and 18). At one site (figure 14) low angle thrusts disrupt bedding within the LFA2b rhythmites and may be rooted in deformation structures in the underlying diamicton. It is as yet unclear if this deformation is due to loading and compaction during deposition of the units or possibly gentle proglacial deformation? Occasionally ice wedge casts can be seen penetrating downwards into this sequence from near the present surface indicating subsequent permafrost development.

Figure 12. Barmston Central TA 1715 5870 (pound coin for scale). Millimetre scale rhythmites with a small dropstone. Bed LFA2b of Evans and Thomson (2010).

Figure 16. Barmston Road End. TA 1710 5935 (walking pole for scale). Compare with figure 5. The sequence from the base is: Skipsea Till is exposed below the beach. Overall coarsening upwards sequence within glacilactustrine basin fill. Clay/silt rhythmites low in the section contain small pebble rich horizons and isolated drop stones. Overlain by gravelly sands with conspicuous load structures. These are overlain by better bedded sands to the cliff top.

Figure 17. Barmston North TA 1708 5940 (spade handle for scale). Rippled sands in the central part of the coarsening upwards glacilucustrine sequence.

Figure 18. Barmston north TA 1705 5968 (section about 5 m high).  Rhythmites and sands overlain by low angle clinoforms. Subaqueous outwash fan terminating the basin fill.

 The LFS2b rhythmites were studied by Rushworth (1998) at a site close to that shown in figure 16, now lost to coastal erosion. He counted 395 laminae, suggested that they were true varves and that the proglacial lake persisted here for more than 350 years. In June of this year [2012] as yet undescribed trace fossils were discovered on bedding planes within these sediments (cf. Gibbard and Stuart 1974) and may provide useful plaeoecological information with further study.

3 Danes’ Dyke. Coastal cliff section at TA 215 692.

Recent work by the Flamborough Quaternary Research Group of the Hull Geological Society has focused on both Danes’ Dyke and South Landing (Figure 19) as Chalk is found close to modern sea level at the base of valley forms now buried beneath substantial thicknesses of glacigenic deposits. Initial work suggested the possibility that Chalk rich gravels resting on the bedrock at both sites could be correlatives of the Ipswichian interglacial marine beach deposit known from the classic site at Sewerby (Catt 2007). However, this has been proved not to be the case. The earliest deposits are significant in providing new chronological information for the initial advance of the North Sea Lobe of the last ice sheet in eastern England during the Dimlington Stadial. Final optically stimulated luminescence (OSL) dates are eagerly awaited for both these sites.

At Danes’ Dyke the presently best exposed section is to the west of the modern valley (figures 20 and 21). Chalk bedrock at the base of the relatively narrow valley form appears tectonised and probably also brecciated by periglacial processes. A lower, coarse, generally angular chalk clast diamicton, with a coarse silt matrix, overlies bedrock (figure22). This passes upwards into a better bedded smaller clast size diamicton, again with a coarse silt matrix and discrete thin beds of coarse silt. Bedding within the units dips at low angles towards the valley centre. This unit has a maximum thickness of 2.5m. The general angular nature of the almost exclusively Chalk clasts in these beds, the position flanking the valley side and the coarse silt matrix suggests that these units are periglacial slope deposits (gelifluctates) with an aeolian silt (loess) matrix. Similar material is widely known from the Chalk dry valleys of the Yorkshire Wolds. Loessic silt from similar deposits at Eppleworth Quarry (west of Hull) and beneath weathered Skipsea Till, has been dated by thermoluminescence to 17.5 +/- 1.6 ka (Catt 2007) and a similar age is proposed within the Dimlington Stadial of the late Devensian for the deposit at Danes’ Dyke. Some large boulders of (probably) Jurassic sandstone are present in the lower, coarser, gelifulction deposit. They may be remnants of an earlier glacial deposit that once filled the valley.

Figure 20. Danes’ Dyke (cliff height about 30m). Valley form cut into Chalk bedrock. The modern valley is excavated through periglacial and glacial sediments.

Figure 21. Danes’ Dyke, west section. Chalk bedrock overlain by periglacial sediments and a complex sequence of glacigenic deposits.

Figure 22. Detail of Danes’ Dyke west section. Locally derived periglacial slope deposits overlain by glacilacustrine muds and sands and Skipsea Till. OSL dating sample site shown.

Overlying the gelifluction diamicton are up to 1.75m of a coarsening upward sequence of laminated muds and laminated and rippled sands. These sediments appear to be the earliest glacial sediments and are interpreted to be proglacial lacustrine deposits. Their presence suggests that ice must have been present to the south, in what is now Bridlington Bay, to block drainage in the valley. A sample of these glacilacustrine sands has been taken for OSL dating.

A succession of about 25 m of further glacigenic sediments overlies the glacilacustrine beds (figure 21) composed of multiple diamicton and gravel/sand units. Whilst these units have yet to be described in detail, the matrix colour of the diamictons is similar to Skipsea Till. Diamicton 1 overlies the glacilactustrine beds and incorporates rafts of sand (figure 23). Clast fabrics in the unit indicate ice flow from the east-north-east, oblique to the likely valley orientation. Further work is required to understand the complex glacial stratigraphy at this site in comparison to adjacent sites.

Figure 23. Danes’ Dyke west section (trowel for scale). Detail of the contact between the glacilacustrine sediments and the overlying Skipsea Till (Diamicton 1).

4 South Landing. Coastal cliff sections between TA 233 692 and TA 230 692.

A much wider valley form is present at South Landing (about 230m wide, see figure 24) just 2km east of Danes’ Dyke and sediments are exposed to both the east and west of the modern valley.

Figure 24. South Landing valley form (cliff height is about 30m). The modern valley has excavated the older valley fill. Note the steep Chalk cliff on the west (left) side of the valley. The white cliff to the east (right) is composed of coarse chalk gravels. Slumping in the upper cliff conceals thick glacigenic sediments.

To the east erratic poor, coarse, chalk gravels are exposed and are up to 10m thick (figure 25). The lower parts are dominantly cobble/boulder gravels, typically unorganised, though occasionally displaying inverse grading. Bedding surfaces dips are towards the southwest. The upper division is less coarse and has more quartz sand beds present. In this unit bedding dips (and probably some medium sized cross beds) suggest plaleoflows towards the north-west. This deposit is enigmatic. The coarseness and thickness of the unit allied to the presence of probable debris flow deposits suggest a provisional interpretation as some form of glacigenic fan. Certainly it does not resemble periglacial deposits know from Danes’ Dyke, Sewerby and other sites on the Yorkshire Wolds. A sample of the quartz rich sands of the upper part of this unit has been taken for OSL dating to help place it in the emerging chronology.

Figure 25. South Landing east section (spade and people for scale). Erratic poor, coarse lower and finer upper Chalk gravels. Pie chart: blue=Chalk, red=flint, green=other erratics. Is this a glacigenic fan deposit? The site of the OSL dating sample from the upper gravels is indicated.

A very different suite of erratic rich Chalk gravels is present to the west of the modern valley (figure 26). Here the gravels rest on a slightly irregular Chalk platform at a similar elevation (approximately 2.5m OD) to that beneath the Ipswichian interglacial beach at Sewerby (Catt 2007). However, the gravels at South Landing do not have any of the characteristics of the beach gravels at Sewerby: the Chalk clasts at South Landing are poorly rounded, erratics make up nearly 50% of the assemblages and no mammalian bones or marine molluscs have been found. The gravels at South Landing are up to 3m in thickness and clast sizes fine upwards and quartz rich sand beds become more common upwards too. A cryoturbated surface is sometimes visible below the top of the gravels (figure 26) and frost shattered Chalk cobbles can be seen in some of the nearby large calcreted gravel and sand blocks on the beach in front of this section. Though field relationships are difficult to evaluate due to slumped material at the cliff base, the erratic rich gravels of the western exposures are believed to be younger than the erratic poor gravels to the east. The erratic rich character of the gravels in the western exposure, allied to the northerly palaeocurrent directions determined from clast imbrications (figure 26) indicates these deposits are glacifluvial outwash gravels derived from an ice margin to the north.

Figure 26. South Landing west section (walking pole for scale). Erratic rich Chalk gravel on Chalk bedrock. Upper portion of the unit is cryoturbated. Glacifluvial outwash gravel. An OSL dating sample was collected from close to the top of the unit at an immediately adjacent section.

An hiatus in the sequence is indicated by the periglacial surface near the top of the gravels. Above these deposits are poorly exposed glacilacustrine facies, gravels and sands, and finally diamictons, the whole being over 20m thick (Lamplugh 1891). Again, the matrix colour of the diamicton, where exposed, suggests it is Skipsea Till, though other units may be present high on the eastern cliff (Lamplugh 1891) and are the subject of further research. Samples have been taken for OSL dating from the top of the western gravels and sands, and from the unit of glacifluvial sand high in the western cliffs. Results are eagerly awaited!

Although Chalk bedrock is exposed close to the modern sea level in the valleys at Danes’ Dyke and South Landing there appears to be no preservation of beach gravels from the last interglacial. Though the Chalk platform beneath the gravels on the west side of South Landing may be a modified marine abrasion platform, no interglacial marine deposits are preserved on it. Despite this, the earliest deposits preserved within the bedrock cut valleys are significant in allowing OSL dating which may provide a closer chronology for the advance of the North Sea Lobe of the last ice sheet as it advanced into eastern England.

5 Acknowledgments.

For help with the ongoing research on Flamborough Head the Flamborough Quaternary Research Group would like to thank all the members of the Hull Geological Society who have helped with fieldwork at the sites; Mark Bateman and Dan Hartmann of Sheffield University for the ongoing OSL dating programme; Al Gemmell and Audrey Innes of Aberdeen University for granulometry; Lynda Howard for searching for beetles and chironimids in samples of the rhythmites; Ian Candy of Royal Holloway University of London for examining carbonate cements; John Catt, Colin Whiteman, John Boardman, Julian Murton, Della Murton and Emrys Phillips for advice and discussions.

6. References.

Bateman M D, P C Buckland, M A Whyte, R A Ashurst, C Boulter & E Panagiotakopolou 2011. Re-evaluation of the Last Glacial Maximum typesite at Dimlington UK. Boreas 40, 573-584.

Evans D J A & S A Thomson 2010. Glacial sediments and landforms of Holderness, eastern England: a glacial depositional model for the North Sea Lobe of the British-Irish ice sheet. Earth Science Reviews 101, 147-189.

Eyles N, A M McCabe & D Q Bowen 1994. The stratigraphic and sedimentological significance of Late Devensian ice sheet surging in Holderness, Yorkshire, UK. Quaternary Science Reviews 13, 727-759.

Gibbard P L & A J Stuart 1974. Trace fossils from proglacial lake sediments. Boreas 3, 69-74.

Heppenstall et al. [m.s.] 2010. Lost and found: the pre-Skipsea Till palaeovalley fill sediments of Flamborough Head and their significance for dating the advance of the last ice sheet in eastern England. [Unfinished draft published on the Hull Geological Society website in 2025.]

Lamplugh G W 1891. On the drifts of Flamborough Head. Quarterly Journal of the Geological Society 47, 384-431.

Penny L F & J A Catt 1967. Stone orientation and other structural features of tills in East Yorkshire. Geological Magazine 104, 344-360.

Rushworth G 1998. Rhythmites from Barmston, East Yorkshire. Quaternary Newsletter 86, 17-21.

[Notes – re-edited 2026. Some figures have not been included because of copyright issues. Some others were not mentioned in the original text, so they have been excluded from this new published version.]

  Flamborough Quaternary Research Group home page

Copyright - Hull Geological Society 2026

Registered Educational Charity No. 229147

Home