Content and licensingview original scan buy a printed copy

Geology of the Rockall Basin and adjacent areas

By K Hitchen, H Johnson and R W Gatliff (editors)

Bibliographical reference: Hitchen, K, Johnson, H, and Gatliff, R W (editors). 2013. Geology of the Rockall Basin and adjacent areas. British Geological Survey Research Report, RR/12/03.

Geology of the Rockall Basin and adjacent areas

K Hitchen, H Johnson and R W Gatliff (editors)

British Geological Survey BGS Research Report RR/12/03

Keyworth, Nottingham, UK: British Geological Survey 2013. ISBN 978 0 85272 707 2 © NERC 2013. All rights reserved.

Copyright in materials derived from the British Geological Survey’s work is owned by the Natural Environment Research Council (NERC) and/or the authority that commissioned the work. You may not copy or adapt this publication without first obtaining permission. Contact the BGS Intellectual Property Rights Section, British Geological Survey, Keyworth, email ipr@bgs.ac.uk. You may quote extracts of a reasonable length without prior permission, provided a full acknowledgement is given of the source of the extract.

Your use of any information provided by the British Geological Survey (BGS) is at your own risk. Neither BGS nor the Natural Environment Research Council gives any warranty, condition or representation as to the quality, accuracy or completeness of the information or its suitability for any use or purpose. All implied conditions relating to the quality or suitability of the information, and all liabilities arising from the supply of the information (including any liability arising in negligence) are excluded to the fullest extent permitted by law.

Example references for individual chapters

Hitchen, K. 2013. Introduction. 1–9 in Geology of the Rockall Basin and adjacent areas. Hitchen, K, Johnson, H, and Gatliff, R W (editors). British Geological Survey Research Report, RR/12/03.

Chapter authors:

• Introduction Hitchen, K
• Structure — Ritchie, D, Johnson, H, Kimbell, G, and Quinn, M
• Precambrian basement — Ritchie, D, Noble, S, Darbyshire, F, Millar, I, and Chambers, L
• Cambrian to Carboniferous — Stoker, M, and Leslie, A
• Permo-Triassic — Johnson, H, and Quinn, M
• Jurassic — Evans, D
• Cretaceous — Smith, K
• Cretaceous and Palaeogene igneous rocks — Hitchen, K, and Johnson, H
• Cenozoic sedimentary rocks — Stoker, M
• Economic geology — Hitchen, K, and Quinn, M
• Physiography and sea-bed sediments — Cotterill, C, and Leslie, A
• Geohazards Long, D

(Front cover) The picture shows the south face of Rockall Island which lies 162.7 nautical miles west of St Kilda in the north-east Atlantic Ocean. The island is approximately 18 m high and has a maximum diameter of 28 m. It is granitic in composition. A sill, 0.5 to 1.0 m thick and dipping at 30° to 35° to the east-north-east intrudes the lower part of the island. It is of a similar granitic composition and, on the south face, can be seen to bifurcate. The island is cut by a series of joints, one set of which is responsible for the north–south orientation of the near-vertical east-facing cliff. The local geology, and associated geophysical anomalies, suggests that Rockall was once part of a large volcano probably active during the latest Paleocene. Lieutenant Basil Hall was the first person to land on Rockall (from HMS Endymion in 1811). The island was annexed for Britain in September 1955 from HMS Vidal. In 1971 the summit was levelled by 39 Regiment (Royal Engineers), reducing its height by 1.5 m, in order to provide a flat surface for a light beacon (the dysfunctional remains of which can still be seen). Rock samples for analysis and study were also collected at this time. The island was formally incorporated into the county of Inverness-shire in 1972. Several small plaques, which record various landings on the rock, are attached to the south face of the summit ridge. Cover photograph courtesy of Andy Strangeway (www.island-man.co.uk)

Acknowledgements

Funding to kick-start the production of the report was provided by the following organisations who were members of the BGS Rockall Consortium in 2004 (BGS and BG, Chevron, DTI (now DECC), ENI, Shell and Statoil). Marathon, who joined the consortium later, also contributed. BGS is grateful to these sponsors and also their representatives who, over the course of many years, have guided the consortium work programme.

We are grateful to Steve Toothill of CGGVeritas and Tony Pedley of Fugro Multi Client Services for permission to use the seismic data shown in (Figure 24) and (Figure 26), respectively. Permission to use the seismic data shown in (Figure 57) was granted by Peter Kennedy of BP and that used in (Figure 82) by Tjeerd van Weering/Henk de Haas of the Nederlands Instituut voor Onderzoek der Zee (Royal NIOZ).The front cover photograph was kindly provided by Andy Strangeway.

Bathymetric contours in various map illustrations are reproduced from the GEBCO Digital Atlas published by the British Oceanographic Data Centre on behalf of the Intergovernmental Oceanographic Commission (of Unesco) (Figure 1), (Figure 2), (Figure 32), (Figure 42), (Figure 44), (Figure 48), (Figure 52), (Figure 53), (Figure 60), (Figure 61), (Figure 63), (Figure 68), (Figure 75), (Figure 76), (Figure 84), (Figure 86), (Figure 87), (Figure 88), (Figure 90), (Figure 91), (Figure 92), (Figure 93), (Figure 95) and (Figure 97) and the International Hydrographic Organization, 2003. Use of the Global Self-consistent Hierarchical, High-Resolution Shoreline Database (GSHHS)/World Vector Shoreline, is courtesy of National Geophysical Data Center (NGDC)/US Geological Survey/US National Imagery and Mapping Agency (Figure 2), (Figure 3), (Figure 5), (Figure 6), (Figure 7), (Figure 16), (Figure 19), (Figure 32), (Figure 38), (Figure 42), (Figure 44), (Figure 48), (Figure 53), (Figure 60), (Figure 61), (Figure 68), (Figure 75), (Figure 76), (Figure 84), (Figure 85), (Figure 86), (Figure 87), (Figure 88), (Figure 90), (Figure 91), (Figure 92), (Figure 93), (Figure 95) and (Figure 97). Multibeam bathymetry data displayed in (Figure 89) are courtesy of a NERC cruise carried out from RRS James Clark Ross during August 2003 and the Offshore Energy Strategic Environmental Assessment programme (SEA7) supported by the UK Department of Energy/Energy and Climate Change (DTI/DECC) in 2005 and DECC and the Department of Environment, Food and Rural Affairs (DEFRA) in 2006.

Thanks are also due to The Geological Society of London who kindly allowed the use within the report of modified reproductions from their published work and also for the use of selected images from their various publications (see appropriate figure caption acknowledgements in the body of the main text). Cambridge Paleomap Services Limited (CPSL) graciously permitted the use of an image from their refit software.

The report is multi-authored and many staff within BGS contributed to its production either by reading early drafts of the chapters or by editing and checking; in particular, the contributions of ‘Cartography and GIS’ and ‘editorial and publications’ sections are acknowledged.

Foreword

The deep-water Rockall Basin extends from south-west Ireland to north-west Scotland and constitutes the exploration frontier for oil and gas on the continental margin to the west of the British Isles. Up to the present time, oil company interest has focused on shallower areas of the UK continental shelf and west of Shetland. Hence the Rockall Basin has not been extensively explored and geological understanding of the basin is limited. Whereas over 10 000 commercial wells have been drilled on the UK continental shelf as a whole, only 12 of these have been drilled within, or on the margin of, the Rockall Basin.

The earliest pioneering exploration of the basin was undertaken in the late 1960s and early 1970s. Exploration was encouraged by the UK government designating ownership of the Hebrides Shelf in 1971 and of the Rockall Trough and Rockall Bank in 1974. This provoked the acquisition of commercial non-exclusive ‘speculative’ seismic datasets. The first hydrocarbon exploration licence was granted in 1974 and the first deep well, a stratigraphical test, was drilled in 1980. The first potential oil well was drilled in 1988.

Most exploration was concentrated on the eastern side, or eastern flank, of the basin. It wasn’t until the British Geological Survey (BGS) established the Rockall Continental Margin Consortium in 1992, comprising BGS and eight sponsoring oil companies, that a systematic basin-wide multidisciplinary exploration programme was developed. Since 1992, and with more companies involved, various geophysics, sea-bed sampling and shallow drilling cruises have been undertaken and the interpretation and analysis of the data collected have formed the basis for this report.

The basin is currently sediment starved and underexplored. Volcanism during the Palaeogene has added to the difficulty of data interpretation but hydrocarbon discoveries, on the eastern margin of the basin in both UK and Irish waters, demonstrate a working petroleum system and suggest that further exploration is warranted.

This report aims to provide a comprehensive summary of the geoscientific knowledge of the Rockall Basin area and we envisage this will be a useful resource for industry, government and academia.

John Ludden, Executive Director, British Geological Survey (NERC)

Chapter 1 Introduction

By Ken Hitchen

This British Geological Survey (BGS) Rockall Basin Offshore Regional Report describes the offshore geology of the area to the west of Scotland (west of the Outer Hebrides). The northern and southern limits of the report area are the UK/Faroe Islands and UK/Ireland median lines respectively. The eastern and western limits have been selected to include the structural highs adjacent to the main basin in order to place the basin into its full geological context (Figure 1) and (Figure 2).The bedrock geology of the area is shown in (Figure 3). Details of the geology outside the report area are also included where appropriate. This report supersedes the previous report by Stoker et al. (1993) which was entitled The geology of the Hebrides and West Shetland shelves, and adjacent deep-water areas and complements Ritchie et al. (2011) The geology of the Faroe–Shetland Basin and adjacent areas.

Data

This new report has been made possible through the acquisition of large volumes of new seismic, gravity, magnetic and bathymetry data, the drilling of a number of shallow boreholes (including the first ever on Rockall Bank) and the collection of many new in situ rotary-drilled short sea-bed cores (up to 5 m below the sea bed) and gravity cores. Much of these data were originally acquired by BGS on behalf of the UK Rockall Consortium (see Foreword) between 1992 and 2000, supplemented by government-funded research cruises in 2001, 2002, 2003 and 2005. Ownership of these datasets now resides with BGS. Other data, such as that acquired by the oil industry for hydrocarbon exploration, have been utilised where available. Published information from the technical literature, including conference proceedings, is also referred to, and referenced, throughout the report. Where no reference is given, the opinion is that of the BGS author.

Bathymetry

Excluding the Outer Hebrides, the report area contains three small areas of land. These are the St Kilda archipelago, the Flannan Isles and Rockall Island. The greatest water depth (approximately 2500 m) is in the central part of the Rockall Trough adjacent to the UK/Ireland median line. Within this vast range there are large areas of shallow shelf (Hebrides Shelf and Rockall Bank) and ‘basin’ floor and three large isolated seamounts which protrude hundreds of metres above the sea bed of the Rockall Trough.

Timescale

Subdivision of the Phanerozoic timescale, including the nomenclature and the absolute ages which define the stage names, is constantly under review. The ICS published a new version of the geological timescale in 2012 but because of the date that the Rockall project was active, the timescale and terminology adopted for this report (Figure 4) are based on Gradstein et al. (2004) but with some amendments.

Terminology

The geographical, bathymetric and structural terminology in the report area has evolved over many years. In the literature, different terms have sometimes been used for the same feature. Hence for this report an attempt is made to distinguish geographical (including bathymetric) terms from geological terms (Table 1). This is consistent with the approach by Naylor et al. (1999) for a similar problem in Irish waters. Hence Rockall Bank is the bathymetric expression of the Rockall High geological structure and Rockall Trough is the bathymetric expression for the deep-water area between the Hebrides Shelf and Rockall Bank which is approximately coincident with the underlying structural Rockall Basin. In places the bathymetric term is quite different to the structural term. Hence the structural Outer Hebrides High, Barra Basin, Flannan Basin and Flannan High all underlie the bathymetric Hebrides Shelf. Terms which are already well established in the literature have been retained (e.g. Wyville Thomson Ridge, West Lewis Ridge etc).

Hydrocarbon exploration

The Rockall Basin and the adjacent areas is still regarded as frontier territory for hydrocarbon exploration. Although a ‘stratigraphical test well’ was drilled in 1980 it did not penetrate the Mesozoic section and yielded little useful information. Five licensing rounds between 1977 and 1997 have offered blocks in the report area resulting in a further ten wells being drilled between 1988 and 2002. The first of these was well 164/25-1ST, drilled in 1988 in the West Lewis Basin. In 2000 well 154/1-1 discovered gas in Upper Paleocene low-porosity sandstones (the ‘Benbecula’ discovery). Subsequently, in 2006, this discovery was appraised by well 154/1-2 but there have been no further announcements. In general the wells have yielded useful stratigraphical information but no commercial hydrocarbons. However, in Irish waters, well 12/2-1 drilled the ‘Dooish’ prospect in the eastern Rockall Basin in 2002, and discovered a significant column of gas condensate so proving a working hydrocarbon system in the eastern Rockall Basin. A follow-up well, located basinward of the original hole, was drilled in 2008.

Potential hydrocarbon source rocks have been proved in BGS boreholes within the report area (Hitchen and Stoker, 1993; Isaksen et al., 2000). They are mainly Jurassic or earliest Cretaceous in age. Carboniferous coals may also be present in the deeper parts of the Rockall Basin and might be a source for gas. Structural hydrocarbon traps, such as tilted fault blocks created during the main Cretaceous rifting episode, are present on both margins of the Rockall Basin. Dip-closed traps created during Cenozoic compression are present in the north. Stratigraphical traps created by changes in sea level and sedimentation styles during the Cenozoic are another potential target of the area (McInroy et al., 2006). Details of potential hydrocarbon ‘plays’ in the area are also listed at https://www.og.decc.gov.uk/UKpromote/index.htm (the Promote section of the Department of Energy and Climate Change (DECC) website).

Gravity and magnetic maps

Gravity and magnetic data provide valuable insights into the geological structure and outcrop pattern of the report area. The gravity image shown in (Figure 5) is based on BGS marine and land surveys in its eastern part (from the Outer Hebrides High eastwards), which were mostly conducted in the 1980s, and data derived from satellite altimetry (Sandwell and Smith, 1997) in the west. The gravity variations revealed by this image are due to a combination of causes. Gravity lows are associated with sedimentary basins because of the relatively low density of the rocks they contain, whereas the major basic volcanic centres typically have high densities and are associated with positive gravity features. The gravity field also includes effects due to variations in water depth and corresponding topography on the Moho, although these can be reduced by the application of isostatic corrections (Figure 19).

The magnetic image (Figure 6) is based mainly on airborne surveys conducted in 1970 and 1971 by Huntings Geology and Geophysics, but also includes data acquired by BGS in the east (between 1962 and 1965) and Fairey Surveys in the west (1973), and an extract from a compilation by Verhoef et al. (1996) in the south. The rights to the Huntings and Fairey Survey results are now owned by BGS. Strong magnetic anomalies occur over the volcanic centres and magnetic disturbances are also evident where Palaeogene lavas or magnetic, crystalline basement are at shallow depth. Regions with thick sedimentary cover are typically characterised by a more subdued magnetic anomaly pattern.

Interpretation of the gravity and magnetic data involves qualitative analysis of the features revealed by images such as these and also by quantitative modelling. This is discussed further in the Crustal structure section.

Previous research

Prior to the late 1960s the offshore geology of the report area was virtually unknown although some information had been published concerning the onshore geology of the Outer Hebrides (Jehu and Craig, 1934) and the St Kilda archipelago (Cockburn, 1935). The latter was considerably updated by Harding et al. (1984).

The first person to set foot on Rockall Island was Royal Navy officer Lieutenant Basil Hall (the son of an Edinburgh geologist) in September 1811 (Sabine, 1960; Fisher, 1956). Some samples were collected and the granitic nature of the island was identified. However, the first specifically scientific non-naval expedition to Rockall was an Irish cruise undertaken in 1896. No landing was possible at that time due to adverse weather conditions. Further samples from Rockall Island were collected by Fisher during the annexation of Rockall Island by Britain in 1955 (Fisher, 1956) and by BGS (then the Institute of Geological Sciences) staff during Royal Navy operations to level the top of the island and install a light beacon in 1971 and 1972 (Harrison, 1975).

The deep crustal structure of the Rockall Plateau, including the Rockall High, was first investigated by Scrutton (1970, 1972) using seismic refraction data. Similar techniques were used by Roberts et al. (1988) and Keser Neish (1993) to determine the velocity structure, and underlying crustal structure, of the Rockall Basin. This type of approach has also been adopted further south across the basin by the Irish RAPIDS project. Jones (1978) and Jones et al. (1984, 1986) used seismic refraction data to investigate the nature of the crust beneath the Hebrides Shelf. Attempts to image the deep structure of the Rockall Basin and Hebrides Shelf, by seismic reflection profiling with unusually long record lengths, have been made by Mobil in 1989 unpublished, but see (Figure 18) and the British Institutions Reflection Profiling Syndicate (BIRPS), which shot the WINCH profile in 1982 to the west of the Outer Hebrides (Hall et al., 1984; Brewer and Smythe, 1986).

Some of the earliest pioneering work in the report area, in the late 1960s and early 1970s, was the investigation of Rockall Bank from which a number of publications followed (e.g. Roberts, 1969, 1972, 1975a, 1975b; Roberts et al., 1972, 1973; Jones et al.,1972). The Anton Dohrn Seamount also attracted early attention (Jones et al. 1974) whereas Rosemary Bank Seamount was studied using geophysical modelling techniques (Scrutton, 1971; Miles and Roberts, 1981). Other volcanic centres such as Geikie (Evans et al., 1989) and Darwin (Abraham and Ritchie, 1991) were not described until much later when hydrocarbon exploration data became available. Hebrides Terrace Seamount was described by Omran (1990) using mainly single channel seismic data collected by Aberystwyth University.

Exploration in the report area received a boost when the UK government designated ownership of the Hebrides Shelf in 1971 and the Rockall Trough and Rockall Bank in 1974. This encouraged regional speculative seismic surveys and hydrocarbon exploration. The interpretation of these new data greatly added to the understanding of the offshore geology. Subsequently the first licences for hydrocarbon exploration here were awarded in 1977 in the 5th UK offshore licensing round. The first deep commercial well to be drilled in the area was well 163/6-1A in 1980 and was, unusually, a stratigraphical test well, in unlicensed acreage, funded by a consortium of 19 oil companies and the UK Department of Energy (Morton et al., 1988). The first oil exploration well (164/25-1) was drilled by BP in 1988. Only twelve deep wells have been drilled in the report area. Most hydrocarbon exploration, and all the wells so far drilled, have been east of 10°W. Numerous publications based on oil industry datasets and exploration for hydrocarbons are contained in conference proceedings edited by Brooks and Glennie (1987), Parker (1993), Fleet and Boldy (1999), Doré and Vining (2005) and Vining and Pickering (2010).

Although geophysical surveys and shallow drilling in The Minch and Sea of Hebrides areas had been undertaken by BGS in the late 1960s and early 1970s, the first regional systematic geophysical surveys across the Hebrides Shelf were not conducted until 1984 and 1985. These were followed in 1985, 1988 and 1990 by shallow drilling operations and subsequently (1988–1992) sea-bed solid geology maps at 1:250 000 scale were published for the Sula Sgeir, Geikie, Lewis, St Kilda and Peach map areas (see inside back cover of this report). Two bedrock geology maps covering the report area have been produced at 1:500 000 scale. The Central Rockall Basin map was published in 2002 in conjunction with the Irish Petroleum Affairs Division. The BGS North Rockall Basin map followed in 2007.

In 1992 the BGS UK Rockall Consortium was established, co-funded by government and the oil industry, to investigate the geology of the Rockall Basin. This boosted all types of data acquisition and resulted in thousands of kilometres of new geophysical data being collected, numerous sea-bed samples being obtained and the first ever shallow boreholes being drilled on Rockall Bank. The Rockall Consortium ceased operations in 2010. BGS has continued its own research in the report area with numerous short sea-bed cores being collected from the Hebrides Shelf and top of Rockall Bank in 2001 (Hitchen et al. 2002) and geophysics cruises in 2002, 2003 and 2005. Detailed multibeam sea-bed bathymetry data were collected in selected parts of the report area in 2005, 2006 and 2009.

Summary of geology

The report area is located on the north-west part of the Eurasian continental plate and continental crust probably underlies all the area. However beneath the Rockall Trough the crust has been stretched, thinned and intruded by igneous material. Whereas the crust may be 25 to 30 km thick beneath Rockall Bank and the Hebrides Shelf, it thins southwards from about 20 km to about 6 km beneath the axis of the Rockall Trough. The depth to the Moho also decreases and the apparent extension factor increases (from β <2 to β >5) in a southerly direction.

The Precambrian crystalline basement comprises two major terranes, of Archaean and Early Proterozoic age, juxtaposed along a suture which is aligned north-west to south-east across the Rockall Basin (the Anton Dohrn Lineament).To the north-east of this is Lewisian basement which formed the foreland to the Caledonian Orogeny and which probably extends northwards beneath the Faroe Islands and the Faroe–Shetland Basin. The rocks have been dated at between 3300 and 2500 Ma. To the south-west of the lineament, is the Rockall Bank/Islay or Rhinns Terrane, samples from which have been dated in the range 1914 to 1750 Ma.

Deposits from the Middle Proterozoic through to the Devonian are completely unknown in the report area. However, Torridonian and Cambro-Ordovician sediments are present west of the Moine Thrust Belt on the north-west Scottish mainland. Devonian sediments, mainly of nonmarine aspect, have been proved by drilling in a significant number of commercial wells in the Faroe–Shetland Basin north of the report area.

During the Early Devonian, at about 420–400 Ma, the region underwent compression in the Caledonian Orogeny. A north-east to south-west structural grain was established which influenced the subsequent geological development of the area though perhaps not as much as in the Faroe–Shetland area to the north. Following the orogeny, tensional stresses gradually became re-established in the Arctic–North Atlantic region due to the initiation of the breakup of Pangaea. Although there are no proven Carboniferous rocks in the report area, the Donegal Basin, just to the south, contains a thick sedimentary succession of this age and Carboniferous rocks crop out, and have been proved in boreholes, in northern Ireland and in the Irish Rockall Basin.

A rifting phase occurred during the Permo-Triassic. In some cases, former Caledonian thrusts were reactivated as normal faults to create half-graben style basins infilled with syntectonic coarse sediments. Permo-Triassic sediments were deposited during a major regression of the sea and hence comprise mainly nonmarine, clastic red-bed sequences which are difficult to date biostratigraphically. There is some debate as to whether the Rockall/Faroe–Shetland/Møre rift axis originated at this time although the main rift phase was not until much later. The proven Permo-Triassic basins (such as the Flannan, West Flannan and West Lewis basins) tend to be marginal to the main rift axis and occur beneath the Hebrides Shelf.

Lower, Middle and Upper Jurassic rocks occur widely, both onshore and offshore, in the Inner Hebrides. In the report area, nonmarine Middle Jurassic mudstones have been proved in the West Lewis Basin and are potential hydrocarbon source rocks. Upper Jurassic muddy sandstones have been drilled in the West Flannan Basin. Jurassic sediments have been proved in similar marginal basins on the eastern side of the Rockall Basin in Irish waters and have also been interpreted to exist on the western side. A minor rift phase, or epeirogenic phase of subsidence, may have allowed Jurassic sediments to accumulate in a proto-Rockall Basin but any sediments of this age are likely to have been fragmented during the later main rift phase.

The main rift development phase of the Rockall Basin was during the Early to mid Cretaceous when the crust beneath the basin was stretched and thinned. Major ‘down to the basin’ faults are imaged on the western side of the basin but the direction of dip of similar faults on the eastern side is more equivocal. This major rift event was followed by passive thermal subsidence and rising sea levels which allowed Upper Cretaceous fine-grained sediments to accumulate. Ryazanian potential hydrocarbon source rocks have been drilled in the West Lewis and West Flannan basins. The thickest Cretaceous sequence drilled in the report area was proven by well 164/7-1 in the North-east Rockall Basin. The well penetrated 1797.7 m of Aptian to Maastrichtian mudstones although a large proportion of this interval comprises igneous intrusion. The Upper Cretaceous is commonly the host interval for Late Paleocene intrusions in the Rockall Basin. The Anton Dohrn Seamount may have originated during the Late Cretaceous.

During the Late Paleocene to Early Eocene an extensional tectonic regime developed across the area which culminated in the complete separation of North American continental material from that of north-west Europe and allowed the initiation of oceanic crust development to the west of Hatton Bank. This was associated with magmatism on a massive scale which may have been related to a mantle plume. The result in the report area was general regional uplift and the extrusion of widespread and voluminous lavas, the emplacement of numerous central volcanic centres and sills, igneous underplating to the base of the crust and small, local volcanic features extruded at the contemporary sea bed. The lavas and sills cause strong reflections on seismic data and degrade the response from deeper horizons. Thus pre-Paleocene structures, and hence the Mesozoic history of the Rockall Basin and its adjacent margins, is difficult to elucidate.

Igneous activity waned after the Early Eocene and passive thermal subsidence allowed basinward-prograding sediment wedges to develop on the basin margins. At about the Eocene Oligocene boundary rapid differential subsidence (the C30 event) caused major deepening of the Rockall Basin, on a kilometre scale, and the establishment of abyssal water circulation patterns. Other parts of the Atlantic Margin underwent uplift and long-wavelength anticlines and domes were formed. Some earlier Cenozoic sediments were eroded but there was also deposition of sediment drifts in the northern Rockall Basin especially during the Miocene. The configuration of the present-day basins began to develop at about this time. Further tectonic uplift in the Early Pliocene (the C10 event), and slight tilting of the Hebridean margin, resulted in major progradation of the shelf edge and further modification of oceanographical circulation patterns. It has been suggested that these events were important in creating the necessary conditions for the subsequent glaciation which affected much of northern Europe from about 0.5 Ma onwards.

Chapter 2 Structure

By Derek Ritchie, Howard Johnson, Geoff Kimbell and Martyn Quinn

Plate setting and tectonic evolution

The plate tectonic setting and evolution of the Rockall and adjacent area (Figure 7) has been the topic of several studies (e.g. Knott et al., 1993, Cole and Peachey, 1999; Roberts et al., 1999; Coward et al., 2003). The area is underlain by crystalline basement terranes, assembled in Palaeozoic times during the Caledonian Orogeny, and subsequently affected by a number of poorly constrained late Palaeozoic to Mesozoic extensional phases (Figure 8). This extension caused substantial thinning of the continental crust and development of sedimentary basins, such as the Rockall Basin. Ultimately, in earliest Eocene times, renewed rifting resulted in continental separation between north-west Europe and Greenland, leading to the development of the north-east Atlantic Ocean. Considerable uncertainty persists regarding the number, timing and duration of the rift phases that affected the Rockall region prior to north-east Atlantic sea-floor spreading. This is due to a number of factors, including the poor calibration of basin stratigraphy, owing to the paucity of well information, and poor imaging of deep structure because of the presence of widespread thick Paleocene lavas and associated intrusive rocks. However, the results of regional tectonic and structural framework studies (e.g. Musgrove and Mitchener, 1996; Naylor et al., 1999; Roberts et al., 1999; Doré et al., 1999; Coward et al., 2003; Naylor and Shannon, 2005) and plate reconstructions (e.g. Cole and Peachey, 1999) suggest that significant rift episodes affected the report area during Permo-Triassic, Jurassic, and Cretaceous times, with the main phase of rifting from Rockall to Vøring (offshore Norway) during the Cretaceous. Significant north-east-, north- and north-west-trending Precambrian and Caledonian shears along the north-east Atlantic margin underwent significant reactivation during the rifting events because they were conveniently orientated to accommodate extension (Doré et al., 1999).

Basement

The Precambrian basement of the north-west European Atlantic margin comprises crystalline rocks derived from at least three tectonic domains (Laurentia, Baltica and Eastern Avalonia), which were accreted during Caledonian collision (Figure 9). The report area occurs completely within the Laurentian domain, which is interpreted to comprise mainly Archaean (2800 to 2700 Ma) and accreted Proterozoic (1800 to 1745 Ma) basement (Chambers et al., 2005; Precambrian basement chapter). Baltica is formed of crystalline crust generated by magmatic-arc accretion during the Mid Proterozoic, followed by Late Proterozoic reworking and possibly plate accretion in southern Scandinavia. Around 460 Ma, the northern Iapetus Ocean was subducted resulting in a collision between Laurentia and Baltica and the Taconic/Grampian phase of the Caledonian Orogeny (Figure 8) and (Figure 9). Eastern Avalonia formed as a magmatic arc on the fringe of the southern continent of Gondwana (Pharaoh, 1999) and probably separated from Gondwana during the Early Ordovician. Following subduction of the southern Iapetus Ocean and the Tornquist Sea, it collided with Laurentia and Baltica between about 430 Ma and 400 Ma resulting in the Acadian phase of the Caledonian Orogeny (Figure 8). The Iapetus Suture marks the boundary between the Laurentia and the Baltica/Eastern Avalonia crustal domains and lies to the south of the report area.

Devonian and Carboniferous

Following Caledonian collision, Middle to Upper Devonian terrestrial basins developed within the orogenic belt, possibly by a combination of gravitational collapse of the thickened crust (e.g. McClay et al., 1986) and strike-slip (e.g. Roberts et al., 1999; Coward et al., 2003) tectonics (Figure 8). Basin development during Devonian and Carboniferous times occurred in parts of the north-east Atlantic margin, including the Clair (e.g. Nichols, 2005), Donegal (Naylor et al., 1999), Slyne (Dancer et al. 2005), Erris (Chapman et al., 1999) and the Clare basins (Croker, 1995). Sedimentary sequences within these basins are poorly resolved seismically, mainly because of overprint by younger events (Doré et al., 1999). Within the Faroe–Shetland region, continental clastic deposition was terminated in the Clair Basin by a Visean marine transgression, which has been cited as evidence that the Faroe–Shetland and contiguous Rockall basins had subsided by the latter part of the early Carboniferous (Haszeldine and Russell, 1987) and that Carboniferous sediments are inferred to be present within the Rockall Basin (Naylor et al. 1999).

Permo-Triassic

By early Permian times, the Variscan Ocean, to the south of the report area, had almost completely closed (e.g. Glennie, 2002), resulting in the grouping together of most of the world’s continents into the supercontinent Pangaea (Figure 8) and (Figure 10)b. Following a phase of regional uplift and peneplanation, two rift systems developed and dominated north-west Europe during Permo-Triassic times (e.g. Roberts et al., 1999; Coward et al., 2003; (Figure 10)a and (Figure 11). The ‘Arctic’ rift system propagated southwards towards western Ireland, reworking many of the Caledonian or earlier structures. The second, northward propagating rift system, termed the ‘Atlantic’ rift, developed initially by collapse of the Appalachian and European Variscan mountain belts.

Thick Permo-Triassic continental red-bed successions occur within the Hebrides and Faroe–Shetland shelf regions (e.g. Stoker et al., 1993), within the marginal basins to the west of Ireland (e.g. Naylor et al., 1999) and within the Irish Rockall Basin (e.g. well 12/2-1Z, see Permo-Triassic chapter). Within the report area, the West Lewis, Flannan and West Flannan basins (Figure 7) and (Figure 11) are considered to represent mainly Permo-Triassic rift-grabens (Stoker et al., 1993; Hitchen et al., 1995a). The preservation of a parallel-bedded Permo-Triassic section in the North Rona Basin, to the north-east of the report area, implies this is the downfaulted remnant of previously more extensive deposits (Kirton and Hitchen, 1987).

Jurassic

The Arctic rift system became relatively inactive in the latest Triassic to Early Jurassic (Figure 8). However, Rhaetian to Early Jurassic ‘Atlantic’ rifting, related to the opening of the Central North Atlantic, is considered to have propagated northward through the basins off Nova Scotia and Newfoundland into the Celtic Sea (Roberts et al., 1999). This rift appears to have inherited both Caledonian and Variscan basement trends (Roberts et al., 1999; Coward et al., 2003). Lower Jurassic rocks are preserved in the North Lewis and Donegal basins (Figure 7); Stoker et al., 1993; Isaksen et al. 2000) and in the North Porcupine Basin where they appear to onlap irregular pre-Jurassic topography (Naylor et al., 1999). A Lower Jurassic post-rift succession is also recorded in the Slyne and Erris basins (Tate and Dobson, 1989; Murphy and Croker, 1992; Chapman et al., 1999; Dancer et al., 1999; Naylor et al., 1999).

Sea-floor spreading commenced south of the Azores Gibraltar Fracture Zone in the Central Atlantic Ocean during the Middle Jurassic (Aalenian) (Srivastava and Tapscott, 1986). At this time, the extensional basins of the Atlantic rift were apparently subjected to thermal subsidence. In contrast, the Arctic rift was reactivated, with extension recorded in the North Sea and Norwegian Sea (Roberts et al., 1999). Middle Jurassic strata are widely developed throughout the Porcupine Basin and are also predicted over much of the Rockall region and most of the Slyne and Erris basins (Naylor et al., 1999). During Middle Jurassic times, the Porcupine, Celtic Sea and Grand Banks area probably developed as a broad thermal sag basin superimposed on minor, narrower Late Triassic–Early Jurassic rifts (Roberts et al., 1999). Further north however, Mid Jurassic rifting (Aalenian to Bathonian) (e.g. Doré et al., 1999; Naylor and Shannon, 2005), associated with a westward branch of the southward propagating Arctic rift, occurred in the Slyne Basin and resulted in the deposition of a very thick, shallow marine and estuarine succession similar to that in the Hebrides basins west of Scotland (e.g. Trueblood and Morton, 1991; Dancer et al., 1999).

Active extension during Late Jurassic to Early Cretaceous times resulted in a fully marine linkage of the Arctic and Atlantic rifts, creating a continuous rift which is now represented by the Vøring, Møre, Faroe–Shetland and Rockall basins (Figure 12). The Arctic rift was mainly active in Oxfordian to Tithonian times (Figure 8). Within the Atlantic rift, Mid and Late Jurassic extension may have been modest, but was sufficient to create restricted marine environments in which organic-rich petroleum source beds were deposited in the Jeanne d’Arc, Porcupine and Faroe–Shetland basins, and possibly also the Rockall Basin (Roberts et al., 1999; De Silva, 1999; Cole and Peachey, 1999).Within the Porcupine Basin, Upper Jurassic synrift deposits include basin-edge alluvial fans and deep marine fans (Croker and Shannon, 1987; MacDonald et al., 1987). Along the western margin of the Erris Basin, footwall uplift associated with a Mid to Late Jurassic rifting event induced uplift and erosion, implying Late Jurassic rifting within the Rockall Basin. (e.g. Naylor and Shannon, 2005). Basin modelling, based on normal incidence and wide-angle seismic profiles from the Rockall Basin, also supports the hypothesis of a significant rift episode there in Late Jurassic times (e.g. Shannon et al., 1999).

Cretaceous

Several authors (e.g. Rattey and Hayward, 1993; Roberts et al. 1999; Doré et al., 1999; Coward et al., 2003) have emphasised that extension within the Arctic rift system, including the North Sea rifts, effectively ceased by Early Cretaceous times and was superseded by north-east-trending, mainly Early to mid Cretaceous Atlantic rifting that extended from the Vøring Plateau to the Bay of Biscay (Figure 13). In Valanginian times, the Atlantic Ocean propagated into the region between the Azores Fracture Zone and the Biscay area. Within the Atlantic margin, two Early Cretaceous regional phases of major fault activity, during Valanginian to Hauterivian and Aptian to Albian times have been recognised by Coward et al. (2003) (Figure 8). From well and seismic evidence however, Musgrove and Mitchener (1996), proposed that the main phase of rifting within the Rockall Basin occurred during the Hauterivian to Cenomanian. Alternatively, Naylor and Shannon (2005) preferred localised Albian to Aptian rifting but also invoked Early Cretaceous non-faulted basin subsidence associated with mantle circulation. Scrutton and Bentley (1988) proposed a Neocomian (Berriasian to Barremian) age for the main rifting in the Rockall Basin, and the Barra Volcanic Ridge system, in the southernmost Rockall Basin, may also have formed at this time. On the basis of its proximity to Cretaceous ocean floor, the Barra Ridge may represent a localised area of incipient sea-floor spreading (Doré et al., 1999). According to Roberts et al. (1999), the Atlantic rift system had become relatively inactive by Albian times and thermal sag subsidence was dominant. Locally, tectonic activity appears to have increased in the Cenomanian and a significant pulse of extension occurred during Late Cretaceous times within the Faroe–Shetland area (Figure 8); e.g. Turner and Scrutton, 1993; Dean et al., 1999). Tectonic inversion and pull-apart basin development in Late Maastrichtian times affected parts of the north-east Atlantic rift, especially to the north of the Anton Dohrn Seamount, particularly within the Faroe–Shetland area (Roberts et al., 1999). This tectonism might be related to transpression and transtension associated with the development of an unstable triple junction in the Labrador Sea off south-west Rockall (Roberts et al., 1999).

Palaeogene and Neogene

The Labrador Sea opened during the Paleocene with the formation of new oceanic crust See (Figure 64); e.g. Chalmers et al., 1993; Doré et al., 1999; Coward et al., 2003). Within the Atlantic margin area, there is only limited seismic evidence for Early to Mid Paleocene rifting in the Porcupine, Rockall, Møre (Roberts et al., 1999) and Faroe–Shetland basins (Figure 8); Smallwood and Gill, 2002). During Mid–Late Paleocene to earliest Eocene times, there was widespread uplift of the Greenland, Vøring, Faroe Platform and Rockall areas in association with thermal doming and the development of the North Atlantic Igneous Province (NAIP) (e.g. Coffin and Eldholm, 1992). The scale of the igneous activity within the NAIP see (Figure 64) has been attributed by many to the effects of the Iceland Plume (e.g. White, 1988). Two igneous pulses have been recognised around the UK and adjacent areas, spanning Mid Paleocene to earliest Eocene times (e.g. Upton, 1988; White and Lovell, 1997; Saunders et al., 1997; Ritchie et al., 1999; Ritchie et al., 2011), corresponding with pre-rift (62 to 56 Ma) and syn-rift (56 to 54 Ma) break-up phases of magmatism (White and Lovell, 1997). However, it should be noted that Jolley et al. (2005) prefer three main phases (i.e. 60.7 to 58.5, 60.6 to 57.5 and 57.4 Ma). Northerly propagating oceanic spreading between eastern Greenland and north-west Europe commenced at approximately 55 to 54 Ma (Figure 8); Saunders et al., 1997; Ritchie et al., 1999) along the Reykjanes, Aegir and Mohns spreading ridges, coincident with the eruption of Balder Formation tuffs (e.g. Roberts and Backman, 1984; Ritchie et al., 1999). Following the separation of Greenland from Europe, the north-east Atlantic margin is regarded as a passive margin (e.g. Musgrove and Mitchener, 1996, Tate et al., 1999). However, the region is known to have experienced phases of tectonism during the Cenozoic and these are manifest as significant departures from a post-rift pattern of thermal subsidence, including episodes of compression, inversion and regional tilting and accelerated subsidence that are at least in part coeval (e.g. Doré et al., 1999, Japsen and Chalmers, 2000; Stoker et al., 2005a, b and c; Johnson et al., 2005; Kimbell et al., 2005; Tuitt et al., 2010).

Four main Cenozoic post-rift tectonic episodes have been recognised. Firstly, growth of the Wyville Thomson Ridge occurred during deposition of the lavas around Late Paleocene times ((Figure 8); Johnson et al., 2005). Secondly, in latest Eocene to Early Oligocene times, an episode of strongly differential subsidence or sagging occurred in the sub-basins underlying the Rockall Trough and the adjacent Hatton and Porcupine basins (e.g. Vanneste et al., 1995, Stoker et al., 2001; McDonnell and Shannon, 2001). An important consequence of sagging was the inception of the present morphological expression of the Rockall Basin as a deep-water trough, and the onset of deep-water current circulation and contourite drift deposition in this basin. This episode resulted in the development of a prominent Late Eocene unconformity, known as C30, which is characterised on seismic profiles by strong onlap. Furthermore, it appears to coincide with a phase of compression and the growth of inversion folds, including the approximately east-north-east-trending North Hatton Bank Anticline and possibly the approximately west-north-west-trending Wyville Thomson Ridge Anticline (Johnson et al., 2005). Thirdly, in the Late Oligocene to Mid Miocene, compressional tectonism, possibly associated with the closing of Tethys, reorganisation of spreading patterns in the ocean basin, mantle drag, intraplate stress etc (e.g. Boldreel and Andersen 1998; Cloetingh et al., 1990; Lundin and Doré, 2002; Mosar et al., 2002; Stoker et al., 2005a, b and c), resulted in the formation of new growth folds in the north-east Faroe–Shetland area (e.g. Ritchie et al., 2003a) and the rejuvenation of existing ones such as the Wyville Thomson Ridge Anticline (Boldreel and Andersen, 1995; Johnson et al., 2005; Lundin and Doré, 2002). This phase of tectonism appears to have instigated a passageway, termed the ‘Faroe Conduit’, for the persistent exchange of intermediate- and deep-water masses between the Atlantic Ocean and Nordic seas, across the Greenland–Scotland Ridge (Stoker et al., 2005a, b and c). Fourthly, from Early Pliocene times, tilting of the continental margin was associated with onshore uplift and accelerated offshore subsidence (Cloetingh et al., 1990; Japsen and Chalmers, 2000; Stoker, 2002), and local inversion folding in the north-east Faroe–Shetland Basin (Ritchie et al., 2003a). This tilting resulted in the development of a prominent unconformity (C10) characterised by basinal progradation of the shelf margins bordering the Rockall Basin, as well as deep-marine erosion during a reorganisation of bottom current patterns (Stoker, 2002). Progradation and development of the Hebridean margin was enhanced by Late Pliocene–Pleistocene glaciation. It should be noted that around the flanks of the Wyville Thomson and Ymir ridges, there are also signs of fairly recent compression, with warping of the sea bed (Johnson et al. 2005).

Deep crustal structure

Important information about deep crustal structure comes from seismic experiments in which long offsets between sources and receivers allow the identification of refractions and wide-angle reflections from deep within the crust and upper mantle. Only a few such experiments have been conducted within the report area however (Figure 14), and the data quality is variable. Deep seismic reflection profiling using multichannel recording and closely spaced (near-normal incidence) sources and receivers can, under favourable circumstances, provide clear images of deep structure which complement the velocity models derived from wide-angle experiments. Deep reflection profiling has been undertaken around the British Isles in the BIRPS surveys (Table 2) and these provide important insights into the structure beneath the Hebrides Shelf along the eastern side of the report area (Figure 15); Klemperer and Hobbs, 1991; Snyder and Hobbs, 1999). The near-normal-incidence method has generally proved less successful at imaging the deep crust further west because of the masking effects of lavas and sills in the shallow sedimentary section. (Table 2) provides a key to the acronyms used for deep seismic experiments in the area.

In the following sections, the crustal structure that underlies the different structural domains within the report area will be reviewed on the basis of the seismic evidence and 3D gravity modelling results (Figure 16); Kimbell et al., 2004, 2005).

Rockall Basin

The shape of the Rockall Trough (see Physiography and sea-bed sediments chapter), and the geometries predicted by the regional 3D model (Figure 16), strongly suggest that the Rockall Basin is offset in a sinistral sense across a north-west-trending zone which passes through the vicinity of the Anton Dohrn Seamount (Doré et al., 1999; Kimbell et al., 2004, 2005).This is most probably a transfer or accommodation zone formed as a result of the interaction between Mesozoic extension and pre-existing basement structures. The basin offset appears to be accommodated across a zone about 100 km wide, although Kimbell et al. (2005) identify three distinct lineaments within this zone, which they called the Anton Dohrn Lineament Complex (Figure 7) and (Figure 16). There is evidence that the location of this accommodation zone has been influenced by a Precambrian terrane boundary.

There is also evidence of a second, north-west-trending accommodation zone—the Wyville Thomson Lineament Complex (Figure 7) and (Figure 16)—close to the northern boundary of the report area, which is associated with the offset between the Rockall Basin and the basins to the north-east. Kimbell et al. (2005) suggested that this comprises a complex of structural lineaments, including the North Orkney Wyville Thomson transfer zone of Stoker et al. (1993) and features along the Ymir Ridge and Bill Bailey's High.

The principal seismic evidence for the deep structure of the northern part of the Rockall Basin comes from experiments reported by Bott et al. (1979), Roberts et al. (1988, profiles 86-002 and 86-005) and Klingelhöfer et al. (2005; profiles AMP-D and E) (Figure 16). Bott et al. (1979) assumed that the basin was floored by thicker than average oceanic crust, although they did not rule out the possibility that it was continental. Roberts et al. (1988) resolved a velocity structure which they interpreted to indicate stretched continental crust, and this interpretation was supported by the results of Klingelhöfer et al. (2005).

Klingelhöfer et al. (2005) resolved a somewhat asymmetric structure for the north Rockall Basin beneath line AMP-E, with the thickest cover sequence (about 5 km) and thinnest crystalline crust (about 12 km) beneath its eastern side (Figure 14)a. The depth to Moho decreases from about 18 km at this point to about 26 km on the east side of Lousy Bank. Farther west, the crust is underlain by a high velocity (7.4–7.6 km/s) layer interpreted to be igneous underplating associated with magmatism at the time of ocean opening (Klingelhöfer et al., 2005). Klingelhöfer et al. (2005) also tentatively identified an underplated zone lying at depths of between 27 km and 32 km beneath the Hebrides Shelf at the eastern end of the profile; the velocity of this zone was not resolved but there was evidence of reflections from its top and base. There was no indication of underplating beneath the basin, where upper mantle with a velocity of 8.0–8.2 km/s lies immediately beneath the stretched crystalline crust (Klingelhöfer et al., 2005).

In the model of Bott et al. (1979), the crystalline crust is approximately 14 km thick beneath the Rockall Basin and the Moho is at a depth of about 18 km, with the underlying upper mantle having a P-wave seismic velocity of 8.20 ± 0.17 km/s. On profile 86-002 of Roberts et al. (1988) the crystalline crust is approximately 10 to 13 km thick (thinning towards the east) and the Moho is at 16 to 19 km (Figure 14); an upper mantle velocity was not determined. As with the AMP-E line further north, there is a degree of asymmetry in the crustal configuration, with the eastern flank of the zone of crustal thinning being steeper than that in the west.

There are some differences between the crustal structure inferred from these wide-angle experiments and that predicted by gravity modelling (Figure 16). Interpreted Moho depths differ by up to 4 km in places, and the modelled sedimentary thickness in the deepest part of the North-east Rockall Basin is significantly greater (about 12 km) in the gravity model than was detected on AMP-D (about 8 km; Klingelhöfer et al., 2005). There is no evidence in the wide-angle seismic interpretation of AMP-E for the source of the north-north-east-trending gravity high which crosses the northern edge of the study area at around 11.6ºW, and which is ascribed to an intrabasin structural high in the gravity model. Despite these differences, there is reas-onable general agreement regarding the overall form of the north Rockall Basin, and both methods predict an extension factor (β) in this part of the basin of less than about 3 (Figure 16)d.

Profile 86-005 of Roberts et al. (1988) crosses the accommodation zone between the northern and southern parts of the Rockall Basin. On this profile, the Moho was interpreted to be at a depth of approximately 24 km beneath the west side of the basin and 17 km in the east (Figure 14)c, but it was not imaged in the intervening region. There was a corresponding decrease in the thickness of the crystalline crust from approximately 17 km to 11 km. There is a general correspondence between this geometry and that of the regional 3D gravity model (Figure 16)b and c, although the latter indicates that it is unlikely (isostatically and in terms of the predicted gravity response) that the shallow Moho extends beneath the western part of the Hebrides Shelf, as indicated in the interpretation of Roberts et al. (1988). Once again the seismic interpretation suggests asymmetry in the overall crustal configuration across the basin, although this may be exaggerated by effects relating to the accommodation zone.

The crystalline crust is significantly thinner (about 6 km) and the Moho significantly shallower (about 14 km) beneath the southern part of the Rockall Basin, to the south of the Anton Dohrn accommodation zone (Figure 16)b and c. This is indicated both by seismic surveys (Joppen and White, 1990; Keser Neish, 1993; O’Reilly et al., 1995; Hauser et al., 1995; Morewood et al., 2005) and by the regional 3D modelling (Figure 16)b and c. The thinnest crystalline crust beneath this part of the basin in the 3D model tends to occur towards its margins rather than its central axis (Figure 16)b, a pattern that is also exhibited by some seismic models (O’Reilly et al., 1995; Mackenzie et al., 2002; Morewood et al., 2005). O’Reilly et al. (1995) suggest that this may occur because syntectonic heat loss in the central part of the basin led to strengthening of the lithosphere and transfer of stretching towards the basin margins (see also Bassi, 1995).

If the high apparent crustal extension factors (β = 4 to 6) inferred for the southern part of the Rockall Basin (Figure 16)d are representative of overall lithospheric stretching, this might have been expected to trigger melting in the upper mantle (McKenzie and Bickle, 1988) and ultimately continental necking and breakup. Some early interpretations (e.g. Roberts et al., 1981) did favour oceanic basement beneath this part of the basin and this idea has been revived recently (Chappell and Kusznir, 2005). Joppen and White (1990) concluded that it is not possible to resolve from velocity structure alone whether the crust is fully oceanic or consists of highly stretched continental crust heavily intruded by syn-rift igneous rocks. Other interpretations (e.g. Makris et al., 1991; O’Reilly et al., 1995; England and Hobbs, 1997) have favoured highly stretched continental crust without major igneous additions. O’Reilly et al. (1995) and Hauser et al. (1995) explain the apparent absence of igneous underplating by a differential stretching model in which the brittle upper middle crust is stretched by a much greater amount than the ductile lower crust and mantle lithosphere. In support of such a hypothesis these authors contrast the three-layer crystalline crust beneath the basement highs on either side of the basin with the two-layer crust beneath the basin itself (e.g. (Figure 14)e, arguing that the upper two layers in the former have been highly attenuated (β = 8 to 10) to produce the single upper layer in the latter. Such a model requires that strain is partitioned at detachment zones formed at the brittle ductile transition at the top of the lower crust. An alternative hypothesis, that does not require differential stretching, is that the upper mantle did not melt to a significant extent because of the speed at which the rifting occurred. Relatively slow rifting allows heat loss by conduction and thus reduces the predicted amount of melt produced (Bown and White, 1995). For example, if β = 5 and the asthenosphere potential temperature is 1300ºC, instantaneous rifting is predicted to generate a melt thickness of 2.1 km, whereas no melt would be produced if such extension took place over a period of about 10 Ma or more (Bown and White, 1995). Such a time-span is well within the Hauterivian to Albian (approximately 35 Ma) rifting interval inferred by Musgrove and Mitchener (1996) for the Rockall Basin.

Relatively low P-wave velocities of 7.5–7.8 km/s are observed in the upper 3 to 10 km of the mantle beneath the southern part of the Rockall Basin (Figure 14)e; O’Reilly et al., 1996. These, together with relatively high values for the P- to S-wave velocity ratio, are interpreted as evidence of partial serpentinisation (O’Reilly et al., 1996; Morewood et al., 2005). The predicted degree of serpentinisation requires a large volume of water that was most probably introduced into the upper mantle from the oceans through fracture systems. In order for this to occur, the whole crust has to behave in a brittle fashion, and such behaviour is supported by rheological modelling by Pérez-Gussinyé et al. (2001), which suggests a transition to brittle behaviour in the lower crust beneath the Rockall Basin at extension factors of between 2.5 and 4, depending on the rift duration. The low coefficient of friction of serpentinites may have allowed a decollement to form at the crust mantle boundary at high extension factors, in a manner analogous to that inferred for west Iberian margin (Pérez-Gussinyé and Reston, 2001).

Outer Hebrides High, Stanton High and Malin High

The PUMA wide-angle experiment (Table 2) provided evidence of a three-layer crystalline crust beneath the Outer Hebrides High (Figure 14), with the Moho at a depth of 27 km and an upper mantle with a velocity of 8.0 km/s (Figure 14)g; Powell and Sinha, 1987. There is good agreement between the Moho identified from wide-angle reflections on the PUMA line and a band of reflectors observed at around 8.6 s two-way travel time (TWTT) on the coincident section of the WINCH near-normal-incidence profile (Table 2); (Figure 15); Brewer et al., 1983; Powell and Sinha, 1987). This is also in accord with the interpreted depth to Moho on the cross-cutting line of Bott et al. (1979). Farther north, the Moho lies at about 26 km depth beneath the western end of the W-reflector wide-angle profile (Figure 16); Morgan et al., 2000, and this agrees with the reflection Moho on the partially coincident GRID-7 seismic reflection profile. There is some disagreement between this evidence and the relatively deep (about 32 km) Moho interpreted by Roberts et al. (1988) at the eastern end of their line 86-002 (Figure 14)b, and the zone of underplating extending to similar depths beneath the eastern end of AMP-E in the interpretation of Klingelhöfer et al. (2005).

Klemperer et al. (1991) interpreted the BIRPS and WIRE seismic reflection profiles (Table 2), which lie close to the south-east corner of the report area, to indicate a Moho at a TWTT of around 9 s (approximately 28 km depth) beneath the Malin High. The COOLE wide-angle seismic experiment (Table 2) indicated a depth to Moho of 28 to 29 km beneath the northern part of Ireland (Lowe and Jacob, 1989). The WIRE seismic sections indicate that the Great Glen Fault and associated strike-slip faults are near-vertical structures that extend through the entire crust into the uppermost mantle (Klemperer et al., 1991).

A series of BIRPS seismic experiments has resolved prominent reflectors within the upper mantle in the region adjacent to the north-west coast of Scotland (e.g. Smythe et al., 1982; Brewer et al., 1983; Brewer and Smythe, 1984; Flack and Warner, 1990; Snyder and Flack, 1990; McBride et al., 1995; Snyder and Hobbs, 1999). The MOIST, WINCH, DRUM and GRID experiments (Table 2); (Figure 17)b mapped out two main reflective zones within the upper mantle beneath the North Lewis and West Orkney basins. The W-reflector occurs at a depth of 40 to 50 km and appears to be truncated at its western end by the eastward-dipping (at about 30º) Flannan reflector (Figure 15)b. These features have been correlated with mantle reflectors farther north, to the west of Shetland, but appear to be disrupted in the intervening region, perhaps by differential extension or shear linked to Mesozoic basin development (Figure 17); McBride et al., 1995; compare with the Wyville Thomson Lineament Complex of Kimbell et al., 2005). The WINCH and WIRE experiments detected further evidence of mantle reflectors farther south. These appear to have been offset by late Caledonian, sinistral, strike-slip displacements on the Great Glen Fault, indicating that they are of Caledonian or older age (Figure 17); Snyder and Flack, 1990; Snyder et al., 1997. The relationship between the Flannan and W-reflectors across this region is unclear. Many authors have implied that the east to south-east-dipping reflections observed along a strike length of about 700 km are associated with the Flannan reflector. Snyder et al. (1997), however, argue that the Flannan reflector is a local feature that only occurs in the area beneath the West Orkney Basin (and is associated with its formation) whereas the W-reflector occurs over a large area to the north of the Highland Boundary Fault.

A number of explanations have been put forward for the mantle reflections beneath north-west Scotland, including shear zones, igneous intrusions and relict subduction zones. Recent experiments indicate that there is a marked increase in acoustic impedance at the top of both the Flannan and W reflectors (Morgan et al., 1994; Warner et al., 1996; Morgan et al., 2000; Price and Morgan, 2000) suggesting that the reflections are not due to shearing or alteration, which would have the opposite acoustic impedance contrast. Warner et al. (1996) and Morgan et al. (2000) proposed that the reflections are caused by fragments of subducted ocean crust that have been metamorphosed to eclogite facies and preserved within the lithospheric mantle. Warner et al. (1996) suggested that the W-reflector might be a relict of low-angle subduction that was superseded by higher angle subduction recorded by the Flannan reflector. Such pre-Caledonian subduction may have been associated with the assembly of north-west Scottish terranes at about 1.9 Ma (Dickin, 1992). The difference between the general north-east trend of the mantle reflectors and the north-west trend of the inferred terrane boundary between the Archaean Hebridean Craton and the Proterozoic Rhinns or Islay Inishtrahull Terrane of Dickin (1992) was seen as an impediment to this hypothesis by Morgan et al. (2000). However, Friend and Kinny (2001) have suggested that a major, north-east-trending Proterozoic boundary exists between the Precambrian terranes of the Outer Hebrides and those of the Scottish mainland, and it is possible that convergence across this boundary was associated with the eastward subduction that now appears to be recorded in the lithospheric mantle.

Chadwick and Pharaoh (1998) noted that the Moho lies at a relatively shallow level beneath the Hebrides Shelf when compared with its expression beneath shelf areas elsewhere around the British Isles. This is seen as a mismatch in Moho depth between the regional 3D model (Figure 16)c and the seismic evidence outlined above. Chadwick and Pharaoh (1998) suggested that the discrepancy might be due to changes in the crustal velocity density relationship across Britain or to a subcrustal mass deficiency. Barton (1992) assumed the first of these hypotheses when employing a laterally varying crustal velocity density relationship in a gravity model along the LISPB seismic profile (Table 2). Kimbell et al. (2004) invoked the latter hypothesis when suggesting that the upper mantle beneath north-west Scotland might have a relatively low density either because of depletion or because of a thermal anomaly. The outcrop evidence suggests that the region could be underlain by Archaean upper mantle, and the highly depleted nature of such rocks (see, for example, Bernstein et al., 1998) is compatible with a reduction in their density (Poudjom Djomani et al., 2001). The possibility of a thermal anomaly is supported by the upper mantle temperatures predicted by inversion of P-wave seismic tomography by Goes et al. (2000), although the anomaly is not evident in their S-wave inversions.

Rockall High

Scrutton (1972) conducted a seismic refraction experiment along the axis of the Rockall High (Figure 14) and Bunch (1979) re-interpreted the data using ray-tracing methods. The data quality was poor towards the north-east end of the line, so the final velocity depth model of Bunch (1979) applies only to its south-west part. The model indicates a three-layer crystalline crust with a transitional Moho at depths of 29.5 to 31 km (Figure 14)f. The estimated upper mantle velocity was 8.2 km/s. On the east-trending RAPIDS profile data quality was poor across Rockall Bank and Vogt et al. (1998) based this part of their RAPIDS model on the previous results from Bunch (1979).

Line CDP87-3 of Keser Neish (1993) crosses Rockall Bank (Figure 14) and detected highly reflective lower crust below about 4 s TWTT. Keser Neish (1993) interpreted the Moho to be at aTWTT of 7.8 s (25.5 km) with reflections between 8.0 and 9.2 s being ascribed to ‘deep layered mantle’. An alternative interpretation would be to place the Moho at the base of the reflective zone, a more conventional interpretation for the location of the seismic reflection Moho that would fit the model of Bunch (1979) more closely.

Summary

The crystalline continental basement underlying the report area was assembled during Proterozoic times, and it is possible that the mantle reflections that have been observed around the north-west coast of Britain are due to relicts of the oceanic subduction associated with this terrane amalgamation. A number of extensional events affected the region during late Palaeozoic and Mesozoic times, but the most major of these was Early Cretaceous in age and caused the substantial crustal thinning that gave rise to the Rockall Basin. During extension, basin offsets developed across north-west-trending accommodation zones in the Anton Dohrn and Wyville Thomson areas. The location of the Anton Dohrn accommodation zone appears to have been controlled by Proterozoic basement structures. To the north of this zone, the crust beneath the Rockall Basin has been extended by a factor (β) of less than about 3, and probably deformed in a brittle fashion in its upper part and in a ductile fashion in its lower part. To the south of this zone, the crust was stretched to a greater degree (β = 4 to 6) and started to behave in a brittle fashion throughout. This enabled sea water to penetrate through to the mantle and react with peridotite to form the partially serpentinised zone that is now detected by its relatively low seismic velocities. Although extension was well advanced, the northward-propagating north-east Atlantic Ocean basin did not extend along this axis, but was directed through the Labrador Sea. This may have been because the Labrador arm of the rift triple junction was more favourably oriented with respect to the direction of oceanic propagation combined with the fact that the lithosphere underlying the Rockall Basin was relatively cool and strong by that time.

Structural elements

The nomenclature used in the definition and description of the various structural elements referred to within this report is shown in (Table 3). A description of each feature is given below. The volcanic centres are described in the Cretaceous and Palaeogene igneous rocks chapter.

Platform areas

Outer Hebrides High

The north- to north-east-trending Outer Hebrides High (including the islands of the Outer Hebrides) forms a major platform approximately 400 km long and 190 km wide that is located within the eastern part of the report area (Figure 7). It has a bathymetric expression that approximately corresponds to the Hebrides Shelf see (Figure 1). The Outer Hebrides High is clearly imaged on seismic (Figure 18); Musgrove and Mitchener, 1996, fig. 11 and gravity data, generally corresponding to a positive free-air and isostatic gravity anomaly see (Figure 6) and (Figure 19). However, it should be noted that there are a number of distinct positive and negative free-air and isostatic gravity anomalies within the high that are caused by the presence of the St Kilda, Geikie and Sula Sgeir volcanic centres, and West Lewis, West Flannan and Flannan basins, respectively.

The western margin of the Outer Hebrides High is interpreted to be separated from the north Rockall Basin by a series of steep, north-trending, west-dipping normal faults (Figure 7), (Figure 18), (Figure 20)b and (Figure 21)a and b. These faults are thought to have been active mainly during Early Cretaceous times, with the development of a 1.5 km thick syn-rift succession in places (e.g. (Figure 20)b. However, alternative interpretations are possible and Musgrove and Mitchener (1996) suggested that the boundary is represented by a simple, west-dipping ramp-like structure (Figure 22). If correct, this would imply that the Rockall Basin as a whole is a half-graben with the main controlling fault located on the conjugate western margin of the basin. To the north and north-east of the Geikie Volcanic Centre, the margin of the Outer Hebrides High is delimited by a combination of the poorly defined Darwin–Geikie High, the North-east Rockall Basin and the Wyville Thomson Ridge. Towards the south, the Outer Hebrides High passes laterally into the Stanton High, with the western part of this boundary defined by the north-west-trending Anton Dohrn Lineament, a fundamental crustal suture that separates the late Archaean Hebridean Terrane to the north from the mid Proterozoic Rhinns Terrane to the south (Dickin, 1992; Dickin and Durant, 2002; Kimbell et al., 2005). The eastern boundary of the Outer Hebrides High occurs outside the report area but is marked by the north-east-trending, south-east-dipping Minch Fault (e.g. Stein, 1988) that bounds the western margins of the North Minch and Sea of Hebrides Little Minch basins (e.g. Fyfe et al., 1993). Within the Outer Hebrides High, the south-west parts of the West Lewis and North Lewis basins appear to be separated from the north-east parts of the West Flannan and Flannan basins by a structural high. Speculatively, this apparent partitioning of the basins may be due to the presence of the Ness Lineament (Figure 7), a generally north-west-trending transfer zone alternatively referred to in part as the Butt of Lewis Fault (BGS, 1990) or Ness Shear/Fault Zone (Stein, 1988; Waddams and Cordingley, 1999).

A number of BGS boreholes have been drilled on the central and north-eastern parts of the Outer Hebrides High within the report area, including BH78/05, BH85/02A and BH85/02B, BH88/07, BH88/09 and BH88/10, BH90/11, BH90/12, BH90/13, BH90/14 and BH90/17 (Figure 7) and (Figure 2). The high mainly comprises Archaean to Proterozoic basement unconformably overlain by thin Permo-Triassic and/or Palaeogene lavas and Cenozoic to Recent sediments (BGS, 1986a, b and c, 1988a, 1989a, 1990, 1991, 1992). The Cenozoic interval demonstrably thickens towards the west, with a significant post-rift succession developed within the Rockall Basin (e.g. (Figure 21)a and b). The western margin of the Outer Hebrides High has not been drilled but the results from the wide-angle seismic refraction experiment profile 86–005 of Roberts et al. (1988) (Figure 14) suggest that it comprises at least 500 m of high velocity (5.3 km/s) Mesozoic to Torridonian rocks resting on metamorphic basement.

Rockall High

The north-north-east-trending Rockall High forms a major, elongate platform feature approximately 650 km long and 120 km wide that straddles the south-west part of the report area (Figure 7). It has a bathymetric expression approximately co-incident with Rockall Bank (see (Figure 1). The Rockall High also extends a considerable distance to the south-west of the report area and in this wider sense, is considered to comprise part of the Rockall Plateau (e.g. Naylor et al., 1999). The high is also clearly defined on seismic (e.g. (Figure 18) and (Figure 23) and potential field data, generally corresponding to a large positive free-air and isostatic gravity anomaly see (Figure 5) and (Figure 19).

The eastern margin of the Rockall High (including the shallow embayment of the East Rockall High Basin) is interpreted as being bounded by a suite of north-east-trending, south-east-dipping, planar normal faults that define a combination of the western margins of the south Rockall and Rónán basins (Joppen and White, 1990; Naylor et al., 1999; BGS and PAD, 2002; (Figure 7), (Figure 18), (Figure 20), (Figure 21) and (Figure 23). The amount of throw on the main basin-bounding fault that defines the margin of the Rockall High and the Rockall Basin is variable with, for example, approximately 1.5 km towards the south-east at the level of top crystalline basement on (Figure 20)c. Though our understanding of the evolution of this flank of the Rockall High is relatively poorly constrained, there is some suggestion from seismic data for the development of syn-rift wedges of probable Lower Cretaceous strata within the hanging-wall blocks of the suite of faults that occur along its length (Figure 23). The western flank of the Rockall High occurs outwith the report area but is not well defined due to the masking presence of thick Palaeogene lavas in that area. Roberts (1975a, fig. 13) considers it to comprise a series of north-west-dipping normal faults that cut the Paleocene lavas, although it is equally likely that these merely represent Paleocene lava escarpments. Recent work has demonstrated that the Rockall High as a whole was uplifted during Eocene times, with the development of contemporaneous submarine fan systems along the western margin of the Rockall Basin (McInroy et al., 2006).

The eastern margin of the Rockall High has been drilled by BGS boreholes BH90/03, BH90/05 and BH90/06, dredging and diving operations (see (Figure 2) and (Figure 7). The results indicate that the high mainly comprises Proterozoic metamorphic basement, unconformably overlain by a variable thickness of intermediate to basic Palaeogene lavas and thin Cenozoic sediments (e.g. Morton and Taylor, 1991; BGS and PAD, 2002; Hitchen et al., 1997; Chambers et al., 2005). Igneous plutonic rocks also intrude the metamorphic basement with, for example, the approximately 18 m high Rockall Island described as an aegerine microgranite of Paleocene to Eocene age (Hawkes et al., 1975). This exposure is interpreted to form a part of the planated Rockall Volcanic Centre (Roberts, 1969).

Stanton High

The Stanton High forms a relatively small platform feature, approximately 75 km from north to south and 100 km wide, that is located within the south-east part of the report area (Figure 7). It has a small, though discernable bathymetric expression that is partly coincident with Stanton Banks (see (Figure 7). The high is associated with a positive magnetic anomaly see (Figure 6).

The northern boundary of the Stanton High is interpreted to be partially defined by the north-west-trending Anton Dohrn Lineament (Figure 7). The transition from the western margin of the high into the south Rockall Basin is marked by east-dipping, planar normal faults (Figure 14)d. Towards the south, Stanton High extends just beyond the report area and is partially separated from the Donegal and Malin basins by a combination of the steep, planar, extensional west-north-west-trending Stanton Bank and north-east-trending Skerryvore faults (BGS, 1983; BGS, 1986b; BGS, 1992). The age of movement on these faults is uncertain, but they were probably mainly active during Carboniferous to Permo-Triassic times, with a minimum throw of approximately 300 m towards the south-west and south-east, respectively. In the east, the boundary between the Stanton High and Stanton Basin is defined by an unnamed steep, north-trending, east-dipping planar Torridonian to Permo-Triassic normal fault, with an estimated throw of approximately 1 km towards the east at the level of top crystalline basement (BGS, 1986b, cross-section 2).

The Stanton High has been drilled by BGS shallow boreholes see (Figure 2). Samples have also been collected by divers. The results indicate that the high mainly comprises Proterozoic metamorphic basement belonging to the Rhinns Terrane that in places, is unconformably overlain by thin Cenozoic sediments (e.g. BGS, 1992, cross-section 2).

Main basinal areas

North-East Rockall Basin

The generally north-north-east-trending North-east Rockall Basin forms an elongate graben approximately 160 km long and up to 90 km wide (Waddams and Cordingley, 1999, figs. 2 and 3) that occurs within the north-east part of the report area (Figure 7). The Cenozoic sedimentary succession is well imaged on seismic data, but the deeper structure is poorly resolved, mainly due to the presence of the thick Palaeogene lavas (Figure 24); Tate et al., 1999; Archer et al., 2005). The basin also corresponds to an elongate, negative free-air and isostatic gravity anomaly (see (Figure 6) and (Figure 19).

The North-east Rockall Basin is bounded on its northern flank by a combination of the north Rockall Basin, Ymir Ridge, Auðhumla Basin and the Wyville Thomson Ridge (Figure 7), though the nature of these contacts is poorly understood. There are conflicting hypotheses regarding the nature of the boundary that marks the south-east margin of the basin with the West Lewis High including a north-west-dipping ramp (Figure 20)a, a south-east-dipping reverse fault (Waddams and Cordingley, 1999, fig. 7) but the most plausible explanation is that it is defined by a north-east-trending, north-west-dipping planar normal fault (Figure 24), with an approximate net displacement of approximately 700 milliseconds TWTT at the level of top crystalline basement/pre-rift (Tate et al., 1999, fig. 5). The western margin of the basin is inferred to be marked by the buried, north-trending Darwin–Geikie High, although the structure and definition of this feature is poor (Figure 25).The southern boundary of the basin is also poorly defined, but is tentatively marked by the presence of a weak, east-north-east-trending, slightly less negative gravity anomaly. It should be noted though that Waddams and Cordingley (1999, fig. 2) preferred the interpretation that the North-east Rockall Basin is continuous to the south-south-west with the West Flannan Basin. According to Waddams and Cordingley (1999), the North-east Rockall Basin is traversed by the north-north-west-trending Ness Lineament (Figure 24). This lineament appears to correspond in part with the location of the Ness Volcanic Escarpment of Tate et al. (1999).

The North-east Rockall Basin has been drilled by a number of commercial wells including 153/05-1, 154/01-1, 154/03-1, 164/07-1, 164/27-1 and 164/28-1, 164/28-1A see (Figure 2) and (Figure 7). These six released wells proved between approximately 1.6 and 2.55 km of Cenozoic to Recent mixed clastic and volcanic rocks. Well 153/05-1 reached total depth within the Cenozoic succession but the remaining five wells proved up to approximately 1.8 km of Upper Cretaceous fine-grained clastic rocks (although at least 50 per cent of this total is due to Palaeogene intrusive rocks in well 164/7-1). Well 164/7-1 (Archer et al., 2005) terminated within sediments of Albian age, which comprise the oldest proven sedimentary rocks within the North-east Rockall Basin. Well 154/03-1 was the only well to reach metamorphic basement, penetrating 25 m or so of late Archaean amphibolite gneiss. In summary, the results of drilling indicate that the Cretaceous to Cenozoic succession has a proven thickness of at least 4.35 km within the North-east Rockall Basin. However, potential field and seismic modelling indicate that the basin could contain a total thickness of between 7.5 km (Figure 25) and 11 to 12 km (Figure 16)a of sediment, volcanic and intrusive rocks. In terms of tectonostratigraphical development, the presence of thick Palaeogene lavas within the North-east Rockall Basin is a significant hindrance regarding any assessment of its pre-Cenozoic evolution. Waddams and Cordingley (1999) considered that the North-east Rockall Basin once formed an integral part of the north-east-trending Faroe–Shetland Basin during Mesozoic and older times, and had only become detached due to Cenozoic growth of the cross-cutting west-north-west-trending Wyville Thomson Ridge inversion anticline. By analogy with the Faroe–Shetland or Rockall basins, the North-east Rockall Basin is thought to represent a mainly Early Cretaceous rift basin ((Figure 20)a; Roberts et al., 1999; Doré et al., 1999, fig. 3; Spencer et al., 1999, fig. 1), with the post-Lower Cretaceous interval representing a dominantly thermal-sag or post-rift succession.

Rockall Basin (North and South)

Basin architecture

The Rockall Basin forms the dominant structural element within the report area (Figure 7). It also extends a considerable distance to the south-west of this, and considered as a whole, forms an elongate, sinuous, north-east- to north-north-east-trending, sediment-starved, deep-water rift basin, approximately 1100 km in length. The width of the basin varies significantly along its length, being approximately 350 km wide at its south-west extremity where it is bounded by a combination of the Late Cretaceous (Santonian) Charlie Gibbs Fracture Zone (e.g. Knott et al., 1993) and the Clare Lineament (e.g. Johnson et al., 2001), before gradually tapering to around 200 km at its north-east margin. The form of this deep-water trough is well defined by bathymetric data, with water depths generally in the range of 1.0 to 2.5 km within the report area, although these increase steadily towards the south-west where truly oceanic depths of 4 km have been recorded close to the Porcupine Abyssal Plain see (Figure 1).

The Rockall Basin coincides with a negative free-air and isostatic gravity anomaly on which large circular or elliptical positive gravity anomalies are superimposed, caused by the presence of mainly Palaeogene volcanic centres including the Anton Dohrn, Hebrides Terrace and Rosemary Bank seamounts see (Figure 5) and (Figure 19). For the purposes of description and discussion of the Rockall Basin as a whole, the informal use of the terms ‘north Rockall Basin’ and ‘south Rockall Basin’ has been adopted, with the central strand of the Anton Dohrn Lineament Complex forming the boundary between them (Figure 7). This central strand of the lineament complex may also represent the boundary between the late Archaean Hebridean and mid Proterozoic Rhinns basement terranes. This subdivision of the basin may be further complicated by the presence of a prominent north-north-east-trending negative gravity anomaly close to the north-west margin of the north Rockall Basin that could be interpreted as indicative of the presence of an additional discrete basin, tentatively named the ?North-west Rockall Basin — a possible mirror image of the North-east Rockall Basin (Figure 19), see ?NWRB). Within or immediately adjacent to the north of the report area, the north Rockall Basin is bounded by the en échelon west-north-west-trending, Wyville Thomson Lineament Complex (Kimbell et al., 2005) that includes the Bill Bailey's High, Ymir and Wyville Thomson ridges (Figure 26), and also the poorly defined Lousy High that separates the north Rockall Basin from the oceanic Iceland Basin (Figure 7).The western margin of the north and south Rockall basins is marked by a combination of the Rockall High, Ladra High, East Rockall High Basin and George Bligh High. The Rockall basins pass laterally into the Hatton and Iceland basins. The boundary with the eastern margins of the East Rockall High Basin and Rockall and Ladra highs is generally interpreted as being defined by a series of north-east-trending, south-east-dipping normal faults (Figure 14), (Figure 18), (Figure 20) and c, (Figure 21) and c and (Figure 23). Similar fault styles have also been identified further north, flanking the eastern margin of the George Bligh High (e.g. Ritchie et al., 1999, fig. 7). However, the transitions with the Hatton and Iceland basins are poorly understood. The conjugate eastern flank of the north and south Rockall basins is defined by a combination of the Darwin–Geikie, Outer Hebrides and Stanton highs. The Darwin–Geikie High, which separates the north Rockall Basin from the North-east Rockall Basin (Figure 7) and (Figure 25), forms a generally north-trending structure but is only defined from gravity data. The nature of the contact between the Outer Hebrides and Stanton Bank highs and the north Rockall Basin is commonly obscured where the Palaeogene lavas are thick and consequently, there are various contrasting interpretations. The boundary is thought by some to be formed by a series of north-trending, west-dipping normal faults (e.g. (Figure 14)c, (Figure 18), (Figure 20)b, (Figure 21)a and b that also partly correspond to the margin of the Hebrides volcanic escarpment (Evans et al., 1989). The faults that define this margin are considered to have been active mainly during Early Cretaceous times with the development of contemporaneous syn-rift successions within their hanging-wall blocks e.g. (Figure 20)b. Musgrove and Mitchener (1996) and England and Hobbs (1997), however, preferred the margin to form an east-dipping, ramp-like structure, with the main controlling faults of the Rockall Basin half-graben occurring on the western basin margin (e.g. (Figure 20)c and (Figure 22)).

Nature of the basin floor

The nature of the basement crust that floors the Rockall Basin as a whole has been a topic of great debate, particularly between the early 1970s and early 1990s. There have been numerous hypotheses and these are summarised in Smythe (1989) and Ritchie and Gatliff (1996). Some of these hypotheses include:

Oceanic crust, formed during Jurassic to Early Cretaceous (e.g. Roberts, 1975a), mid to Late Cretaceous (e.g. Kristofferson, 1978; Srivastava, 1978; Roberts et al., 1981; Scrutton, 1986), early Permian (e.g. Russell and Smythe, 1978), or late Carboniferous (Haszeldine and Russell, 1987) times

1. Quasi-oceanic crust formed during late Carboniferous to early Permian (Smythe, 1989) times
2. Stretched continental crust with zones of oceanic crust (e.g. Megson, 1987)
3. Highly stretched continental crust (e.g. Roberts et al. 1988; Shannon et al., 1994; 1999; Hauser et al., 1995; O’Reilly et al., 1996 and Morewood et al., 2005)
4. Highly stretched continental crust with concomitant igneous intrusion (Joppen and White, 1990)
5. Arguably, the current favoured hypothesis for the type of crust beneath the Rockall Basin is that it is mainly continental in nature, albeit heavily intruded, and suffered major extension in the Early Cretaceous, although earlier phases of extension may have occurred in the Permo-Triassic and Jurassic (see rifting history below). The possibility of some oceanic crust, particularly within the highly extended south Rockall Basin, cannot be completely ruled out.

Basin infill

Within the report area, only a small number of BGS boreholes i.e. BH94/01 and BH94/04 and commercial wells i.e. 163/06-1, 132/06-1 and 132/15-1 have tested the stratigraphy of the north Rockall Basin see (Figure 2) and (Figure 7). Well 163/06-1 was drilled close to the northern margin of the basin and tested the north-west flank of the Darwin Volcanic Centre (Abraham and Ritchie, 1991, fig. 3; (Figure 7), (Figure 20)a and (Figure 24), proving approximately 1.25 km of Eocene to Recent sediments resting on 1.03 km of Paleocene basic and acidic lavas. Wells 132/06-1 and 132/15-1 were drilled to the north-east of the Hebrides Terrace Seamount, with the former proving approximately 2.2 km of Cenozoic sediments resting on 250 m of Upper Cretaceous strata whereas the latter encountered approximately 2.8 km of Cenozoic sediments resting on 500 m of Upper and Lower Cretaceous strata before terminating in late Archaean metamorphic basement (Musgrove and Mitchener, 1996, fig. 5). The Lower Cretaceous (Hauterivian to Albian) rocks encountered within well 132/15-1 represent the oldest proven sedimentary rocks within the north Rockall Basin, although down-dip to the west, Jurassic and older rocks have been inferred from seismic data (e.g. Spencer et al., 1999, fig. 11). Within the south Rockall Basin, Irish wells 5/22-1 and 12/2-1Z see (Figure 2) and (Figure 7) have been drilled on the eastern flank of the basin. The latter well terminated in Carboniferous Westphalian sandstones and claystones which are the stratigraphically oldest sedimentary rocks proved so far within the Rockall Basin as a whole.

The nature, age, distribution and tectono-stratigraphical development of the post-Palaeogene lavas succession in particular within the Rockall Basin as a whole, is relatively well understood from conventional seismic mapping, calibrated with BGS and commercial drilling results (e.g. Shannon et al., 1999; Stoker, 2002; STRATAGEM partners, 2002; Stoker et al. 2005b;). However, the presence of thick Palaeogene lavas and sills within the north and south Rockall basins degrade the definition of the deep structure. Consequently, estimates of the nature, age and distribution of the pre-Palaeogene stratigraphical intervals are largely inferred from velocity data that is derived from the results of wide-angle seismic reflection/refraction experiments (e.g. Roberts et al., 1988; Keser Neish, 1993; Shannon et al., 1994; O’Reilly et al., 1995; Shannon et al., 1999; Mackenzie et al., 2002; Morewood et al., 2005), and to a lesser extent from integrated potential field and seismic modelling (Kimbell et al., 2005). A summary of inferred stratigraphical intervals and thicknesses derived from these geophysical experiments is given in (Table 4). Within the north Rockall Basin, the typical sediment thickness appears to vary between approximately 4.5 and 6.5 km, with 5 km representing a reasonable average. In the south Rockall Basin, the sediments vary between 4.5 to 7.0 km thick although the sedimentary succession generally thickens towards the western and eastern margins of the basin (Figure 20)b and (Figure 22), where the inferred early Mesozoic basins are better developed (e.g. Naylor et al., 1999; Naylor and Shannon, 2005; Mackenzie et al., 2002, fig. 5).

Rifting history

The Rockall Basin as a whole is considered to represent a large, sediment-starved, highly extended rift basin. The amount of extension due to rifting within the Rockall Basin is thought to vary considerably along its length, with a significant increase in stretching factors occurring in a south-westerly direction see (Figure 16)d. For example, present-day stretching factors of β = 3.2 and β = 6.4 have the suggested for the areas close to the Wyville Thomson Ridge and Charlie Gibbs Fracture Zone, respectively (Cole and Peachey, 1999). More specifically, within the area straddling the north and south Rockall basins, analysis of regional seismic profile M89-WB-02 (Figure 20)b has revealed a bulk stretching factor value of β = 2.2 (Nadin et al., 1999). Within the south Rockall Basin, larger bulk stretching factors of β = 4 to 6 have been reported by Shannon et al. (1999) from the RAPIDS 2 experiment, though it should be noted they prefer a non-uniform or depth-dependent stretching model with extension values varying between different crustal layers i.e. the upper/middle and lower crustal layers are considered to have stretching factors of β = 9 and 4, respectively (Morewood et al., 2005). In general terms, the bulk stretching factors for the Rockall Basin as a whole are broadly similar to those derived from the results of potential field modelling see (Figure 16)d.

Though it is clear that the Rockall Basin has undergone significant extension, there is still great uncertainty regarding the timing and number of phases of rifting associated with the formation of the basin (e.g. Smythe, 1989; Ritchie and Gatliff, 1996; Nadin et al., 1999). This uncertainty is caused by a combination of poor calibration of the basin stratigraphy, due to a lack of well information, and poor imaging of the deep structure of the basin because of the seismic-masking effect of thick Palaeogene lavas and intrusive rocks. There are very few published interpretations based on commercial seismic data across the Rockall Basin that demonstrate the classic rift-related stratal geometries that are considered typical of extensional grabens or half-grabens (e.g. see Musgrove and Mitchener, 1996, fig.5; Spencer et al., 1999, fig. 11; Walsh et al., 1999). Some interpretations even suggest that the axial part of the basin forms a relative structural high or plinth (e.g. (Figure 20)b and c and (Figure 22)). However, to the south of the study area, results from the RAPIDS 3 deep seismic experiment (e.g. profile 33) have indicated some topography at the base of the sedimentary succession across the entire south Rockall Basin (e.g. Mackenzie et al., 2002; Morewood et al., 2005).

There are several hypotheses regarding the timing and number of rift episodes (i.e. single or poly-phase). From the integration of seismic interpretation and analysis of data from well 132/15-1, on the eastern flank of the north Rockall Basin, Musgrove and Mitchener (1996, fig. 5) proposed that the main phase of rifting occurred during Early Cretaceous times (Hauterivian to Albian), with unfaulted Hidra Formation (Cenomanian) limestone, marking the transformation from rift to post-rift succession development. Further to the south, within the south Rockall Basin, Shannon et al., (1999, fig. 7) illustrated seismic data that suggested the possibility of Mesozoic rifting close to the eastern margin of the basin. A number of the hypotheses regarding the rifting history of the Rockall Basin are derived from the results of deep, near-normal incidence seismic data (e.g. England and Hobbs, 1997), seismic refraction/wide-angle reflection experiments (e.g. Roberts et al., 1988; Keser Neish, 1993; Shannon et al., 1994; O’Reilly et al., 1995; Shannon et al., 1999; Mackenzie et al., 2002; Morewood et al., 2005), quantitative structural and stratigraphical modelling (e.g. Nadin et al., 1999), integrated potential field and seismic modelling (Kimbell et al., 2005), plate tectonic reconstructions (e.g. Knott et al., 1993; Cole and Peachey. 1999) and regional geological studies (e.g. Doré et al., 1999; Spencer et al., 1999; Roberts et al., 1999). Some of the suggestions regarding the timing of rifting are:

1. Early Cretaceous (Hauterivian to Albian) (Musgrove and Mitchener, 1996; England and Hobbs, 1997)
2. Mainly Late Cretaceous (Santonian to Coniacian) (Hanisch, 1984) but also in Permo-Triassic, Jurassic and Early Cretaceous (Knott et al., 1993)
3. Cretaceous to Cenozoic but with the possibility of additional older pre-Cretaceous phases (Cole and Peachey, 1999; Nadin et al., 1999)
4. Neocomian (Ryazanian to Barremian) (Scrutton and Bentley, 1988)
5. Jurassic to Early Cretaceous (Corfield et al., 1999)
6. Jurassic to Cretaceous (MacKenzie et al., 2002)
7. Triassic to Late Jurassic–Cretaceous (Shannon et al., 1999)
8. Latest Triassic to Cretaceous (Roberts et al., 1999)
9. Permo-Triassic to Jurassic (Shannon et al., 1994)
10. Permo-Triassic, Late Jurassic and localised Early Cretaceous (Albian to Aptian) (Naylor and Shannon, 2005)
11. Permo-Triassic (Bott and Watts, 1971)
12. Carboniferous (e.g. Haszeldine and Russell, 1987)

In summary, the probable general consensus of opinion would be that the Rockall Basin as a whole is a mainly Cretaceous rift basin, with earlier phases during Permo–Triassic and Jurassic times.

Post-rift history

Following Cretaceous rifting, the Rockall Basin underwent thermal cooling and post-rift subsidence during Cenozoic times. Indeed in very general terms, the Cenozoic interval (and to a certain extent the Late Cretaceous interval too) superficially appears to form a stratal geometry pattern typical of a mainly post-rift succession (Figure 18), (Figure 20)b and c and (Figure 21)b and c; Stoker et al., 2005a, b and c; STRATAGEM partners, 2002.). The thickness of this Cenozoic post-rift interval within the north and south Rockall basins is summarised in (Table 4). It should be noted, however, that the development of this post-rift succession does not merely reflect passive infilling of Cretaceous block-faulted rift topography within the basin, as there have been a number of events that have interrupted the pattern of thermal subsidence. For example, during Early to Mid Paleocene times, regional dynamic uplift over an area 2000 km in diameter, caused by the developing Iceland Plume (e.g. White and McKenzie, 1989), was responsible for the transition from deposition of relatively deep water Upper Cretaceous argillaceous sediments to widespread continental and shallow water basic volcanic rocks and sediments in and around the Rockall Basin area. Denudation of the Rockall High was responsible for the widespread development of Eocene fan systems along the western margin of the Rockall Basin (McInroy et al., 2006). However, there is no evidence from the deep geophysical experiments that any of this uplift in the basin was of a permanent nature due to the presence of underplating (compare with Clift and Turner, 1998). The development of the sedimentary succession that postdates the Palaeogene lavas has also been influenced by episodes of accelerated subsidence and inversion (e.g. Japsen and Chalmers, 2000; Stoker et al. 2005a, b and c; Johnson et al. 2005). Stoker et al. (2005b) and Johnson et al. (2005) have listed four main tectonic events that have substantially modified the development of the post-rift succession, preserving it as a series of unconformity-bounded megasequences see (Figure 62). These are: (i) growth of the Wyville Thomson Ridge during Late Paleocene times e.g. (Figure 26), with thinning of Palaeogene lavas onto the ridge; (ii) strong differential subsidence (sagging) in latest Eocene to Early Oligocene times within the Rockall Basin coupled with inversion on the Hatton High; (iii) Late Oligocene to Mid Miocene compression and growth of inversion structures including the Wyville Thomson and Ymir ridges; and (iv) Early Pliocene onshore uplift, accelerated offshore subsidence and tilting of the continental margin.

Marginal basinal areas (with associated highs or ridges)

Auðhumla basin

The north-west-trending Auðhumla Basin occurs mainly to the north of the report area, where it forms a narrow, synclinal structure about 80 km in length that broadens in a south-easterly direction from zero to approximately 30 km (Figure 7); Keser Neish, 2003. It is well defined from seismic data, occurring between the Wyville Thomson and Ymir ridges (Figure 26).Towards the north-west and south-east, the margins of the basin are poorly defined, though in the latter case, it may be that the Auðhumla Basin represents a north-westerly extension of the North-east Rockall Basin.

The basin has not been drilled but Keser Neish (2003, Enclosure 2 Profile J) speculated that it comprises approximately 3sTWTT of earliest Neogene to Jurassic sediments and lavas that have been deformed by post-Eocene compressional tectonic activity. The Drekaeyga Volcanic Centre (e.g. Keser Neish, 2005; Keser Neish and Ziska, 2005) has been inferred to intrude the basin (Figure 7).

Barra Basin

The north-north-east-trending Barra Basin forms a small, narrow, fault-bounded graben approximately 45 km long and 15 km wide located within the Outer Hebrides High close to the south-east margin of the report area (Figure 7); BGS 1986b; 1992.

The Barra Basin has been drilled by BGS borehole BH90/16 see (Figure 2) and (Figure 7), proving sediments of presumed Permo-Triassic age close to the sea bed. The basin is thought to contain a minimum of 300 m of Permo-Triassic strata, unconformably overlain by less than 100 m of Cenozoic sediments (BGS 1992, cross-section 1). In terms of its the tectonostratigraphical development, the basin is considered to represent the remnant of a mainly westerly tilted half-graben (e.g. Stoker et al., 1993) of probable Permo-Triassic age (Doré et al., 1999, fig. 1).

Donegal Basin

The generally east-north-east-trending Donegal Basin forms a small, irregularly shaped basin that only just encroaches into the extreme south-east margin of the report area (Figure 7); Dobson and Whittington, 1992; Naylor et al., 1999; Naylor and Shannon, 2005). Its general morphology has been defined from BIRPS WIRE seismic profiles (e.g. Dobson and Whittington, 1992, fig. 4).

The basin is bounded to the north and south by the Stanton and Islay–Donegal basement highs, respectively, with the former high partly defined by the west-north-west-trending, south-south-west-dipping, listric-normal Stanton Bank Fault (Figure 7). Approximately 150 m of Neogene to Recent and 300 m of Carboniferous to Jurassic strata have been interpreted to occur within the hanging-wall block of the fault within the Donegal Basin (BGS, 1992, cross-section 2). Towards the west, the Donegal Basin is flanked by the narrow, north-north-east-trending mainly Mesozoic Erris Basin whereas in the east, it is separated from the Malin Basin by the north-east-trending, south-east-dipping Skerryvore Fault (BGS 1986b, cross-section 2).

The basin has been drilled within Irish waters by well 13/03-1 see (Figure 2 and (Figure 7), proving thin Cenozoic to Recent sediments resting on approximately 950 m of upper Carboniferous (Stephanian to Westphalian) clastic rocks (Tate and Dobson, 1989; Dobson and Whittington, 1992). The Donegal Basin is thought to have evolved as a pull-apart basin in response to dextral strike-slip fault movement between the Islay–Donegal and Stanton highs during Carboniferous times (Dobson and Whittington, 1992). However, the potential presence of Permo-Triassic and Jurassic strata within parts of the basin (Naylor et al. 1999) could suggest intervals of subsequent rejuvenation.

East Rockall High Basin

The north-east-trending East Rockall High Basin is inferred to comprise an elongate structure approximately 120 km long and a maximum of 40 km wide that is located towards the south-west margin of the report area (Figure 7). Its presence was first inferred from potential field data by Roberts and Jones (1978, fig. 13), displaying a less positive free-air and isostatic gravity anomaly when compared with that of the surrounding Rockall High see (Figure 5) and (Figure 19). Due to the presence of Palaeogene lavas, the deep structure and morphology of the East Rockall High Basin is not resolved from seismic data (Figure 18) and (Figure 23).The Palaeogene lavas that overlie the inferred basin have been proven in BGS boreholes BH94/02 and BH94/03 (Figure 7). The nature and age of any pre-Palaeogene strata are unknown.

Flannan Basin and Flannan High

The north-east-trending Flannan Basin forms a narrow, north-west-tilted half-graben approximately 135 km long and up to 40 km wide that is located close to the eastern margin of the report area (Figure 7); BGS, 1990. The Flannan High flanks the north-west margin of the basin and is approximately 80 km long and up to 20 km wide. Both the basin and high are well imaged from shallow seismic (Figure 27) and gravity data — see (Figure 5) and (Figure 19).

The Flannan Basin is bounded to the north-west by a combination of the Outer Hebrides and Flannan highs (Figure 7), the latter defined by the Flannan Fault, a south-east-dipping listric normal fault with an inferred throw of approximately 2 km at the level of top crystalline basement (Figure 27). Towards the south-east, the central part of the basin is bounded by the Gallen Head Fault, a north-east-trending, north-west-dipping planar normal fault, with a throw of approximately 1 km at top crystalline basement level. From potential field data, the Flannan Basin is thought to extend further to the north-east than that originally defined by the BGS (1990), with the boundary now lying close to the Ness Lineament, the possible location of a reactivated Mesozoic transfer zone. The south-west margin of the Flannan Basin is marked by a series of small north-west- to north-north-west-trending faults that are considered to be related to similarly trending basement shears observed within the Outer Hebrides (e.g. Jones, 1981).

The Flannan Basin has been drilled, with sediments of presumed Permo-Triassic age recovered from BGS borehole BH88/08 - see (Figure 2) and (Figure 7). Within the northern part of the basin, a small remnant patch of Palaeogene lavas has also been proved by BGS short sea-bed cores (BGS, 1990). The Flannan High is inferred to comprise mainly late Archaean crystalline basement similar to that described from the Flannan Isles (Stewart, 1933) and from a small number of BGS shallow sample sites in the adjacent offshore area (BGS, 1990). On the basis of magnetic anomaly patterns, Proterozoic gabbroic and anorthositic rocks associated with the South Harris Igneous Complex mapped within the Outer Hebrides are considered to extend offshore towards the north-west and the Flannan Basin and Flannan High - see (Figure 6); McQuillin and Watson, 1973, Westbrook, 1974; Stoker et al., 1993. Seismic experiments have predicted the presence of 500 m to 1.7 km of strata (with velocities of about 4.4 km/s) of mainly Permo-Triassic age within the basin (Jones, 1981; Powell and Sinha, 1987). In terms of its tectonostratigraphical evolution, the Flannan Basin is considered to form a small mainly Permo-Triassic (e.g. Stoker et al., 1993; Hitchen et al., 1995a; Doré et al., 1999; Coward et al. 2003) or Triassic (Spencer et al., 1999) rift-graben.

Iceland Basin

The Iceland Basin is a deep oceanic basin that encroaches within the extreme north-western part of the report area (Figure 7). The basin is floored by oceanic crust, initially formed during earliest Eocene (magnetic anomaly chron 24r) times as a result of ?subaerial sea-floor spreading (e.g. see Nunns, 1983; Srivastava and Tapscott, 1986; Smallwood and White, 2002) associated with the developing Reykjanes Ridge which remains active today.

North Lewis Basin and Sula Sgeir High

The north-east-trending North Lewis Basin generally comprises a pair of westward-tilted half-grabens approximately 90 km long and 50 km wide that mainly lie outwith the north-east margin of the report area (Figure 7). The more easterly of the half-grabens occurs within the hanging-wall block of the north-east-trending, south-east-dipping, Outer Isles Fault or thrust zone. The basin as a whole is well defined from seismic (Figure 28) and gravity data, corresponding with a general negative free-air and isostatic gravity anomaly - see (Figure 5) and (Figure 19). The Sula Sgeir High is interpreted to form a north-north-east- to east-north-east-trending, basement block approximately 90 km long and 15 km wide that is also defined from seismic (Figure 28) and gravity data, tentatively corresponding to an elongate, sinistrally offset, positive isostatic gravity anomaly (Figure 19). However, other configurations are possible and it cannot be ruled out that the Sula Sgeir High is not sinistrally offset, but extends in an east-north-east direction towards the West Rona High.

The North Lewis Basin and Sula Sgeir High are bounded to the north-east by a combination of the Wyville Thomson Lineament Complex (Figure 7), North Rona Basin and Solan Bank High. Towards the north-west, a north-east-trending, south-east-dipping, low angle listric-normal fault, with several kilometres of Early Jurassic (or younger) displacement, separates the basin from the high (Figure 28).The south-west margin of the North Lewis Basin is defined by a combination of the north-west-trending Butt of Lewis Fault and Ness Lineament, whereas the Sula Sgeir High is separated from the West Lewis Basin to the north-west by either a ramp-like structure (Figure 20)a; Musgrove and Mitchener, 1996, fig. 3; Tate et al., 1999, fig. 8; Isaksen et al., 2000) or a steep, north-east-trending, north-west-dipping normal fault with a very large throw of approximately 9 km at the level of top crystalline basement (Figure 25).

The North Lewis Basin has been drilled by BGS boreholes BH72/36 and BH77/08, see (Figure 2) and (Figure 7), proving Upper Triassic to Lower Jurassic strata at the sea bed. The basin also contains the island of Rona, part of a Precambrian intrabasinal horst referred to as the West Rona High (Nisbet and Bowes, 1961). The Sula Sgeir High has been drilled by BGS borehole BH88/02 which proved basement of Lewisian aspect at the sea bed. Seismic interpretation suggests that the North

Lewis Basin may contain up to 4 km of north-west-dipping Permo-Triassic and Lower Jurassic sediments (Figure 28). In terms of its tectonostratigraphical evolution, the basin is considered to represent a mainly westward-tilted Permo-Triassic (e.g. Kirton and Hitchen, 1987; Stoker et al., 1993; Doré et al., 1999) or Triassic to Jurassic (Spencer et al., 1999) half-graben complex that has subsequently been intruded by Paleocene igneous rocks (e.g. Ofoegbu and Bott, 1985). However, the presence of Torridonian strata in well 156/17-1 within the adjacent North Minch Basin (Fyfe et al., 1993) suggests that a pre-Devonian precursor of the basin cannot be discounted (Klemperer and Hobbs, 1991; Stoker et al., 1993).

Rónán Basin

The north-north-east-trending Rónán Basin forms a narrow, elongate, westward-tilted half-graben at least 85 km long and 15 km wide, most if not all of which occurs to the south-west of the report area ((Figure 7); Corfield et al., 1999; Walsh et al., 1999; Naylor et al., 1999). The basin is defined from seismic data (Naylor et al. 1999, Enclosure 2A; (Figure 21)c).

The Rónán Basin is bounded to the north-west and south-east by the steep, north-east-trending, normal faults that define the eastern and western flanks of the Rockall and Ladra highs, respectively ((Figure 7), (Figure 21)c; Naylor et al., 1999, Enclosure 2A). The controlling fault that forms the north-west margin of the Rónán half-graben is estimated to have a displacement of about 1 s TWTT at the level of top crystalline basement (Figure 21)c. The basin is thought to be truncated to the south-west by an unnamed north-north-west-trending transfer zone that separates it from the Conall Basin (Naylor et al., 1999). Towards the north-east an increasing thickness of Paleocene lava hinders resolution and consequently, the nature and position of this boundary is somewhat speculative but could conceivably encroach within the report area.

No wells or boreholes have been drilled within the Rónán Basin so the nature and age of the sedimentary fill is inferred (Figure 21)c. By analogy with the Erris and Slyne basins located on the conjugate margin of the south Rockall Basin (Figure 7), the Mesozoic fill of the basin could comprise at least 5 km of Permo-Triassic, Jurassic and Cretaceous sediments which, in the deeper parts, may rest on Upper Palaeozoic rocks (Naylor et al., 1999, Enclosure 2A; Corfield et al., 1999). The basin is not well resolved in the gravity models of (Figure 10), probably because of the high density of its sedimentary fill. In terms of its tectonostratigraphical development, the Rónán Basin is considered to represent a mainly Late Jurassic to Early Cretaceous half-graben, particularly on account of the westward thickening of these inferred successions towards the basin-bounding fault ((Figure 21)c; Naylor et al., 1999, Enclosure 2A; Corfield et al., 1999). Thin Palaeogene strata unconformably overlie Upper Cretaceous and Palaeozoic rocks of the Rónán Basin and Ladra High, respectively.

West Flannan Basin

The north-east-trending West Flannan Basin forms an elongate, easterly-tilted half-graben approximately 115 km long and up to 60 km wide that is located within the eastern part of the report area (Figure 7). The basin is poorly defined from seismic data, with the Mesozoic and older structure of the basin only better resolved where the Palaeogene lavas are thin or absent (Figure 29). With the exception of its northern flank, the West Flannan Basin is generally well imaged from potential field data, corresponding to an elliptical, negative isostatic gravity anomaly (Figure 19).

The northern boundary of the West Flannan Basin is poorly defined, possibly because the basin is continuous to the north with the North-east Rockall Basin (Waddams and Cordingley 1999, (Figure 2); (Figure 7)). The south-east margin of the basin is separated from the Flannan High by a north-west-dipping normal fault with a downthrow of approximately 1 km towards the north-west at top crystalline basement level (Figure 29). However, due to the presence of thick Palaeogene lavas, the nature of the boundaries that define the western and southern flanks of the basin are largely obscured.

The south-east margin of the West Flannan Basin has been drilled, with for example, Palaeogene lavas, Upper Jurassic, and Permo-Triassic sediments proved close to the sea bed in BGS boreholes, BH90/07, BH90/08 and BH88/04, respectively, see (Figure 2) and (Figure 7). Towards the south-west of the basin, BGS borehole BH90/10 proved Palaeogene lavas. Estimates for the thickness of the succession within the West Flannan Basin can be derived from wide-angle seismic reflection/refraction experimental data. For example, the eastern part of profile 86-002 (Figure 14)b and the W-reflector experiment (Morgan et al., 2000) have indicated that nearly 5 km of mainly Permo-Triassic sediments and Palaeogene lavas (with velocities interpreted to range between approximately 2.0 and 5.77 km/s) occur within the basin. In terms of its tectonostratigraphical development, the West Flannan Basin is considered to represent a mainly Permo-Triassic rift-graben (e.g. Stoker et al., 1993; Hitchen et al., 1995a), overlain in places by thin Jurassic and Cretaceous strata, Palaeogene lavas and an south-eastward tapering Cenozoic sedimentary succession that is interpreted to be associated with contemporaneous thermal sag or post-rift subsidence focused within the Rockall Basin (Figure 29).

West Lewis Basin and West Lewis High

The north-east-trending West Lewis Basin, in the north-east corner of the report area (Figure 7), comprises a westward-tilted half-graben approximately 75 km long and 40 km wide. The basin is clearly imaged from both seismic (Figure 24) and gravity data, corresponding to a general negative free-air and isostatic gravity anomaly, see (Figure 6) and (Figure 19).The north-west boundary of the West Lewis Basin is defined by the north-east-trending West Lewis High, a fault-bounded basement block approximately 85 km long and 10 km wide (Figure 7). This is well defined from seismic (Figure 24) and potential field data, corresponding with an elongate positive free-air and isostatic gravity anomaly, see (Figure 6) and (Figure 19).

The boundary between the West Lewis Basin and High (Figure 7) is a north-east-trending, south-east-dipping planar normal fault, with apparent downthrows towards the south-east of approximately 1.4 s TWTT (Figure 24), 3 km (Figure 20)a and 5 km (Figure 25) suggested at the level of top crystalline basement. However, this fault was reactivated in a reverse sense during Palaeogene times, with a displacement of 0.15 s (Tate et al. 1999, fig. 5) or 0.25 s (Figure 24) TWTT at the level of top Palaeogene lavas. It should be noted that there are also conflicting hypotheses as to the nature of the boundary between the western flank of the West Lewis High and the North-east Rockall Basin. It may be a steep, west-dipping normal fault (Figure 24) or a ramp-like structure (Figure 20)a. The nature of the boundary between the West Lewis Basin and the Sula Sgeir High is also disputed, being either a steep, north-west-dipping, planar normal fault with a displacement of about 9 km at the level of top basement being suggested (Figure 25) or a simple ramp-like structure, e.g. (Figure 20)a and (Figure 30). The southern margin of the West Lewis Basin is not particularly well defined but is probably marked by the Ness Lineament. Towards the north-east, the West Lewis Basin and High are bounded by the Outer Hebrides High and Sula Sgeir Volcanic Centre although the details of these contacts are poorly understood.

The western margin of the West Lewis Basin and the crest of the West Lewis Ridge have been drilled by commercial wells 164/25-1,1Z and 164/25-2, respectively, see (Figure 2), (Figure 7), (Figure 20)a, (Figure 24), (Figure 25) and (Figure 30). Well 164/25-1, 1Z proved approximately 2700 m of Cenozoic clastic sediments and volcanic/intrusive rocks resting on 225 m of Cretaceous clastic sediments (intruded by Palaeogene sills) and 675 m of Permo-Triassic strata. Well 164/25-2 proved approximately 1925 m of Cenozoic clastic and volcanic rocks resting on late Archaean metabasic gneiss. The eastern margin of the West Lewis Basin has also been drilled by a number of BGS boreholes, with mainly Palaeogene to Permo-Triassic strata recovered at, or just below the sea bed, in boreholes BH88/01, BH90/05, BH90/02 (Figure 30), BH88/03, BH90/04, BH85/07, BH90/01 and BH90/06, see (Figure 2) and (Figure 7). The results of potential field modelling have indicated that the West Lewis Basin contains 8 km (Figure 16)a to 11 km (Figure 20)a of strata. In terms of its tectonostratigraphical development, the West Lewis Basin represents an north-west-tilted half-graben that formed mainly during Permo-Triassic times ((Figure 24); Stoker et al., 1993; Hitchen et al., 1995a; Doré et al. 1999, fig. 1). The overlying thin Jurassic and Cretaceous successions within the basin may have developed in response to subsequent phases of fault-controlled and/or thermal subsidence. The thick south-east-thinning Cenozoic sedimentary succession that drapes the West Lewis High and Basin (Figure 24) is interpreted to be associated with a contemporaneous phase of thermal sag or post-rift subsidence that was focused further towards the west within the Rockall Basin.

Other major structural features

Anton Dohrn Lineament Complex

The Anton Dohrn Lineament Complex comprises three major north-west-trending, approximately 600 km-long transfer elements, that are interpreted to traverse the Rockall Basin and beyond ((Figure 7); Kimbell et al., 2005). These features are defined from a combination of potential field data, and isotopic/geochemical analysis of basement terrane types, gross sediment thickness variations and stretching factor patterns.

The most northerly strand of the lineament complex (Figure 7) is identified by the disruption and left-lateral offset of the sediment thickness pattern (Figure 16)a and alignment in magnetic features (see (Figure 6); Kimbell et al., 2005). The origin of the inferred transfer element or zone is unknown but speculatively, it could represent the location of a former pre-Caledonian Laxfordian shear zone, similar to those described from the Lewisian Complex on the north-west Highlands of Scotland (e.g. Park et al., 2002).

The presence of the central or Anton Dohrn Lineament (Figure 7) was originally inferred from isotopic analysis of basement and igneous rock samples recovered from mainland Scotland, Ireland, and the offshore area around Blackstones, Rockall, George Bligh and Outer Hebrides highs and the Hebrides Terrace Seamount (Dickin, 1992; Dickin and Durant, 2002). Its existence has also been supported by regional structural interpretations (e.g. Doré and Lundin, 1996; Doré et al., 1999; BGS and PAD, 2002) and was interpreted by Dickin (1992) to represent a long-lived terrane boundary, which separates the late Archaean Hebridean Terrane from the mid Proterozoic Rhinns Terrane. The location of this lineament has been refined using potential field imaging and the 3D modelling results (Kimbell et al., 2005), and is associated with a marked change of the thickness of the crust beneath the north and south Rockall basins (Figure 16)b, with larger apparent extension factors occurring in the south Rockall Basin (Figure 16)d. Outwith the study area to the west, the Anton Dohrn Lineament cross-cuts the axial trace of the North Hatton Bank Anticline, before terminating at the continent–ocean boundary (Kimbell et al., 2005).

The most southerly strand of the Anton Dohrn Lineament Complex is interpreted to pass through the Mammal, Swithin, Rockall and Hebrides Terrace Seamount volcanic centres (Figure 7) and (Figure 16). It can also be correlated with the south-westerly truncation of magnetic features on the margins of the south Rockall Basin, see (Figure 6). Similar to that described for the most northerly strand of the lineament complex, this element could represent the location of a former Laxfordian shear zone.

Darwin–Geikie High

The north-trending Darwin–Geikie High is inferred to form an elongate buried basement high approximately 135 km long and 35 km wide that is located within the north-east part of the report area (Figure 7). The high is defined using potential field data, corresponding to an elongate, partially discontinuous positive isostatic gravity anomaly (Figure 19).

The Darwin–Geikie High is flanked to the east and west by the North-east Rockall and north Rockall basins, and to the north and south by the Ymir Ridge and Outer Hebrides High respectively, though the nature of these boundaries is poorly understood. The high incorporates the Darwin Volcanic Centre at its northern extremity.

The Darwin–Geikie High has not been drilled but according to Waddams and Cordingley (1999), this deeply buried structural culmination acted as a major control during Mesozoic times, effectively separating the Jurassic North-east Rockall Basin from the mainly Cretaceous north Rockall Basin (Figure 7). Since Mesozoic times however, the high does not appear to have been a positive structural feature, and consequently, exerted little influence on the widespread development of the Cenozoic post-rift succession within Rockall Basin and surrounding area (Figure 25).

George Bligh High

The George Bligh High straddles the boundary of the north-west part of the report area where it is interpreted to represent a small, broadly elliptical bathymetric high about 80 km wide in an easterly direction, and up to 40 km long in a northerly direction, see (Figure 1) and (Figure 7). The high is well defined from seismic (Ritchie et al. 1999, fig. 7) and gravity data, corresponding with distinct twin-peaked, positive free-air and isostatic gravity anomalies, see (Figure 5) and (Figure 19).

The George Bligh High forms a structural culmination that partially separates the north Rockall Basin in the east from the Hatton Basin in the west (Figure 7). It is dominated by the inferred presence of the West and East George Bligh volcanic centres, with the Palaeogene lavas associated with these centres obscuring the deep structure and the nature of the boundaries defining its margins (e.g. Ritchie et al. 1999, fig. 7).

The George Bligh High has been drilled by BGS borehole BH94/07, see (Figure 2) and (Figure 7). Samples from this, and other sea-bed sites on the crest of the structure, have proved a variety of lavas (including benmoreite and hawaiite) of presumed Palaeogene age (Hitchen et al., 1997).

Lousy High

The Lousy High occurs within the extreme north-west part of the report area where it is interpreted to form an elongate, north-east-trending bathymetric high (Figure 7). The high is well defined at top Palaeogene lavas level on seismic data (Vanneste et al., 1995, fig. 3; Tuitt et al., 2010), forming part of a general structural high that separates the north Rockall Basin (and the ?North-west Rockall Basin) from oceanic crust of the Iceland Basin to the north-west. The internal structure of the Lousy High is poorly understood, but a negative isostatic gravity anomaly (Figure 19), coupled with its positive bathymetric expression, see (Figure 1), could indicate that it represents an inverted sedimentary basin.

Minch Fault

The north-north-east-trending Minch Fault forms a large, approximately 340 km long, east-south-east-dipping normal fault that occurs adjacent to the eastern margin of the report area (Figure 7). The fault is clearly imaged from conventional seismic data and deep seismic experiments including GECO MP-1 and WINCH ((Figure 15)a and c).

The Minch Fault bounds a combination of the North Lewis Basin and the Outer Hebrides High, and the North Minch and Sea of Hebrides–Little Minch basins within footwall and hanging-wall blocks of the fault, respectively (Figure 7). It is interpreted as a mainly Permo-Triassic extensional fault but with significant post-Early Jurassic movement (e.g. Stein, 1988, figs. 5 and 6). There is an estimated net displacement of about 3 km at the level of top Permo-Triassic (BGS, 1989b, cross-section 2).

Ness Lineament

The north-west-trending Ness Lineament (Tate et al., 1999), or Ness Shear Zone (Stein, 1988; Waddams and Cordingley, 1999) is considered to represent a Mesozoic transfer fault that runs for up to 190 km between the Butt of Lewis Fault and the Ymir Ridge (Figure 7). The south-east part of the lineament, on the Outer Hebrides High, was originally defined as the Ness Shear Zone (Stein, 1988; Fyfe et al., 1993) or Butt of Lewis Fault (BGS, 1990). The resolution of the north-west part of the lineament, within the deeper water area of the North-east Rockall Basin, is more conjectural (Waddams and Cordingley 1999, fig. 7 A-A’) but is considered to represent a deep-seated crustal structure (Tate et al., 1999) that exerted a significant influence on the development of a major Paleocene basaltic escarpment and subsequent sedimentation patterns (Waddams and Cordingley, 1999; (Figure 24). On the Outer Hebrides High, the Ness Lineament could represent the site of a former Laxfordian shear zone similar to those described on the mainland of Scotland (e.g. Park et al., 2002), which subsequently acted as a Mesozoic transfer element, offsetting the Outer Isle Fault (e.g. Stein, 1988). Additional support for the hypotheses that the Ness Lineament represents a transfer element is derived from the juxtaposition of the North and West Lewis basins and the Flannan and West Flannan basins on the north-east and south-west flanks of the lineament respectively (Figure 7).

Outer Isles Fault

The Outer Isles Fault occurs just within, or immediately adjacent to, the eastern margin of the report area and forms a large, north-north-east-trending fault zone that extends along the eastern margin of the Outer Isles between north Lewis and south of Barra island for approximately 200 km (Figure 7). It has also been interpreted to extend at least a further 75 km or so into the offshore area to the north-east of Lewis, where it has been identified for example, on the WINCH, MOIST and GECO MP-1 deep seismic profile data (Figure 15)a. According to Stein (1988, fig. 7), the trace of the Outer Isles Fault appears to be sinistrally offset by the Ness Lineament.

Results from both onshore and offshore mapping indicate that the Outer Isles Fault dips at 20–30 degrees towards the east-south-east, before detaching at approximately 10 s TWTT (Morgan et al. 2000, fig. 2). The fault zone is described as comprising brecciated gneiss, mylonite, pseudotachylite and phyllonite (see Park et al., 2002). Its history of movement is complex, with several phases recognised including ductile thrust, brittle thrust, ductile sinistral strike-slip and finally, extensional movements (Butler et al., 1995). The initial ductile thrust movements are considered to be of Grenvillian age (Park et al., 2002). Other suggestions for the age of inception of the Outer Isles Fault include late Laxfordian (Fettes and Mendum, 1987) and Inverian (Lailey et al., 1989). Results from Ar-Ar analysis of a pseudotachylite (Kelley et al., 1994) have suggested a late Caledonian age of 430 ± 6 Ma for brittle thrust movements along the fault. The component of strike-slip movement has also been attributed to late Caledonian events whereas the extensional movement is associated with late Palaeozoic to Mesozoic opening of the Minch Basin (e.g. Stein, 1988).

Wyville Thomson Lineament Complex

The Wyville Thomson Lineament Complex is interpreted to comprise two north-west- to west-north-west-trending lineaments that straddle the north-east margin of the report area, but extending from the Iceland Basin in the north-west to the junction of the Outer Hebrides and Sula Sgeir highs in the south-east (Figure 7). Kimbell et al. (2005, fig. 6) prefer a model that is interpreted to comprise three lineaments. These lineaments may have acted as a transfer zone during Mesozoic times, separating basins with differing polarities (e.g. Stoker et al., 1993; Kimbell et al. 2005). The interpretation of Kimbell et al. (2005) modified and extended previously published versions of transfer zones in this area such as the Orkney–Faroe Alignment (Earle et al., 1989), the North Orkney/Wyville Thomson Transfer Zone (Stoker et al., 1993) or the Wyville Thomson Transfer Zone (Waddams and Cordingley, 1999). The individual lineaments or transfer zones that underlie the Wyville Thomson Ridge and Ymir Ridge are interpreted to have been reactivated as part of a ramp anticline complex during Palaeogene times (Figure 31). The part that coincides with the Wyville-Thomson Ridge is interpreted to have an element of sinistral strike-slip faulting associated with it, since there is evidence for the development of smaller, oblique and approximately north-trending anticlines on its southern flank during Cenozoic times (Johnson et al., 2005). However, the lack of a significant offset on the Outer Hebrides High militates against the possibility of any large-scale strike-slip fault activity (Figure 7). It should be noted that the results of potential field modelling by Kimbell et al. (2005) suggested that there is a major dextral offset in the vicinity of the Wyville Thomson Ridge between the Rockall Basin and the basins to the north-east.

Wyville Thomson Ridge

The west-north-west-trending Wyville Thomson Ridge forms a large, broadly symmetrical basalt-covered anticline approximately 200 km long and up to 20 km wide that encroaches within the north-east margin of the report area (Figure 7). The anticline is located in a geologically complex area; occurring at the confluence of the south-west Faroe–Shetland, Faroe Bank Channel, Munkur, Auðhumla, North-east Rockall and north Rockall basins, but is clearly imaged on bathymetric, see (Figure 1), seismic (Figure 26); Boldreel and Andersen, 1993; Johnson et al., 2005) and potential field data, corresponding to a positive free-air gravity anomaly, see (Figure 5). The nature of the boundaries between the Wyville Thomson Ridge and surrounding structures is poorly understood, mainly due to the presence of Palaeogene lavas that mask the deep structure of the area.

The Wyville Thomson Ridge has not been drilled but basalt samples have been recovered by dredging the sea bed around the crest of the anticline (Waagstein, 1988, fig. 1; (Figure 26)). The ridge was originally considered to comprise a 12 km thick pile of Palaeogene lavas that was deposited from a deeply buried fissure running along its axis (Roberts et al., 1983). However, the results of potential field modelling by Waddams and Cordingley (1999, fig. 5) suggest the presence of as much as 10 km of Paleocene sediments and older strata below a relatively thin 600 m thick cover of folded Palaeogene lavas. In contrast, Tate et al. (1999, fig. 3) preferred a model that suggested that basement is no deeper than 4 to 6.5 km depth, whereas Smith et al. (2009) suggested a variation of between 500 m to 1.2 km and 4 to 6 km, in the respective thickness of the Palaeogene lavas and pre-volcanic sediment intervals along the central part of the ridge. The results of a wide-angle seismic experiment by Klingelhöfer et al. (2005, fig. 9) across the south-east part of the ridge indicated the presence of 1 km of Palaeogene lavas and 5 km of older sediments and lavas resting on metamorphic basement at 6 km depth.

The Wyville Thomson Ridge is generally considered to represent a Cenozoic inversion structure (e.g. Boldreel and Anderson, 1993; Lundin and Doré, 2002; Johnson et al., 2005; Smith et. al., 2009), although there are conflicting views regarding its origin. For example, (Figure 31) Structural model for the formation of the Munkagrunnur, Wyville Thomson and Ymir ridges (modified from Tate et al., 1999). it could form part of a system of ramp anticlines (that include the Ymir and Munkagrunnur ridges) caused by compressional reactivation of a northward-dipping extensional fault plane (Boldreel and Andersen, 1993; Doré and Lundin, 1996;Tate et al., 1999), a reactivated transfer zone (Duindam and van Hoorn, 1987; Stoker et al., 1993; Musgrove and Mitchener, 1996; Waddams and Cordingley 1999, Keser Neish, 2003; Kimbell et al., 2005) or even be the location of a north-west-trending transient Paleocene rift system that was infilled with basaltic lavas (e.g. Lundin and Doré, 2005a; Ziska and Varming, 2008) and then subsequently inverted. Growth of the Wyville Thomson Ridge anticline is thought to have started in Late Paleocene times, with significant phases in Eocene and particularly Miocene times (Johnson et al., 2005; Stoker et al., 2005a and b; Tuitt et al., 2010).

Ymir Ridge

Thenorth-west-trending Ymir Ridge is a complexly faulted and asymmetrically folded and inclined basalt-covered anticline at least 50 km long and up to 15 km wide that encroaches within the north-east part of the report area (Figure 7). According to Ziska and Varming (2008), it comprises northern, central and southern segments that are bounded to the south-west and north-east by the north Rockall and Auðhumla basins, respectively. It is clearly imaged on bathymetric, see (Figure 1), seismic ((Figure 26); Boldreel and Andersen, 1993; Johnson et al., 2005; Stoker et al., 2005a) and potential field data, coincident with an elongate positive free-air and isostatic gravity anomaly, see (Figure 5) and (Figure 19).

The Ymir Ridge is bounded to the north-east and south-west by the Auðhumla and north Rockall basins, respectively (Figure 7). It has not been drilled but eroded basaltic lavas are inferred to crop out at the sea bed over its crest (Figure 26). There are different models regarding its genesis, including suggestions that it represents a frontal ramp anticline of a deeply buried system of Cenozoic fault-controlled ramp anticlines (Boldreel and Andersen, 1993), a segmented series of transpressional anticlines, possibly buttressed against the Darwin–Geikie High (Smith et al., 2009) or the location of a north-west-trending rift zone that has been infilled with lavas (Lundin and Doré, 2005a; Ziska and Varming, 2008) that share a similar inversion and fold growth history to the more extensively studied Wyville Thomson Ridge (see above).

Chapter 3 Precambrian basement

By Derek Ritchie, Stephen Noble, Fiona Darbyshire, Ian Millar and Lynne Chambers

Precambrian (Archaean and Proterozoic) basement is widely exposed on the Outer Hebrides Isles and the Flannan Islands that are located close to the eastern margin of the Rockall report area (Figure 32). Although Archaean and Proterozoic basement is considered to be present throughout the offshore part of the report area, its nature and age is poorly understood due to burial beneath varying thicknesses of Cenozoic, Mesozoic and Palaeozoic rocks within the Rockall Basin and flanking highs. However, analysis of cored basement material from a limited number of commercial wells and BGS boreholes/shallow drill sites on the Outer Hebrides, Stanton, West Lewis, Sula Sgeir and Rockall highs and in the south Rockall and North-east Rockall basins (Figure 32), together with the results of deep geophysical experimental data, have provided significant information regarding its nature, age and affinity.

Archaean to Proterozoic basement within the Rockall report area was originally part of Laurentia, which collided with Baltica approximately 430 to 400 Ma ago during the Scandian phase of the Caledonian Orogeny (e.g. Oliver, 2002; Coward et al., 2003; (Figure 9)). As far as is known, Archaean and Proterozoic basement appears to be mainly restricted to the north-east and south-west parts of the Rockall report area, respectively (Figure 32). These areas of basement rocks are assigned to the so-called late Archaean Hebridean (e.g. see Park et al., 2002) or late Palaeoproterozoic Rhinns (e.g. see Muir et al., 1989, 1992, 1994) terranes that are separated by the north-west-trending Anton Dohrn Lineament (Figure 32). The presence of the Anton Dohrn Lineament was inferred from isotopic analysis of basement and igneous rock samples recovered from mainland Scotland, Ireland, and the offshore area (Dickin, 1992; Dickin and Durant, 2002) and supported by regional structural interpretations based on seismic and potential field data (e.g. Doré and Lundin, 1996; Doré et al., 1999; BGS and PAD, 2002; Kimbell et al., 2005). It was interpreted by Dickin (1992) to represent a long-lived terrane boundary.

For the purposes of discussion, the distribution, nature, age and affinity of the Precambrian basement rocks within the report area are described within three broad subdivisions, namely Lewisian, Rhinns Terrane and Torridonian. Moinian and Dalradian rocks of the North-west and Grampian Highlands (see Strachan et al. 2002) respectively, occur immediately to the south-east of the Rockall report area (Figure 32). Based on current data, rocks of Moine and Dalradian affinity are not predicted to occur within the Rockall Basin or on immediately adjacent flanking highs (e.g. Roberts et al., 1999, fig. 2) and have not been encountered in any sampled locations to date.

Lewisian Gneiss Complex

The term Lewisian Gneiss Complex of mainland north-west Scotland and the Outer Hebrides Islands is used to describe an exposed Archaean to Proterozoic lower crustal section mainly comprised of grey quartzofeldspathic gneisses of tonalitic-trondhjemitic-dioritic (TTG) affinity (formerly igneous plutonic rocks) that was intruded into metasedimentary and mafic to ultramafic bodies, and cut by later basic dykes/sheets, acidic and other intrusive bodies. Subsequently, the complex has undergone high-grade metamorphism and deformation before approximately 1600 Ma (e.g. see Kinny et al., 2005). The Outer Hebrides Isles form the western extremity of the Lewisian Gneiss Complex, straddling the boundary of the Rockall report area. A geological summary of islands, including the results of detailed mapping is provided by Fettes and Mendum (1987), Fettes et al. (1992) and Park et al. (2002).

The long-held view of the evolution of the Lewisian Gneiss Complex is that it represents a single block of Archaean crust that was influenced by two orogenic cycles i.e. an Archaean Badcallian event with granulite facies metamorphism at about 2700 Ma, and a Proterozoic Laxfordian event with amphibolite facies metamorphism at about 1700 Ma including reworking associated within north-west-trending shear zones. These two events are considered to be separated by intrusion of basic Scourie dykes at about 2400 Ma (e.g. Sutton and Watson, 1951; Park et al., 1994, 2002). An additional significant retrogressive and deformational metamorphic event that took place after the main Badcallian granulite facies metamorphism and before intrusion of Scourie dykes is termed the Inverian event (Evans, 1965).

Rocks on the Outer Hebrides Isles and north-west Scottish mainland are lithologically similar (mainly TTG gneiss) and the subdivision of geochronological events initially recognised on the mainland by Sutton and Watson (1951) were subsequently correlated with apparently similar events on the Outer Hebrides by Dearnley (1962, 1963). For example, the Laxfordian granite injection complex of west Lewis and north Harris and the granite sheets of the ‘northern region’ of the mainland (e.g. Dearnley and Dunning, 1968). However, work over the past decade, particularly U-Pb geochronology, has cast significant doubts regarding both these earlier classifications of the Lewisian Gneiss Complex and the validity of this mainland/Outer Hebrides correlation model (e.g. Friend and Kinny, 2001; Love et al., 2004; Kinny et al., 2005; Park, 2005), although as yet a final consensus has not yet emerged (e.g. Park et al., 2005). Friend and Kinny (2001) and Kinny et al. (2005) suggested that the Lewisian Gneiss Complex can be subdivided into nine different terranes and a single undifferentiated block, which together represent a welded assemblage of Archaean cratonic blocks and Proterozoic arcs separated by north-west to north-north-west-trending shear zones (Figure 32), inset b. On the basis of structural, metamorphic and igneous characteristics, Park, (2005) critically reviewed the scheme of Friend and Kinny (2001) and Kinny et al. (2005), suggesting that the Lewisian Gneiss Complex can be subdivided into fourteen separate crustal blocks, some of which may be terranes and others fragmented and displaced blocks (or suspect terranes) derived from a few larger terranes. Furthermore, Kinny et al. (2005) stated that the terranes observed on mainland Scotland and the Outer Hebrides do not correlate well (e.g. Figure 32), inset b, with the area to the west of the Outer Isles Fault in particular having potentially more in common with the Archaean and Proterozoic assemblages of East Greenland (see Friend and Kinny, 2001, fig. 9). However, on the hanging wall of the Outer Isles Fault, the granulite facies gneiss described on Barra, and the Corodale gneiss (Figure 32) of South Uist may have generic links with similar facies rocks on the mainland area (Kinny et al., 2005).

Outer Hebrides

Archaean to Proterozoic basement rocks on the Outer Hebrides have been divided into a western and eastern series, separated by the north-east-trending, north-west-verging Outer Isles Fault (Fettes et al., 1992) (Figure 32). The history of the Outer Hebrides Fault is complex, but it is generally considered to have been initiated as a ductile thrust possibly associated with Inverian, Laxfordian or Grenvillian events, followed by brittle thrusting and strike-slip movements during the Caledonian Orogeny. In late Palaeozoic to Mesozoic times, the sense of fault movement was reversed, acting as a south-east-dipping extensional fault associated with the formation of the Minch Basin (e.g. Stein, 1988).

In recent times, the Archaean and Proterozoic basement on the footwall block of the Outer Isles Fault has been subdivided into three terranes and a single undifferentiated block, namely the Nis, Tarbet and Roineabhal terranes and the Uist/Barra Block (Figure 32), inset b. Summaries of these structural elements are provided below from Friend and Kinny (2001) and Kinny et al. (2005).

Nis Terrane

The smallest and most northerly of the terranes on the Outer Isles comprises amphibolite facies plutonic rocks such as anorthosite and diorite, separated from the Tarbet Terrane to the south by the Alasdair Shear Zone (Figure 32), inset b. These rocks are considered by Kinny et al. (2005) to be entirely Palaeoproterozoic, as constrained by whole-rock Sm-Nd data suggesting a protolith age of about 2.2 Ga for the Ness anorthosite (Whitehouse, 1990) with a 1.86 Ga Pb-Pb whole-rock isochron age constraining the timing of high grade metamorphism. This age of metamorphism is supported by zircon overgrowth formation in a diorite from the Butt of Lewis at 1859 ± 10 Ma (Whitehouse and Bridgwater, 2001), but the zircon core ages in this rock suggest that the protolith may be as old as 2.7-2.8 Ga.

Tarbert Terrane

The predominant rock types in this terrane comprise amphibolite-facies TTG gneisses with subordinate gabbro. The terrane is bounded to the north and south by the Alasdair and Langavat–Finsbay shear zones, respectively (Figure 32), inset b. Protolith zircon ages of 3125 ± 14 Ma (Friend and Kinny, 2001) and about 2770 ± 10 Ma (Pidgeon and Aftalion, 1972) have been recorded from the Scrap and Carlabhagh gneiss around the west coast of Harris. This was followed by the injection of basic dykes at 2140 ± 38 Ma (Cliff et al., 1998). The Tarbet Terrane was affected by amphibolite-facies metamorphism at about 1675 Ma, as constrained by zircon ages for granite sheet intrusion and neosome development in the Archaean TTG gneisses, particularly associated with the Harris Granite Complex (Fettes et al., 1992; Friend and Kinny, 2001).

Roineabhal Terrane

This terrane comprises the South Harris Complex (SHC), which is composed of the Langavat Belt in the north-east and the Harris Granulite Belt in the central and south-west part of the terrane. The Langavat Belt is itself composed of deformed amphibolite-facies metasediments and highly deformed metamorphosed igneous rocks (Mason et al., 2004b). The Harris Granulite Belt can be subdivided into a suite of variably retrogressed granulite-facies igneous rocks of anorthositic, noritic, gabbroic and dioritic composition, also known as the South Harris Igneous Complex (Baba, 1999; Dearnley, 1963; Mason et al., 2004a) and the Leverburgh Belt metasedimentary rocks (Cliff et al., 1998, Mason et al., 2004a). The terrane is separated from the Tarbet Terrane to the north by the Langavat–Finsbay shear zone and from the Uist/Barra Block to the south by the Ensay Shear Zone (Figure 32), inset b, the former shear zone initiated prior to 1645 Ma (Mason et al., 2004b). The Leverburgh and Langavat metasedimentary units include pelitic, psammitic and calcareous metasediments, (Mason et al. 2004b). The two belts are distinct in that they have different detrital zircon populations, indicating that they are not correlative (Mason et al., 2004b). Anorthosite, norite and diorite have yielded U-Pb zircon ages of 2491 +31/-27, 1890 + 2/-1, and 1888 ± 2 Ma, respectively (Mason et al., 2004a), with an associated tonalite dated at 1876 ± 5 Ma (Whitehouse and Bridgwater, 2001) and additional evidence of 1.8 Ga activity provided by 1.86 ± 0.05 Ga Sm-Nd model ages for the calc-alkaline rocks (Cliff et al., 1983). The Proterozoic diorite and norite intrusive rocks are interpreted to be associated with juvenile arc magmatism but the relationship of the older anorthosite bodies with these and the surrounding TTG country rock are unclear. Ultra-high temperature granulite-facies metamorphism particularly affected the Leverburgh unit and occurred throughout the terrane at about 1830 to 1870 Ma (Cliff et al., 1983, 1998; Baba, 1999; Whitehouse and Bridgwater, 2001).

Uist/Barra Block

This comprises Archaean gneisses to the south of the Ensay Shear Zone (Figure 32 inset b). Granulite facies TTG gneisses yielded a protolith age of 2834 ± 9 Ma and amphibolite facies rocks with ages of about 2770 to 2750 Ma (Whitehouse and Bridgwater, 2001; Love, 2004). To the east of the Outer Isles Fault, the Corodale gneiss yielded Sm-Nd and Pb-Pb whole rock isochron ages of 2770 ± 140 Ma and 2900 ± 100 Ma, respectively (Whitehouse, 1993). This granulite facies metamorphism is dated at 2730 Ma (Love, 2004). In the south of Uist, granitic intrusions have been dated at 1700 Ma (Love, 2004).

Outer Hebrides High — North-east Rockall Basin (Hebridean Terrane)

Lewisian-like rock samples have been recovered from the offshore area that have similar characteristics (e.g. nature and age) to metamorphic basement described from the Lewisian Gneiss Complex in north-west Scotland and the Outer Hebrides Isles. Within the study area, there are thirteen commercial wells, BGS boreholes and shallow drill sites that have recovered Lewisian basement from the Outer Hebrides High (including the Flannan, Sula Sgeir and West Lewis highs) and eastern margin of the North-east Rockall Basin (Figure 32) and (Table 5). Various lithologies have been described including quartzofelpathic, metabasic and other gneisses, amphibolite, schist, granitic pegmatite and granite (Figure 33) and (Figure 34); (Table 5) (Chambers et al., 2005). Generally, Lewisian rocks are considered to be restricted to the area to the north-east of the Anton Dohrn Lineament.

To the north and west of the Outer Hebrides, Archaean Sm-Nd model ages ranging between 3080 and 2800 Ma were recorded from well 154/03-1, BGS boreholes BH88/02, BH90/14 and BGS shallow drill sites 58-08/228, 57-09/536, 56-08/920 and 56-08/921 (Figure 32) and (Table 5). Approximately 40 to 50 km to the west of South Uist, an Archaean U-Pb age of 2709 ± 6 Ma has also been derived from BGS shallow drill site 57-09/537. To the south of this borehole is BGS shallow drill site 56-08/921 which yielded concordant to nearly concordant U-Pb ages of 2713 ± 5.2 and ± 2.9 Ma. To the north-east of 57-09/537 is BGS borehole BH90/14 which contained zircons that also indicated Archaean and Proterozoic growth episodes, at 2838 ± 15, 2767 +14/-15 and 1797 ± 8 Ma. One interpretation is that these rocks are essentially Archaean in age (56-08/921 has Pb isotope characteristics typical of Lewisian basement) but have suffered significant partial melting or high-grade metamorphism during Proterozoic times, with the latter hypothesis for 56-08/921 favoured on mineralogical grounds (Chambers et al., 2005). Further south, it is interesting to note that granite neosomes on the western side of South Uist have yielded ages of about 1700 Ma (Love, 2004).

A mid Silurian U-Pb age of 427.4 ± 0.3 Ma has been recorded from a Caledonian syenite recovered from BGS shallow drill site 58-08/230 on the Flannan High (Figure 32) and (Table 5). Also, approximately 110 km to the north-north-west of Lewis, U-Pb zircon and Sm-Nd model ages of 1633.5 ± 3.3 and 2110 Ma respectively, have been obtained from metabasic gneiss at granulite-facies grade recovered from well 164/25-2 on the West Lewis High. An ε_Nd value of -0.5 at the crystallisation age indicates a mantle-derived magma with only a modest involvement with older crust (Chambers et al., 2005). Although Proterozoic Sm-Nd model ages of about 1630 Ma have also been recorded from two granulite-facies metabasic intrusions within the Harris Granite Complex about 150 km to the south on Harris (Cliff et al., 1998), they are very different to the offshore metabasic gneiss, given the mainland gneisses have radically different Nd isotope characteristics suggestive of significant Archaean crustal input (ε_Nd -11.5 and -13.7, Cliff et al., 1998).

Rhinns Terrane

Rhinns Terrane basement was initially defined as comprising a complex of deformed late Palaeoproterozoic syenitic gneiss and gabbro that occurs on Colonsay, western Islay and Inishtrahull (e.g. Marcantonio et al., 1988; Muir et al., 1992; 1994). On Islay and Inishtrahull, the basement yielded Sm-Nd model ages of 1980 to 1910 Ma (Marcantonio et al., 1988) and 2030 to 1910 Ma (Daly et al., 1991; Dickin and Bowes, 1991) and U-Pb ages of 1782 ± 5 Ma (Marcantonio et al., 1988) and 1779 ± 3 Ma (Daly et al., 1991) respectively. The U-Pb ages are considered to represent igneous crystallisation ages (Muir et al., 1992). Similar U-Pb ages of about 1750 Ma have also been recorded for the Annagh Gneiss Complex of North Mayo, Ireland (Daly et al., 1995). The high μ = 7.9 (ratio of 238U/204Pb) value for Islay (normal mantle values of 8 are typical for Proterozoic rocks) compared to μ =7.3–7.5 for the Laxfordian granites, indicate that Rhinns Terrane rocks cannot be derived from reworking of older Lewisian rocks (Morton and Taylor, 1991). Rhinns basement is considered to represent juvenile material derived from a depleted mantle in Palaeoproterozic times. Its major and trace element geochemistry also suggests a magmatic arc setting (Muir et al., 1994).

Stanton High and Eastern Flanks of the South Rockall Basin

Approximately 110 km to the north-west of Islay, seven commercial wells, BGS boreholes and shallow drill sites are located on the Stanton High and eastern margin of the south Rockall Basin (Figure 32). A wide variety of rocks types have been described including a highly deformed cataclastic granite, monzonite, monzodiorite, gneiss and amphibolite, some of which have protomylonitic characteristics (Table 5). Sm-Nd data of the recovered samples yielded Archaean to Proterozoic model ages that range from 2980 to 2430 Ma (Table 5). U-Pb ages of 1794.4 ± 4.2 and 1797.8 ± 3.2 Ma were also recorded from BGS shallow drill sites 56-09/384 and 56-08/924 towards the north-west and south-east margins of the Stanton High. Scanlon et al. (2003) also report U-Pb age ranges of about 1830 to 1790 Ma from five unspecified sample locations on the high. These Proterozoic U-Pb ages are similar to those recorded from Islay and Inishtrahull, indicating that parts of the Stanton High have a strong affinity with the Rhinns Terrane. However, the variation in the Sm-Nd model ages when compared with Islay and Inishtrahull seems to suggest that there is a variable component of Archaean material (Table 5). For example, well 132/15-1 within the south Rockall Basin and BGS shallow drill sites 56-09/386 and 56-09/388 on the south-west flank of the Stanton High have Sm-Nd model ages that range between 2980 to 2750 Ma i.e. intermediate between Archaean and Proterozoic age, and thus are suggestive of mixed Archaean-Proterozoic sources. Their Pb isotope characteristics are also consistent with either derivation entirely from an Archaean source, or mixing of Lewisian Gneiss Complex and Rhinns Terrane-type sources (Chambers et al., 2005).

Rockall High

One BGS shallow drill site and five dive sites have recovered samples along a 130 km long tract of exposed metamorphic basement close to the crest or on the north-west flank of the Rockall High (Figure 32). Various lithologies have been described from these investigations including granite and mafic to granitic gneiss, with metamorphic grade up to granulite-facies. (e.g. two-pyroxene gneisses, (Table 5). Proterozoic Sm-Nd model ages determined from all the samples recovered from the Rockall High range between about 2140 and 1890 Ma (Table 5). These almost completely overlap with Sm-Nd model ages of 2030 to 1910 Ma for Islay and Inishtrahull (Marcantonio et al., 1988; Daly et al., 1991; Dickin and Bowes, 1991). A U-Pb age of 1744.9 ± 2.2 Ma was obtained from 56-15/18 (Chambers et al., 2005). This result, when compared with U-Pb data from Islay, Inishtrahull and the Stanton High, indicate that the granulite-facies metamorphosed intrusive rocks on the Rockall High crystallised later than mainland Rhinns lithologies, though the reasons for this are unclear. It is interesting to note that the age is also approximately coeval with the thermal disturbance on the mainland recorded by titanite growth in the Laxford area, interpreted as the age of docking of the Assynt and Rhiconich terranes along the Laxford Shear Zone (Kinny et al., 2005). Additional support for the hypothesis that the Rockall High forms part of the Rhinns Terrane is derived from a high μ = 8.3 from samples, comparing favourably with values of μ = 7.9 for Islay (Morton and Taylor, 1991). It should be noted that an Ar-Ar Grenville cooling age of 987 ± 5 Ma was obtained from Sample B (Miller et al., 1973).

Terrane correlations

There have been numerous correlations produced of Archaean and Proterozoic basement terranes observed within the Canadian maritime provinces, Greenland, Ireland, UK, Norway and Finland over the last few decades (e.g. Bridgwater et al., 1990; Winchester, 1988; Dickin, 1992; Park et. al 1994; Park, 1995; Baba, 1999; Buchan et al., 2000; Friend and Kinny, 2001; Park, 2005).

These correlations have been somewhat hampered by a lack of samples within the intervening offshore areas between the now-dispersed components of Laurentia and Baltica. Nevertheless, there is a common theme of Archaean to Palaeoproterozoic (~2–1.7 Ga) to late Palaeproterozoic (e.g. Gothian <1.7–1.5 Ga) terranes along the margin of a formerly amalgamated Laurentia–Baltica continent (c.f. Buchan et al., 2000; Park, 2005). A relatively recent terrane correlation between the Outer Isles and East Greenland has been made by Friend and Kinny, (2001). Limited drilling results in areas to the west and north of the Outer Hebrides are mainly inconclusive regarding the proposed correlation. However, well 164/25-2 (Figure 32) yielded a U-Pb age of 1633.5 ± 3.3 Ma which is comparable to ages of 1650 to 1620 Ma recorded for the Vyborg intrusive suite related to Gothian orogenesis in the Baltic Shield (Åhäll et al., 2000) and in the Labradorian Orogeny recorded in the Grenville Province of eastern Labrador, which involved terrane accretion and crustal thickening at 1650 to 1630 Ma (Gower et al., 1992). The most compelling correlation may lie with East Greenland, where a pre-north-east Atlantic opening refit suggests that well 164/25-2 is juxtaposed with the predominately Proterozoic Ammassalik Mobile Belt (an accretionary arc system) (Figure 35). Here, a post-tectonic suite of granites (that also includes gabbroic bodies) has yielded a U-Pb age of about 1680 Ma (Kalsbeek et al., 1993).

The Rhinns Terrane Proterozoic rocks present on Islay, Inishtrahull and the Stanton and Rockall highs are considered to form part of a single juvenile Palaeoproterozoic crustal province that also extends to include the Ketalidian Terrane of south-east Greenland (e.g. Dickin, 1992; Morton et al., 2009) (Figure 35) The Ketalidian Terrane is of broadly similar age to the Rhinns Terrane and comprises intrusions of the calc-alkaline Julienhab pluton at about 1854 to 1795 Ma followed by metamorphism and intrusion of I–type granites at about 1795 to 1785 Ma and the Rapikivi granite sheets at about 1755 to 1732 Ma (Garde et al., 2002).

Torridonian

Within north-west Scotland, Torridonian strata occur within a north-east-trending zone that extends for approximately 200 km from the Cape Wrath Peninsula in the north to the island of Rum in the south (see Park et al., 2002) (Figure 32). The Torridonian succession is subdivided into the Stoer (oldest), Sleat and Torridon groups, that are mainly comprised of relatively unmetamorphosed red to grey fluviatile sandstone up to 2 km, 3.5 km and 5 km thick, respectively (see Park et al., 2002). The Torridon Group is considered to have been deposited as an orogen-parallel foreland basin to the Grenville Orogeny (Krabbendam et al., 2008). Radiometric age-dating indicates Neoproterozoic depositional ages of 1200 and 1000 Ma for the Stoer and Torridon groups, respectively (Turnbull et al., 1996; Rainbird et al., 2001).

Within the offshore area immediately adjacent to the eastern margin of the study area, undifferentiated Torridonian red sandstone, arkose, siltstones and mudstones have been proved in BGS borehole BH72/11 within the Sea of Hebrides–Little Minch Basin and in borehole BH78/03 and well 156/17-1 in the North Minch Basin (Figure 32), inset b. Well 156/17-1 bottomed after drilling 130.9 m of Torridonian strata. The Torridonian is inferred to have widespread distribution between north-west mainland Scotland and the Outer Hebrides Isles, with up to 6 km possibly developed within the hanging wall of the Minch Fault (Stein, 1988; 1992), Within the study area however, there are no known occurrences of Torridonian, although a deep seismic refraction, wide-angle reflection experiment conducted within the northern part of the Rockall Basin by Roberts et al., (1988) suggested that a high-velocity sediment layer (5.3 km/s) close to the eastern margin of the basin could conceivably be caused by the presence of deeply buried Torridonian sediments.

Chapter 4 Cambrian to Carboniferous

Martyn Stoker and Alick Leslie

Rocks of Cambrian to Carboniferous age have not been recovered in the report area, although Upper Palaeozoic strata have been postulated to occur within the Rockall Basin. In contrast, Palaeozoic rocks are preserved on adjacent areas of the UK and Irish Atlantic continental margin. It is a common assumption on most palaeogeographical reconstructions for the Palaeozoic interval that the area west of Britain was a largely emergent ‘source’ area floored by an amalgamation of basement terranes (e.g. Ziegler, 1988; Roberts et al., 1999; Coward et al., 2003) (Figure 36) and (Figure 37). However, it has been shown that basement rocks on the Rockall Plateau may be covered by a substantial thickness of ?Palaeozoic and younger strata (Keser Neish, 1993; Jacob et al., 1995; Shannon et al., 1999; Hitchen, 2004). Thus, as the structural configuration and geological history of the Atlantic margin deep-water basins remains poorly understood, so the possibility remains that Palaeozoic strata may be present within the report area. This chapter presents a brief summary of what is known about the Cambrian to Carboniferous development in the vicinity of the report area.

The Palaeozoic record preserved adjacent to the report area includes rocks of Cambro-Ordovician, Siluro-Devonian and Carboniferous age (Figure 38) and (Figure 39). The Cambro-Ordovician rocks are part of the Hebridean basement terrane of the cratonic foreland province (Figure 35), and accumulated prior to the Caledonian orogeny. Middle Silurian to lowest Carboniferous strata — generally termed the Old Red Sandstone — represent post-orogenic molasse deposits (Trewin and Thirlwall, 2002). An increasing deltaic to marine influence is reflected in later early Carboniferous deposits, whilst fluviodeltaic facies comprise the upper Carboniferous rocks (Coward et al., 2003).

The Caledonian orogenic cycle spanned the Cambrian to earliest Devonian interval (Ziegler, 1988), which was dominated by continental collision and plate accretion as the continents of Laurentia, Baltica and Avalonia ultimately converged during Mid/Late Ordovician to late Silurian times (Strachan et al., 2002; Coward et al., 2003) (Figure 36). In Cambro–Ordovician times, the report area was located on the south-east passive margin of Laurentia. A thick clastic–carbonate succession that accumulated on the Hebridean Terrane was part of a contiguous sequence that extended from north-east Greenland to the southern Appalachians of North America (Park et al., 2002). Although a large part of the British Isles was accreted on to the Laurentian margin during the Caledonian orogeny, the effects of deformation were minimal in the Hebridean Terrane. There is little sign of the subsequent history of the foreland province preserved in the rock record until the development of latest Palaeozoic–Mesozoic extensional basins along the flanks of the Rockall Basin, though the possibility of Devono-Carboniferous basin development in the report area cannot be discounted (e.g. Coward et al., 1989; Naylor et al., 1999). In the Devonian to early Carboniferous, strike-slip tectonism together with the effects of orogenic collapse, resulted in the development of pull-apart basin systems, such as the Great Glen–Møre–Trøndelag and Midland Valley–Solund fault systems between Scotland and Norway, and extending south-west to Ireland (Coward et al., 2003 and references therein) (Figure 37). According to Coward et al. (1989), this pull-apart system may have extended into the foreland province where the North Lewis and North Minch basins, controlled by the extensional reactivation of the Outer Isles thrust, are inferred to be areas of Old Red Sandstone basin development. Tectonic inversion during the late Carboniferous resulted in large inversion-related fold structures, especially in the Midland Valley of Scotland and east Shetland region (Coward et al., 1989; Roberts et al., 1999; Ritchie et al., 2003b). This may have been associated with north-west to south-east directed compression related to Variscan tectonics in north-west Europe (Coward et al., 2003). Sedimentation was predominantly terrestrial (Figure 37)c.

In a palaeogeographical context, the report area lay south of the equator for much of the Cambrian to Carboniferous interval. During the Early Palaeozoic, north-west Scotland experienced a tropical climate just south of the equator (Park et al., 2003). In the Early Devonian, northern Britain lay about 30º south of the equator, moving to 20º south at the end of the period (Trewin and Thirlwall, 2002). Deposition of the Old Red Sandstone occurred under semi-arid conditions. At the beginning of the Carboniferous, the region lay in low southern latitudes on the fringes of the southern arid climatic belt, crossing the equator and experiencing humid tropical conditions in the late Carboniferous, though by the end of this period northern Britain occupied a position about 10º north of the equator, and had reverted to a semi-arid climate (Anderton et al., 1979; Read et al., 2002).

For the purpose of this report, the Palaeozoic is described with regard to the Cambro–Ordovician, Old Red Sandstone (mid Silurian to earliest Carboniferous) and Carboniferous successions. Information is presented both from mainland Scotland and the Atlantic continental margin.

Cambro–Ordovician

Mainland Scotland

A Cambro-Ordovician succession crops out in north-west Scotland and extends 250 km from Loch Eriboll on the northern Scottish coast to Skye, east of the report area (Figure 38). The succession consists of approximately 1 km of sedimentary rocks dated from early Cambrian to Mid Ordovician (Llanvirn) in age (Figure 39). These rocks unconformably overlie and onlap Torridonian and Lewisian strata in the west, whilst to the south-east the succession is truncated by the Moine Thrust and associated thrust sheets (Park et al., 2002).

The sedimentary succession is divided into two groups. The Eriboll Group consists of about 250 m of quartzites and clastic sedimentary rocks and is overlain by 750 m of dolostones and limestones that comprise the Durness Group (Wright, 1985; Park et al., 2002). The development of a carbonate shelf was favoured by the tropical equatorial climate, stable passive margin and apparent low run-off from land. This marine shelf succession appears to be conformable from early Cambrian to Mid Ordovician, though it is uncertain whether deposition was continuous throughout this interval. It has been proposed that time gaps may be present in the sequence, possibly associated with a phase of uplift and erosion during the mid to late Cambrian (Palmer et al., 1980). Although mid and late Cambrian faunas have not been recognised, a barren 200 m-thick section may cover this interval. Thus, there is no accepted evidence for a major unconformity within this succession (Nicholas, 1994; Trewin and Rollin, 2002).

Farther east, the Dalradian Supergroup of the Grampian Highland Terrane, part of the Caledonian orogenic fold belt (see Precambrian basement chapter and (Figure 35)), extends south-west from the Aberdeenshire and Moray coast across the Scottish Highlands for 300 km to Kintyre and Islay, and beyond into Ulster. Although the majority of the Dalradian Supergroup is Precambrian in age it is probable that the upper part of the Southern Highland Group forms a deeper-water equivalent to the Cambro-Ordovician rocks in north-west Scotland (Daly, 2001; Strachan et al., 2002).

Offshore

There are no proven occurrences of Cambro-Ordovician rocks within or immediately adjacent to the report area, although Brewer and Smythe (1984) have inferred their presence beneath sediments of the West Orkney Basin for up to 80 km north of the Scottish mainland, and extending some 60 to 70 km east of the Moine thrust zone. Strong similarities between the Cambro-Ordovician successions in north-west Scotland and western Newfoundland (Park et al., 2002 and references therein) suggest that their occurrence at depth within the report area cannot be discounted.

Old Red Sandstone (Mid Silurian to earliest Carboniferous)

The Old Red Sandstone (ORS) represents a facies characterised by distinctive continental red beds. It is mostly of Devonian age, and Lower, Middle and Upper ORS subdivisions are broadly equivalent to Lower, Middle and Upper Devonian marine sequences of southern Britain and continental Europe. However, the base and top of the ORS cannot generally be defined in Scotland. The lower limit of ORS facies is probably as old as the mid Silurian (Marshall, 1991), whilst the top of the ORS facies is diachronous across the Devonian–Carboniferous boundary (House et al., 1977) (Figure 39).

Mainland Scotland

There are two main areas of preserved ORS deposition. These are the Orcadian Basin area that extends from the Moray Firth region to Shetland, and the Midland Valley (Trewin and Thirlwall, 2002) (Figure 37)a and (Figure 39). The earliest ORS sediments were deposited in the Midland Valley during the Silurian as the pull-apart basin system was initiated, and resulted in the accumulation of a thick succession of Middle Silurian to Lower Devonian alluvial sandstones, lacustrine shales and volcaniclastic sediments and lavas (Marshall, 1991). In the Orcadian Basin, deposition started slightly later (in the mid Early Devonian) with the emplacement of sandstones and conglomerates of alluvial and aeolian origin, and lacustrine mudstones in numerous half-graben sub-basins. In excess of 1000 m of Lower Old Red Group sediment accumulated in both basins (Marshall and Hewett, 2003).

In the Mid Devonian, there is no evidence of Middle ORS deposition in the Midland Valley. Indeed, Lower ORS sediments were folded and the area was uplifted and eroded (Trewin and Thirlwall, 2002). North of the Highland Boundary Fault, the Lower ORS sub-basins coalesced into a single lacustrine-dominated sedimentary system that was the main Mid Devonian development of the Orcadian Basin (Marshall and Hewett, 2003). Lacustrine facies largely consist of playa-lake deposits with subordinate laminated siltstones rich in organic matter. The Achnacarras Fish Bed, deposited about halfway through the Mid Devonian, represents an important stratigraphical marker, both onshore and in offshore wells. The lacustrine environment was terminated by a return to fluvial and aeolian sandstone deposition in the later part of the Mid Devonian. The Middle Old Red Group in the Orcadian Basin exceeds 1500 m in thickness (Marshall and Hewett, 2003).

During latest Mid Devonian and Late Devonian times, the Orcadian Basin evolved to a more-open drainage system, and the dominant sedimentary process became fluvial with an inundation of coarse-grained sandstones (Stuart et al., 2001). The southward spread of the sandstones was diachronous, and Upper ORS sediments were not deposited in the Midland Valley until the later part of the Late Devonian (Marshall and Hewett, 2003). In the Midland Valley, the Upper Old Red Sandstone Group rests with angular unconformity on the Lower Old Red Sandstone Group. The uppermost part of the Upper Old Red group can be as young as Carboniferous in age, as ORS facies sedimentation in the Midland Valley was eventually terminated by a marine transgression from the east in the early Carboniferous (Trewin and Thirlwall, 2002) (Figure 37)b. The Upper Old Red Group in both the Orcadian Basin and Midland Valley exceeds 800 m in thickness (Marshall and Hewett, 2003).

Offshore

Although there are no proven occurrences of ORS deposits in the report area, it should be noted that the westward and southward extent of ORS basins remains unknown. Outcrops of probable Lower ORS breccias, conglomerates and fluvial sandstones, adjacent to major faults, occur in the Loch Linnhe–Lorne region of western Scotland (Stoker, 1982; Trewin and Thirlwall, 2002 and references therein) and at Ballymastocker Bay in north-west Ireland (Pitcher et al., 1964; Graham, 2001). Furthermore, the Lower ORS is unconformably overlain by Upper ORS in the Machrihanish Basin on Kintyre (Stephenson and Gould, 1995) (Figure 38). These occurrences indicate that the presence of ORS sediments on the southern Hebrides Shelf and Irish Shelf cannot be discounted, particularly adjacent to major faults.

To the north of Scotland, a predominantly Middle Devonian succession between 3000 and 5000 m thick underlies much of the Orkney–Shetland Platform (Bott and Browitt, 1975). This sequence forms the western edge of the Mid Devonian Orcadian Basin, and the sediments consist of fluviatile sandstones and lacustrine siltstones and mudstones (Stoker et al., 1993). To the north-west of the Orkney–Shetland Platform, Middle to Upper ORS strata are preserved on the Rona High, which separates the West Shetland Basin and the Faroe–Shetland Basin Complex. These sediments are termed the Clair Group and are remnants of the Clair Basin, a Mid Devonian to early Carboniferous depocentre (Blackbourn, 1987; Allen and Mange-Rajetzky, 1992; Nichols, 2005) (Figure 37)a and (Figure 39). Offshore wells have proved in excess of 900 m of fluviatile conglomerates and sandstones, aeolian sandstones and lacustrine mudstones. Heavy mineral assemblages indicate provenance areas to include northern Scotland, Scandinavia and Greenland (Allen and Mange-Rajetzky, 1992), the latter perhaps implying that the area north of the Rockall region may have been largely a source area (Trewin and Thirlwall, 2002). Visean miospores at the top of the Clair Group, together with the presence of acritarchs, indicates a marine influence (Meadows et al., 1987) consistent with the termination of the ORS facies by an early Carboniferous marine transgression (as noted above).

Reworked Devonian microfossils have been found in sediments ranging from Late Jurassic to Palaeogene in age to the north and west of Shetland, suggesting a more widespread occurrence of Devonian strata than is currently proven (Hitchen and Ritchie, 1987). Interpretation of seismic reflection data from the Faroe–Shetland Basin led Duindam and van Hoorn (1987) to propose that Devono–Carboniferous sediments, analogous to the Clair Basin, may be preserved on other basement highs, such as the Westray High. This was subsequently confirmed by well 213/23-1, drilled on the Corona High, which proved 314 m of Middle to Upper ORS sandstones, siltstones and claystones (Smith and Ziska, 2011) (Figure 38) and (Figure 39).

Coward et al. (1989) considered that the Orcadian Basin was part of a much larger, linked, basin system that extended farther west and included the West Orkney, North Minch and North Lewis basins adjacent to, and within the north-eastern part of, the report area (Figure 37) and (Figure 38). A thickness of between 3000 and 4000 m of Devonian infill, beneath Permian and Mesozoic sediments, has been estimated for the West Orkney Basin (Coward and Enfield, 1987; Enfield and Coward, 1987). Earle et al. (1989) further support a Lower Devonian to Lower Carboniferous component to the West Orkney Basin succession, albeit restricted to the eastern part of the basin. To date, this interpretation remains speculative as well data and adjacent onshore exposures (along the north-west Scottish coastline) reveal Torridonian and Permo–Triassic strata, but no Devonian sediments (cf. Fyfe et al., 1993 and Stoker et al., 1993 and references therein). Farther west, in the report area, seismic refraction data from the Rockall Basin have been interpreted as indicating that Upper Palaeozoic (Devonian to ?Permian) deposits may form a basal sediment wedge along the eastern margin of the basin (Roberts et al., 1988).

Carboniferous

The Carboniferous stratigraphical terminology utilised below retains the western European regional series and stage nomenclature, which remains prevalent throughout the Scottish Carboniferous literature. However, in order to connect with the newly developing international classification of the Carboniferous (e.g. Gradstein et al., 2004) correlation between regional and global terminology is indicated in (Figure 40). Essentially, Mississippian and Pennsylvanian have replaced the terms lower Carboniferous (Dinantian) and upper Carboniferous (Silesian), respectively, though it should be noted that the Mississippian/Pennsylvanian boundary is not coincident with the Dinantian/Silesian boundary. In terms of stage names, Tournaisian and Visean are the only stages to be retained in the new classification.

Mainland Scotland

Marine sedimentation did not significantly affect northern Britain until the late Visean (mid early Carboniferous) when ORS facies were replaced by a more diverse assemblage of marine, fluvial, deltaic and continental sediments (Bruce and Stemmerik, 2003). In Scotland, Carboniferous deposits are largely confined to the Midland Valley, which is part of a larger depocentre that extends north-east into the Forth Approaches and Moray Firth regions, and south-west into Ireland (Glennie, 2000; Read et al., 2002) (Figure 37)b. Predominantly grey fluviodeltaic and shallow-marine sediments comprise the bulk of a 1500 m-thick upper Dinantian to Middle Namurian succession in the Midland Valley, overlain by up to 600 m of Westphalian Coal Measures (Cameron and Stephenson, 1985) (Figure 39). Additionally, up to 900 m of basaltic lava was erupted during the Carboniferous, mostly during the early to mid Visean, becoming more localised in the Namurian and Westphalian. No Stephanian sediments or volcanics are preserved in the Midland Valley (Bruce and Stemmerik, 2003).

Further to the north-west, in the Northern Highlands, Carboniferous strata occur in the Morvern Peninsula where about 100 m of Upper Carboniferous sandstone with thin mudstone bands and coal seams, belonging to the Westphalian Coal Measures, are preserved at Inninmore Bay (Richey, 1961; Johnstone and Mykura, 1989). Other Carboniferous rocks in the Highlands occur to the south and include the Pass of Brander near Loch Awe, and the Machrihanish Basin in Kintyre (Figure 38). In the latter region, the sequence consists of Dinantian, possibly Visean, basalts, overlain by Namurian sandstones, mudstones, limestones and coals with interbedded volcanics, in turn overlain by Westphalian clastics including several coal seams (Fyfe et al., 1993) (Figure 39). This basin fill is about 700 m thick. All these outliers most probably represent small basins of accumulation that developed as the main basin of the Midland Valley intermittently extended into the bordering Highland massif (Francis, 1983). To the south-west, Visean to Namurian sediments and volcanics, up to 1000 m thick, infill the Rathlin Trough in Northern Ireland, which may be linked to the Machrihanish Basin (Fyfe et al., 1993).

Offshore

Glennie (2000) has suggested that it is possible that deposition took place adjacent to, and within, the report area, but that late Carboniferous (Variscan) uplift and erosion resulted in the removal of much of the Carboniferous succession on the north-west UK margin. As noted above, a residual outlier of Lower Carboniferous strata is preserved in the Clair Basin, west of Shetland (Figure 37)b and (Figure 39), where 27 m of Visean marine siltstones and thin sandstones containing acritarchs were proved in well 206/8-2 (Meadows et al., 1987; Allen and Mange-Rajetzky, 1992) (Figure 38). More recently, a 142 m-thick succession of sandstone, siltstone and claystone assigned to the Visean, overlain by 120 m of Namurian conglomerate, sandstone and coal, was proved in well 213/23-1 from the Corona High (Smith and Ziska, 2011) (Figure 38) and (Figure 39).

Reworked Carboniferous miospores are found in Permo-Triassic rocks in the North Lewis and West Orkney basins. These include Lycospora spp., which has a Tournaisian to late Westphalian age range (Neves et al., 1973). A similar flora has been recovered in sea-bed samples east of the Outer Hebrides (Eden et al., 1973; Owens and Marshall, 1978; Chesher et al., 1983), and Carboniferous erratics have been found on the Outer Hebrides (Jehu and Craig, 1934). The occurrence of so much reworking implies a formerly more extensive distribution of Carboniferous strata in the Hebridean region, though its identification as part of the deeper basin fills in this area remains ambiguous. In the Sea of Hebrides–Little Minch Basin (Figure 38), up to 500 m of Carboniferous sediment has been suggested on seismic evidence (Stein, 1988), although drilling has not proven a Carboniferous succession (Fyfe et al. 1993). In the North Minch Basin, for example, well 156/17-1 (Figure 38) penetrated a Torridonian succession overlain by Triassic sandstone.

On the Irish margin, the Donegal, Erris and Main Porcupine basins preserve a record of upper Carboniferous sedimentation and volcanism (Tate and Dobson, 1989; Stoker et al., 1993). In the Donegal Basin (Figure 38) and (Figure 39), well 13/3-1 penetrated 949 m of upper Carboniferous (Westphalian to Stephanian) deltaic to marginal marine sediments overlying, and variably metamorphosed by, an undated tholeiitic gabbro. Common volcaniclastic intervals in the upper 60 m of the section are Stephanian in age, and are broadly contemporaneous with the late Carboniferous–early Permian volcanism of southern Scotland and north-east Ireland. Farther south, well 19/5-1 (Figure 38) proved over 1400 m of Visean to Westphalian sedimentary rocks in the Erris Basin. This is comparable with the adjacent Mayo Basin, which is inferred to contain up to 1 sec TWTT thickness of Carboniferous strata (Klemperer et al., 1991; Naylor et al., 1999). In general terms, the Erris and Main Porcupine basins preserve a succession of upper Tournaisian to lower Namurian deposits unconformably overlain by upper Namurian to Stephanian sediments.

Significantly, Upper Carboniferous rocks have recently been proven in the adjacent Rockall Basin, where well 12/2-1Z (Figure 38) penetrated 154.5 m of interbedded sandstone, claystone and coal of Westphalian age. The sandstone is dark grey to pale olive grey, fine- to medium-grained and very well sorted; the claystone is varicoloured dark brownish red, greenish grey and dark grey, silty and slightly dolomitic. Beds and seams of black, hard, fissile coal are present towards the base of the section. Stringers of pale to dark brownish grey argillaceous limestone are common in the upper part of the sequence. This discovery lends some credence to previous suggestions, based largely on the interpretation of seismic refraction and reflection data, that pre-Permian Upper Palaeozoic strata may underlie many of the Mesozoic basins developed along the Atlantic seaboard, including the Rockall–Hatton region (Roberts et al., 1988; Naylor et al., 1999; Naylor and Shannon, 2005).

The recognition of coherent, relatively unfolded reflection packages, commonly subparallel to the overlying Mesozoic (Permo-Triassic to Cretaceous) beds, has been used to infer the occurrence of Upper Palaeozoic strata along the faulted western margin of the Rockall Basin, immediately south of the report area, in the Rónán and Conall basins (Naylor et al., 1999; Naylor and Shannon, 2005) (Figure 38). By analogy, this may also apply to the East Rockall High Basin, which extends into the south-western part of the report area, though Hitchen (2004) favours a Permo-Triassic to Cretaceous infill. This is comparable with basins in the north-east part of the area, such as the West Lewis Basin where well 164/25-1 terminated in Permo-Triassic strata towards the base of the infill (Tate et al., 1999). However, the age of the oldest part of the basin fill remains to be proved.

In the south-east part of the report area, a deep seismic refraction line running east–west across the Rockall Basin was interpreted by Roberts et al. (1988) to reveal a sediment package at depth within the basin fill, on the eastern margin of the basin, with average interval velocity of 5.3 km/s. Although this layer was interpreted by Roberts et al. (1988) to consist of ‘late Palaeozoic, early Mesozoic or fractured Torridonian’ sedimentary rock, well 132/15-1, approximately 39 km south of the transect line (Figure 38), proved Cretaceous sediments overlying Precambrian basement. Nevertheless, rocks infilling half-graben style basins on the Hatton High, west of the report area, display similar acoustic velocities (5.1–5.3 km/s) and have been tentatively interpreted to indicate ‘older’ (Upper Palaeozoic to Early Mesozoic) sedimentary rocks (Keser Neish, 1993; Hitchen, 2004). A comparable interpretation has been made from the RAPIDS seismic profile that was acquired across the Hatton Basin in Irish waters (Shannon et al., 1994, 1995, 1999). Significantly, perhaps, the Carboniferous miospore Lycospora spp. has been found in easterly prograding deltaic wedges of Early to Mid Eocene age on the east side of George Bligh High (Stoker, 1995a) (Figure 38), which implies a derivation from the west, including the Rockall Plateau (Rockall High–Hatton Basin–Hatton High). Thus, the possibility cannot be discounted that Carboniferous strata may be present at depth in the Rockall–Hatton region.

Chapter 5 Permo-Triassic

Howard Johnson and Martyn Quinn

By early Permian times, the Variscan Ocean, to the south of the report area, had almost completely closed, resulting in the grouping together of most of the world’s continents into the supercontinent of Pangaea (Figure 10)b. However, the supercontinent was unstable (Ziegler, 1990) and consequently continental assembly and the beginnings of breakup were virtually simultaneous. Following a phase of regional uplift and peneplanation, two rift systems developed and dominated north-west Europe during Permo-Triassic times (e.g. Roberts et al., 1999) (Figure 10)b and (Figure 11). The ‘Arctic’ rift system propagated southwards towards western Ireland, reworking many of the Caledonian or earlier contractional structures. The second, northward propagating rift system, termed the ‘Atlantic’ rift, developed initially by collapse of the Appalachian and European Variscan mountain belts. Reconstructions showing the former extent of the Permo-Triassic rift basins within and immediately adjacent to the report area vary. For example, whereas Coward et al. (2003) (Figure 10)b and (Figure 11) and Ziegler (1990) postulate the development of an extensive rift throughout the Rockall Basin during Permian and Triassic times, Doré et al. (1999) restrict Permo-Triassic basin development mainly to the Erris Trough, Sea of the Hebrides and West Shetland areas (Figure 41).

Štolfová and Shannon (2009) reviewed Permo-Triassic basin development from Ireland to Norway, and noted that Permo-Triassic basin fills, despite being erosional remnants, are of sufficient extent and thickness to illustrate considerable variation in large-scale basin geometries. They postulate that basin geometries and orientations are controlled by basement rheology (e.g. crustal thickness, composition and temperature), inherited basement structures and pre-rift palaeotopography. A range of large-scale basin architectures has been recognised including:

• localised small Permian half-graben basins in the Celtic Sea region, which developed in response to the early gravitational collapse of the Variscan Mountains
• extensive, uniformly thick Triassic strata characteristic of basins in the Celtic Sea, the Irish Atlantic margin and parts of the UK margin, interpreted as the results of wide rift extension
• large-scale fault-controlled depocentres, best illustrated in the Norwegian and parts of the inner UK Atlantic margin, and interpreted as the products of narrow rift extension. The wide rift basins are thought to have developed where crust was thicker but generally weaker. Narrow rifts probably developed where the crust was thinner but stronger.

During the early Permian, northern Britain drifted out of an equatorial position and into the northern tradewind belt (Ziegler, 1990). Between the late Permian and the Late Triassic, the report area moved from about 15°N to approximately 35°N (Lovell, 1991) and an arid, desert climate prevailed throughout most of Permo-Triassic times (Ziegler, 1990). The Permo-Triassic sediments were deposited at a time of overall eustatic regression, and are represented almost entirely by terrestrial red beds. These were derived from adjacent uplifted areas, including the Scottish mainland, which remained a mostly stable, upland area (Lovell, 1991; Glennie, 2002). Coarse-grained, water-lain breccias and conglomerates were deposited adjacent to active fault scarps and as valley-fill accumulations in the deeply dissected landscape (Steel and Wilson, 1975). Away from the upland areas, sandstones, siltstones and mudstones were deposited in fluviatile, lacustrine and aeolian environments. Whether the Hebrides and west Shetland areas were partly submerged by Arctic Seas during the late Permian Zechstein transgression is debated (Ziegler, 1988, 1990; Glennie, 2002), but fully marine conditions were only established over large areas of the UK during the latest Triassic Rhaetian transgression, which advanced from the south (Anderton et al., 1979). Leleu and Hartley (2010) reviewed the tectono-stratigraphical development of central and north Atlantic Triassic basins, including the Sea of the Hebrides–Little Minch Basin and Inner Hebrides Basin. They recognised that an upward transition from fluvial to playa or lacustrine deposits occurs in all of the basins, but this transition is diachronous. They interpret the lack of synchronicity to indicate that the transition reflects a decrease in source area relief related to a decline in regional tectonic activity, rather than to eustatic control.

Thick Permo-Triassic continental red-bed successions are widely distributed beneath the Hebrides Shelf and West Shetland shelf regions (Stoker et al., 1993) and to the west of Ireland (e.g. Naylor et al., 1999, 2002) (Figure 42). However, the present-day distribution is considered to represent a remnant of a previously more extensive Permo-Triassic sedimentary cover. Within the Rockall Basin, the occurrence of Permo-Triassic strata remains largely a matter of conjecture and is mainly based on plate reconstructions (e.g. Cole and Peachey, 1999) and the interpretation and modelling of deep seismic and other geophysical data (e.g. Nadin et al., 1999; Shannon et al., 1999). However, Irish well 12/2-1 and associated sidetrack well 12/2-1Z, (Figure 42) and (Figure 43) were drilled in 2002 and 2003, respectively, and discovered gas condensate (known as the ‘Dooish’ discovery) within a thick Permian sandstone reservoir close to the southern margin of the report area and within the eastern flank of the south Rockall Basin.

Offshore western Scotland, terrestrial strata assigned to either the Permian or the Triassic are generally indistinguishable. They consist predominantly of locally conglomeratic sandstones, siltstones and mudstones. Within these successions, biostratigraphical evidence is extremely sparse and rarely enables even the broadest subdivision between deposits of Permian and Triassic age. The term ‘New Red Sandstone’ has therefore continued to be applied to these successions of predominantly terrestrial facies in the Scottish region (Warrington et al., 1980). More marine-influenced, Permian evaporite-bearing successions are known from offshore Ireland, for example in the Slyne Basin (e.g. Dancer et al., 2005; Chapman et al., 1999), and offshore Norway (e.g. Roberts et al., 1999), and the Slyne Basin is also known to contain Triassic evaporites (Corcoran and Mecklenburgh, 2005). In the Slyne Basin, the Triassic succession drilled within the Corrib Gas Field has been informally subdivided into an upper mudstone-dominated Mercia Mudstone Group, including a thick halite unit at its base, and an underlying sandstone-dominated Sherwood Sandstone Group (Dancer et al., 2005).

Due to sparse well control and limited biostratigraphical resolution, the timing of Permo-Triassic extension within the north-east Atlantic margin region is generally poorly dated. In East Greenland a major phase of normal faulting culminated in the mid Permian and further block faulting took place in the early Triassic (Surlyk, 1990). In contrast, in the Irish Sea area, major rift phases are postulated in early Permian and Early to Late-Triassic times (Chadwick et al., 2001). Within the report area, the West Lewis, Flannan and West Flannan basins are considered to represent mainly Permo-Triassic rift-graben, on the basis of the local imaging of Permo-Triassic sedimentary wedges with divergent internal seismic reflection patterns (Kirton and Hitchen, 1987; Stoker et al., 1993; Hitchen et al., 1995a; Doré et al., 1999). However, the preservation of a parallel-bedded Permo-Triassic section in the North Rona Basin, to the north-east of the report area, suggests that this is a downfaulted remnant of previously more extensive deposits (Kirton and Hitchen, 1987). Štolfová and Shannon (2009) assert that UK Atlantic margin basins show a combination of local initial palaeotopographical infill, narrow half-graben synsedimentary fault-controlled geometries and wide, uniformly thick sedimentary fill with no obvious synsedimentary controls. Close to the Scottish coast, controlling faults dip eastwards and are probably reactivated Caledonian and/or Proterozoic structures.

Igneous rocks of Permian and Triassic age are unknown within the report area. However, lavas thought to be of late Permian age (250 Ma) have been drilled by well 12/2-1Z in the Irish Rockall Basin. In the oil company final well report, the lavas are reported to conformably overlie the sediments below and comprise several thin, stacked flows of pale olive grey to medium dark greenish-grey basalt with calcite-filled vesicles. The top of the basalt unit is weathered and brecciated. The lavas are flow-textured with chilled margins and contain distinct layers of different sized vesicles that are mostly calcite and clay mineral cemented. The lavas are now highly altered with olivine phenocrysts preserved as clay pseudomorphs.

During the Carboniferous, and continuing into the Permian, widespread volcanism occurred across the Midland-Valley region of central Scotland, and a number of vents, sills, dykes and lavas were emplaced here during the early Permian (Read et al., 2002; Glennie, 2002). Locations of this activity include Arran, the small island of Glas Eilean, between Islay and Jura, where alkali-olivine basalts were reported by Upton et al. (1987), and onshore Northern Ireland in the Larne No. 2 borehole (Penn et al., 1983). Offshore, to the north and west of Shetland, thin early Permian lava successions were drilled in three wells (205/27a-1, 210/4-1 and 220/26-2) (Hitchen et al., 1995a) and in a fourth well (210/13-1) fragments of Permian lava were contained in a Triassic sandstone (Hitchen and Ritchie, 1987). The early Permian volcanism was probably related to the first phase of post-Variscan continental disintegration (Ziegler, 1988). To the south of the report area, volcaniclastic sandstones containing fragments of basic igneous rocks have been drilled by Irish wells 12/13-1A and 19/5 1 in the Erris Basin (Figure 42), and by Irish well 26/22-1A in the North Porcupine Basin. Tate and Dobson (1989) consider these occurrences to represent an episode of extrusive igneous activity probably during the later part of the Triassic, although this cannot be proved directly. Igneous activity in Scotland ceased by the Zechstein and there is no onshore occurrence of Triassic igneous rocks.

Basins on the Hebrides Shelf

West Lewis Basin and adjacent areas

The greatest drilled thickness of presumed Permo-Triassic strata within the report area is a 664.5 m-thick succession, which includes two thin dolerite units, in well 164/25-1Z, near the western margin of the West Lewis Basin (Figure 42) and (Figure 43). Indeed, much of the Mesozoic and Paleocene succession is intruded by dolerite sills of presumed Paleogene age. The basin itself has been interpreted as a westerly tilted half-graben (Earle et al., 1989). The Permo-Triassic strata are buried beneath about 3000 m of Cenozoic and Cretaceous deposits, and comprise thick units of predominantly reddish brown, very fine- to coarse-grained, variably micaceous, lithic and subarkosic sandstone and argillaceous sandstone interbedded with reddish brown, locally anhydritic siltstone and mudstone. Although the Permo-Triassic succession within the West Lewis Basin is largely obscured beneath an extensive cover of Palaeogene basalts, seismic reflection data from the eastern margin of the basin indicate a north-westerly dipping succession of unknown thickness resting unconformably on the Lewisian of the Sula Sgeir High. This succession was sampled by BGS borehole BH90/01, which penetrated up to 11 m of undated red-brown to grey-green, poorly sorted, fine- to coarse-grained sandstone and muddy sandstone, which was probably deposited in a fluviatile environment. It is in part massive, but also displays cross-bedding. Carbonate nodules of probable pedogenic origin are common. Similarly, BGS boreholes BH78/05 and BH90/17, to the south of the West Lewis Basin on the adjacent Outer Hebrides High, penetrated an undated succession of probable Permo-Triassic age comprising up to 26 m of undated red, yellow and grey, very fine- to fine-grained sandstone, with scattered carbonate nodules, interbedded yellow-brown to dark brown and grey-green slightly micaceous siltstone and mudstone and a few plant fragments. The sandstone has been described as an immature arkose comprising quartz, feldspar and some rock fragments. Cementation is limited, resulting in many friable beds with porosity from four samples ranging from 13 to 37 per cent. This is mainly primary, intergranular porosity enhanced by leaching of accessory minerals and secondary dissolution of sodic feldspars.

West Flannan Basin

Up to 4000 m of mainly westerly dipping, Permian and Mesozoic deposits are estimated to be present in the largely basalt-covered West Flannan Basin (Figure 42). The bulk of the sediments are considered to be of Permo-Triassic age (BGS Lewis and Geikie solid geology sheets) and were probably deposited in a half-graben analogous to the West Lewis Basin (Roberts, 1989). Close to the eastern margin of the basin, BGS borehole BH88/04 penetrated 8 m of undated, white to greenish-grey, poorly sorted, fine- to very coarse-grained and slightly pebbly sandstone, interpreted to be of fluviatile origin.

Flannan Basin

The Flannan Basin, to the west of Lewis (Figure 42), is considered to form a mainly Permo-Triassic rift-graben (e.g. Stoker et al., 1993; Hitchen et al., 1995a; Doré et al., 1999) or Triassic rift-graben (Spencer et al., 1999). Results from seismic refraction work suggest that it contains 0.5 km to 1.7 km of sediments (with velocities of about 4.4 km/s) probably of mainly Permo-Triassic age (Jones, 1981; Powell and Sinha, 1987). This is in broad agreement with thickness estimates of 0.5 to 1.5 km derived from the results of potential field modelling (Figure 16)a. BGS borehole BH88/08, drilled in the western part of the basin, encountered 3 m of reddish brown, fine-grained, micaceous sandstone with intraformational mudstone clasts. A fluvial origin has been proposed for the sandstone (Hitchen et al., 1995a).

North Lewis Basin

The North Lewis Basin is situated to the west of the Minch Fault north of Lewis (Figure 42). It consists of several westerly tilted, half-graben sub-basins containing over 2500 m of Permo-Triassic sediments, the base of which occurs at about 4 to 5 km depth (BGS, 1990). The maximum thickness of the Permo-Triassic succession occurs adjacent to contemporaneously active faults (Kirton and Hitchen, 1987), although later fault movements have also affected the succession. Outside the report area, the southern part of the basin extends onshore at the east coast of Lewis, where the infill comprises the Stornoway red-bed succession. This consists of thick, red-brown conglomerates, with subordinate sandstones, rare siltstones, and incipient cornstones. The succession was deposited as a series of overlapping alluvial fans along the foot of actively retreating scarps (Steel and Wilson, 1975; Johnstone and Mykura, 1989). Smith (1976) suggested that the succession is of lower Permian aspect, but palaeomagnetic work (Storetvedt and Steel, 1977) indicates a late Permian to Triassic age. The nature of the offshore sediments in the North Lewis Basin is less well known. From the continuation of the onshore outcrop of the Stornoway Formation, west of the Minch Fault, BGS borehole BH72/32 drilled 8 m of red, poorly cemented, fine-grained sandstone, possibly deposited in a floodplain environment (Hitchen et al., 1995a). Red calcareous mudstone was recovered from the northern part of the North Lewis Basin in borehole BH72/36 (Figure 42). A sparse miospore assemblage from borehole BH72/36 is indicative of a latest Triassic to Jurassic age (Stoker et al., 1993).

Barra Basin and adjacent areas

The Barra Basin is an isolated, fault-bounded outlier within Lewisian basement on the southern part of the Outer Hebrides High (Figure 42). BGS shallow seismic profile show folded, predominantly westerly dipping, parallel-layered reflections suggesting that postdepositional faulting is responsible for preservation of these strata, which formerly may have been more widespread over the Outer Hebrides High (BGS, 1992). BGS borehole BH90/16 penetrated 13 m of red sandstone and pebbly sandstone in stacked, upward-fining cycles; carbonate nodules also occur. No biostratigraphical age has been obtained for these sediments. The basin is thought to contain a minimum of 300 m of Permo-Triassic sediments that are unconformably overlain by less than 100 m of Cenozoic sediments (BGS, 1992, cross-section 1).

Further west on the Outer Hebrides High, rocks unconformably overlying Lewisian basement have been mapped as Permo-Triassic, cropping out in a thin north south strip (BGS, 1992). However, the results of subsequent BGS drilling have shown that these sediments, which underlie Paleocene lavas, are of Paleocene age (Hitchen et al., 1995a).

Erris Basin

The Erris Basin is situated in Irish waters just south of the report area. It forms a Mesozoic half-graben, probably underlain by Upper Palaeozoic strata, perched on the eastern flank of the Rockall Basin (Naylor et al. 1999). Although the total thickness of Permo-Triassic sediments in the basin is unknown, well 12/13-1A drilled 686 m of Upper Permian to Upper Triassic strata, mainly comprising reddish brown mudstone and sandstone units, with minor anhydrite, dolomite, limestone and scarce volcaniclastic detritus (Tate and Dobson, 1989). The sedimentation probably occurred in terrestrial settings, including low-lying playa lakes and sabkhas (Stoker et al., 1993).

Rockall Basin and North-East Rockall Basin

As noted above, Irish well 12/2-1 and sidetrack well 12/2-1Z (Figure 7) were drilled on a tilted fault-block prospect termed ‘Dooish’, located within the eastern flank of the south Rockall Basin. The wells drilled condensate-bearing Permian reservoir rocks which, together with the underlying Westphalian sediments, are the oldest reported sedimentary rocks to have been drilled within the Rockall Basin as a whole. Within the North-east Rockall Basin, the oldest sedimentary rocks drilled are Albian strata encountered in well 164/7-1 (Figure 7).

Wells 12/2-1 and 12/2-1Z drilled a tilted pre-rift succession that is sealed by dominantly argillaceous post-rift Turonian mudstone and younger rocks. A unit of high-gamma (sodic feldspar-rich) yellowish brown, greenish grey, pale grey and reddish brown arkosic sandstone is recognised in both wells and lies directly below the post-rift succession. In well 12/2-1Z, this 171 m thick unit mainly comprises sandstone and is biostratigraphically barren, but is tentatively regarded as Permian on the basis of a radiometric age date (reported to be 250 Ma) from intercalated extrusive basalt. The arkosic sediments are interpreted as distal sheet flood and aeolian deposits that formed in a semi-arid sandflat setting. Soil processes and the formation of calcrete have modified much of the sediment.

A major change in sediment character and composition at 4087 m in well 12/2-1Z suggests an unconformity and probably a significant time gap. At this stratigraphical break, a 228 m thick unit of ‘clean sandstone’ underlies the high-gamma sandstone in both wells. This unit dominantly comprises pale brownish grey, friable, mainly fine- to medium-grained, texturally more mature, quartz-rich sandstone with variable amounts of pinkish orange feldspar, interbedded with numerous beds of varicoloured greenish grey, greyish black and occasionally reddish brown mudstone with desiccation cracks. A mostly dry, terrestrial, aeolian sandflat depositional setting is envisaged and, in contrast to the high gamma sandstone unit above, soil processes seem to have been relatively minor, possibly reflecting either a more rapid rate of deposition or conditions too harsh for significant vegetation. The clean sandstone unit is assigned an early Permian (Asselian) age on the basis of the co-occurrence of the miospores Potoniesporites spp. and Cordaitina spp. between 4138 and 4255 m in 12/2-1Z (Millennia Limited, 2004). The first indications of Carboniferous occur at 4327 m within 12/2-1Z, with the occurrence of Endosporites globiformis and Calamospora breviradiata, which indicate a Westphalian C age.

The presence of thick Palaeogene lavas and sills within the North-east, and north and south Rockall basins degrades the definition of the deep structure. Consequently, estimates of the nature, age and distribution of the pre-Palaeogene stratigraphical intervals are largely inferred from velocity data, which is derived mainly from the results of wide-angle seismic reflection/refraction experiments (e.g. Roberts et al., 1988; Keser Neish, 1993; Shannon et al., 1994; O’Reilly et al., 1995; Shannon et al., 1999) (see Structure chapter). The succession in the main part of the Rockall Basin has been interpreted as Jurassic to Recent in age, but is commonly assigned largely to the Cretaceous and Cenozoic. However, ‘perched’ basins along the flanks of the Rockall Basin, and best imaged within the Irish sector, contain several kilometres of undrilled, pre-Cenozoic sedimentary rocks, interpreted as Permo-Triassic to Early Cretaceous in age (Naylor et al., 1999; Morewood et al., 2004, 2005). The presence and distribution of Permo-Triassic strata on the western flanks of the north Rockall Basin and within the Hatton Basin areas remains very uncertain.

Chapter 6 Jurassic

By Dan Evans

Direct evidence for the presence of Jurassic rocks in the report area is restricted to a few boreholes on the Hebrides Shelf (Figure 44). Their wider distribution is not well defined from any other available data source. However, Jurassic strata occur extensively to the east in the Hebridean region despite significant postdepositional erosion that has influenced present-day distribution patterns (Fyfe et al., 1993). The possibility of their existence in the Rockall Basin and its flanks is the subject of debate (see also Chapter 2).

During the Jurassic Period the report area lay at a latitude of about 37°N and its western part lay at the confluence of the Boreal (?Arctic) rift system that separated Norway and Greenland and extended towards western Ireland, and the Central Atlantic rift zone that may have extended into the southern Rockall Trough/Porcupine Basin during the Triassic (Doré et al., 1999; Coward et al., 2003). Early Jurassic sedimentation was post-rift infilling of the Triassic rifts, including the North Porcupine Basin (Naylor et al., 1999) and the Slyne and Erris basins (Tate and Dobson, 1989; Dancer et al., 1999; Naylor et al., 1999) to the south of the report area.

A generally emergent Scottish landmass during the Jurassic included most of the Highlands and extended northwards to an occasionally emergent Shetland Platform (Hudson and Trewin, 2002). This landmass probably had low relief at the start of the Jurassic and was well vegetated, but marine conditions generally existed between the Scottish landmass and the Hebrides Platform that had variable extent (Hudson and Trewin, 2002). The Hebrides Platform is bounded to the east by the Minch Fault that approximates to the likely location of its eastern coastline during Jurassic times (Hudson and Trewin, 2002). This platform may have covered part or much of the present-day outer shelf, with a seaway in the Rockall rift to the west. To the west of the Rockall rift, the Rockall Plateau and adjacent areas may have been land, and is shown as such by Coward et al. (2003), although its western extent is uncertain.

Coward et al. (2003) and Roberts et al. (1999) suggest that Liassic sedimentation may have been widespread across Britain as a result of marine transgressions at around the start of the Jurassic that brought about a change from terrestrial to marine facies over extensive areas, and elements are preserved in the Hebridean region where they are locally thick (Fyfe et al., 1993; Hudson and Trewin, 2002).

During the Mid Jurassic, the palaeogeography was broadly similar to Early Jurassic times, but there was recurrent uplift of the Scottish landmass leading to thick accumulations of relatively coarse-grained sediments (Hudson and Trewin, 2002). Local tectonism was an important control on sedimentation in the Hebridean region (Trueblood and Morton, 1991), and depositional environments were commonly brackish (Hudson, 1963). There was inversion and erosion in the area west of Shetland and in the Hebrides (Haszeldine and Russell, 1987; Morton, 1989), and to the north there was thick sedimentation in the Faroe Rift that is considered to be deep-marine facies (Haszeldine and Russell, 1987). To the south of the report area Middle Jurassic strata are widely developed throughout the Porcupine Basin and a thick, shallow marine to estuarine succession is found in the Slyne Basin (Naylor et al., 1999). Towards the end of this time the seas in the region were becoming more extensive in response to rising sea levels that continued in Late Jurassic times (Hudson and Trewin, 2002).

During the Late Jurassic the North Sea graben system opened to the east of Scotland (Fraser et al. 2003), and rising sea levels led to a slightly more restricted land area, but the major land masses continued to be emergent (Hudson and Trewin, 2002). Argillaceous sedimentation was dominant during the Oxfordian to Kimmeridgian in Skye, and also west of Shetland (Hitchen and Ritchie, 1987), although there was some fault-controlled coarse-grained sedimentation in the Faroe–Shetland Basin (Haszeldine and Russell, 1987). There is general agreement that the Rockall Basin region was probably by now a seaway linking the Arctic and Tethyan seas, and some authors have suggested that widespread Late Jurassic rifting in north-west Europe may have included faulting in the Rockall Basin leading to the accumulation of deep-marine sediments (Roberts et al., 1999; Coward et al., 2003; Naylor and Shannon, 2005).

Basins on the Hebrides Shelf

Jurassic strata have been sampled only in the West Lewis and Flannan basins (Figure 45), but their presence in some other basins remains a possibility.

West Lewis Basin and adjacent areas

To the north-west of the Sula Sgeir High, north-westerly younging Mesozoic rocks crop out in a small area to the east of the extensively overlying Palaeogene and Neogene strata (Hitchen and Stoker, 1993; BGS, 2007). Isaksen et al. (2000) show these rock units dipping north-westwards towards the buried West Lewis Ridge, although the Jurassic section pinches out before reaching the ridge. Jurassic rocks on this eastern flank of the West Lewis Basin have been sampled in three boreholes.

BGS borehole BH88/01 drilled 15.8 m of massive to faintly laminated, dark grey, very hard mudstone with traces of shell debris, and borehole BH90/02 encountered 12.5 m of similar mudstone with sparse siltier bands. Biostratigraphical evidence implies a lacustrine or brackish water depositional environment. Although originally designated as Aalenian to Bathonian on the basis of miospore assemblages (Stoker et al., 1993), Hitchen and Stoker (1993) have redefined the ages of both borehole successions as Bathonian based on comparison with lagoonal and deltaic palaeo-environments from the Hebridean region. Total organic carbon analysis of the mudstones ranges from 3.40 to 7.48 per cent, and Isaksen et al. (2000) indicate that although they are presently immature, these rocks would generate oil upon maturation.

Nearby, in the same area of Mesozoic outcrop, BGS borehole BH88/03 proved 17.2 m of greenish grey interbedded sandstone, pebbly sandstone and conglomerate. The sequence is massive to cross-bedded and is considered to be fluviatile in origin. No biostratigraphical dating has been possible, but the presence of comminuted plant tissue has been taken as an indication of a Mid Jurassic age, and Hitchen and Stoker (1993) suggest, on lithological grounds, that a latest Bajocian to Bathonian age is most likely, although an earlier age cannot be ruled out.

West Flannan Basin

Much of the West Flannan Basin is covered by Palaeogene lavas that extend as far east as the Flannan Ridge against which they are faulted, but near the northern end of the ridge there are uncovered rocks of Permo-Triassic to Cretaceous age to the west of the ridge (Hitchen and Stoker, 1993). In this westerly dipping succession, BGS borehole BH90/08 proved 8.8 m of mainly fine-grained muddy sandstone with organic debris and thin coaly bands. The sediments were deposited in a nearshore marine environment between late Oxfordian and Tithonian times. The sub-basalt extent of the Jurassic is unknown.

Other basins

Regional mapping and borehole evidence suggests that Jurassic rocks do extend a short distance into the report area north of Lewis in the North Lewis Basin but that only Permo-Triassic rocks are present in the Flannan and Barra basins (Stoker et al., 1993; BGS, 2007; BGS, PAD, 2002). Near the southern margin of the report area, Irish well 12/13-1A has tested a predominantly argillaceous Hettangian to Sinemurian succession in the Donegal Basin, but the interpretation shown on the BGS Central Rockall Basin Map would indicate that these rocks do not extend into the report area.

The eastern flank of the Rockall Basin and much of the North-east Rockall Basin include extensive Palaeogene or older lavas that commonly obscure the underlying strata on seismic records. On the western flank of the basin immediately to the south of the report area in the Rónán Basin and also farther south in the Conall Basin, the presence of Jurassic rock is proposed by Naylor et al. (1999) on the basis of seismic interpretation and regional considerations. Naylor et al. (1999) also propose the presence of Jurassic strata in the Erris and Slyne basins on the eastern flank of the Rockall Basin in the Irish sector, and Jurassic strata have been proven in the Slyne Basin (Scotchman and Thomas, 1995; Scotchman, 2001) as well as farther south in the northern Porcupine Basin (Spencer and MacTiernan, 2001).

With regard to the Rockall Basin, Nadin et al. (1999) and Cole and Peachey (1999) proposed the need for a pre-Cretaceous rift basin, and Naylor and Shannon (2005) suggested that there was Late Jurassic rifting here. This implied that Jurassic rocks might be expected in the basin but it was unproven. However, in 2008 Shell drilled Irish well 12/02-2 (see (Figure 7) for its location) targeted at a large tilted fault block within the basin (the West Dooish prospect). Jurassic shales were penetrated in this well (Geir Jansson, pers. comm.) before the well was terminated at 4456 m depth. This discovery confirms that Jurassic rocks occur at depth within the Rockall Basin although their total extent remains speculative.

Chapter 7 Cretaceous

Kevin Smith

Evidence of the Cretaceous geology of the Rockall Basin is scanty. Deep boreholes are rare and much of the late Mesozoic geological history of the report area remains conjectural. Several hydrocarbon wells, together with shallow stratigraphical boreholes drilled by BGS, provide a partial geological framework for the Cretaceous rocks of the adjoining platform areas, but the deep geology of the centre of the basin is poorly constrained.

This chapter will initially describe geophysical evidence for the geological development of the Rockall Basin during the Cretaceous. It will go on to outline the likely Cretaceous stratigraphy of the basin and its flanks based partly on the sparse borehole data, and partly on regional comparisons with neighbouring parts of the North Atlantic area, where knowledge of the Cretaceous interval has benefited from the results of prolonged hydrocarbon exploration.

During the Cretaceous, the process of continental rifting extended northwards into the report area (Figure 46). Different interpretations of the regional structure and stratigraphy depend largely on whether or not this rifting process culminated in Cretaceous sea-floor spreading within the Rockall Basin (see discussion in Structure chapter). A well-developed pattern of magnetic stripes, in the form of alternating linear areas of normally and reversely magnetised basaltic rocks, usually provides the best evidence for the presence of oceanic crust at depth. The observed absence of magnetic stripes within the Rockall Basin could be interpreted as an indication that the underlying crust is of continental origin. However, it is known from studies elsewhere in the world that the mid Cretaceous interval was almost devoid of magnetic reversals. Sections of mid Cretaceous oceanic crust are usually uniformly magnetised and define a magnetic ‘quiet zone’, which lasted approximately from 118 to 84 Ma (Early Aptian to latest Santonian: Driscoll et al., 1995; Wold, 1995). This crust can be difficult to distinguish magnetically from continental crust. On the basis of magnetic evidence alone, the floor of the Rockall Basin could be continental, or alternatively consist of oceanic crust created entirely within the mid Cretaceous. Seismic experiments designed to distinguish between these alternatives have yet to reach a robust conclusion.

Some early geophysical studies did infer that the Rockall Basin was underlain by oceanic crust of possible Cretaceous age (Scrutton, 1972, 1986; Bott, 1978; Roberts, 1975a; Roberts et al., 1981). An alternative suggestion that the basin in the UK sector was underlain by highly stretched continental crust (Talwani and Eldholm, 1972; 1977) was supported by Roberts et al. (1988), although Smythe (1989) subsequently criticised the quality of the wide-angle seismic reflection and refraction data on which this interpretation was based. Joppen and White (1990) agreed that the seismic velocities and velocity gradients obtained by Roberts et al. (1988) were too poorly defined to discriminate unequivocally between continental and oceanic crust, but suggested that the thickness of the crust in the northern part of the Rockall Basin did probably favour a continental origin. While the total crustal thickness in the centre of the northern Rockall area does exceed that of normal oceanic crust, the diagnostic significance of thickness is undermined by the wide range of oceanic crustal thicknesses that are developed within the north-east Atlantic area.

In their interpretation of deep seismic data from the Irish sector, Joppen and White (1990) conclude that the southern part of the Rockall Basin consists of highly stretched and thinned continental crust. They suggest that this crust is so pervasively intruded by basic igneous rocks that it forms a transitional crustal type between continental and oceanic. Summarising Joppen and White’s (1990) interpretation prior to its publication, Smythe (1989) termed this form of crust ‘quasi-oceanic’ and pointed out that the classification of true oceanic crust must then depend upon the presence of a critical percentage of oceanic igneous rock. In the case of the Rockall Basin, the possibility that the continental crust has been stretched by up to six times its original length suggests that a very high proportion of the basement must consist of new igneous rock.

If the Rockall crust is of continental origin and has thinned by stretching, it is significant that it shows little sign of faulting and is characterised by parallel seismic reflectors (the ‘acoustic layering’ of Joppen and White, 1990). Typically, the overlying sedimentary succession is also largely devoid of faults. These observations could be explained if the acoustic layering formed by extension of continental basement before the overlying sediments were laid down.

Re-examination of the seismic velocity data from the expanding spread profile ESP-12 (Joppen and White, 1990), suggests that there is little difference between the crust of the southern Rockall Basin and true oceanic crust. Joppen and White (1990) themselves observe that both the thickness of the crust in this area, and its velocity profile, are fairly typical of normal oceanic crust elsewhere, and can be readily distinguished from the stretched continental crust of the Atlantic margin in the Bay of Biscay (Whitmarsh et al., 1986). It seems likely that isotopic evidence of contamination by continental crust obtained from some igneous samples (Morton et al., 1988), the absence of a firm indication of oceanic crust in the form of magnetic stripes, and the pre-existing interpretation of Roberts et al. (1988), prompted Joppen and White (1990) to conclude that continental crust is present throughout the basin.

While Joppen and White’s (1990) proposal that the Rockall Basin is underlain by crust that combines oceanic and continental characteristics probably represents the present consensus, the evidence from deep seismic data still remains consistent with the possibility that Cretaceous oceanic crust is developed locally at the centre of the Rockall Basin. This possibility was envisaged by Megson (1987, fig. 1) based on an initial interpretation of the seismic data subsequently described by Joppen and White (1990) and has been re-considered more recently by Chappell and Kusznir (2005).

In the Irish sector, the WESTLINE deep seismic profile across the Rockall Basin shows clear evidence of tilted blocks close to the flanks of the basin, but the structure of a 140 km wide central section of the profile below 6 s TWTT is largely unresolved and provides no indication of the top of the crust or its composition (England and Hobbs, 1997; Corfield et al., 1999). It is possible that much of the basement in this part of the area is of oceanic origin. If the inferred northward increase in crustal thickness does reduce the chances of it being oceanic, then a potential boundary may exist that separates the oceanic basin from an area of thinned continental crust to the north. The Anton Dohrn Lineament, which is characterised by significant changes in geology and bathymetry, marks the most likely location of this potential transition along the axis of the basin.

If part of the floor of the Rockall Basin is composed of oceanic crust, it follows from the absence of magnetic stripes that it was likely to have been created during the Cretaceous Normal Quiet interval, which lasted from early Aptian to end Santonian times. Near the end of the Santonian, at the time of magnetic anomaly 34 (~82–84 Ma), sea-floor spreading was initiated to the south of the Rockall Basin between Orphan Knoll and Porcupine Bank (Driscoll et al., 1995). Since anomaly 34 is absent within the Rockall Basin, any episodes of localised spreading to the north of the Charlie Gibbs Fracture Zone must have ended by the start of the Campanian.

New structural models of lithospheric tectonics based on the process of extreme continental stretching (or hyperextension) may help to resolve the prolonged dispute about the continental or oceanic origin of the Rockall Basin. These models, which emphasise the role of mid-crustal detachment surfaces, mantle exhumation and associated serpentinisation in the evolution of the continent–ocean transition, have been developed largely from seismic investigations and drilling results. These were obtained from magma-poor conjugate continental margins, such as Western Iberia–Newfoundland in the North Atlantic (Whitmarsh et al., 2001; Pérez-Gussinyé et al., 2001; Lavier and Manatschal, 2006; Reston, 2007; 2009a, b; Péron-Pinvidic and Manatschal, 2010).

If the Rockall Basin originated by hyperextension, its ‘acoustically layered basement’ (Joppen and White, 1990) may be composed entirely of sheared and altered lower crust, following the removal of the brittle upper and middle crustal layers along low-angle detachment surfaces that dip towards the basin margins. The top basement surface beneath the middle of the basin would then correspond to one or more flat-lying fault planes, which are in tectonic contact with an overlying, essentially unstretched, basin fill. If stretching took place predominantly during the Early to mid Cretaceous, then the basin fill would consist solely of post-rift Cretaceous and younger sediments. In these circumstances, pre-rift sediments can be removed completely from the basin centre along gently dipping zones of omission and be preserved only on the basin flanks, as the eroded cover of widely-spaced, highly rotated blocks.

The main difference between the processes of hyperextension and oceanic opening lies in the origin of the mantle lithosphere underlying the basin. In the case of hyperextension, the mantle consists of partially uncovered pre-existing continental lithosphere, whereas during oceanic opening it forms largely from uplifted, cooling asthenosphere. Thinning the continental crust by hyperextension enables downward percolating fluids to serpentinise the underlying mantle through dense fault networks. This process provides an alternative explanation (instead of igneous underplating) for the anomalously low seismic velocities observed at uppermost mantle level beneath parts of the Rockall Basin (O’Reilly et al., 1995).

Although structural models based on extreme continental stretching have been widely adopted, some authors have acknowledged that the Rockall Basin remains slightly anomalous geophysically. Pérez-Gussinyé et al. (2001) suggest that the section interpreted as crust in southern Rockall area may actually consist of exhumed serpentinised mantle in places. On the other hand, given its velocity structure (Joppen and White, 1990), the possibility of partial continental separation accompanied by the local formation of oceanic crust cannot yet be entirely ruled out. Reston (2009a) strongly advocates hyperextension, but still includes the Rockall Basin in a list of marginal areas where crustal separation may have taken place, or was at least imminent.

Seismic interpretation of the sedimentary fill of the Rockall Basin has been complicated by the inconsistent use of reflector nomenclature. Regional seismic reflectors were originally correlated, assigned a colour identifier and provisionally dated by Masson and Kidd (1986), but the various modifications and applications of this scheme have introduced a series of errors, some of which are catalogued by Smythe (1989). Joppen and White’s (1990) seismic velocity model of the Rockall Basin within the Irish sector is particularly relevant to the interpretation of the Cretaceous fill of the basin. These authors correctly identified a near-top Paleocene event, but by mistakenly correlating it with the ‘brown’ event of Masson and Kidd (1986), introduced a redefinition of the age of that reflector. Stoker et al. (2001) re-labelled Masson and Kidd’s (1986) ‘brown’ reflector C30 (Cenozoic 30) and confirmed their original suggestion that it is of Late Eocene to Early Oligocene age. Underlying the brown event, there is a widespread high amplitude reflector, which was not assigned a colour by Masson and Kidd (1986). This event is commonly taken to correspond to the top of the Balder Formation and is equivalent to the earliest Eocene or ‘top Paleocene’ event (R50) of Smythe (1989). It is this deeper reflector that forms the misidentified ‘brown’ event of Joppen and White (1990).

Masson and Kidd (1986) did not recognise a seismic reflector corresponding to the top of the Cretaceous, but described a ‘blue’ intra-Cretaceous event, as possibly equivalent to the top Campanian. In subsequent analyses of seismic data from the Rockall Basin, the ‘blue’ reflector label was assigned to a prominent seismic reflector underlying the Balder Formation by Smythe (1989), Joppen and White (1990), Shannon et al. (1993) and England and Hobbs (1997), but whether they are all describing the same event is difficult to establish. It is possible that some of the reflectors that are designated as ‘blue’ by these authors actually correspond to either the top of the Cretaceous or the base of Paleocene volcanic rocks.

Although basic intrusions are widely developed in parts of the Mesozoic succession, well data from the Faroe–Shetland Basin indicate a preferential intrusion of early Cenozoic sills close to the Cretaceous–Cenozoic boundary. In the absence of well control, the convention widely adopted by seismic interpreters in the Faroe–Shetland Basin is to pick the top Cretaceous below the Balder Formation, at the top of an interval that consists predominantly of igneous sills. In the Rockall Basin, much of the Cretaceous reflectivity can be attributed to the presence of basic sills. In well 164/07-1, in the northern part of the Rockall Basin, Paleocene sediments rest unconformably upon a truncated Cretaceous succession capped by Coniacian mudstones. Nearly half the proven Cretaceous interval in this well consists of sills of probable Palaeogene age (Figure 47) (Archer et al., 2005).

Published seismic reflection profiles and seismic velocity data from the Rockall Basin do not provide firm constraints on the thickness of the Cretaceous sedimentary succession.

Joppen and White (1990) propose that the interval between the ‘blue’ reflector and their ‘acoustically layered basement’ comprises both Cretaceous and older sediments. In the Irish sector, several different attempts have been made to subdivide this sedimentary succession on the basis of seismic velocity. On the RAPIDS deep seismic profiles, the sediments have been divided into three main intervals. The upper unit consists of early Cenozoic to Recent sediments. The middle interval, which has an average seismic velocity of approximately 3 km/s, is thought to include both Cretaceous and Paleocene strata (Shannon et al., 1993; 1995). It is, however, the composition of the underlying, deepest sedimentary layer that is critically important for the structural and stratigraphical history of the whole Rockall Basin during the Cretaceous. The seismic velocity of this layer was originally defined as 4.1 km/s (Makris et al., 1991; Shannon et al., 1993). This was subsequently increased to an average seismic velocity of 4.5 km/s (Shannon et al., 1994; 1995; Jacob et al., 1995; Hauser et al., 1995), while O’Reilly et al. (1995) describe the same interval as having an even higher velocity (ranging from 4.5 to 4.7 km/s). In the absence of well control, the variety of ages proposed for these sediments has included; Upper Palaeozoic to Cretaceous (Makris et al., 1991); Upper Palaeozoic to Lower Cretaceous (Shannon et al., 1993); largely Permo-Triassic to Jurassic, with Upper Carboniferous strata also likely to be present (Shannon et al., 1994; 1999); late Palaeozoic through to Jurassic, with the bulk of the layer being Jurassic (Shannon et al., 1995); late Palaeozoic to Jurassic (Jacob et al., 1995); Jurassic and older (O’Reilly et al., 1995) and largely Permo-Triassic through Jurassic (Hauser et al., 1995). Some of these correlations include Cretaceous sediments while others do not. In their interpretation of the equivalent interval, Joppen and White (1990) show a layer of basic igneous sills with a seismic velocity of 5.2–5.6 km/s interleaved with sediments whose seismic velocity increases with depth from 3.3 to 4.2 km/s. From their interpretation, it is clear that it is the presence of the sills that raises the average seismic velocity of the deepest sedimentary interval to the value of 4.5 km/s adopted by Shannon and his co-workers. If the seismic velocities related to igneous rocks are excluded, the remaining values of 3.3 to 4.2 km/s are entirely consistent with the velocities and densities of Cretaceous and Cenozoic successions buried at these depths (Kimbell et al., 2002; Archer et al., 2005). Although the presence of pre-Cretaceous sediments cannot be ruled out completely, it is clear that current velocity profiles do not provide unequivocal support for the presence of Jurassic and older strata along the axis of the Rockall Basin (England and Hobbs, 1997; Musgrove and Mitchener, 1996). In contrast, the deep seismic data combined with the results of the velocity analysis indicate that much of the section presently identified as Jurassic and older on the RAPIDS seismic data by Shannon and his co-workers is equally likely to be composed of Cretaceous sediments intruded by sills of Palaeogene age. The potential absence of older sedimentary strata, with the Cretaceous interval resting directly on high seismic velocity basement, is consistent with the possibility that the southern part of the Rockall Basin is floored either with oceanic crust of mid Cretaceous age, or exhumed and altered continental lower crust, resting upon a thin layer of serpentinised uppermost mantle.

In the remainder of this chapter the Cretaceous is subdivided chronologically into Early and Late Cretaceous intervals. The actual distribution of Upper and Lower Cretaceous strata within the report area remains largely conjectural (Figure 48), but it is likely that there is a strong structural control. It is proposed that Lower Cretaceous sediments are mostly preserved in half-graben structures on the flanks of the main Rockall Basin. These structures demonstrate the onset of Cretaceous extension in the Rockall area. Analogous Cretaceous half-grabens are found in many of the marginal basins around the North-east Atlantic, including the Faroe–Shetland Basin, the Erris Basin in the Irish sector (Chapman et al., 1999) and the Jeanne D’Arc Basin (Driscoll et al., 1995).

The regional tectonic history of the Rockall area (see Structure chapter) suggests that a period of moderate extension and half-graben formation probably culminated in a major episode of continental stretching during the mid Cretaceous. Consequently, in the centre of the southern Rockall Basin, Upper Cretaceous strata may rest directly upon high velocity basement rocks (England and Hobbs, 1996), in an area where localised sea-floor spreading may have developed. The symmetrical nature of the later Cretaceous and Cenozoic basin fill, its lack of faulting and pattern of overlap on the basin flanks are consistent with deposition in a tectonic episode dominated by thermal subsidence, following mid Cretaceous extension.

Stratigraphy

Lower Cretaceous

The stratigraphy of the Rockall Basin at the beginning of the Cretaceous is of crucial importance for its future hydrocarbon potential. There is some evidence from drilling on the flanks of the basin that local deposition of the oil-prone Kimmeridge Clay Formation persisted from the Jurassic into the earliest Cretaceous. Elsewhere, within the main part of the basin, the current preservation of the Kimmeridge Clay Formation is largely dependent on the nature and amount of later, mid Cretaceous extension.

In block 165/23 of the West Lewis Basin, the BGS shallow borehole BH90/05 proved 21 m of hard, dark grey to black, massive to poorly bedded mudstone. The mudstone is organic-rich with thin bands and fragments of lignite, together with plant remains. A Berriasian age is suggested by the presence of dinoflagellate cysts such as Batioladinium pomum, B. radiculatum and Egmontodinium torynum, while other wider-ranging species, such as Endoscrinium pharo, Gochteodinia villosa and Tubotuberella apatela are broadly consistent with a Berriasian to Valanginian age.

The BGS shallow borehole BH90/09, in block 154/17 in the West Flannan Basin, also proved a Berriasian succession that consists of 11.5 m of grey, fine- to medium-grained, arkosic sandstone overlying 9 m of dark grey to black, organic-rich mudstones. The proportion of terrestrial organic debris in these sediments implies a relatively nearshore location (Hitchen and Stoker, 1993). The mudstones are thermally immature, but contain a mixture of sapropelic and humic kerogen, which has the potential to be a good source for gas and oil (Isaksen et al. 2000). The Berriasian age is indicated by the dinoflagellate cysts Gochteodinia villosa and Batioladinium radiculatum, and deposition probably took place in an open marine environment, with the sporadic development of anoxic conditions.

Although Roberts et al. (1999) and Naylor and Shannon (2005) suggest that similar oil source beds were deposited throughout the Rockall Basin, and that the anoxic rift basin extended northwards into the Faroe–Shetland Basin, the predicted distribution of similar rocks within the Rockall Basin depends largely on the selected mode of basin formation. If oceanic crust of mid Cretaceous age is present within the centre of the basin, Jurassic and earliest Cretaceous sediments will clearly be absent, since that part of the basin floor had yet to form when those sediments were laid down. In the alternative models of highly extended crust, the floor of the Rockall Basin must be composed of isolated fault-bound blocks, with the pre-existing sedimentary cover separated by wide zones of omission. After stretching, the area of the basin floor covered by the zones of omission must be 4 to 6 times the area covered by the original sedimentary cover. In Joppen and White’s (1990) interpretation of the basin, any pre-stretching sediments are probably envisaged to form part of the ‘acoustically layered basement’, but with seismic velocities ranging from 5.7 to 6.3 km/s, the proportion of sediments in this interval with typical Cretaceous seismic velocities must be very small.

At the southern margin of the Rockall Basin, the Barra Volcanic Ridge System is thought to have been initiated in Valanginian–Barremian times (Bentley and Scrutton, 1987; Scrutton and Bentley, 1988) and compares closely with the Porcupine Median Volcanic Ridge, which flanks the Clare Lineament in the middle of the Porcupine Basin (Tate, 1993). These features demonstrate how Early Cretaceous extension and associated lithospheric thinning may have controlled linear zones of magma generation or deep crustal detachment within these basins, even if the presence of oceanic crust is not firmly established. The transverse orientation of the Barra Volcanic Ridge System suggests that an element of longitudinal stretching may have been involved in the extension of the Rockall Basin at this time.

At the Early Cretaceous platform margin of the Faroe–Shetland area, thick early Cretaceous (Hauterivian and older) sandstones are developed alongside the Shetland Spine Fault, and have also been interpreted further west on seismic data in the East Solan Basin, alongside the Solan and Otter Bank Faults. Proximal packages of similar, syn-rift, scarp-derived sandstones have been inferred in the Valanginian–Hauterivian succession of the south Porcupine Basin (Tate, 1993; Johnson et al., 2001). In the Erris Basin, Irish well 12/13-1A proved a Berriasian to Hauterivian succession resting unconformably upon Permo-Triassic and consisting of 167 m of interbedded calcareous sandstone, shale and thin, dolomitic limestone overlain by 173 m of calcareously cemented Hauterivian sandstone (Musgrove and Mitchener, 1996; Chapman et al., 1999) (Figure 49). It is likely that similar sandstones will be preserved in analogous half-graben structures along the eastern flank of the Rockall Basin in the UK sector, but they have yet to be proved by drilling. On the north-western flank of the Rockall Basin in the Irish sector, two series of faults (north-east to south-west-trending in the north and east-north-east to west-south-west-trending in the south) control the margin of the basin and are possibly associated with linear half graben of Early Cretaceous age (Walsh et al., 1999). The overall structure suggested by these limited observations is of a set of Lower Cretaceous half-grabens on opposing flanks of the basin, symmetrically tilted away from its centre. Since most of the extension in the basin probably post-dated the formation of these structures, they were much more closely juxtaposed at their time of origin.

In the West Lewis Basin, the BGS shallow borehole BH88/01 in block 165/20 proved 4 m of Lower Cretaceous glauconitic sandstone overlying 15.8 m of Middle Jurassic mudstone. The Lower Cretaceous sandstone is poorly sorted and coarse grained, with matrix-supported pebbles up to 3 cm in length and shell fragments scattered throughout, although they decrease in abundance down core. Grains are composed mainly of quartz and lithic fragments. Both marine and terrestrially derived palynomorphs have been recovered from the sandstone, which is dated as Barremian on the basis of the overlapping ranges of the dinoflagellate cysts Sirmiodinium grossii and Odontochitina operculata (Hitchen and Stoker, 1993). In the hydrocarbon exploration well 132/15-1, mudstones of similar age (Late Barremian to Early Aptian) rest directly upon the Precambrian granitic basement of a tilted block on the basin flanks (Musgrove and Mitchener, 1996) (Figure 50). The basal Cretaceous unconformities in these boreholes suggest that active rifting during the Early Cretaceous was followed by onlap of the basin margins. The evidence of Early Cretaceous basement erosion increases the possibility that contemporaneous basin floor fans, analogous to the Scapa sandstones of the Central North Sea and Moray Firth areas may be developed downdip within the Rockall Basin itself (Knott et al., 1993, fig. 13).

The mid Cretaceous Normal Quiet interval lasted from early Aptian to end Santonian times (118 to 84 Ma) and probably coincided with the major episode of continental stretching and possible sea-floor spreading that formed the Rockall Basin. During the Early Aptian, sea-floor spreading was initiated in the Bay of Biscay-Grand Banks rift system in central Iberia to the south of the Figueiro–Newfoundland Seamounts Fracture Zone (Driscoll et al., 1995). If sea-floor spreading developed contemporaneously within the southern part of the Rockall Basin then these two areas of oceanic crust must have been separated by a continental north-west-trending dextral transfer zone in the Bay of Biscay, an area where sea-floor spreading did not originate until later in the Cretaceous. Even if oceanic crust were not formed in the Rockall area at this time, these plate tectonic movements would have been associated with major continental extension in the area. The development of a Late Aptian unconformity in the Erris Basin shows that the flanks of the Rockall Basin were further onlapped during this stretching event (Chapman et al., 1999). In well 12/13-1A, thick Hauterivian sandstone is succeeded by 310 m of Hauterivian to Albian shale with thin limestone and sandstone. The uppermost 167 m of Lower Cretaceous sediments are Albian, and consist of marl with limestone overlain by argillaceous and dolomitic limestone (Figure 49). Mudstones of possible Albian age in well 164/07-1 show that the centre of the Rockall Basin within the report area continued to subside rapidly near the end of the Early Cretaceous (Archer et al., 2005).

Upper Cretaceous

The sparse seismic and well data suggest that the Upper Cretaceous is developed fairly uniformly throughout the area, with most of the stratigraphical variation confined to marginal basins and highs alongside the main Rockall Basin. In places, continued onlap of the basin margins provides evidence of high eustatic sea level during the Late Cretaceous.

In the north-east of the report area, wells 154/03-1 and 164/25-2 show that Cretaceous sediments are absent from the West Lewis Ridge, which forms a structural high between the West Lewis Basin and the North-east Rockall Basin. Well 164/25-1Z, which is located close to the inverted western edge of the West Lewis Basin, proved an Upper Cretaceous succession (225 m thick) separated from the underlying Triassic sediments by 2 m of possible Lower Cretaceous strata.

In the North-east Rockall Basin, which forms a narrow marginal basin to the west of the West Lewis Ridge, well 164/28-1 drilled 136.6 m of Maastrichtian claystones before terminating in the Upper Cretaceous (Tate et al., 1999; Waddams and Cordingley, 1999). Closer to the axis of the North-east Rockall Basin, the mudstone-rich Upper Cretaceous succession penetrated by well 164/07-1 is pervasively intruded by approximately 700 m of basic igneous sills of Late Cretaceous or Paleocene age, which have thermally metamorphosed much of the sediment into dark grey hornfels. In this part of the area, the top of the Cretaceous is deeply truncated by a basal Cenozoic unconformity, with Coniacian mudstones overlain by Upper Paleocene sediments. This erosion may be associated with magmatic-related uplift or localised tectonic deformation in the vicinity of the Wyville Thomson transfer zone. Truncated intra-Cretaceous reflectors beneath the Darwin Volcanic Centre may be linked to the same episode of end Cretaceous erosion (Abraham and Ritchie, 1991;Waddams and Cordingley, 1999). On the south-east flank of the main Rockall Basin, well 132/15-1 proved a complete succession of Upper Cretaceous (308 m thick), consisting of Maastrichtian and Campanian mudstones with thin interbedded limestones overlying a condensed interval of Turonian to Santonian age. A thick Cenomanian limestone at the base of the succession gives rise to a prominent reflector (Musgrave and Mitchener, 1996) and is possibly equivalent to the Hidra Formation elsewhere. In the adjoining part of the Irish sector, well 5/22-1 terminated in a succession of Maastrichtian calcareous claystones interbedded with thin limestones and a trace of sandstone (>79 m thick). The nearby Dooish discovery well (12/02-1, 1Z) penetrated the Cretaceous and proved a 282.8 m thick Upper Cretaceous succession consisting largely of dark grey calcareous claystones, with some thinly interbedded argillaceous limestones and chalks. The Dooish Cretaceous section ranges in age from Maastrichtian to Turonian and acts as a seal for the underlying hydrocarbon-bearing Permian interval, upon which it rests unconformably (see Permo-Triassic chapter). Close by in the Irish sector, well 12/13-1A is located in a similar structural position in the Erris Basin and proved 152 m of Cenomanian to Maastrichtian sediments, consisting of white to pale grey, partly dolomitic and cherty limestone overlying more argillaceous limestones of Albian age.

In general, the Upper Cretaceous succession thickens rapidly away from these marginal locations towards the centre of the Rockall Basin, where it may overstep the Lower Cretaceous and rest directly upon high seismic velocity basement. Well 132/06-1, which lies in almost two kilometres of water closer to the centre of the basin, terminated within the Upper Cretaceous, proving 252.1 m of calcareous claystones with interbedded marls and micritic limestones of Maastrichtian and late Campanian age. Seismic interpretation suggests that the Upper Cretaceous is up to 2.5 km thick beneath the well and in most of block 132/6, with the base of the succession pervasively intruded by igneous sills of probable Paleocene age. The seismic velocity analysis of Joppen and White (1990) shows that similar thicknesses are widely developed within the Irish sector near the centre of the basin, perhaps indicating that the Upper Cretaceous basin is saucer-shaped, with most of the thickness variation confined to its margins. Despite the lack of evidence for widespread Late Cretaceous faulting in the Rockall area, these Upper Cretaceous thicknesses are broadly comparable with those of the more obviously fault-controlled Faroe–Shetland Basin. The sparse well data confirm that Upper Cretaceous strata within the report area are predominantly argillaceous. The proportion of Upper Cretaceous chalk-dominated sedimentary facies on the UK continental shelf diminishes northwards, where the more argillaceous successions are generally assigned to the Shetland Group, as in the Dooish discovery well. On the western margin, chalk is poorly developed to the north of the Porcupine Basin, although calcareous facies and thin limestones are common throughout the area.

Evidence of Cretaceous sediments from seamounts in the Rockall Basin

Some potential additional evidence for Cretaceous deposition has been obtained from the geological sampling of seamounts within the Rockall Basin.

Anton Dohrn

The Anton Dohrn Seamount rises to 649 m below sea level. Chalk was discovered in boulders dredged from its steep eastern flank by Jones et al. (1974). The chalk either occupies vesicles or occurs as a matrix between basaltic clasts. Nannofossils recovered from the chalk yielded a late Maastrichtian assemblage (Stoker et al., 1993).

About 60 per cent of the material recovered from the original dredge of Jones et al. (1974) consisted of corals and exotic blocks of glacial origin. The recovery site lay at a depth of approximately 1500 m. Attempts to take cores at depths of 700, 850 and 1200 m, failed to penetrate a cover of stiff glacial or postglacial shelly and silty clays. In 1998, further dredges were recovered from depths between 875 and 1480 m (O’Connor et al., 2000). These later samples included a group of igneous specimens from which Ar-Ar age dates ranging between 40.2 + 2.2 and 48.6 + 2.0 were obtained (see Igneous chapter). Two Late Cretaceous ages were also measured, but the results giving old ages are much more uncertain (that is, have larger values of 2σ) than the younger samples.

BGS short sea-bed core 57-12/18 was drilled on the top of Anton Dohrn at a water depth of 705 m. A core consisting of rounded basaltic pebbles in a white, carbonate biosparite matrix containing shell fragments, sponge spicules and foraminifera was recovered. The basalt pebble gave Ar-Ar ages of 45 + 1 and 46 + 1.4 Ma, while the carbonate matrix yielded a late mid Eocene to late Eocene age.

These observations suggest that at least some of the volcanic rocks on the top of Anton Dohrn are of possible mid Palaeogene age (as suggested by radiometric dating), and were eroded and redeposited in a carbonate-rich matrix during the Eocene (as revealed by the biostratigraphical data). At the very least, this evidence raises potential doubts about the significance of the earlier dredged material of inferred Mesozoic origin.

Rosemary Bank

The possibility that volcanism on Rosemary Bank is of Cretaceous age (see Igneous chapter) is based partly on calcareous nannofossils observed in core from BGS shallow borehole BH90/18, which was drilled near the top of the seamount. This borehole drilled through approximately1.5 m of recrystallised bioclastic limestone containing basaltic clasts, immediately overlying the basaltic core of the seamount. The identification of nannofossils indicating a late Maastrichtian age would appear to confirm that this is one of the most northerly proven occurrences of Cretaceous chalk lithofacies on the western side of Britain (Stoker et al., 1993).

Dietrich and Jones (1980) recovered dredge samples from Rosemary Bank between depths of 2160 and 1830 m. They noted that 65 per cent of the samples were clearly glacial erratics. Calcareous ooze was preserved in small pockets on the surface of some lava samples. Coccoliths in this material show signs of intensive reworking, and include Quaternary, Late Eocene and Oligocene, and Mesozoic (including Late Cretaceous) forms. Dietrich and Jones (1980) attributed the Mesozoic forms to reworking by the action of Oligocene bottom currents.

It is clear that several sedimentary horizons on Rosemary Bank include a mixed assemblage of Cretaceous and later fossils. The possibility remains that the key Cretaceous indicator specimens are all redeposited and are associated in some way with glaciation or current reworking on the flanks of the seamount. However, since the Mesozoic fossils may have only been subjected to local reworking, Dietrich and Jones (1980) did not entirely rule out a pre-Cenozoic age for Rosemary Bank.

The present top of Rosemary Bank is at a shallow depth (370 m) compared to the nearby Late Paleocene Darwin volcano (top at approximately 1700 m depth). Darwin itself may be constructed from more than 2000 m of Paleocene lavas, which would place the top of the Cretaceous at a present depth of >3700 m below the centre of the volcano. Accepting the borehole biostratigraphical evidence of Cretaceous sediments as it stands means that, at their current levels of preservation, Maastrichtian rocks can be found at depths ranging from 478 m in borehole BH90/18 on top of Rosemary Bank to a possible 3700 m beneath Darwin. Since Darwin was constructed at least in part as a subaerial volcano a possible implication of these figures is that there was a huge drop in relative sea-level between the Maastrichtian and the Late Paleocene and a subsequent rise to re-bury Rosemary Bank by 370 m at the present day. It would also suggest that the eroding Cretaceous volcanic centre at the site of Rosemary Bank towered more than 3 km over the active Darwin volcano, as it erupted during the Paleocene. Uplift and erosion at the end of the Cretaceous is widespread in the UK, but these figures seem to be unrealistically large for a Mesozoic and Cenozoic basin area.

The observations of potential Cretaceous sediments on Rosemary Bank can now be compared to the drilling results of the BGS short sea-bed core 57-12/18, which was drilled on the top of Anton Dohrn. There the Eocene age obtained from the lava pebbles and the carbonate sediments seems more compatible with the present-day submarine topography of the Rockall Trough than the Cretaceous age inferred from the earlier dredge hauls (Jones et al., 1974). As with Rosemary Bank, it is much easier to envisage an Eocene volcanic centre, rather than the stump of an eroded Cretaceous structure, rising above the Paleocene Darwin centre to these shallow depths.

A BGS seismic profile across Rosemary Bank shows that if the bank originated as a Cretaceous volcanic structure then at least a part of its structural relief is a result of post-Cretaceous tectonics. On the seismic profile it is noticeable that the lava escarpment around Rosemary Bank is better preserved at its south-eastern flank. This suggests that the bank forms a tilted block at base Eocene level and it is possible that Cenozoic tilting has exposed the Cretaceous foundation of the bank to erosion at its north-western flank. In this respect the bank would be closely analogous to the Palaeogene volcanic centre of Mull, where a thin relict Cretaceous succession is exposed beneath the Palaeogene lava pile.

Chapter 8 Cretaceous and Palaeogene igneous rocks

Ken Hitchen and Howard Johnson

The Palaeogene igneous rocks of the report area formed over a relatively short interval of time (approximately 64–54 Ma) and represent just a part of the extensive North Atlantic Igneous Province. Before Eocene–Recent North Atlantic opening, this province spanned an area of approximately 2000 km in diameter (Figure 51) (White, 1992). The passive continental margins within the province are commonly described as ‘volcanic’ margins and they include characteristic igneous features such as widespread and thick flood lavas and associated lava delta deposits, sill/dyke-complexes, volcanic centres (e.g. Planke et al., 2000), lower crustal lenses interpreted as underplated igneous rocks (e.g. Fowler, 1988), and seaward-dipping reflectors (e.g. Eldholm et al., 2002).

The volume of igneous material within the North Atlantic Igneous Province is very large, but hard to quantify (White, 1988; Coffin and Eldholm, 1992). Outcrop and borehole information from the Faroe Islands indicate that a cumulative total of over 6 km of flood basalt are present. Generally, however, the base of the flood lava succession is difficult to image on seismic data, and estimates of the extent and amount of underplated igneous material remain uncertain. On the basis of subsidence anomalies in a number of oil exploration wells offshore Britain and Ireland, Clift and Turner (1998) proposed that igneous underplating reaches over 5 km in thickness beneath the Faroe Islands, but reduces to between 4 and 2 km in the Rockall Basin area. However, such calculations are complex and may need revision in the light of improved understanding of a number of other Cenozoic tectonic processes that have affected the region, such as inversion tectonics associated with compressional stress (e.g. Boldreel and Andersen, 1993; Lundin and Doré, 2002; Johnson et al., 2005; Tuitt et al., 2010) and kilometre-scale Neogene differential uplift and subsidence (Stoker et al., 2005b). Theoretically, the amount of melt produced depends on the temperature of the asthenospheric mantle and on the rate of continental rifting (White, 1992). Much of the magmatic material (perhaps 60–80 per cent) may have frozen in the lower crust, while the rest was extruded as lava flows or erupted as pyroclastic deposits, commonly after undergoing fractionation in crustal magma chambers (Smallwood and White, 2002). Hence, although most of the lavas are basaltic in composition, some are more evolved and have also been contaminated by crustal material en route to the surface (Hitchen et al., 1997).

The ocean crust juxtaposed with the continental portions of the North Atlantic Igneous Province displays a number of interesting features and patterns (White, 1997) (Figure 52). For example, the presence of V-shaped ridges flanking the active spreading axis to the south-west of Iceland is thought to indicate along-ridge asthenospheric flow (Vogt, 1971). The broader aseismic, bathymetric ridges that extend to the west-north-west and south-east of Iceland are known as the Greenland–Iceland and Faroe–Iceland ridges respectively. These occur on either side of the present-day subaerial ‘sea-floor’ spreading in Iceland (Figure 52) and are generally considered to comprise anomalously thick (20–35 km) oceanic crust which contrasts markedly with normal oceanic crustal thickness of 7 km (Richardson et al., 1998; Smallwood and White, 2002). There are also areas of oceanic crust that formed without fracture zones and other areas of oceanic crust with fracture zones. The thickness and style of oceanic crust formation is thought to reflect the temperature of the mantle, with thicker and more ductile/less fractured crust developing above hotter mantle (White, 1997; Kimbell et al., 2004).

The development of the North Atlantic Igneous Province is spatially and temporally associated with continental breakup, west of Hatton Bank, during the Early Eocene. The pattern of magmatic activity on the continental margins and within the adjacent ocean basins has been interpreted to reflect the controlling influence of a mantle plume, known as the Iceland Plume. Such plumes are considered to comprise buoyant, hot material that originates from deep within the mantle (e.g. Morgan, 1971; White, 1988). After a succession of rift phases within the North Atlantic region, the initiation of the Iceland Plume, and its impingement on the base of the lithosphere at approximately 62 Ma, is thought to have generated a large volume of magma by adiabatic decompression of anomalously hot mantle (approximately 150–200º C above normal) that rose beneath progressively thinning lithosphere. The plume is also considered to have ultimately triggered continental separation about 5 Ma later, by introducing regional uplift. The plume is thought to underlie Iceland now and may be providing dynamic support to the ‘sea-floor spreading taking place there so making it subaerial (e.g. White, 1988; White and McKenzie, 1989; Barton and White, 1997).

The anomalously thick oceanic crust beneath the Faroe–Iceland and Greenland–Iceland ridges is considered by some to have formed directly above the core of the Iceland Plume, similar to that being formed now at the spreading centre in Iceland (e.g. Smallwood and White, 2002). On these ridges, a lack of linear, ocean floor magnetic anomalies is interpreted to be the result of a number of processes, including subaerial ‘sea-floor spreading and associated complex lateral flow of the lavas and their subsequent erosion (Smallwood and White, 2002). However, it is important to note that the Iceland Plume model remains controversial, with dispute about factors such as a lack of extensive evidence for high temperatures, a volcanic track and a seismic anomaly in the lower mantle (e.g. Foulger, 2002). Alternative hypotheses include the suggestion that relatively shallow mantle processes and structures associated with plate tectonics can account for the observations from Iceland and the development of the North Atlantic Igneous Province (e.g. Foulger, 2002).

Within the report area and surrounding region, uncertainty remains regarding the precise correlation between biostratigraphical ages and radiometric ages of igneous events. Ages from within the report area obtained by the Ar-Ar method are summarised in (Table 6). From these, and other selected ages, two main phases of magmatic activity have been described (White and Lovell, 1997; Smallwood and White, 2002; Jolley and Whitham, 2004). The first of these, during Early to Mid Paleocene times (approximately 62–58 Ma), includes some picritic basalts, indicative of high temperature mantle sources and, within the report area, the emplacement of the Hebrides Terrace Seamount. This first phase also includes nearly all the onshore British Tertiary Volcanic Province (e.g. Emeleus and Goyopari, 1992). During the second main phase of igneous activity, at approximately 57–54 Ma, both the rate of volcanic extrusion and total melt generation were much larger than during the first phase, and the volcanism was accompanied by rapid thinning of the continental lithosphere and the start of sea-floor spreading in the North Atlantic (Saunders et al., 1997). Within the report area, this second phase of activity includes emplacement of the Rockall and probably the Darwin Volcanic Centre. It should be noted that the Late Cretaceous (Maastrichtian or older) age of origin attributed to the Rosemary Bank and Anton Dohrn seamounts remain somewhat equivocal and consequently uncertainty persists regarding whether a separate, earlier, phase and mechanism (rift-related?) of igneous activity is required to explain their development, or if they could in fact be included in the Palaeogene phases. Multiple pulsing of the Iceland Plume has also been invoked to explain widely-spaced ages obtained from samples dredged from the Rosemary Bank, Anton Dohrn and Hebrides Terrace seamounts in the Rockall Basin (O’Connor et al., 2000).

The report area includes many features typical of the North Atlantic Igneous Province, such as extensive and thick flood lavas and associated lava delta deposits, sill/dyke-complexes, volcanic centres and possibly lower crustal lenses interpreted as underplated igneous rocks (Figure 53). Seaward-dipping reflectors are developed further to the west, close to the continent–ocean boundary on the Hatton Bank (White et al., 1987; Morgan et al., 1989; Kimbell et al., 2004).

Volcanic centres

The report area contains eight major volcanic centres (Anton Dohrn Seamount, Darwin, Geikie, Hebrides Terrace Seamount, Rockall, Rosemary Bank Seamount, St Kilda and Sula Sgeir) (Figure 53). Several other confirmed centres occur just outside the report area (Drekaeyga, Sigmundur and Swithin) and two major igneous intrusions also affect the George Bligh High. Other centres occur in the Inner Hebrides (mostly onshore), and to the north-east and west of the report area. Several have also been reported in Irish waters.

The Anton Dohrn, Hebrides Terrace and Rosemary Bank centres are generally referred to as seamounts although technically the crust here is thinned continental rather than true oceanic (see Structure chapter). The seamounts are massive structures which protrude hundreds of metres above the sea floor and have diameters measured in tens of kilometres. In contrast, the Darwin centre, retains its original circular plan and dome shape, but is completely buried by younger sediments. The Geikie, Rockall, St Kilda and Swithin centres are all deeply eroded. Geikie and Swithin do not crop out at the sea bed. The tiny Rockall island (approximately 18 m high and 25 m in diameter) is the only subaerial representation of the Rockall centre whereas the St Kilda archipelago has some of the highest sea cliffs in Scotland.

Although ranging in size, the volcanic centres in the report area are generally much larger than those of the Inner Hebrides (Ardnamurchan, Arran, Blackstones, Mull, Rum and Skye). This may be a function of the thickness and type of crust into which the centres were intruded or through which the lavas passed. Each centre is coincident with a positive gravity anomaly presumed to be caused by an associated large dense plutonic intrusion. In the case of the three seamounts it may also reflect the fact that they are large prominent bathymetric features surrounded by water.

Rosemary Bank Seamount

Rosemary Bank is centrally situated in the northern Rockall Trough. The highest point of the seamount is reportedly only 316 m below sea level (Pudsey et al., 2004) whereas water depths in the moat on the south-west side of the feature exceed 2200 m. The lower flanks of the seamount are buried by younger sediments. A seismic line across the seamount is shown in (Figure 54). Samples of igneous rocks have been recovered from Rosemary Bank by dredging and coring (Figure 53). Geochemically they fall into two categories, being either basaltic lavas or highly–potassic lavas and tuffs. Dietrich and Jones (1980) analysed blocks dredged from the south-western flank of the seamount. The rocks were highly altered but the original composition was considered to be transitional to mildly alkaline, olivine-rich porphyritic basalts with high Mg, Ni and Cr contents (Hitchen et al., 1997). The olivine phenocrysts were altered to chlorite-serpentine and haematite. The original rocks were considered to have been fairly primitive olivine-tholeiites with oceanic affinities.

BGS borehole BH90/18, drilled on top of the seamount, proved a 16.72 m section of altered basalts between 23.53 m below sea bed and TD of the borehole at 40.25 m below sea bed. The sequence mainly comprises a number of fine-grained, phyric or sparsely-phyric (usually plagioclase) basaltic lava flows with some reddened flow tops. There is little chemical variation within the sequence and the lack of any major differences in the degree of fractionation indicates that there was no significant hiatus in extrusion. However a poorly-sorted clastic unit occurs at 37.50–37.85 m and is considered to be the result of local subaerial reworking of the basalt sequence (Morton et al., 1995). As a whole, the lavas are slightly more evolved, and less tholeiitic, than the Dietrich and Jones (1980) dredge samples from the south-west flank of the bank. Neither the dredge samples, nor the samples from the borehole, show any indication of crustal contamination.

Highly alkaline, potassium-rich lapilli tuffs, with K₂O values ranging from 1.86 to 3.20 per cent (Hitchen et al., 1997), have been dredged from the top of the bank. The tuffs are vitric, strongly palagonitised and cemented by zeolite and calcite (Waagstein et al., 1989). Some fresh glass material has a composition of phonolitic nephelinite which contrasts with the MORB-like dredged tholeiites of Dietrich and Jones (1980). The tuffs contain basaltic fragments and blocks which suggests they may represent a late-stage explosive event that incorporated earlier basaltic country rock as xenoliths.

BGS short sea-bed core 59-11/12 penetrated 4.05 m of highly potassic vesicular lavas (Figure 55) before terminating at 5.1 m below sea bed. The core comprises five, highly vesicular thin flows considered to be submarine in origin. A subsample of the core has 4.83 per cent K₂O and a phono-tephrite composition (Hitchen et al., 1997). Both the tuffs and the lavas of 59-11/12 have rare earth element (REE) and trace element patterns which suggest that they were derived by small degrees of partial melting of ancient subcontinental lithospheric mantle similar to that underlying the Outer Hebrides and identified by studies of the Loch Roag dyke (Menzies et al., 1987). Hence these potassium-rich rocks lend support to the idea that continental crust underlies the northern part of the Rockall Basin.

According to Scrutton (1971) the residual free-air positive gravity anomaly associated with the seamount is 90 mGal and this could be explained by an underlying pluton shaped like a truncated cone and extending down to a depth of 22 km. However this model requires the crust beneath Rosemary Bank to be 23 km thick whereas it is likely to be much thinner than this, see (Figure 16). Palaeomagnetic evidence indicates that, although the seamount is not uniformly magnetised, the bulk of it formed during a period of magnetic reversal (Miles and Roberts, 1981) (Figure 6) and at lower latitudes than its present location (Irving, 1964). High-frequency positive magnetic anomalies have also been observed to be associated with Rosemary Bank. These suggest that a phase of igneous activity occurred after the formation of the bulk of the seamount and that the rocks concerned overlie or intrude the pre-existing feature (Miles and Roberts, 1981). Evidence for a sustained or multiphase development of the feature also derives from palaeomagnetic studies of lavas recovered from the top of the seamount. Eight samples of basaltic lavas from BGS borehole BH90/18 were analysed for primary thermal remanent magnetisation. The lower four samples (below about 38 m below sea bed) retained normal polarity whereas the upper four samples showed ‘shallow’ polarity indicating a possible magnetic excursion or imminent change in polarity (Carter, 1994; Morton et al., 1995). Somewhat conversely five samples were taken from several individual lava flows penetrated by short sea-bed core 59-11/12. Each showed reversed polarity. Inclinations were in the range -44.5° to -52.3° (Bannister, 1995). Hence both normally and reversely magnetised lavas have been recovered at, or close to, the sea bed on the top of Rosemary Bank.

The formation age of the Rosemary Bank Seamount has proved equivocal. The dredged and drilled basaltic lavas are generally too altered to be considered ideal for radiometric age dating. The K-Ar age of 22 Ma is a minimum age (Dietrich and Jones, 1980) and the K-Ar range of 24.2+0.6 to 61.9+1.1 Ma (ages increasing with depth down borehole BH90/18) (Hitchen and Ritchie, 1993) also suggests that the oldest age is a minimum. The potassic lavas and tuffs are not suitable rocks for dating. O’Connor et al. (2000) have reported Ar-Ar ages of 42.5+0.4 (re-analysis of the Dietrich and Jones dredge material) and also 52.4+0.7 and 52.7+0.5 Ma from dredge samples recovered from the lower western slopes of the bank.

Immediately overlying the basalts in BGS borehole BH90/18 is 1.53 m of white, recrystallised bioclastic limestone containing basaltic clasts. A limited calcareous nannofossil assemblage has been recovered from the limestone which includes a specimen of Lithraphidites cf. grossopectinatus which can be attributed to nannofossil zone CC25 to CC26 (late Maastrichtian). Assuming the nannofossils are in situ and not reworked this suggests that the basalts are early Maastrichtian or older in age (Morton et al., 1995).

Recent swath bathymetry data have revealed areas of mounded sea-bed topography and isolated pinnacles around the flanks, and on the central part, of the Rosemary Bank plateau. These features have been interpreted as small, late-stage parasitic volcanic cones (Pudsey et al., 2004) which may explain the Early to Mid Eocene Ar-Ar ages of O’Connor et al. (2000) and the recrystallisation of the limestone proved in borehole BH90/18. Hence all the current geological observations of Rosemary Bank Seamount can be explained by a large Late Cretaceous eruption which formed the bulk of the feature followed by erosion, deposition of the limestones and then a smaller Eocene volcanic event which produced both lavas and tuffs.

Anton Dohrn Seamount

The seamount is centrally situated in the Rockall Trough and is comparable in overall dimensions to Rosemary Bank. The upper plateau surface has a diameter of approximately 40 km and the highest point confirmed to date is about 650 m below sea level. This is considerably lower than Rosemary Bank and may reflect in part the greater subsidence of the central part of the Rockall Trough compared to that further north. However the present depth is about half that predicted from subsidence models and may reflect some former, or continued, thermal support mechanism (Jones et al., 1994). The Anton Dohrn Seamount is situated on a major structural line (the Anton Dohrn Lineament) which marks the north-west to south-east aligned boundary between the Archaean (Lewisian) Hebridean Terrane and the Early Proterozoic Rhinns Terrane (see Structure and Basement chapters). The formation of the seamount at this location may have been influenced by the coincidence of this terrane boundary and the thin crust underlying the centre of the Rockall Basin.

Very little material has been obtained from the Anton Dohrn Seamount. Jones et al. (1974) reported highly altered basic lavas and tuffs recovered by dredging the steep eastern flank of the seamount. Selected thin sections commonly show trachytic textures commonly defined by plagioclase laths. Original olivine phenocrysts are pseudomorphed by serpentine, clay and carbonate. Geochemical and isotopic data indicate the basalts have ‘within plate’ characteristics and exhibit no evidence for crustal contamination (Jones et al., 1994). Significantly, unmetamorphosed chalk was discovered either as vesicles in the lavas or as the matrix between the basalt clasts and yielded an abundant Upper Maastrichtian nannofossil assemblage (Jones et al., 1974).

BGS borehole BH90/15 and short sea-bed core 57-12/18 were both drilled on the top of the seamount. The borehole did not reach the volcanic pile before being terminated but did penetrate a bioturbated, bioclastic limestone which yielded a sparse Lower Paleocene to Lower Eocene nannofossil assemblage. The short core comprises 1.0 m of a similar limestone containing angular basaltic clasts (Figure 56). The limestone yielded Middle to Upper Eocene (NP17 to 19) microfossils and one of the basalt clasts has been Ar-Ar dated at 54.01+0.17 Ma (Ypresian, Early Eocene) (Chambers, 2000). Further Ar-Ar ages have been obtained from samples dredged from the steep flanks of the seamount (Table 6). These range from 42.0+0.3 to 70.4+0.9 Ma but are concentrated into four phases at 70–69, 62, 48–47 and 42–41 Ma (O’Connor et al., 2000).

Gravity data for Anton Dohrn show a large positive gravity anomaly associated with the seamount but no model across the feature has ever been published. Magnetic data show a complex interplay of positive and negative anomalies but with no coherent pattern (Figure 6). Some shipboard magnetic profiles are shown by Jones et al. (1974).

If all the dredged material is indeed from the Anton Dohrn Seamount, and not ice-rafted debris, then both the nannofossil data and older Ar-Ar dates indicate that at least part of the seamount is Maastrichtian or older in age. The ages of O’Connor et al. (2000) suggest that activity here extended over a 29 million-year period. This is very prolonged for a volcanic centre. They suggest that activity was episodic in nature with volcanism occurring in discrete phases, at intervals of 5 to 10 Myr, due to a hot, pulsing Iceland Plume.

Hebrides Terrace Seamount

This seamount is situated at the foot of the Hebrides Slope on the eastern side of the Rockall Trough and adjacent to the southern limit of report area (the UK/Irish median line). The seamount protrudes over 1000 m above the ocean floor and its summit is approximately 1000 m below sea level, the lowest of the three seamount summits in the Rockall Trough. The flanks are partially covered by deposits of the Barra and Donegal submarine fans which prograde westwards from the shelf and slope areas. The seamount is slightly elongate in an east–west orientation and has a very flat top, probably due to planation during the Eocene (Omran, 1990).

Only dredge samples have been recovered from the seamount (Figure 53) and most of these come from the steep southern flank. One example comprises a fresh olivine basalt, where the olivine is Fe rich, and has a low Cr content (27 ppm) indicating some fractionation has occurred. The dredge sample from the northern flank is a coarse-grained altered dolerite where the olivine has largely been replaced by chlorite. Both samples are contaminated by continental crustal material. Interpretation of the Pb isotope data suggests the contamination is derived from the Rhinns Terrane (Hitchen et al., 1997).

The value of 140 mGal for the positive free-air gravity anomaly is the largest published residual value for any of the Hebridean volcanic centres or those further offshore (Abraham and Ritchie, 1991, table 1). However many of the centres discovered in the last ten years have not been modelled due to insufficient data being available to constrain the modelling process. Buckley and Bailey (1975) suggested the gravity anomaly could be accounted for by a large dense cylinder-shaped body extending to the base of the crust at 23 km depth. Remodelling by Omran (1990) satisfied the anomaly by assuming a slightly upward-tapering body but extending downwards to the Moho at only 17 km depth. The bulk of Hebrides Terrace Seamount was emplaced during a period of magnetic reversal. Local magnetic anomaly variations over the summit area may be produced by a heterogeneous association of local dykes, feeder pipes, pillow lavas, pyroclastic and weathered, partially demagnetised, rocks (Buckley and Bailey, 1975).

Unlike Rosemary Bank and Anton Dohrn seamounts, there are no biostratigraphical data to suggest that the Hebrides Terrace may, at least in part, be Late Cretaceous in age. Omran (1990) reported K-Ar ages of 67 to 60 Ma for the samples of dredged lavas described above. The fresher southern sample was redated by the Ar-Ar method at 62.43+0.06 and 65.35+0.21 Ma (Table 6). Both these sets of ages support an origin at about the Cretaceous–Paleocene boundary. O’Connor et al., (2000) report Ar-Ar ages, obtained from dredge samples (south-western flank), which apparently support episodic volcanic activity at 61.6, 51.2 and 48–47 Ma (Table 6). The oldest and youngest of these coincides with almost identical ages for possible activity on the Anton Dohrn Seamount.

Rockall

The idea that Rockall island was the sole subaerial expression of a Palaeogene volcanic centre was first proposed by Roberts (1969) largely on the basis of high-amplitude arcuate magnetic anomalies to the north and east of the island. Lewis et al. (1975) have described samples from the island as being reversely magnetised. Rockall is now recognised as being a deeply eroded volcanic centre of which the island itself, Helen’s Reef and Hasselwood Rock are all constituent parts.

Rockall island is approximately 25 m across and 18 m high. It is semi-circular in shape with a near-vertical east-facing cliff extending from sea level to summit. This has undoubtedly been influenced by the north-south set of fractures within the granite. Shallow reefs are developed around much of the island. The bulk of the island comprises medium to coarse-grained aegirine-riebeckite granite. Xenoliths have been recognised within the granite and a sheet intrusion, dipping at 30–35° to the east-north-east cuts through the lower part of the island (Hawkes et al., 1975). The granite is considered to be derived by melting of the Rhinns Terrane material which makes up the bulk of the Rockall High (Dickin, 1992). BGS short sea-bed core 57-14/58 was drilled 9 km north-north-east of Rockall island. It consists of 1.81 m of coarse-grained weakly altered or metamorphosed gabbro with an indistinct foliation. Within the immediate vicinity of Rockall island examples of granite, microgranite, syenite, dolerite and olivine-basalt have been recovered mainly by dredging or diving. Other samples from Rockall Bank are lavas or basement rocks and are discussed elsewhere.

Helen’s Reef (3 km to the east-north-east of Rockall) comprises mainly olivine-microgabbro and microporphyritic olivine-dolerite. In thin section the dolerite can be seen to have crystallised later. Local troctolitic cumulates of olivine and pyroxene occur on the northern part of the reef (Roberts et al., 1974; Harrison, 1975).

Many attempts have been made to date the age of the Rockall Volcanic Centre by radiometric methods. A wide spread of ages has been derived from various lithologies, collected by dredging, drilling and in situ sampling and analysed by several different techniques. A summary of the ages is given in Ritchie and Hitchen (1996). The youngest age of 44.3+1.8 Ma has been obtained by K-Ar analysis of a dredged basalt lava from 14 km east of Rockall island (Jones et al., 1972). The oldest is 83+3 Ma for a troctolitic gabbro recovered by divers from Helen’s Reef and obtained by K-Ar analysis (Roberts et al., 1974) subsequently recalculated by Pankhurst (1982). BGS short sea-bed core 57-14/53 was drilled 21 km east-south-east of Rockall island and consists of 2.56 m of flow-banded trachyte lava. A subsample from 0.77 m below sea bed has yielded an Ar-Ar age of 56.5+0.1 Ma (Hitchen, 2004). If this lava is considered to have derived from the Rockall centre this may be the best estimate of the true age.

Geikie

The Geikie Volcanic Centre is situated 80 km north-west of St Kilda, at a point where the Hebrides Shelf bulges westwards into the Rockall Trough, and forms the southern end of the Darwin–Geikie High (Figure 7). This bulge in the shelf may in part be due to the presence of the centre. Geikie was first recognised from its gravity response (Himsworth, 1973) but was not named until much later (Evans et al., 1989). The centre is deeply eroded, probably due to the subsequent uplift and planation of the Outer Hebrides High into which it was emplaced. There is no remaining volcanic dome.

On seismic data, the ‘top lava’ reflector in the vicinity of Geikie appears as a flat, slightly irregular surface which truncates underlying short, dipping, radially disposed reflectors thought to represent the attitude of the lavas that flowed outwards from the original vent. These prograde from a central ‘rugged zone’, approximately 3 km in diameter, which may represent the location of the central vent of the original volcano and which is coincident with the centre of the gravity anomaly (Evans et al., 1989, figs. 4 and 6). The whole centre is overlain by Palaeogene and Neogene sediments. No samples of igneous rocks have been recovered from close to Geikie although undoubtedly some of the lavas on the shelf here were derived from it.

The Geikie gravity anomaly is the second largest of any in the Hebrides Shelf area including those associated with the volcanic centres of the Inner Hebrides (see Abraham and Ritchie, 1991). The peak Bouguer anomaly exceeds 180 mGal (Evans et al., 1989) and the residual free-air anomaly measures 106 mGal (Himsworth, 1973). The anomaly, which is broadly circular, has been explained by the presence of a large dense plutonic body, with a density contrast of 350 kg/m3 compared to the adjacent crust. Both Evans et al. (1989) and Himsworth (1973) used the same basic model of two superimposed co-axial cylinders of dense rock to estimate the size of the pluton.

Both models show the top of the upper cylinder is at a very shallow structural level which suggests that, in the central part of the structure, the overlying sediments rest almost directly on the intrusion (Table 7).

Darwin

The Darwin Volcanic Centre, originally centre ‘A’ of Roberts et al., (1983) is situated at the northern end of the Geikie–Darwin High (Figure 7) It consists of a buried former Palaeogene volcano and a basic pluton. Seismic data show it to be the best preserved centre in the report area with the former volcanic dome being almost completely intact beneath a cover of younger sediments (Abraham and Ritchie, 1991, figs. 3 and 4; Tate et al., 1999, fig. 5). The lavas dip away radially from a central vent 1.6 km in diameter and over 100 m deep, situated on the crest of the dome. The flows form a series of asymmetric arcuate scarps around the former volcano (Abraham and Ritchie, 1991) which, if they represent lavas freezing at the contemporary shoreline, record the gradual subsidence of the volcano below sea level.

Well 163/06-1A drilled the north-west flank of the former Darwin volcano. The well proved 689 m of basalt overlying 356 m of dacite but did not penetrate the base of the volcanic sequence. The basalt is olivine-tholeiite and can be subdivided into an upper sequence, which exhibits alkalic and picritic tendencies, and a lower sequence which is nearer in composition to normal mid-ocean ridge basalt (N-type MORB) (Morton et al., 1988). The basalt overlies cordierite-phyric dacite with high Al, Ni and Cr content. The Sr and Pb isotopic composition indicates it is not a basalt differentiate but is likely to have originated from melting of argillaceous, and possibly arenaceous, rocks present at depth in the Rockall Basin. The unusual peraluminous geochemistry of the dacite has been used to suggest that organic-rich black shales may have been involved and that these might be a source for hydrocarbons in this basin (Morton et al., 1988).

The residual free-air gravity anomaly associated with the Darwin centre is 52 mGal (Himsworth, 1973) whereas the stripped anomaly is 72 mGal (Abraham and Ritchie, 1991). These authors have modelled and accounted for the anomaly by the presence of variously shaped, dense plutons beneath the Darwin volcano (Table 8). In both cases the pluton is regarded as asymmetric due to the north-north-east to south-south-west elongation of the anomaly.

Various K-Ar ages have been obtained for both the basalts and dacites of the Darwin Volcanic Centre (Morton et al., 1988) but most can be discounted. An average age for the most acceptable of these ages is 55.2 Ma (Hitchen and Ritchie, 1993). An Ar-Ar plateau age of 55.9+0.3 Ma was obtained by Sinton et al. (1998) from a dacite sample from approximately 3535 m depth. However this was a ‘junk basket’ sample and the depth cannot be determined accurately.

St Kilda

The St Kilda archipelago is centrally situated in the Outer Hebrides High 65 km west of North Uist. It comprises the main islands of Hirta, Soay and Boreray plus numerous other smaller islets and rocks. The archipelago is the remaining subaerial expression of the St Kilda Volcanic Centre. The geology has been described in detail by Cockburn (1935) and Harding et al. (1984).

The centre is deeply eroded so that there are no Cenozoic lavas or sediments preserved on the islands. However lavas crop out at the sea bed, and also subcrop younger sediments, over large areas of the Hebrides Shelf and Outer Hebrides High to the north and west of St Kilda (Figure 53). The rocks of St Kilda were emplaced during two main phases. The first phase comprised gabbros and dolerites and the second phase was mainly gabbros, dolerites, granite and other mixed rocks. Felsite, diorite and breccias are also present. In order of intrusion the principal formations which form the bulk of the archipelago are the Western gabbro, igneous breccia, Glen Bay gabbro, Glen Bay granite, the four-phase Mullach Sgar intrusive complex and the Conachair granite. Late stage dykes and sheets cut some of the above formations and represent the final phase of igneous activity on the islands (Harding et al., 1984).

A large 80 mGal free-air residual gravity anomaly is associated with St Kilda. Its centre is 3 km north of Soay. The anomaly has been modelled by Himsworth (1973) by assuming an underlying basic pluton extending down to 21 km below sea level and having a diameter increasing with depth from 25 km at the top to 35 km at the base. Oriented cores for palaeomagnetic studies were taken from a variety of lithologies at 19 sites across Hirta, the main island of St Kilda. Samples from seventeen sites responded well to alternating field demagnetisation. All these showed good stability with southerly declinations and moderately steep negative inclinations. The fact that all stable magnetisations on St Kilda are of reverse polarity suggests that intrusion of the complex occurred over a fairly short time period during a single polarity interval. Furthermore the position of the palaeomagnetic pole implies that St Kilda, and therefore most of Britain, has drifted northwards about 19° during the last 55 million years (Morgan, 1984).

The age of the St Kilda centre was first addressed by Miller and Mohr (1965) who obtained several dates at around 57–56 Ma using the K-Ar whole rock method. Pankhurst (1982) subsequently recalculated these at around 60+3 Ma. More recently Brook (1984) obtained a Rb-Sr age of 55+1 Ma from the Conachair granite and this appears to be the preferred age (Mussett et al., 1988).

Sula Sgeir

The Sula Sgeir Volcanic Centre is located at the south-east end of the Wyville Thomson Ridge. It is the least well defined of any of the volcanic centres within the report area, its presence mainly being inferred from a large, near-circular positive gravity anomaly (Ritchie et al., 1999). The extent of the lavas emanating from the centre is uncertain although it is likely that the lavas proved in BGS borehole BH85/07, drilled approximately 25 km to the south of the centre, were derived from the Sula Sgeir centre. These basaltic lavas yielded disparate K-Ar ages that range from about 63 to 46 Ma (Hitchen and Ritchie, 1993).

Other volcanic centres on the margins of the report area

Drekaeyga

The Drekaeyga centre is located in the Auðhumla Basin, a synclinal fold between the Wyville Thomson and Ymir ridges, and is completely buried by younger Cenozoic sediments. At the level of the top Palaeogene lavas, the centre appears to form a caldera-like crater (Keser Neish, 2005, fig 6) which has an area of approximately 225 km². The lavas which define the structure are folded. Uniquely amongst the volcanic centres in the report area there is no associated positive gravity anomaly which might have been indicative of a possible underlying pluton. The origin of the present configuration of the Drekaeyga structure remains problematic. It may represent an eroded former volcano, but it has also been described as an intrusion (Keser Neish, 2005) or an intrusive centre (Keser Neish and Ziska, 2005).

George Bligh High

The George Bligh High is a broad anticlinal feature covered in lavas. It is associated with two circular positive gravity anomalies (Figure 5) assumed to be caused by basic intrusions from which the lavas may have been extruded. No models for these intrusions have been published. Potential field data also suggest the presence of a large low-density magnetic intrusion within the dome, possibly a granitic body with a diameter of 40 km at 10 km depth (unpublished data).

Sigmundur

This is situated just north of the report area (Figure 53). It was first recognised by the small positive free-air gravity anomaly, in excess of 40 mGal, and labelled centre ‘B’ (Roberts et al., 1983). It was named the Sigmundur Seamount by Andersen (1988) but the term Sigmundur complex was preferred by Ritchie et al. (1997) as the feature is largely buried and there is only a limited bathymetric expression. It does however retain its original volcanic dome shape indicating only limited erosion.

Seismic data show the Sigmundur centre can be defined by a continuous, nearly circular volcanic scarp which has a diameter in excess of 30–40 km. Within the volcanic pile there are prograding lava flows (Andersen, 1988; Ritchie et al., 1997).

Modelling of the gravity anomaly suggests the presence of a large mafic pluton immediately underlying the volcanic pile of the former volcano. This intrusion is shaped like a truncated asymmetric cone which extends from 2.5 to 25 km below sea level and has a diameter which increases downwards from 4 to 21 km (Ritchie et al., 1997).

No rock samples have been recovered from Sigmundur. Lithologies are likely to be basaltic but may include acidic rocks similar to those proven by well 163/06-1A which drilled the adjacent Darwin centre. The precise age of the centre remains unknown.

Swithin

The Swithin Volcanic Centre is situated in the north-west part of the Rockall High and its presence may have influenced the shape of the high and the bathymetric expression of Rockall Bank. Seismic data show the centre to be deeply eroded with an irregular top basalt reflector. The centre does not crop out at the sea bed and no samples have been recovered from it (Hitchen, 2004). There is an associated large positive gravity anomaly but no modelling has been performed in order to estimate the size of the underlying intrusion.

Extrusive volcanic rocks

These comprise mainly basaltic lavas and tuffs. In the literature the term basalt is generally used for all the lavas but a small number of more evolved lavas (basalt differentiates) has also been recovered, mainly from the banks in the west of the report area, indicating that the parent basalt magma may have been ponded in the upper crust for some time prior to extrusion. The lavas, as recognised in wells and on seismic data, comprise numerous thin stacked flows and not a single massive extrusive event. Most of the commercial wells on the eastern side of the Rockall Basin drilled into or through the extrusive volcanic interval. However in wells 154/01-1 and 164/28-1A lavas were absent and the extrusive interval comprises tuffs, tuffaceous claystones, volcaniclastic sandstones and volcanic breccias or a combination of these lithologies. The sedimentary volcanic lithologies sometimes dominate the extrusive interval on the Outer Hebrides High and indicate extensive reworking of volcanic material during the Late Paleocene and earliest Eocene.

Regional interpretation, tied into well and borehole data, shows that extrusive volcanic rocks crop out over a large part of the report area (Figure 53). Lavas crop out at the sea bed on Rockall Bank, the Hebrides Shelf and on the three seamounts. Elsewhere they subcrop younger Cenozoic sediments. Lavas are largely absent on the inner Hebrides Shelf, from which they have probably been eroded, and locally from parts of the Rockall High. In parts of the northern Rockall Basin the lavas occur at depths greater than 3000 m below sea level. As most of the lavas are considered to have been extruded at, or near, sea level this illustrates the amount of differential subsidence that has occurred since their extrusion. However Boldreel and Andersen (1994) consider that some of the lavas in the northern Rockall Basin may have been extruded in a submarine environment (based on ‘hummocky’ seismic character). BGS borehole BH94/03 terminated at 209.65 m below sea bed after penetrating 1.71 m of interbedded pillow lavas (three individual flows) and dark grey shelly mudstones.

On seismic data, the top surface of the lavas produces a prominent reflection. Within the lava succession there is usually only limited structure visible and below the lavas, seismic data commonly fail to resolve structure in any detail. Hence the Mesozoic and older geology is ‘invisible’ to conventional seismic interpretation over large parts of the report area. In the centre of the Rockall Basin the exact limits of the lava subcrop are difficult to constrain due to the resolution of the seismic data. In this area it is difficult to determine whether the high-amplitude reflection on seismic data is from the top of the lavas or from extensive flat-lying sills which are also present here. Where the lavas are thin, and tuffaceous and volcaniclastic sediments dominate the Upper Paleocene interval, the seismic response comprises a package of numerous closely spaced high and low amplitude reflectors.

The large volcanic centres undoubtedly acted as source areas for many of the extrusives although it is not always possible to map the extent of the lavas from each individual centre. However the subcrop limit of the lavas around the Anton Dohrn and Hebrides Terrace seamounts (Figure 53) may indicate the extent of the flows from these centres. Surface flows were also undoubtedly fed from dykes leading to fissure eruptions but such features are extremely difficult to recognise on conventional 2D seismic data.

Structural character of the extrusive volcanic succession

The upper limit of the succession is a prominent seismic reflector usually attributed to the top of the lava sequence. In places this surface coincides with the sea bed but elsewhere it generally dips into the Rockall Basin. Prominent scarps are developed on this surface. These are generally assumed to be primary depositional features caused by lavas reaching the position of the contemporary shoreline and freezing. Beneath the scarps prograding lava foresets are sometimes imaged on seismic data (Figure 57) (Ritchie et al., 1999). The foresets are thought to comprise ‘flow-front breccias’ caused by the reworking of fragmented volcanic material generated by hot lavas entering the sea. The largest foresets are in the vicinity of the Geikie Volcanic Centre where they are in the order of 1500 m vertical extent measured from toe to crest. Foresets a few hundred metres high are more typical.

Scarps are present around the Darwin Volcanic Centre (Abraham and Ritchie, 1991) and are also associated with the south-east flank of George Bligh High. The major Hebridean Escarpment (Figure 53) and (Figure 57) can be traced for over 400 km along the eastern margin of the Rockall Basin (Smythe, 1989) and, in places, especially in the vicinity of the Geikie Volcanic Centre, exceeds 600 m high. The escarpment subcrops in an arcuate form around the Geikie centre on its basinward side and has clearly been influenced by it. Eocene subsidence and post extrusion faulting may be partially responsible for the height of the escarpment here (Evans et al., 1989).

A large scarp in the top lava surface marks the eastern boundary of the Rockall High. This appears to be largely the result of post-extrusion faulting rather than being a primary emplacement feature.

On seismic data short dipping reflectors within the upper part of the extrusive sequence are commonly truncated by the top lava surface. This is best imaged around the Geikie Volcanic Centre (Evans et al., 1989, fig. 4) and on the top of Rockall High. These short reflectors are interpreted as representing the attitude of the original lava flows which have been subsequently planed off by erosion.

Extrusive rocks of the Rockall High

Extrusive igneous rocks from Rockall High have been recovered in four boreholes, 12 sea-bed short cores and also by divers and dredging (Figure 53). Most lithologies are alkaline or sub-alkaline basalt lava flows. Some flows, however, are more evolved (basalt differentiates) and can be classed as hawaiite (in borehole BH94/05) or trachyte (sea-bed short cores 57-14/53 and 58-14/51) or trachyandesite (57-15/15). The conglomerates in sea-bed short cores 58-14/31 and 58-14/32 recovered basalt clasts in carbonate matrix and core 57-13/54 was a basaltic volcanic agglomerate. A tuffaceous interval was proved in borehole BH94/03.

In BGS borehole BH94/03 basalt lavas were recovered at two levels, the higher flow at 47.26–51.00 m below sea bed and the lower flows (interbedded with mudstones) between 207.94 and 209.65 m. Their mineralogy is similar. However a basalt pebble from a conglomeratic interval between the flows differs petrographically and in clinopyroxene composition and may have been derived from a different local source. The tuffaceous interval in this borehole comprises colourless or pale yellow, highly vesicular andesitic volcanic glass, best described as pumice. The deposit is probably the result of rapid cooling, degassing and fragmentation of andesitic lavas flowing into water. The tuffs are probably local in origin as they bear no resemblance to the regionally widespread basaltic Balder Formation tuffs of north-west Europe.

All of the basaltic rocks from Rockall High possess similar trace and REE compositions. What variations there are probably reflect the degrees of partial melting in the mantle source. Most of the basalts also have similar trace element data with characteristic positive Ba and Rb anomalies considered to be caused by continental crustal contamination.

The trachyte lavas in cores 57-14/53 and 58-14/51 have trace element data and similarity in REE patterns to the nearby basalts suggesting they originated through fractional crystallisation from a parent basalt magma. Both trachytes have been dated by the Ar-Ar method (Table 6) but yielded ages 3.7 Ma apart.

The geochemical and isotopic data show that most of the extrusive rocks recovered from the Rockall High have evidence of crustal contamination although 57-13/66 is an exception to this. Contamination has generally been by granulite facies material although the lavas in borehole BH94/05 may have been affected by rocks of amphibolite facies. Both facies types have been proved within the basement of Rockall High. The Pb isotopic data are consistent with contamination from Rhinns Terrane material rather than from the Lewisian Hebridean Terrane showing that the Rockall High is situated to the south of the Anton Dohrn Lineament.

Extrusive rocks of the George Bligh High

Samples of lavas obtained from the top of George Bligh Bank by dredging have now been supplemented by those obtained by BGS drilling and sea-bed short cores (Figure 53).

BGS borehole BH94/07 was terminated at 23.55 m below sea bed after drilling 2.58 m of a dark grey extrusive igneous rock comprising upper and lower flows. Subsamples from both flows have been described as hawaiite and exhibit variations in grain size, texture (some are trachytic) and amount of alteration. Four short sea-bed cores (58-14/8, 58-14/42, 58-14/57 and 58-14/58) recovered conglomerate of reworked igneous clasts in a carbonate matrix. In short core 58-14/42 the conglomerate overlies 0.23 m of lava which is described as a benmoreite in composition. Regional stratigraphical considerations suggest the carbonates, and therefore the underlying lavas, are NP6 or older in age (mid Late Paleocene).

All the non-dredged samples comprise relatively evolved basalts or basalt derivatives with high Fe/Mg ratios and low Cr and Ni contents (Hitchen et al., 1997). The rocks are alkaline with high Nb/Y and Ti/V ratios and show within-plate affinities, probably originating by small degrees of partial melting (less than for the extrusives of Rockall Bank) at relatively deep levels within the mantle. The Pb isotopic characteristics indicate that the lavas have been contaminated by Archaean basement rocks of the Lewisian Hebridean Terrane suggesting that George Bligh High is situated to the north of the Anton Dohrn Lineament.

Extrusive rocks of the Wyville Thomson and Ymir ridges

The Wyville Thomson Ridge is a north-west to south-east oriented structural and bathymetric ridge which separates the Faroe–Shetland Channel (to the north-east) from the northern Rockall Trough (to the south-west). Although originally modelled as a very thick pile of lavas (Himsworth, 1973; Roberts et al., 1983) it is now recognised as a structural feature with the folded Upper Paleocene lavas probably being in the order of only 2 km thick (Boldreel and Andersen, 1993). The associated Ymir Ridge is oriented in a similar direction but is shorter in length and more extensively faulted. Both ridges are compressional in origin (see Structure chapter).

Seismic interpretation shows that Palaeogene lavas crop out at the sea bed on both ridges but no samples from the extrusive volcanic interval have been acquired from either ridge within the report area. However, in Faroese waters, samples of basaltic lava been recovered by dredging from the north-west end of the Wyville Thomson Ridge (Waagstein, 1988) and olive-green to grey calcareous tuffs were acquired, also by dredging, from steep slopes just north of the report area. The tuffs have been dated as Early Eocene in age on the basis of calcareous nannofossils (Jones and Ramsay, 1982).

Extrusive rocks of the North-east Rockall Basin and Outer Hebrides High

On the east side of the Rockall Basin the extrusive volcanic interval has been drilled by six BGS boreholes and nine commercial wells (excluding well 163/06-1A). Most of these are north and west of Lewis. There are two wells east of the Hebrides Terrace Seamount (Figure 53). The BGS boreholes were terminated after drilling just a few metres into the lavas whereas the wells drilled through the volcanic interval and, between them, proved underlying Paleocene and Cretaceous sediments and metamorphic basement.

The extrusive volcanic interval comprises a range of lithologies including submarine and subaerial lavas, tuffs, volcanic breccias and volcaniclastic and tuffaceous sediments. Correlation between wells is very difficult due to the variable lithologies and thicknesses encountered and no formal lithostratigraphical terminology has been established. This has resulted in a variety of terms being employed. However, upper and lower (Hebridean) volcanic units (or members) have been recognised in several wells (e.g. 132/06-1, 164/25-1ST and 164/25-2).

The thickness of the extrusive interval is very variable. Seismic interpretation across areas of lava subcrop suggests the thickest interval in the report area may occur on the Outer Hebrides High just south of the Geikie Volcanic Centre where 2000–2500 m of lavas, tuffs and breccias may be present.

The thickest drilled interval is in well 164/07-1 (Figure 58), in the North-east Rockall Basin, which proved 1165.5 m of basalt lavas, interflow breccias and tuffs overlying 122 m of tuffs and tuffaceous volcanic breccias. The lava pile comprises between 60 and 70 individual flows with individual thicknesses ranging from 3 m to approximately 45 m. Both the frequency and thickness of weathered lavas, and the frequency of reddened and oxidised drill cuttings, increases upwards. Distinctive hyaloclastite textures occur only within the basal part of the volcanic sequence (below about 2950.3 m measured depth). Hence the upper lavas were probably extruded subaerially whereas the lower lavas and tuffs were deposited into a submarine environment (Archer et al., 2005).

In well 154/03-1, the extrusive interval is 911.6 m thick and comprises basaltic and andesitic lavas, interbedded with tuffs and tuffaceous siltstones. A boundary has been recognised at approximately 1575 m to 1590 m below sea level. Below this the lavas are less weathered, there are fewer siltstones, the log character changes and there is a change of particle type as described from drill cuttings. This lower interval can be correlated with volcanic foresets on seismic data and is interpreted to be part of a hyaloclastite delta whereas the lavas of the upper interval were extruded in a subaerial environment (Archer et al., 2005).

The extrusive volcanic intervals in wells 164/25-1ST and 164/25-2 are very similar. In well 164/25-1ST upper and lower units are 75.2 m and 62.5 m thick respectively separated by 45 m of tuffaceous claystones whereas in well 164/25-2 the corresponding thicknesses are 104 m and 133 m separated by 175 m of tuffaceous sandstones. The principal lithologies are basaltic flows (some weathered) and tuffs. Thin tuff units also occur immediately above the upper lavas in this well.

There are two wells in which lavas appear to be absent. Well 154/01-1 drilled through several tuff units (including one 150.4 m thick), volcaniclastic sandstones and possible volcanic breccias (identified from logs only). In well 164/28-1A the main extrusive interval comprises 132 m of interbedded volcanic breccias and claystones overlying 232 m of tuffs and tuffaceous claystones. There are also thin volcanic breccias and interbedded volcaniclastic units higher in the section. However nearby well 154/01-1 encountered lavas at 1620.8 m below sea bed and was terminated without drilling through the complete volcanic interval, having proved 61 m of lavas overlain by tuffaceous claystones. The extrusive ‘volcanic’ interval in the two wells east of the Hebrides Terrace Seamount is dominated by sediments rather than by lavas flows. In well 132/06-1 upper and lower ‘Hebridean volcaniclastic units’ have been designated. The upper unit is 278.6 m thick and comprises mainly volcaniclastic argillaceous sandstones and tuffaceous claystones. Only three thin lava flows are present in the unit with a combined total thickness of 15 m. The lower unit is 124.6 m thick and consists predominantly of volcaniclastic argillaceous sandstone. No flows are present in this unit. Well 132/15-1 has 38 m of interbedded lavas and tuffaceous mudstones and 8 m of tuffs and tuffaceous sandstone. Between these thin intervals is 330 m of locally tuffaceous mudstones and siltstones.

The age of the lavas in the North-east Rockall Basin and on the Outer Hebrides High has been estimated at 63–50 Ma (Ritchie et al., 1999a, b). However this is based mainly on K-Ar radiometric dating (Hitchen and Ritchie, 1993; Ritchie and Hitchen, 1996). No Ar-Ar ages have been obtained from the lavas to date. Hence best age estimates for the extrusive volcanic interval are gained from biostratigraphical dating of the overlying, underlying and interbedded sediments in the wells. Here, the interval is consistently dated as Upper Paleocene (61–54.8 Ma) with some, mainly tuffaceous events, extending into the earliest Eocene. According to Archer et al. (2005) the thick lavas encountered in well 164/07-1 were probably extruded over a short time frame of 1.5 Ma spanning the Paleocene–Eocene boundary at about 55 Ma. The Balder Formation (earliest Eocene), so widely recognised in the North Sea by its characteristic bell-shaped gamma and sonic log responses to widespread tuff deposition, is not similarly well developed west and north-west of the Scottish mainland.

The lavas in BGS boreholes are well constrained geochemically as they are cored and can be analysed. Lavas in the wells are generally not cored and there is less geochemical information available. Most lavas are weathered and altered to some extent by processes active either at the time of extrusion or after burial. The majority of the lavas in the Outer Hebrides High boreholes (and those in well 164/07-1) are alkaline or subalkaline basalts in composition and are slightly to moderately evolved. In particular those in boreholes BH85/05B, BH88/10 and BH90/10 have low Cr and Ni contents and high Fe/Mg ratios indicating extensive fractional crystallisation (Hitchen et al., 1997). Lavas in borehole BH90/10 are best described as hawaiite.

The lavas of boreholes BH85/05B, BH85/07 and BH90/04 and well 164/07-1 show no isotopic evidence of crustal contamination and are considered to be pristine mantle material. Analysis of lavas from boreholes BH90/07, BH90/10, BH88/10 and well 164/25-1ST suggest they have been contaminated by Lewisian Archaean basement.

Intrusive igneous rocks

The term ‘Atlantic Igneous Supersuite’ has been suggested by Gillespie et al. (2008) for all intrusive igneous rocks resulting from magmatism that was a precursor to, and accompanied, the opening of the North Atlantic Ocean. This includes the Hebridean Igneous Province which contains ‘almost all of the intrusions in the UK and the north-west continental shelf.’

Within the Rockall Basin intrusive igneous sills are widespread within the sedimentary succession. However their exact geographical extent is difficult to establish precisely due to recognition problems in the subsurface (compare Musgrove and Mitchener (1996) and Ritchie et al., (1999a, b)). Sills are either absent or not recognised on the Rockall and Outer Hebrides highs.

On seismic data sills are characterised by high-amplitude seismic reflections (Figure 59) which are commonly associated with diffraction patterns, on unmigrated data, at their limits. Recent 3D seismic surveys indicate that sills are complex, multi-component structures rather than single-sheet intrusions. Individual components may be designated ‘leaves’ which can be saucer shaped (concave upwards) and sourced from a deep-seated feeder system (Thomson, 2005; Thomson and Hutton, 2004). See also Smallwood and Maresh (2002) and Bell and Butcher (2002) for examples from the Faroe–Shetland Basin.

The sills are very variable in size. Those in the middle of the Rockall Basin appear flat lying and may cover tens of square kilometres. These can be confused with lavas on seismic data as some occur at approximately the same stratigraphical level. Sills towards the margins of the basin are usually smaller in extent and are more likely to be inclined and cross-cutting the sedimentary bedding. The attitude of these may possibly have been influenced by fault planes and tilted fault blocks.

Lavas and sills can also be difficult to distinguish in wells. However sills do not possess reddened flow tops and are generally less weathered. Thick sills may be coarser grained in the middle and are likely to generate thermal aureoles both above and below the intrusion which can be recognised on log data. Seismic interval velocities may also be higher than those for a succession of multiple thin lava flows.

Within the report area well 164/07-1 penetrated over 70 sills intruded into ?Albian to ?Coniacian mudstones (Figure 58). Many of these sills are likely to be linked in the subsurface and may represent multiple intrusion events. The sills are dolerites and range in thickness from 1.5 m to 151.8 m. They comprise more than half of the drilled section below the lavas in the well. Geochemically the sills are slightly enriched in light REE probably due to contamination by fluids derived from the host mudstones (Archer et al., 2005). Ages in the range 62.2+6.8 to 64.2+0.4 Ma (Early Paleocene) have been obtained from the sills by Ar-Ar dating and implies that the sills were emplaced in the order of 8 Ma before the overlying lavas were extruded despite the fact that they are compositionally similar.

Well 164/25-1ST proved a number of sills in the West Lewis Basin intruded into various stratigraphical intervals including the Permo-Triassic and Upper Cretaceous. However two sills also intrude the Upper Paleocene. The upper sill is 188 m thick, described as doleritic and appears to have generated little or no thermal aureole. The lower one is 210 m thick and appears to be part of complex multi-leaved intrusion comprising dolerite, gabbro, hornfels and meta-siltstones. In well 154/01-1 the well log describes the 35.4 m interval from 2539.1 m to 2574.5 m TVDss as a dolerite sill intruded into Upper Paleocene claystones. This must represent either very late stage sill emplacement or possible misidentification of lavas in the well. Further south well 132/15-1 penetrated two thin sills of 5 m and 7 m thickness intruded into Campanian mudstones. The lower thicker sill is described as microgranite, the only sill of this lithology anywhere in the report area.

No sills have been radiometrically dated apart from those in well 164/07-1 (see above). Hence the age of the intrusions has to be estimated by dating the host rock (the sills must be younger), by theoretical modelling and from regional considerations. Musgrove and Mitchener (1996) suggest that some of the sills may be Cretaceous in age and were intruded in areas of maximum crustal thinning (i.e. in the Rockall Basin). Although it seems possible that some sills in the vicinity of the Cretaceous seamounts may be Cretaceous in age, the vast majority are likely to be Mid to Late Paleocene associated with the major phase of volcanism on the UK Atlantic Margin.

Chapter 9 Cenozoic sedimentary rocks

By Martyn Stoker

The Cenozoic Era began at 65.5 Ma (Gradstein et al., 2004) and marked the onset of an interval of major global change, including the rearrangement of tectonic plates in the Arctic–North Atlantic region, the realignment of oceanic circulation and the transition from a warm ‘greenhouse’ to a cold ‘icehouse’ climate. This legacy of change is recorded in the geological record of the Rockall Basin where, following the Early Eocene initiation of sea floor spreading between north-west Europe and Greenland (see Cretaceous and Palaeogene igneous rocks chapter), a succession of post-rift tectonic episodes drove regional changes in the patterns of sedimentation and deep-ocean circulation along the ocean margin (Praeg et al., 2005; Stoker et al., 2005b, c, 2010; Holford et al., 2009). Although the post-rift development of the continental margins bordering the north-east Atlantic is generally classed as passive (i.e. tectonically quiescent), the evidence for a pattern of stepwise subsidence and recurrent uplift preserved in the Cenozoic succession within and adjacent to the Rockall Basin indicates a margin evolution that has been anything but passive.

The Cenozoic is divided into Palaeogene, Neogene and Quaternary periods (Berggren et al., 1995a, b; Gradstein et al., 2004; Gibbard et al., 2010). It should be noted that the Neogene/Quaternary boundary has only recently been formally ratified by the International Union of Geological Sciences, and dated at 2.58 Ma (Gibbard et al., 2010). This boundary is adopted in this report.

The present-day physiographical configuration of the Rockall Basin as a large underfilled deep-water basin (Figure 60) was largely established during late Palaeogene (late Eocene–Oligocene) differential subsidence of the north Rockall, south Rockall and North-east Rockall basins (herein collectively referred to as the Rockall sub-basins). Prior to this event, the syn- to early post-rift (mid Cretaceous–Palaeogene) history of the Hebrides–Rockall region evolved in a compartmentalised (segmented) manner, controlled to some extent by north-west-trending transfer structures, such as the Anton Dohrn Lineament, which separates the north and south Rockall basins (Kimbell et al., 2004; see Structure chapter). Marginal half-grabens, such as the West Lewis Basin (Figure 60), separated from the North-east Rockall Basin by the north-north-east-trending West Lewis High (Figure 60) and (Figure 61), may have remained active through the Paleocene interval. The similarly trending Darwin–Geikie High, see (Figure 7), separates the North-east Rockall Basin from the north Rockall Basin. In the subsequent late post-rift (late Palaeogene–Quaternary) period, the region developed as a more integrated ocean margin.

The structural terminology used in this chapter reflects this change in structural development. Structural elements such as the north and south Rockall basins, North-east Rockall Basin, and the Outer Hebrides High, are most applicable to the syn- to early post-rift (early Palaeogene) phase of margin development rather than the late post-rift (late Palaeogene–Quaternary) expression of the margin, for which the more unified structural/physiographical terms such as Rockall Basin (replacing Trough) and Hebrides Shelf apply to an ocean margin that is still being shaped today by bottom-current and sporadic mass-flow processes. In this chapter, the term Hebridean margin is also utilised as an informal geographical term when discussing the late Neogene–Quaternary (Pliocene–Holocene) development of the Hebrides shelf and slope.

Cenozoic strata are widespread throughout most of the report area; their absence from parts of the Hebrides Shelf (Figure 60) is largely the result of mid- to late Pleistocene glacial erosional processes on the shelf. The outcrop of Palaeogene strata on the Rockall High and George Bligh High contrasts with the Outer Hebrides High, where its landward limit is marked by erosional truncation on the outer part of the present-day Hebrides Shelf, and burial beneath Neogene and Quaternary strata. This relationship reflects a difference in the development of the eastern and western flanks of the Rockall Basin since Late Eocene–Oligocene times. Late Neogene–Quaternary seaward progradation and build-out of the Hebridean margin (Figure 61) contrasts with the Rockall and George Bligh regions, which have remained essentially starved of sediment input since the late Palaeogene. The axial seamounts of Rosemary Bank, Anton Dohrn and Hebrides Terrace have also remained largely isolated since the mid Cenozoic, accumulating only a veneer of Neogene and Quaternary strata.

It has previously been noted that the landward limit of pre-glacial (pre-Mid Pleistocene) Cenozoic strata on the Hebrides Shelf (as delineated by the broadly coincident subcrop limits of the Palaeogene strata and the pre-glacial late Neogene–Quaternary strata on Figure 60) approximately correlates with, and is probably related to, the zone marking the transition from ‘normal’ thickness continental crust beneath the shelf to highly attenuated crust beneath the Rockall Basin (cf. Stoker et al., 1993 and references therein) (see Structure chapter and (Figure 14) and (Figure 16)b. The largely offlapping geometry of the Palaeogene succession is similarly preserved on the outer part of the Rockall High and George Bligh High, beneath which there is a comparable change in crustal thickness (Roberts et al., 1988).

Recognition of the base of the Cenozoic succession is problematic across much of the area, as the high impedance contrast of lower Palaeogene volcanic strata commonly obscures or reduces the quality of the image of the underlying geology on seismic profiles. Beneath the outer Hebrides Shelf and Rockall High, the base of the Palaeogene locally occurs at a depth of about 200 m below sea level. This deepens into the Rockall Basin. In areas not obscured by the volcanic rocks the base lies at a depth of 3000–3500 m below sea level in the north-east part of the basin (the North-east Rockall Basin: Tate et al., 1999; BGS, 2007), increasing to 4000–5000 m in the central part of the basin (north and south Rockall Basin: BGS, PAD, 2002) (Figure 61). The total Cenozoic thickness, including volcanic rocks, is estimated to range from 1500 to 3000 m, with a maximum thickness preserved in the area underlain by the North-east Rockall Basin (Figure 60).

The nature of the basal contact is reportedly unconformable and erosional in the Hebrides–Rockall and adjacent regions (e.g. McDonnell and Shannon, 2001). This is likely the result of Late Cretaceous (Maastrichtian) inversion and uplift of the proto-North-east Atlantic region (Figure 62), forced by strike-slip tectonics in response to the northward propagation of the Atlantic spreading system into the Labrador Sea (Coward et al., 2003). Late Cretaceous inversion and uplift has been recorded on the Porcupine High (Haughton et al., 2005) (Figure 63), and in the North-east Rockall Basin (Tate et al., 1999), as well as the West Shetland (Roberts et al., 1999) and Mid Norwegian (Doré et al., 1999) regions.

The Labrador Sea opened in the Paleocene (Chalmers, 1997), and an extensional regime persisted in the proto-North-east Atlantic region. During Late Paleocene to earliest Eocene times, widespread uplift of the proto-North-east Atlantic was accompanied by extensive and voluminous magmatism, both driven by the North Atlantic hot spot (Figure 64). Some workers (e.g. White and Lovell, 1997; Jones et al., 2002; Maclennan and Lovell, 2002; Smallwood and Gill, 2002) interpret the Iceland hot spot as a mantle plume beneath the lithosphere, which provided support until continental breakup. The extensive and coeval volcanism has also been linked to the interaction between rifting and an Iceland mantle plume (Smallwood and White, 2002 and references therein). An alternative view is that the hot spot anomaly is an upper mantle response to plate breakup, of which the volcanism is a by-product of this extension (Doré et al., 1999; Foulger and Anderson, 2005; Lundin and Doré, 2005a, b). Whatever the cause, this tectonic episode was the precursor to continental breakup between Greenland and north-west Europe, which occurred at about 54 Ma (Chron 24r) (Ritchie et al., 1999) forming the Iceland Basin to the north-west of the Rockall Plateau (Figure 65). As the Iceland Basin widened, the Iceland volcanic province was formed, including the Greenland–Scotland Ridge and Iceland itself (Figure 65) to (Figure 67) (Foulger and Anderson, 2005).

The post-rift development of the Hebrides–Rockall region has been classed as tectonically quiescent (Musgrove and Mitchener, 1996; Tate et al., 1999). However, in common with the north-east Atlantic region in general, the Hebrides–Rockall region is known to have experienced tectonic movements during the Cenozoic, manifest as significant departures from the post-rift pattern of subsidence, including episodes of accelerated subsidence and inversion/exhumation episodes that were, at least in part, coeval (Doré et al., 1999; Japsen and Chalmers, 2000; Stoker et al., 2005a, b, 2010; Praeg et al., 2005; Holford et al., 2009, 2010). Three main tectonic episodes have had an important influence on the Cenozoic post-rift stratigraphical record, which preserves a series of unconformity-bounded megasequences (Figure 62).

During the Late Eocene–Oligocene, strongly differential subsidence (sagging) in the Rockall sub-basins (and adjacent Hatton and Porcupine basins) outstripped sedimentation, driving a deepening of well over a kilometre (Vanneste et al., 1995; Stoker, 1997; Stoker et al., 2001; McDonnell and Shannon, 2001) (Figure 65). This event appears to coincide with a phase of compression on the North Hatton High and Lousy Dome (Johnson et al., 2005).

In the Late Oligocene–Mid Miocene, compressive tectonism resulted in the formation of inversion structures, including the Wyville Thomson and Ymir ridges and complementary synclines, e.g. the present-day Faroe Bank Channel (Boldreel and Andersen, 1995; Lundin and Doré, 2002; Johnson et al., 2005; Stoker et al., 2005b; Ritchie et al., 2008) (Figure 66).

From the Early Pliocene, onshore uplift was accompanied by accelerated offshore subsidence and the progradation of sedimentary wedges: a tilting of the continental margin (Cloetingh et al., 1990; Japsen and Chalmers, 2000; Stoker, 2002) (Figure 67).

Recent studies of the thermal history of the British Isles and its surrounds are consistent with this pattern of episodic deformation across the continental margin. Apatite fission-track analysis of basement and cover rocks from north-west Scotland, southern Britain and the Irish Sea reveal episodes of regional post-rift uplift and exhumation in both the Mid and Late Cenozoic (Green et al., 2002; Holford et al., 2005, 2009, 2010; Hillis et al., 2008). From these studies, it has been proposed that crustal compression linked to plate-boundary forces was the major cause of Cenozoic uplift.

The variations in uplift, subsidence and sediment supply associated with these tectonic movements had a pronounced effect on the patterns of both shelf-margin to deep-water sedimentation and of palaeoceanographical circulation. Late Eocene to Early Oligocene sagging cut off sediment supply to Eocene prograding wedges that had developed in the north, south and North-east Rockall basins (Stoker, 1997; Stoker et al., 2001; McInroy et al., 2006). An important consequence of sagging was the inception of the present morphological expression of the Rockall Basin (Trough) as a deep-water basin, and the onset of deep-water current circulation and contourite drift deposition in this basin (Figure 65). Early to Mid Miocene compression in the area of the Wyville Thomson–Ymir Ridge and in the Faroe–Shetland region modified the physiography of the northern margin of the Rockall Basin, such that the Faroe Bank Channel was formed, which together with the Faroe–Shetland Basin instigated a passageway (herein referred to as the Faroe Conduit) for the persistent exchange of intermediate- and deep-water masses between the Atlantic Ocean and Nordic Seas, across the Greenland–Scotland Ridge (Stoker et al., 2005b, c) (Figure 66). The other significant deep-water passageway across the Greenland–Scotland Ridge, the Denmark Strait, may also have developed at about this time, as the Greenland–Scotland Ridge became fully submerged (Thiede and Eldholm, 1983). The Early Pliocene event records a seaward tilting of the Outer Hebrides High, and possibly Rockall High, in which uplift and erosion was accompanied by increased offshore subsidence, resulting in basinal progradation of the shelf-margins bordering the Rockall Basin, as well as deep-marine erosion during a reorganisation of bottom current patterns (Stoker, 2002; Stoker et al., 2005b). Progradation and development of the Hebridean margin was enhanced by Late Pliocene–Pleistocene glaciation (Figure 67). This climatic extreme represents the culmination of a progressive cooling of Northern hemisphere climate since the Mid Miocene, although global cooling and the change from greenhouse to icehouse conditions began earlier, in the Late Eocene (Zachos et al., 2001) (Figure 62).

Seismic stratigraphy

No formal stratigraphy exists for the Cenozoic succession in the Hebrides–Rockall region, though a lithostratigraphy-based scheme has recently been developed for the late Neogene–Quaternary deposits (Stoker et al., 2011). For the most part, regional investigation of the Cenozoic succession has been largely based on a seismic-stratigraphy approach, albeit supported by biostratigraphical control (foraminifera, dinoflagellate cysts and calcareous nannofossils), much of it unpublished, from BGS boreholes and short, rock and sediment cores, together with released commercial well data. For correlation purposes, the biostratigraphical data are calibrated to the calcareous nannoplankton zonal scheme of Martini (1971).

On the basis of seismic stratigraphy, the Cenozoic succession in the Rockall Basin has been divided into four megasequences consisting, in ascending stratigraphical order, of: 1) RPd (Paleocene–Eocene); RPc (Oligocene–Lower Miocene); 3) RPb (Middle Miocene–Lower Pliocene); and, 4) RPa (Lower Pliocene–Holocene) (Stoker et al., 2005a) (Figure 62). The ‘RP’ notation was introduced and defined as a common stratigraphical nomenclature for the Rockall–Porcupine region, and reflects the integration and unification of formerly separate schemes in the Rockall (Stoker et al., 2001) and Porcupine (McDonnell and Shannon, 2001) basins (Figure 63). Although the megasequences are of informal stratigraphical status, they represent physically mappable units throughout the Rockall Basin, on the basis of four major correlation surfaces: (1) a Paleocene unconformity (C40); (2) a Late Eocene unconformity (C30); (3) a late Early/early Mid Miocene marker (C20); and (4) an Early Pliocene unconformity (C10) (Figure 62). This stratigraphy is most clearly observed within the Rockall Basin with marginal areas commonly showing condensed or eroded remnants of the succession (Figure 61). Other important, albeit more localised, breaks occurring within the Cenozoic succession, include the Base Neogene Unconformity in the north-eastern part of the basin, and the Intra-Pleistocene Glacial Unconformity on the Hebridean margin (Figure 62). The latter divides the recently defined Hebrides Margin Group (late Early Pliocene–Mid Pleistocene) and overlying Eilean Siar Glacigenic Group (Mid Pleistocene–Holocene) (Stoker et al., 2011). All of these key surfaces are summarised in (Figure 61) and detailed below, in ascending stratigraphical order.

Key Cenozoic correlation surfaces

C40 unconformity

The C40 unconformity is best defined from the south Rockall Basin, where it is identified as a basinwide, very high-amplitude, continuous, reflector, which is onlapped by lower-amplitude, less-continuous reflectors (McDonnell and Shannon, 2001). In the south Rockall Basin, together with the south-eastern part of the north Rockall Basin (well 132/15-1: see (Figure 68)) and the Porcupine Basin, this boundary is envisaged to mark the top of the Cenomanian (Upper Cretaceous) to Danian (Lower Paleocene) chalk sequence (Corfield et al., 1999; McDonnell and Shannon, 2001). This represents the change from carbonate to clastic deposition. Over most of the report area, however, the boundary is not so clearly defined due largely to the masking effect on seismic profiles of the Upper Paleocene–Lower Eocene lavas (Figure 61). In the North-east Rockall Basin, Tate et al. (1999) describe the base of the Cenozoic succession as an angular unconformity marked by Lower Paleocene sediments overlying faulted, syn-rift, Lower Cretaceous and older strata. It remains unclear how this surface relates to C40.

C30 unconformity

The C30 unconformity is expressed as a high-amplitude reflection around the margin of the Rockall Basin, where it forms a prominent angular unconformity that is onlapped by the overlying RPc–RPa succession (Stoker et al., 2001) (Figure 61). A similar relationship is observed adjacent to the seamounts of Rosemary Bank, Anton Dohrn and Hebrides Terrace. Traced basinwards, the C30 unconformity locally loses expression within the more parallel-bedded basin fill, though commonly it can be followed across the basin (Figure 61). This boundary is the expression of the Late Eocene/Early Oligocene rapid, differential, subsidence (sagging) event that created the present-day deep-water basin (Stoker, 1997; McDonnell and Shannon, 2001; Stoker et al., 2001, 2005a, b) (Figure 62). Equivalent surfaces occur in the Hatton Basin (R4 reflector of Roberts, 1975a; Stoker, 1997), the Porcupine Basin (McDonnell and Shannon, 2001), and on the southern margin of the Edoras High (reflector III of Bull and Masson, 1996) (Figure 63).

Base Neogene Unconformity

Over most of the report area, and throughout the Rockall Basin in general, the boundary between the Palaeogene and Neogene strata lies within the RPc megasequence, and does not correspond to a seismically defined boundary (Figure 61). However, a latest Oligocene–Mid Miocene hiatus is present along the north-eastern margin of the basin, particularly adjacent to the Wyville Thomson Ridge, where pre-C20 Lower Miocene sediments (upper part of megasequence RPc) are absent, resulting in an angular unconformity between Palaeogene and Neogene strata (Figure 61) and (Figure 62). This boundary, informally termed the Base Neogene unconformity, is inferred to be a composite unconformity spanning the Early Miocene interval up to the formation of the C20 marker (Figure 62). In the Faroe–Shetland region, an equivalent interval bounded by the Top Palaeogene and Intra-Miocene (see below) unconformities represents the convergence of these surfaces on the northern flank of the Wyville Thomson Ridge (Stoker et al., 2005a, b). These composite unconformities are interpreted to have formed in response to the Late Oligocene to Mid Miocene compressive tectonism, focused in the north-east Rockall and Faroe–Shetland regions (Johnson et al., 2005; Stoker et al., 2005b, c; Ritchie et al., 2008) (Figure 62).

C20 marker

The C20 marker is a submarine surface that formed on the margin of the Rockall Basin in the area to the west and south of Rosemary Bank Seamount (Stoker et al., 2001) (Figure 68). Unlike the Base Neogene unconformity, the C20 marker is an important regional surface that is recognised throughout the Rockall Basin. The C20 marker is characterised by a subhorizontal, high-amplitude acoustic signature that varies from a sharp, single, reflection to a reflective zone several tens of milliseconds thick that onlaps the basin margin. At DSDP site 610, in the southern part of the Rockall Basin (Figure 63), the reflective zone has been attributed to the diagenetic alteration of clay minerals, and the development of massive smectite, originally derived from volcanic ash (Dolan, 1986). This is consistent with the observation that the highest amplitude reflectors within this zone commonly vary with space and time, and concurs with the proposition of Dolan (1986) that the spread of smectite across this zone may have occurred with sporadic hiatuses and periods of deposition. The C20 marker correlates with the ‘green’ reflector of Masson and Kidd (1986) at site 610, which is dated as late Early to early Mid Miocene in age. This surface marks the start of an intense phase of sediment-drift growth throughout the basin (Stoker et al., 2001), preserved in the RPb megasequence. In a regional context, the C20 marker is equivalent to North-east Atlantic reflector R2 of Miller and Tucholke (1983), the Intra-Miocene Unconformity in the Faroe–Shetland and Mid Norwegian regions (Johnson et al., 2005; Stoker et al., 2005a, b), and reflector II of Bull and Masson (1996) to the south of the Edoras High (Figure 63).

C10 Unconformity

The C10 unconformity is a well-defined surface that extends from the outer part of the Hebrides Shelf into the Rockall Basin, where it forms a high-amplitude, continuous, reflection that is characteristically angular and cut into older Neogene strata by submarine erosional processes (Stoker et al., 2001) (Figure 61). On the outer Hebrides Shelf, and locally along the eastern flank of the Rockall High, the surface is progressively downlapped by clinoforms of the seaward prograding RPa megasequence; in deeper water, the surface is onlapped by high-amplitude reflections. On the western flank of the Rockall Basin, the C10 unconformity is, itself, commonly eroded and truncated by the present-day sea bed (Figure 61). In the Rockall Basin, the C10 unconformity has been dated as Early Pliocene in age, between 3.85 and 4.5 Ma, based on biostratigraphical data from ODP site 981, DSDP site 610 and BGS borehole BH88/07, BH88/07A (Stoker et al., 2001; Stoker, 2002) (Figure 60) and (Figure 63). This unconformity correlates with the intra-Neogene unconformity in the Faroe–Shetland region and the base Naust unconformity off Mid Norway, all of which are inferred to have formed in response to intra-Pliocene tilting (Stoker et al., 2005a, b) (Figure 62).

Intra-Pleistocene Glacial Unconformity

The Intra-Pleistocene glacial unconformity is a planar to irregular erosion surface preserved on the Hebrides shelf and slope (Figure 61). On the shelf, it characteristically truncates older Pliocene–Pleistocene prograding strata, and is overlain by a flatter lying sequence that displays an aggrading geometry, and preserves glacial moraines. On the slope, it is overlain by deposits associated with prograding, glacially-fed, trough-mouth fans, such as the Sula Sgeir and Barra-Donegal fans (Figure 67). In the newly defined lithostratigraphy for the late Neogene–Quaternary succession, the glacial unconformity separates the Hebrides Margin Group (below) from the Eilean Siar Glacigenic Group (Stoker et al., in press), and is considered to mark the switch from restricted (down to coastline) to expansive (shelf-wide) glaciation. Sediments cored from above the glacial unconformity are younger than 0.5 Ma (Stoker et al., 1994). Equivalent surfaces are present in the Faroe–Shetland and Mid Norwegian regions (Stoker, 1999; Rise et al., 2002; Stoker et al., 2005a).

Palaeogene

Sedimentary and volcanic rocks of Palaeogene age occur throughout most of the report area, and are absent only on the inner parts of the Outer Hebrides High, the flanks of the Rosemary Bank and Anton Dohrn seamounts, and locally on the Rockall High (Figure 68). Subdivision of the Palaeogene succession to series level, as shown on (Figure 68), remains tentative, as is evident from the assignment of these strata to the two regional megasequences of RPd (Paleocene–Eocene) and RPc (Oligocene–Early Miocene) (Figure 62). Series definition is best established on the flanks of the various Rockall sub-basins, where most short cores, shallow boreholes and commercial wells are sited (Figure 68). However, the marginal succession tends to be thin relative to the basinal succession. The available lithological and age data are summarised in (Figure 69) and (Figure 70). The age data are based on biostratigraphical and radiometric information.

During the Early to Mid Paleocene, renewed tectonic movement controlled deposition adjacent to major faults, but Late Paleocene to Early Eocene events were dominated by regional uplift and massive volcanism driven by processes associated with North Atlantic rifting (Figure 63). Sediments deposited during the latter interval are commonly typical of nonmarine to shallow-water nearshore environments. Marginal or shallow-marine sedimentation prevailed throughout much of the Eocene, interspersed with pulses of tectonic uplift and erosion, especially on the flanks of the Rockall sub-basins. This tectonic instability is manifest by the widespread development of prograding sediment wedges that built out from the Outer Hebrides and Rockall highs into the adjacent basins (Stoker et al., 2010), which form part of a larger system of shelf-margin progradation in the North Atlantic region (Figure 64). A reduction in the supply of coarse-grained clastic material into the North-east Rockall Basin may be associated with a deepening of the basin from the Mid Eocene (Figure 65). A deeper-water environment may have prevailed to the south of the report area, in the south Rockall Basin, for much of the Paleocene–Eocene interval.

The initial development of the present bathymetric configuration of the Rockall Basin began in Late Eocene–Early Oligocene times concomitant with the sagging of the Rockall sub-basins (Figure 65). On a regional scale, this event appears to coincide with the cessation of spreading in the Labrador Sea (Lawver et al., 1990) and the Pyrenean Orogeny (Lundin and Doré, 2002) (Figure 62). An important consequence of sagging was the onset of deep-water current circulation and contourite drift deposition in the Rockall Basin (Masson and Kidd, 1986; Stoker, 1997; Stoker et al., 2001) (Figure 62) and (Figure 65), which a number of workers have suggested records the onset of overflow of Norwegian Sea Deep Water across the Greenland–Scotland Ridge from the Late Eocene–Early Oligocene (Miller and Tucholke, 1983; Davies et al., 2001). However, Wold (1994, 1995) and Stoker (1997) have proposed that late Palaeogene circulation in the Rockall Basin was southerly derived, with no significant connection across the Greenland–Scotland Ridge at this time. This is supported by C13 and O18 data from numerous DSDP sites that show bottom circulation patterns in the North Atlantic Ocean prior to the Neogene were driven by southerly derived water masses, including Tethyan Outflow Water, as far north as the Greenland–Scotland Ridge and including DSDP site 610 (Ramsay et al., 1998) (Figure 63). The Late Eocene to Oligocene initiation of bottom-current activity and contourite drift accumulation in the Rockall Basin is therefore consistent with a northward extension of activity that had been ongoing farther south since the Mid Eocene (Ramsay et al., 1998; Norris et al., 2001). Late Eocene–Early Oligocene sagging also cut off the regional sediment supply that had fed the extensive, Eocene, prograding sediment wedges, although discrete, Oligocene, lowstand fans developed locally along the flanks of the Rockall Basin (Stoker, 1997; Egerton, 1998; Stoker et al., 2010). In addition to the deep-water contourites and lowstand fans, other sediments deposited at this time are indicative of nonmarine (in the inner Hebridean region) and outer shelf/upper slope environments (Outer Hebrides High).

Paleocene

Paleocene sedimentary rocks are largely restricted to the area of the Rockall sub-basins and the West Lewis Basin (Figure 61) and (Figure 68), although details of their distribution, thickness and lithology are poorly known. Nonmarine or marginal marine sediments, commonly containing volcaniclastic layers, have been proved on the flanks of the basins, whereas marine siliciclastic sediments have been recovered from within the basins, both above and below volcanic rocks. Although an unconformity has been proved between the Paleocene and Eocene rocks on the flanks of the West Lewis Basin, available well-log data imply apparent conformity across this boundary in the deeper-water wells (Figure 69).

The topography at the basin margin might not have been significant, with a shallow-marine environment covering much of the northern part of the area. South of Anton Dohrn Seamount, water depths were probably in excess of 200 m. In the latest Paleocene the crest of Anton Dohrn Seamount may have been exposed, and fringed by a high energy, nearshore–shallow-marine environment, largely starved of clastic sediment, but including locally reworked igneous fragments incorporated into a carbonate matrix (see below). The Rockall High and George Bligh High accumulated a more mixed clastic–carbonate succession, including calcareous conglomerates with basalt clasts, interpreted as conglomeratic hardgrounds.

On the southern part of the Outer Hebrides High, three BGS boreholes (BH90/11, BH90/12, BH90/12A and BH90/13), drilled just to the east of, and underlying, the feather-edge of the main basalt horizon, proved, collectively, 4.6 m of Lower Paleocene (Danian) dark grey mudstones, and 45.26 m of Middle to Upper Paleocene (Selandian to Thanetian) muddy and pebbly sandstones consisting largely of eroded volcanic debris (Figure 69). Palynological considerations suggest an open-marine depositional environment, but the association of graded and cross-bedded sedimentary structures, the presence of abundant plant fragments (including leaf debris), and the carbonaceous nature of the sediments, are more indicative of a deltaic or lagoonal environment.

West of the Outer Hebrides High, two commercial wells (132/15-1 and 132/06-1) have recovered Paleocene sedimentary and interbedded volcanic rocks from the northern margin of the south Rockall Basin (Figure 68). Well 132/15-1 tested a 746 m-thick succession of Lower (Danian) to Upper (Thanetian) Paleocene strata. The lithologies are mudstone dominated and tuffaceous, with some sandstone and limestone units, particularly in the Danian section (Figure 69). Farther west, well 132/06-1 recovered 856 m of Paleocene rocks punctuated by an unconformity. The Danian succession lies unconformably on Cretaceous strata and consists of 42 m of mudstones with thin limestone and sandstone beds, deposited on a marine outer shelf or slope. This is unconformably overlain by 814 m of Thanetian claystone, siltstone and volcaniclastic rocks, including thin basalt lavas. These wells lie to the west (seaward) of the Hebridean volcanic escarpment (Figure 61) and (Figure 68), which has been interpreted to mark a shoreline during the Paleocene (Musgrove and Mitchener, 1996). The intra-Paleocene unconformity development was coeval with an increased clastic input that implies uplift of the adjacent highs (Corfield et al., 1999). Seismic records suggest that this Paleocene marine succession extends throughout the south Rockall Basin (McDonnell and Shannon, 2001), perhaps with input from the southern end of the Rockall High via deltaic systems (Walsh et al., 1999). A comparable succession of marine, cherty, Danian limestone and claystone overlain by Upper Paleocene claystones, resting with minor unconformity on Maastrichtian limestone, has also been recovered in well 12/13-1A, in the Erris Basin, south of the Outer Hebrides High (Figure 63).

Farther north, commercial wells and BGS shallow drilling has yielded information from the West Lewis and North-east Rockall basins (Figure 68). The fault-bounded West Lewis High separates these basins (Figure 61) and (Figure 68). Movement on the bounding faults reflects an episode of accelerated subsidence in the Paleocene, which resulted in a marked thickening of the sediments adjacent to the footwall block (Tate et al., 1999).

Well 164/25-1Z penetrated the Paleocene infill of the West Lewis Basin adjacent to the West Lewis High, and proved 61.5 m of Danian sandstone and mudstone unconformably overlain by 1256.2 m of Upper Paleocene basalt, with interbedded sandstone and mudstone (Fig. 69). An unconformity separates the Paleocene and Eocene rocks in this well. Basin-floor fan deposits have been interpreted to comprise the Lower Paleocene clastic succession (Tate et al., 1999), which are unconformable on Upper Cretaceous strata. BGS borehole BH85/07 occurs in the north of the basin, adjacent to the intersection of the Sula Sgeir High and the south-east end of the Wyville Thomson Ridge (Figure 68), and proved 2 m of pale grey, fine-grained sandstone with lignite bands, overlying Upper Paleocene amygdaloidal basalt.

Wells 164/25-2 and 154/03-1 were drilled above the West Lewis High and its extension towards the Outer Hebrides High (Figure 68) and (Figure 69). Well 164/25-2 contains 407.8 m of interbedded sandstone, limestone, mudstone, lava and tuff dated as Late Paleocene to Early Eocene in age, unconformably overlying Precambrian basement. Well 154/03-1 contains 909 m of Upper Paleocene to Lower Eocene basalts with thinner interbeds of red-brown, tuffaceous siltstone. The base of the sequence is marked by a 19 m-thick section of poorly differentiated Upper Cretaceous–Upper Paleocene siltstones overlying Precambrian gneiss.

Wells 153/05-1, 154/01-1, 164/28-1A and 164/07-1 tested the Paleocene succession on the eastern and northern flanks of the North-east Rockall Basin (Figure 68) and (Figure 69). Well 153/05-1 penetrated 70.1 m of Upper Paleocene claystone with thin tuffaceous beds overlying 61 m of basalt. In well 154/01-1, 720 m of Upper Paleocene sandstone, mudstone and subordinate limestone interbedded with tuff and breccia unconformably overlies Cretaceous rocks. This is overlain by 155 m of sandstone and claystone (including a Nummulites-rich bed) and tuff dated as Paleocene to Early Eocene. In well 164/28-1A, the Lower Paleocene consists of 16 m of mudstone and limestone, which is unconformably overlain by 863 m of Upper Paleocene interbedded sandstone, mudstone, basaltic breccia and tuff. The lower 300 m of the Upper Paleocene section is sandstone dominated with thin mudstone beds; above this, tuff and volcaniclastic breccia form the majority of the succession, albeit capped by claystone and sandstone. Well 164/07-1 proved 121 m of Upper Paleocene volcaniclastic rocks lying unconformably on Upper Cretaceous strata. The Upper Paleocene section consists of 32.6 m of volcanic conglomerate, comprised of basaltic pebbles in a tuffaceous matrix, overlain by 88.4 m of dark grey tuff (Archer et al., 2005).

The Paleocene succession underlying basalt is interpreted to thicken westwards into the central part of the North-east Rockall Basin. Waddams and Cordingley (1999) have modelled up to 1 km of Paleocene sediments underneath the basalt, whilst Tate et al. (1999) imply a thicker succession adjacent to the Wyville Thomson Ridge.

On the Anton Dohrn Seamount, BGS borehole BH90/15, BH90/15A recovered 3.5 m of limestone overlying 8.9 m of gravel (Figure 69). The limestone is bioclastic and contains abundant hardgrounds; it has been dated as Early Paleocene to Early Eocene in age. The underlying shelly gravel is assumed to be of Paleocene age and is interpreted to be a winnowed lag deposit. Both rocks are envisaged to have formed in shallow water, with the gravel indicating relatively high-energy conditions.

Along the western margin of the north and south Rockall basins, Paleocene sediments have been recovered in a number of BGS boreholes and short rock cores (Figure 68) and (Figure 70). BGS core 57-13/54 recovered 0.66 m of Upper Paleocene agglomerate from the fault scarp on the east side of Rockall High, and cores 58-14/31 and 32 recovered 0.48 m of ?Upper Paleocene–Lower Eocene calcareous conglomerate, with basalt clasts set in a matrix of biosparite, from the escarpment on the northern flank of Rockall High. On the crest of George Bligh High, cores 58-14/8, 58-14/42, 58-14/57 and 58-14/58 similarly recovered up to 0.85 m of Upper Paleocene calcareous conglomerate with basalt clasts up to boulder grade. Core 58-14/42 terminated in basalt; the latter was also proved in adjacent BGS borehole BH94/07, which included a thin layer of sandstone with shell and plant material. Short rock cores 58-14/10, 58-14/11, 58-14/43 and 58-14/55 proved up to 1.2 m of Upper Paleocene to Lower Eocene deltaic to shallow-marine, calcareous sandstone and mudstone, with thin limestones and rare plant fragments, from a prograding sediment wedge on the south-east flank of George Bligh High, which continued to develop through the Eocene (Stoker et al., 2001).

Eocene

Eocene sedimentary rocks predominantly occur on the flanks of, and within, the Rockall sub-basins, as well as atop the Anton Dohrn Seamount (Figure 61) and (Figure 68). They commonly crop out along the western margin of the present-day Rockall Basin, and around the base of Anton Dohrn (Figure 61) and (Figure 71). Following extrusion of the Late Paleocene–Early Eocene volcanic succession, much of the Eocene sedimentary record indicates phases of nearshore or shallow-marine deposition, interspersed with periods of relative uplift and erosion of the marginal areas (Figure 69) and (Figure 72).

This has resulted in a predominantly clastic succession, commonly tuffaceous in the Lower to lower Middle Eocene section, and characterised by the development of prograding sediment wedges that built out both from the Hebridean and Rockall (including George Bligh) margins (Stoker et al., 2001, 2010; McInroy et al., 2006) (Figure 61), (Figure 64) and (Figure 72). In the north-eastern part of the region, the West Lewis Basin may have been less of a discrete basin from Early to Mid Eocene times, as the Eocene sediment wedge advanced into the North-east Rockall Basin, prograding by downlap across the West Lewis Basin and High (Figure 61). Southward progradation from the Wyville Thomson–Ymir ridge system implies the early stages in the development of this high (Tate et al., 1999). Late Eocene differential subsidence of the Rockall sub-basins, possibly incorporating tilting of the adjacent highs (up to 4°), resulted in extensional deformation and slumping of the sediment wedges, followed by basin-margin erosion linked to the formation of the C30 deep-water unconformity (Stoker, 1997; Corfield et al., 1999; Stoker et al., 2001; Haughton et al., 2005).

In the report area, the Eocene succession has been tested, to varying degrees, in all of the Rockall sub-basins by a number of commercial wells and BGS boreholes (Figure 69). At many of these sites, the Eocene succession is dominated by calcareous siltstones with intervals of other clastic rocks. Exploration wells in the region generally begin their logging within the Eocene succession, which means that the upper part of the Eocene succession is commonly not recovered.

In the south Rockall Basin, well 132/06-1 proved an almost complete Eocene marine succession, which consists of 621 m of claystones and siltstones. However, a possible unconformity is noted at the Ypresian–Lutetian boundary (Figure 69). According to Corfield et al. (1999) this represents a regional hiatus at the base of the Middle Eocene in the south Rockall Basin, and is interpreted to mark the onset of significant shelf-margin progradation. The succession in well 132/15-1 is coarser grained, and contains 1111 m of predominantly siltstones with minor claystone and sandstone. The uppermost 294 m is dated as Eocene–Oligocene. Above the level where logging commenced, drillers encountered a ‘gravel’ layer that impeded penetration, possibly related to the uppermost Eocene C30 unconformity. In the adjacent Erris Basin, well 12/13-1A (Figure 68) recovered glauconitic claystone overlain by sandy glauconitic limestone stringers dated as Early to Mid/Late Eocene.

On the western flank of the south Rockall Basin, BGS borehole BH94/01 cored 31.55 m of Middle to Upper Eocene organic-rich mudstones and pebbly sandstones (Figure 69). The palynofacies and abundance of plant debris is indicative of a nearshore marine setting, though the presence of a Rhabdammina biofacies (see BGS borehole BH94/03, below) may be indicative of restricted water circulation. Further south, on the south-eastern flank of the south Rockall Basin, Lower and Middle Eocene clastic, pelagic carbonate and hemipelagic rocks, deposited in an outer shelf/upper slope setting, have been recovered in boreholes 16/28-sb01 and 83/24-sb01, 83/24-sb02 collected by the Irish Rockall Studies Group (Haughton et al., 2005) (Figure 63). The sedimentary rocks recovered in borehole 16/28-sb01 are disturbed by failure along low-angle faults, compatible with regional seismic evidence for shelf-margin slumping and progradation in the Mid to Late Eocene (Corfield et al., 1999; Haughton et al., 2005).

Seismic sections across the south Rockall Basin suggest an Eocene succession up to 600 m in thickness extending across much of the basin (McDonnell and Shannon, 2001), though the section is generally thinner and more condensed in the central part of the basin (Corfield et al., 1999).

In the north-east part of the region, BGS boreholes and commercial wells have tested the Eocene prograding wedge, which builds out from the Sula Sgeir High (Figure 61) and the Wyville Thomson–Ymir ridge system into the North-east Rockall Basin. BGS boreholes BH88/10, BH90/02 and BH90/06 are located at the landward limit of the preserved Eocene strata (Figure 68). Borehole BH88/10 recovered 10.53 m of fine-grained, shallow-marine, Middle–Upper Eocene sandstone overlying Lower Eocene basalt (Figure 69). Borehole BH90/02 cored 31.9 m of interbedded sandstones, pebbly sandstones, siltstones and mudstones, tentatively assigned an Early Eocene age, unconformable on Middle Jurassic mudstones (Figure 69). The presence of fragmentary lignite beds and terrestrially sourced palynological debris, combined with an absence of in situ marine fossils, suggests a nonmarine origin. Borehole BH90/06 recovered 15 m of Middle Eocene shallow-marine sandstone and calcareous siltstones that appear to form a small outlier of Eocene rocks to the south-east of Paleocene basalt (Stoker et al., 1993). At the south-east end of the Wyville Thomson Ridge, BGS borehole BH85/02B recovered 4.95 m of fine-grained sandstone consisting largely of reworked basaltic debris. The sandstone was dated as earliest Eocene on the basis of the foraminifer Nummulites rockallensis Hinte and Wong, and records the age of the marine inundation of the ridge (Stoker et al., 1988). Immediately to the east of the report area, a 91.5 m-thick sequence of Lower Eocene deltaic sandstones and claystones has been reported from borehole BH90/03 (Hitchen et al., 1995b) (Figure 68).

Commercial wells 164/25-1,1Z, 164/25-2 and 154/03-1 are located at the more distal (basinal) end of the prograding wedge. Well 164/25-1,1Z proved 465 m of clastic rocks lying unconformably on Paleocene strata (Figure 69). The Lower to Middle Eocene section consists of sandstones and claystones, with claystones and siltstones dominating the upper Middle and Upper Eocene strata. Siltstones with thin limestones overlying sandstone interbedded with volcanic rocks comprise the 391 m-thick Lower to Upper Eocene succession in well 164/25-2. In well 154/03-1, 152 m of Lower Eocene siltstones and claystones overlie Upper Paleocene–Lower Eocene volcanic rocks. A deltaic to shallow-marine environment is envisaged in the earliest Eocene, succeeded by a regressive fan system in the later Eocene as the shelf margin prograded into the basin (Tate et al., 1999).

Farther west, wells 164/28-1, 154/01-1, 153/05-1 and 164/07-1 tested the infill of the North-east Rockall Basin, which consists predominantly of a marine clastic succession overlying igneous rocks that shows a general upward-fining pattern through the Eocene (Figure 69). Well 164/28-1 penetrated 705 m of Lower Eocene claystone, which includes a 70 m-thick basal unit of interbedded volcaniclastic breccia and sandstone, unconformably overlain by 148 m of Middle to Upper Eocene claystone. Well 154/01-1 cored 851 m of Lower to Upper Eocene claystone and subordinate limestone, with thin sandstones and tuffs in the Lower Eocene section. Well 153/05-1 penetrated 844 m of Eocene marine strata, which consists of a 466 m-thick claystone-dominated Lower Eocene section, but with a sandy conglomerate 94.5 m thick, overlain by about 378 m of Middle and Upper Eocene siltstones. Thin sandstones and limestones occur throughout the sequence, with tuffs near the base. In the northern part of the basin, well 164/07-1 proved 1165 m of lowermost Eocene tuffs, basalts, volcaniclastic conglomerates and breccias, and thin beds of claystones (Archer et al., 2005). The basalt lavas are dated at about 55 Ma. Overlying the volcanic succession is at least 453 m of Lower to Upper Eocene claystone and siltstone, deposited in a marine environment.

In the north Rockall Basin, well 163/06-1A penetrated at least 739 m of predominantly Eocene claystone and siltstone overlying an uppermost Paleocene volcanic sequence on the north-west flank of the Darwin Volcanic Centre. On the northern flank of the basin, just to the north of the report area, dredge samples (S1— recovered Lower Eocene tuffs from the Ymir Ridge (Jones and Ramsay, 1982) (Figure 68).

BGS boreholes and short rock cores have tested Eocene strata in the western part of the report area (Figure 68), (Figure 69) and (Figure 70). A prograding sediment wedge initiated in the Late Paleocene on the south-eastern margin of the George Bligh High, continued to develop during the Eocene (McInroy et al., 2006). BGS borehole BH94/07 recovered 10.37 m of Middle to Upper Eocene bioclastic limestone from the uppermost part of the sediment wedge; the limestone is capped by a cemented and stained horizon indicative of a hardground, and possibly of subaerial exposure and meteoric diagenesis. On the north-east flank of the George Bligh High, the Eocene fill of the north Rockall Basin has been partly exhumed by the action of post-Eocene deep-water currents (Howe et al., 2001), and the basinward-dipping strata have been sampled by BGS short rock cores along the length of the exposed stratigraphical section (Figure 71)a. Collectively, cores 59-14/5, 59-14/6, 59-14/7, 59-14/8, 59-14/9 and 59-14/10 proved a Lower to Middle Eocene succession of shallow-marine mudstones and claystones with thin sandstones and siltstones (Figure 70).

On the south-west margin of the north Rockall Basin, between the George Bligh High and the Rockall High, the Eocene succession displays enhanced thickening due to the post-depositional collapse of the shelf-margin system, which resulted in much internal deformation, manifest as faulting and slumping (Stoker et al., 2001) (Figure 71)b. The top of the deformed section is commonly at or near to sea bed, due to the erosive action of deep-water currents, and has been targeted by BGS short rock cores 58-14/34, 58-14/44, 58-14/45, 58-14/53 and 58-14/54 (Figure 60) and (Figure 69). These cores recovered Lower to Middle Eocene, red to orange–brown mudstone and claystone, with sporadic thin-bedded sandstone, though core 58-14/54, closer to George Bligh High, sampled sandstone with thin-bedded pebbly sandstone (Figure 70). A shallow-marine environment is envisaged. According to Ferragne et al. (1984), the colour of the argillaceous rocks reflects the high smectite content, as deduced from analysis of an Lower Eocene claystone recovered in their core, 7710, from the southern flank of George Bligh High (Figure 68). The source for this sediment is attributed to subaerial weathering of local igneous rock, most probably from the George Bligh High. This implies that there may have been exposure of the Rockall High and George Bligh High during the Early to Mid Eocene (Figure 64).

Eocene prograding sediment wedges are exposed (or close to sea bed) along the eastern and northern flank of the Rockall High (McInroy et al., 2006). BGS boreholes BH94/02, BH94/03 and BH94/06 tested the sedimentary succession at various locations on the High, and short rock cores 57-13/65, 57-13/77, 57-14/43 and 58-14/29, 30 sampled the strata exposed along the upper slope escarpment (Figure 68). Borehole BH94/03 was drilled into the East Rockall Wedge and recovered the most complete succession (209.65 m thick). Although a previous study of this wedge has stated that its base was cored by BH94/03 (McInroy et al., 2006), re-inspection of the seismic profile data suggests that only about two-thirds of the wedge was penetrated (Figure 72). By combining the seismic data with the lithological and biostratigraphical information from borehole BH94/03, four main depositional packages (1–4) have been defined with the borehole terminating at the top of package 4. Reflection configurations, including erosional truncation and stratal downlap, imply that the packages are bounded by unconformities, and this has been proved by the biostratigraphy of packages 1–3, which are separated by unconformities of intra-Early and intra-Mid Eocene age (Figure 72). The dating of these packages is based on foraminifera and dinoflagellate cysts, which reveal an NP10–17 range for the entire sampled wedge. In contrast, calcareous nannofossils consistently indicate Mid- to Late Paleocene species throughout the entire core, including the Pleistocene veneer; this is most likely a consequence of reworking, but may have significance for the age range of the wedge below the base of the borehole.

The basal 1.71 m in borehole BH94/03 (207.94–209.65 m below sea bed (bsb) comprises three basaltic pillow lavas interbedded with black to dark greenish grey, shelly, shallow-marine mudstone with sand-filled burrows. Correlation of the borehole with the seismic profile suggests that these pillow lavas represent only the top of a more extensive shingled sequence (package that appears to continue for up to another 100 m or so overlying undifferentiated, faulted, igneous and/or metamorphic basement (Figure 72). The pillow lavas are sharply overlain by, and partly intermixed with, a shallow-marine bioclastic sandstone at the base of package 3 (Figure 73)a. This overlying section (135.92–207.94 m bsb) consists predominantly of several stacked sequences of prograding sandstone with interbedded conglomerate and sporadic mudstone, near the top of the package (Figure 72). Bioclasts indicate a predominantly shoreface marine environment; however, fining and upward-coarsening successions, evidence of channelisation, and soft sediment deformation may be indicative of sporadic deposition from subaqueous mass-flow processes. The uppermost sandstone bed in package 3 is heavily bioturbated and its top is stained yellow-brown indicative of subaerial exposure. Dinoflagellate cysts, such as Deflandrea oebisfeldensis, Microdinium cf. ornatum, Adnatosphaeridium multispinosum, Adnatosphaeridium robustum, Achilleodinium biformoides, Achomosphaera alcicornu and Cordosphaeridium gracile, and miospores Inaperturopollenites hiatus and Tricolpites cf. hians, together with a foraminiferal fauna that includes Operculina sp, Guttulina communis, Cibicides simplex, Nutallides truempyi, Anomalinoides howelli, Nummulites sp. and Gavelinella danica suggest an Early Eocene (Ypresian: NP10–11) age for both package 4 and 3 in the borehole. However, it is clear from the seismic profile and the borehole record that the transition between the packages reflects a marked environmental change from a submarine volcanic sequence to a clastic-dominated unit that is envisaged to have accumulated in a nearshore–marine environment, incorporating both shoreface and subaqueous mass-flow deposits prior to subaerial exposure.

Package 2 of the wedge is 99.92 m thick (36.0–35.92 m bsb) and downlaps the eroded top of package 3 (Figure 72). It consists predominantly of sandstones and mudstones, tuffaceous in the lower part (88.85–135.92 m bsb), with one basalt horizon (47.26–51.0 m bsb) near the top of the package. The sandstones contain organic (plant) fragments and abundant bioclasts; where grain-size changes occur, bioturbation (Planolites and Chondrites) is commonly observed (Figure 73)b. In the middle of the succession, between 88.62 and 88.85 m bsb, the top of the volcaniclastic sandstone/tuff sequence appears rubbly with common subrounded igneous clasts set in a yellow-brown to yellowish-red muddy matrix. This may be indicative of subaerial exposure, and correlates with a discordant reflector on the seismic profile that essentially marks the end of this phase of volcaniclastic activity. The overlying sandstones are uniformly shelly and bioturbated and downlap onto the tuffaceous section (Figure 72). An increase in conglomeratic material between 60.0 and 50.0 m bsb precedes the thin basalt layer near the top of the package; the final expression of volcanic activity at this site. The basalt is aphyric and sparsely vesicular; unfortunately its contacts with the surrounding sandstones were not recovered. An overall nearshore–marine setting is envisaged for package 2, accompanied by extrusive volcanism. Dinoflagellate cysts, such as Deflandrea phosphoritica, and Deflandrea oebisfeldensis, Microdinium cf. ornatum, Cordosphaeridium gracile and Wetzeliella lobisca plus miospores Tricolpites cf. hians, Caryapollenites veripites, Sequoiapollenites polyformosus and Platycaryapollenites platycaryoides, together with foraminifers such as Planulites costata, Globigerina linaperta, Neoeponides karsteni and the radiolarian Cenosphaera sp. suggest a late Early to early Mid Eocene (NP12–14) age for package 2.

The upper package (1) of the wedge is about 24.45 m thick (11.55–36. m bsb), and consists of shoreface sandstone (Figure 72). Well-preserved bivalve shells, possibly in life position, are locally preserved in the sandstones (Figure 73)c. A late Mid Eocene age (NP16–17) is indicated by the presence of dinoflagellate cysts such as Glaphyrocosta intricate, Phthanoperidinium geminatumi, and Rottnestia burussica, the foraminifers Psammosiphonella gr. discreta, and Trochammina cf. subvesicularis, and the radiolarian, Cenosphaera sp. The foraminifera and radiolarian represent the Rhabdammina biofacies of Gradstein and Berggren (1981) from the North Sea. This assemblage is entirely facies controlled and commonly indicative of restricted water circulation. An equivalent faunal assemblage was recovered in BGS borehole BH94/01 in the south Rockall Basin (see above). The occurrence of the Early Eocene dinoflagellate cyst species Wetzeliella lunaris and Wetzeliella cf. meckelfeldensis is indicative of intra-Eocene reworking.

The association between sporadic volcanic activity and coarse clastic sedimentation implies a tectonically active setting throughout the Eocene. Clasts within the sediments are predominantly volcanic in origin; this suggests contemporary uplift and erosion of the Rockall High, and is consistent with evidence in borehole BH94/03 for subaerial exposure. Although the main phase of breakup volcanism (57–54 Ma: see Cretaceous and Palaeogene igneous rocks chapter) predates the formation of the East Rockall Wedge, the borehole record indicates that sporadic phases of Early to early Mid Eocene volcanic activity, including explosive volcanism and lava extrusion, may have persisted for up to 6–7 Ma after breakup.

To the south-west, BGS borehole BH94/02 cored packages 1 and 2 of the East Rockall High overlying basalt (Figure 68) and (Figure 69). The borehole proved 14.52 m of Middle to Upper Eocene sandstone and pebbly sandstone with a thin basal conglomerate overlying 1.61 m of uppermost Lower to lower Middle Eocene, tuffaceous, pebbly, muddy sandstone; this, in turn, overlies basalt. The Middle–Upper Eocene sandstone is glauconitic and contains intact bivalve shells, possibly in life position (Figure 73)c.

Packages 1 and 2 of the prograding wedge are also exposed along the escarpment that runs along the eastern upper slope of the Rockall High. Sandstones and mudstones from package 2 were cored in BGS short cores 57-14/43 and 57-13/65, respectively, and sandstones equivalent in age to package 1 were recovered in core 57-13/77 (Figure 68) and (Figure 70). On the western side of the Rockall High, DSDP site 117 proved Upper Paleocene to Lower Eocene (NP9–12) conglomerates, sandstones and mudstones on the flank of the Hatton Basin (Figure 63), deposited in a coastal to shallow-marine environment (Laughton et al., 1972; Berggren and Schnitker, 1983).

A smaller wedge, located on the northern flank of the Rockall High, was tested by BGS borehole BH94/06 (Figure 68) and (Figure 69), which proved 8.9 m of Middle Eocene sandstone overlying basalt, and capped by an eroded hardground related to exposure of the High. On the adjacent scarp slope, BGS short cores 58-14/29 and 58-14/30 proved a coarsening-upward sequence of mudstones and strongly bioturbated sandstones dated as Early to Mid Eocene (Figure 68) and (Figure 70), which probably represents the basinward advance of the deltaic shelf-margin.

On the Anton Dohrn Seamount, an Upper Paleocene to Lower Eocene shallow-water limestone was proved in BGS borehole BH90/15, BH90/15A, and BGS short core 57-12/18 recovered 1.0 m of Middle to Upper Eocene conglomerate (Figure 68), (Figure 69), (Figure 70). The conglomerate consists of basaltic and carbonate clasts within a biosparite matrix, and is interpreted as representing a conglomeratic hardground that accumulated on what was probably an isolated seamount. Whilst no Eocene strata have been recovered from the top of the Rosemary Bank Seamount, their presence is inferred on the basis of seismic interpretation on the eastern flank of this feature (BGS, 2007; Howe et al., 2006).

Oligocene

There has been relatively little recovery of Oligocene rocks in the report area (Figure 69). The bulk of the Oligocene strata are preserved in, and on the flanks of, the Rockall Basin, where they form part of the RPc megasequence (Figure 61) and (Figure 68). Occurrences on the adjacent highs are restricted to several, small, terrestrial basins of Late Oligocene age in the Minch and the Sea of the Hebrides, east of the report area (Evans et al., 1979, 1991; Fyfe et al., 1993) (Figure 68).

The present bathymetric expression of the Rockall Basin was initiated during Late Eocene/Early Oligocene times, a consequence of the strongly differential subsidence that deepened the existing framework of sub-basins, of the order of 1 km (Vanneste et al., 1995; Stoker, 1997; Praeg et al., 2005) (Figure 65). This rapid and dramatic change in basin configuration is manifest by the cessation of the sediment supply to the Eocene prograding wedges, which locally collapsed and deformed (Stoker et al., 2001; Haughton et al., 2005) (Figure 71)a. The subsidence allowed southerly derived currents to enter the basin, which instigated the onset of deep-water circulation and bottom-current-influenced sedimentation that led to the accumulation of the predominantly onlapping basin fill preserved as the RPc megasequence (Figure 61). No significant deep-water connection with basins to the north is envisaged at this time (Figure 65). Contourite deposits form the bulk of the basin fill, with restricted deposition of clastic and carbonate sediments on the outer shelf and slope of the adjacent highs. The erosion of the basin margin by the action of bottom currents may have been one of the sediment sources for the contourite deposits. Sediment input points to the basin from the highs are poorly known, though the region between the northern Hebrides Shelf and the Wyville Thomson Ridge may have been a provenance area throughout the Oligocene, whilst the north Rockall High was a source area particularly during the Late Oligocene (Stoker, 1997; Egerton, 1998).

Upper Oligocene terrestrial–paralic sediments in The Minch–Sea of the Hebrides region are indicative of a general lowering of sea level during the Late Oligocene and exposure of the platform areas. The onset of compressional tectonics towards the end of the Oligocene exerted a major influence on the growth of the Wyville Thomson and Ymir ridges and related inversion domes (see Structure chapter) (Johnson et al., 2005; Ritchie et al., 2008). The pattern of distribution of the Palaeogene strata on these ridges (Figure 68) reflects latest Palaeogene–early Neogene folding.

At the edge of the Hebrides Shelf, a marked bathymetric scarp, termed the Geikie Escarpment ((Figure 61), (Figure 68) and (Figure 74), has been eroded into an upper slope wedge of Early Oligocene (NP21–22) strata dated by a short core (44: see (Figure 68), (Figure 70) and (Figure 74) that retrieved a cream-coloured chalk from the exposed scarp face (Jones et al., 1986; Evans et al., 1989; Stoker, 2002). A series of irregular, aggrading, mounded seismic reflectors that comprise the front of the wedge have been tentatively interpreted as reefal carbonates (Stoker et al., 2001), whereas parallel, continuous, landward-converging reflectors characterise the back-reef area (Figure 74). Whilst the Lower Oligocene chalk contains bryozoan fragments (Jones et al., 1986), framework carbonates are not reported in the sample. It is possible that the lower Oligocene wedge consists of carbonate-rich build-ups, without a significant framework component, formed in a subtidal environment. Adjacent to the Geikie Escarpment, the wedge is up to 350 m thick, but landward, within 10–15 km, the wedge is only half that thickness (Figure 74). The asymmetry of the wedge suggests that there was contemporaneous subsidence of the outer margin, with the shelf-margin carbonates keeping up with the changing sea level, whereas shallow-marine stratified sediments infilled the back-reef area. Jones et al. (1986) attributed the formation of the steep erosional escarpment to a major drop in eustatic sea level in the mid-Oligocene, though Evans et al. (1989) considered the cutting mechanism to be related to contour-current activity. Whilst the latter process has undoubtedly been important in shaping the present-day escarpment and upper slope (Stoker et al., 1994), the aggradational development of the carbonate wedge implies that a significant depositional scarp most probably developed at the Early Oligocene shelf-edge, regardless of any subsequent erosional modification.

The Lower Oligocene wedge is unconformably overlain by a thinner wedge of sediment initially interpreted as Miocene strata (Evans et al., 1989), but subsequently proved, by BGS borehole BH88/07, BH88/07A, to consist of Upper Oligocene sediments, in turn, unconformably overlain by Middle to Upper Miocene strata (Stoker et al., 1994) (Figure 74). The Lower and Upper Oligocene sequences have been differentiated along the upper Hebrides Slope, above the Geikie Escarpment (Figure 68). The Upper Oligocene section is generally less than 50 m thick, and less extensively preserved. It is commonly eroded adjacent to the Geikie Escarpment — the action of Neogene alongslope currents — and pinches out landwards by low-angle onlap onto Lower Oligocene strata (Figure 74). The mid Oligocene unconformity that separates these packages is expressed as a high-amplitude reflector that marks an erosional hiatus.

BGS borehole BH88/07, BH88/07A cored into the eroded tongue of Upper Oligocene strata close to the Geikie Escarpment, and recovered 11.2 m of Upper Oligocene mudstone containing both calcareous and siliceous bioclasts, with thin sands and flints (Figure 69) and (Figure 74). The flints prevented further penetration into the Oligocene succession. The Upper Oligocene sediments contain a varied fauna including molluscs but there is no indication of framework building organisms. A well-constrained Late Oligocene (NP24) age is indicated on the basis of the calcareous nannofossils Reticulofenestra bisectas and Zygrhablithus bijugatus, and the planktonic foraminifera Globorotalia opima (Stoker et al., 1994). The sediments were probably deposited in a low energy, outer shelf–upper slope setting.

In the south Rockall Basin, Oligocene strata have been tested in well 132/06-1 (Figure 69), which recovered 119 m of Oligocene marine claystone with thin sandstone beds unconformable on Eocene strata. This relatively thin sequence is a common feature of the basinal section over much of the central and southern part of the Rockall Basin south of the Rosemary Bank Seamount, where the combined Oligocene to Lower Miocene RPc megasequence is generally less than 300 m thick in the central part of the basin, and is interpreted to represent deep-marine contourite deposits (Corfield et al., 1999; McDonnell and Shannon, 2001; Stoker et al., 2001).

In the adjacent Hatton Basin, comparable bottom-current-influenced deep-water sediments were cored at DSDP site 117 (Figure 63) on the flank of the basin, which recovered 145 m of Oligocene (NP23–25) chalk. In the central part of this basin, DSDP site 116 proved a more punctuated sequence of Upper Eocene to Lower Oligocene (NP19–22) chalk unconformably overlain by Upper Oligocene to Neogene (NP25–NN21) oozes. The contrast in siliciclastic content between the basins may be indicative of relative proximity to major terrigenous source areas.

In contrast to the rest of the Rockall Basin, the north east corner of the basin is interpreted to have accumulated up to 600 m of Oligocene sediments (Tate et al., 1999). Well 164/25-2 proved 360 m of Lower Oligocene sandstone with thin beds of calcareous mudstone unconformably overlain by 56 m of Upper Oligocene calcareous mudstone. The Lower Oligocene strata are associated with a discrete depocentre adjacent to the Hebrides Shelf that may include lowstand fan deposits (Egerton, 1998; Tate et al., 1999); the Upper Oligocene deposits form part of a westward thickening basinal package that was mostly influenced by bottom currents (Figure 61). In the centre of the basin, the Oligocene strata may be locally absent above inverted Eocene strata (Figure 61). In other wells in this area, the distinction of Oligocene strata is less well established (Figure 69). Wells 153/05-1 and 154/01-1 recorded approximately 129 m and 111 m, respectively, of undifferentiated ‘Upper Eocene to Oligocene’ siltstone with thin limestone, claystone and sandstone beds on the composite well logs. Lowest Oligocene siltstones may have also been penetrated in well 163/06-1A, though the Eocene–Oligocene boundary is similarly poorly recognised in this well. The top of the Oligocene is not defined in any of these wells. In well 164/28-1, 142 m of undifferentiated ‘Oligo-Miocene’ sand and mudstone were recorded.

A discrete lowstand fan deposit of Late Oligocene age has been identified on the north-east slope of the Rockall High (BGS and PAD, 2002) (Figure 68). The fan deposit was cored by BGS borehole BH94/04, which recovered 33 m of coarse-grained bioclastic limestone comprised almost entirely of molluscan, echinoderm and partially micritised bryozoan debris. The limestone contains crude lamination and fining-upward sequences, is highly porous as it is consolidated only by sparse calcite cement (Figure 73)d, and commonly contains no fine-grained material. The limestone was initially dated as Late Eocene to Early Oligocene in age (Stoker, 1997; Stoker et al., 2001), although the microfossil assemblage was generally poorly constrained. A subsequent Sr isotope analysis provided a Late Oligocene (26.6 ± 0.3 Ma) age (BGS and PAD, 2002), which suggests that the Eocene–Oligocene microfossil assemblage is most probably reworked, in common with the macrofauna. The fan deposit is interpreted to have resulted from the reworking of carbonate material on the Rockall High, in a high-energy environment, and its subsequent downslope transportation by mass-flow processes into the basin.

Neogene to Quaternary

Neogene to Quaternary sediments are best preserved in the Rockall Basin and along its eastern flank, the Hebridean margin. In contrast, they are absent or represented only by a veneer on the highs, domes and ridges flanking the western and northern margins of the basin, as well as on the tops of the Rosemary Bank, Anton Dohrn and Hebrides Terrace seamounts (Figure 60), (Figure 61), (Figure 75) and (Figure 76). The total thickness of the Neogene to Quaternary succession above C20 (Mid Miocene and younger strata) commonly exceeds 600 milliseconds (ms) TWTT in the Rockall Basin. In general terms, sound velocities in the Neogene to Quaternary strata are in the range of 1.5–2.0 km/sec, giving a maximum estimated thickness of 600 m to the Neogene to Quaternary succession, based on a sound velocity of 2.0 km/sec (STRATAGEM Partners, 2002, 2003; Stoker et al., 2005a). The thickness of the lowest part of the Neogene (upper part of the RPc megasequence) remains unknown, though its distribution is limited by the C20 basinal onlap surface, beyond which Lower Miocene strata are not preserved (Figure 75).

The bulk of the Neogene to Quaternary succession is assigned to the two regional megasequences of RPb (Middle Miocene to Lower Pliocene) and RPa (Lower Pliocene to Holocene), whose distribution and thickness are shown in (Figure 75) and (Figure 76). The available lithological and age dates for these megasequences are summarised in (Figure 77) and (Figure 78). The age data are based on biostratigraphical information. On the Hebridean margin, the RPa megasequence comprises a number of higher-resolution, seismic-stratigraphical sequences previously defined (informally) during the course of the BGS 1:250 000 offshore mapping programme (cf. Stoker et al., 1993); however, these units have recently been ratified as a formal lithostratigraphical scheme, incorporating groups and formations (Stoker et al., 2011) (Figure 79).

The Neogene–Quaternary tectonic development of the region has involved both compressive and epeirogenic movements in the early Neogene (Miocene) and late Neogene–Quaternary (Pliocene–Pleistocene), respectively, which have influenced patterns of palaeoceanographical circulation and sedimentation (Praeg et al., 2005; Stoker et al., 2005a, b, c). The most obvious expression of early Neogene compression is the development of elongate anticlinal domes in the north of the region, including the Lousy High and Bill Bailey’s domes, the buried Alpin Dome, and the Wyville Thomson and Ymir ridges (Figure 60), (Figure 61) and (Figure 66). The anticlinal structures are up to 2 km in amplitude and tens of kilometres across (Boldreel and Andersen, 1998; Johnson et al., 2005). Although these structures may have been developing since the Mid Eocene (Figure 65), an intensification of contractional deformation is interpreted to have occurred during the Early–Mid Miocene (Johnson et al., 2005; Stoker et al., 2005c). Regional flexure across the continental margin may have preceded this phase of deformation, reflecting the build-up of stress (Stoker et al., 2005b). Although the amplitude of flexural deflection may have been relatively small (tens to a few hundreds of metres), it may have been sufficient to change the bathymetric threshold in the Rockall Basin (Cloetingh et al., 1990). A modification to the pattern of deep-water circulation is expressed by erosion associated with the formation of the Base Neogene Unconformity in the north of the basin (Figure 61).

This discrete phase of early Neogene compressional tectonism was coeval with a local reorganisation of the adjacent North-east Atlantic plate system, which accommodated the transfer of the spreading ridge across the Jan Mayen Microcontinent between about 23 and 15 Ma (Brekke, 2000; Mosar et al., 2002; Stoker et al., 2005c; Holford et al., 2009). Regional flexure is envisaged to be a response to the prolonged build-up of compressive stresses transferred from the plate boundary into the plate interior during this interval, which was followed by the rapid release of stress in specific areas in the form of the anticlinal structures. The formation of deformation structures in the early Mid Miocene had a very significant change in deep-water circulation patterns off north-west Britain. Differential vertical movements on the northern boundary of the report area, notably the development of the Wyville Thomson and Munkagrunnur ridges and the complementary syncline of the Faroe Bank Channel, created the Faroe Conduit (Stoker et al., 2005c) (Figure 66). The formation of this deep-water passageway, concomitant with a general deepening of the Greenland–Scotland Ridge (Thiede and Eldholm, 1983), facilitated the onset or acceleration of the exchange of intermediate and deep waters across the Greenland–Scotland Ridge from about 15 Ma. This resulted in an expansion of contourite drift accumulation in the Rockall and Faroe–Shetland basins, which experienced significant onlap of deep-water sediments along their margins, e.g. (Figure 80), as well as the North-east Atlantic region in general (Laberg et al., 2005). The preservation of a Middle to Upper Miocene transgressive systems tract on the Hebridean margin is further indicative of a general subsidence of the region in Mid Miocene to earliest Pliocene times (Stoker et al., 1994) (Figure 66) and (Figure 74).

Late Neogene to Quaternary epeirogenic movements instigated a regional change, from about 4 Ma (Early Pliocene), in the patterns of contourite sedimentation (submarine erosion, new depocentres) in the Rockall Basin, and the onset of rapid progradation of the Hebridean margin by up to 50 km (Stoker, 2002; Stoker et al., 2005a, b). The sedimentary response to change is well illustrated by comparing the distribution of the RPb and RPa megasequences in (Figure 75) and (Figure 76), which indicate a late Neogene to Quaternary shift in the focus of sedimentation from the Rockall Basin (RPb) to the Hebridean margin (RPa). The growth of the slope apron is especially well developed in two depocentres — the Sula Sgeir and the Barra–Donegal fans — where Lower Pliocene to Holocene sediment accumulation locally exceeds 600 ms TWTT (Figure 76) and (Figure 81). The build-out of the shelf and slope is inferred to record a marked increase in sediment supply to the Hebridean margin in response to kilometre-scale uplift and tilting of the margin (Praeg et al., 2005; Stoker et al., 2005b, 2010; Holford et al., 2010). Although glacially derived sediments form a significant component of the prograding sediment wedges, the initiation of the wedges pre-dates the onset of glaciation in this region (influx of ice-rafted detritus at about 2.5 Ma) by up to 1 Ma (Stoker, 2002). The Sula Sgeir and Barra–Donegal fans form part of a larger domain of shelf-margin progradation extending the length of the north-west European Atlantic margin, indicative of the regional extent of this tectonic pulse (Praeg et al., 2005) (Figure 62) and (Figure 67).

Uplift of onshore and shallow-shelf areas bounding north-west Britain may have been accompanied by accelerated subsidence of several hundred metres in the Rockall Basin (e.g. Cloetingh et al., 1990) (Figure 67), defining a large-scale tilting of the Hebridean margin (Stoker et al., 2005b). This distortion of the margin was likely to be a factor in the modification of the deep-water circulation pattern in the Rockall Basin, manifest by the widespread development of the C10 unconformity (Figure 61) and (Figure 80) (Figure 81), (Figure 82). Whilst, on a global scale, Early Pliocene submarine erosion may have been driven by an increase in the volume and strength of Norwegian Sea Deep Water passing across the Greenland–Scotland Ridge (via the Faroe Conduit and Wyville Thomson Ridge) — following the closure of the Central American Seaway (Haug and Tiedemann 1998; Lear et al., 2003) — the relatively local intensity of current flow in the Rockall Basin is likely to have been enhanced by the change in its palaeobathymetry induced by tilting (Stoker et al., 2005a, b). Although subsequent patterns of deep-water sedimentation continued to be dominated by bottom-current activity, there was a shift in depocentres. For instance, the Rockall Basin has undergone prolonged erosion along its north-west flank (Figure 71) and (Figure 80), with contourite drift accumulations having migrated to the north-east onto the Wyville Thomson Ridge and Hebridean margin (Stoker et al., 2001) (Figure 83).

In the following sections, the Neogene to Quaternary is described broadly in terms of its Miocene and Pliocene to Holocene successions, though the boundary between the RPb and RPa megasequences is of Early Pliocene age. This subdivision most aptly corresponds to the two-stage tectonic evolution of the region summarised above.

Miocene

Miocene sediments are mainly preserved in the Rockall Basin, with the thickest accumulations (locally exceeding 600 ms TWTT) located in the western half of the basin (Figure 75). In shallower settings, such as the Hebrides Shelf and upper slope, and the Rosemary Bank and Anton Dohrn seamounts, a thinner cover of sediment (locally eroded) is more commonly developed. Although mostly buried beneath the Lower Pliocene to Holocene (RPa) megasequence, Miocene to Lower Pliocene (RPb megasequence) deposits are exposed along the western flank of the Rockall Basin (Figure 80). The deep-marine sedimentary system that was established in the Rockall Basin during the

Oligocene continued into the Early Miocene, with no obvious change in style of sedimentation. Consequently, there is a lack of a distinctive boundary separating Oligocene and Lower Miocene strata over much of the basin, and the inclusion of the latter within the RPc megasequence. The exception to this is in the north-east part of the basin where the latest Oligocene to Mid Miocene, Base Neogene Unconformity, may reflect contemporary deformation (flexure) and erosion along the northern flank of the Rockall Basin that culminated with the inversion of the Wyville Thomson and Ymir ridges and related domes. The most dramatic change in Miocene sedimentation patterns occurred immediately following stress release (through inversion), manifest by the C20 marker and overlying sediments of the RPb megasequence, which reveal a massive expansion of contourite sediment drifts across the basin, with significant onlap and upslope migration of drifts onto the basin margins (Stoker et al., 2001).

In general, the basinal sediments, including the Lower Miocene component of the RPc megasequence, display a ponded basin-floor-fill geometry but with locally significant morphological variation acquired especially during the deposition of megasequence RPb (Figure 61), (Figure 71) and (Figure 80). This variable geometry reflects three main styles of contourite sediment-drift accumulation within the Rockall Basin (Howe et al., 1994; Stoker et al., 1998, 2001): (1) broad, flat-lying to gently domed, sheeted drifts which occupy a large part of the axial region of the basin floor (Figure 61) and (Figure 81); (2) elongate mounded drifts onlapping and migrating up the margins of the basin (Figure 80); and, (3) the giant, elongate, Feni Ridge in the southern Rockall Basin (Figure 63) and (Figure 82), south of the report area. It is important to note, however, that these distinctive morphologies are simply type members within a continuous spectrum that forms the Rockall Basin drift complex. Adjacent to the Hebridean margin, the contourite deposits are mostly buried beneath younger sediments (megasequence RPa), but along the western flank of the basin they are commonly exposed at the sea bed and have suffered extensive erosion since the Early Pliocene (Figure 80). The basinal sediments also underwent extensive erosion during the formation of the Early Pliocene, upper bounding surface, reflector C10, which is commonly an erosional, angular unconformity (Figure 80), (Figure 81), (Figure 82).

Seismic profiles have revealed a predominantly parallel-bedded internal reflection configuration for the basinal sheeted drifts over most of the area, though the lower Neogene component of megasequence RPc varies from stratified to reflection free ((Figure 71), (Figure 81), (Figure 82), (Figure 83). The stratified reflections are mostly continuous and vary in expression from flat lying or gently domed to undulatory and waveform, which characteristically onlap the margin of the Rockall Basin (Figure 71). Large-scale sediment waves, up to 40 m high with a wavelength of 2–3 km, are locally preserved adjacent to the Anton Dohrn Seamount (Stoker, 1998). The continuity of individual reflections is locally offset by small-scale normal faults (Figure 81)b.

Miocene elongate mounded drifts are best developed on the western flank of the Rockall Basin, where the basinal drift deposits of the RPb megasequence have migrated a significant distance from the basin floor onto the adjacent slope, terminating at the foot of the Rockall Escarpment (Figure 75) and (Figure 80). These drifts are commonly separated from the adjacent slope by an erosional moat several tens of metres deeper than the crest of the mounded drift. Internally, these drifts display both onlapping and downlapping reflector terminations onto the underlying strata, but with a net upslope migration of the contourite deposit. The relict nature of the contourite drift illustrated in (Figure 80) is evident from the erosional, irregular, upper surface with internal reflectors clearly truncated at, or near to, sea bed and mantled only by a veneer of younger deposits correlated to the RPa megasequence. On the middle part of the slope, erosion has been so intense that parts of the section have been removed, exposing older, Eocene, strata. This has resulted in parts of the upslope-accreting contourite deposit now being locally detached from its basinal counterpart.

The Feni Ridge is a giant elongate drift, up to a maximum of 125 km wide, and separated from the Rockall High by a wide erosional moat, several hundred metres deeper than the crest of the drift (de Haas et al., 2003) (Figure 82). It can be traced for about 400 km parallel to the western margin of the southern Rockall Basin, but loses its classic morphological expression northwards towards the report area, between 55° and 56° north (Stoker et al., 2001), becoming incorporated into the sheeted and elongate-mounded drift succession of the northern Rockall Basin. A similar, albeit smaller, drift occurs on the eastern flank of the basin adjacent to the Porcupine High, where evidence of internal erosion surfaces is preserved (Figure 82). The size and morphology of the Feni Ridge is more characteristic of open-ocean drifts, such as the Hatton, Gardar and Bjorn drifts in the Iceland Basin (Figure 63). It may be no coincidence that the Feni Ridge has developed in water depths deeper than 2000 m, which is mostly deeper than the floor of the basin in the report area.

All of these drift forms have developed since the Mid Miocene establishment of a linked deep bottom water current system across the Greenland–Scotland Ridge (Figure 66), which resulted in the interaction of northerly (e.g. Norwegian Sea Deep Water) and southerly (e.g. North Atlantic Deep Water) derived water masses. The broad sheeted drift forms are generally associated with large areas swept by currents of lower velocity than those which formed the elongate mounded drifts, in some cases where major bottom current gyres develop in a semi-enclosed basin, such as the Rockall Basin (Stoker et al., 1998). The mounded drift forms, including the Feni Ridge, have accumulated where there is a lateral gradient of decreasing velocity away from the core of a relatively strong, high-velocity bottom-current system, which typically develops on basin margins where there is a change in bathymetric profile (Stow et al., 2002). Whilst there is some evidence locally for internal erosion surfaces within the Middle Miocene to Lower Pliocene succession (Figure 82), which may reflect episodic changes in bottom-current strength, the steady development, growth and migration of the drift complex as a whole attests to the stability of the current system during the Mid Miocene–Early Pliocene interval.

The nature of the deep-water sediments has been tested at several sites within and beyond the report area (Figure 77). In the southern Rockall Basin, DSDP site 610 sampled the Feni Ridge, including the upper part of megasequence RPc, and proved Lower–Middle Miocene to Lower Pliocene (NN3–12) nannofossil chalk and ooze (Hill, 1987). Lower Pliocene (NN12/13) nannofossil ooze was recovered from a basinal, sheeted drift at ODP site 981, onlapping onto the western flank of the basin (Jansen et al., 1996) (Figure 63). On the eastern flank of the southern Rockall Basin, borehole 83/20-sb01 cored Middle to Upper Miocene muds onlapping onto the slope of Porcupine Bank (Haughton et al., 2005) (Figure 63). Farther north, BGS borehole BH94/01 proved Middle Miocene/Lower Pliocene (NN10/12–15) bioclastic sand and gravel from the moat of an elongate mounded drift on the slope of Rockall High (Figure 80) (Stoker et al., 2001). Whilst the dinoflagellate cyst Unipontidinium aquaeductum and the bolboformid Bolboforma badensis are particularly characteristic of the Mid Miocene (NN6–7), the presence of medium-sized nannofossil placoliths, such as Dictyococcites antarcticus, favours a Late Miocene to earliest Pliocene age. The lack of a more precise age designation in this borehole may be as expected from a moat, which is an area of bottom-current reworking. About 40 km to the north-east, a 1.3 m-thick section of Lower Pliocene (NN13–15) bioclastic limestone was sampled in short core 57-14/48 (Figure 78) from the same elongate mounded drift. A basinal chalky facies consisting of white, massive, bioturbated micrite was recovered in core 57-11/33, near the foot of Anton Dohrn Seamount. The chalk is dated as Early Pliocene (NN15) though its affiliation to the RPb or RPa megasequences is unclear on seismic data.

On the eastern flank of the Rockall Basin, Middle Miocene siltstone was recovered from well 132/06-1 (Figure 75). In the North-east Rockall Basin, a sandy and muddy sheeted-drift succession was reported from well 164/25-2 (Figure 81)a (Egerton 1998; STRATAGEM Partners, 2002). West of the report area, DSDP site 116 and ODP site 982 proved a Lower Miocene to Lower Pliocene (NN2/3–NN15) succession of nannofossil chalk and ooze from the adjacent Hatton Basin (Laughton et al., 1972; Jansen et al., 1996; Flower, 1999) (Figure 63). The predominance of biogenic sediment implies a high biological productivity in both the Rockall and Hatton basins during the Miocene to Early Pliocene interval.

At shallower depths, Miocene strata correlated with the RPb megasequence have been proved on the upper slope of the Rockall High and George Bligh High, and from the axial seamounts of Rosemary Bank and Anton Dohrn. On the north-east slope of the Rockall High, pale yellow, bioturbated, bioclastic limestone of Early Pliocene (NN12–15) age was sampled in short rock core 58-14/32 (Figure 78), unconformably overlying Lower Palaeogene conglomerate. The context of this carbonate occurrence is unclear from seismic data. In contrast, an isolated knoll on the eastern slope of George Bligh High, cored at site 58-14/56, proved 1.84 m of lithified bioclastic limestone, but no conclusive age has been obtained from this site. BGS borehole BH90/15 proved a veneer of Upper Miocene (NN8–9) bioclastic sands and muds from the top of Anton Dohrn Seamount, whereas BH90/18 recovered at least 5 m of Middle to Upper Miocene (NN7–9) bioclastic sand and mud, with subordinate nannofossil ooze and limestone from the top of Rosemary Bank (Stoker et al., 1993) (Figure 77). The Mid to Late Miocene age for these sediments is based on a highly varied nannoflora that includes Calcidiscus leptopora, Coccolithus miopelagicus, Dictycoccites minutula, Reticulofenestra haqii and Sphenolithus grandis. The distribution and maximum thickness of Miocene strata on both seamounts remain uncertain. Seismic profiles reveal probable Neogene sediment drifts, tens of milliseconds (TWTT) thick, on the flank on the Rosemary Bank (Howe et al., 2006), but the distinction between Miocene and Pliocene–Holocene strata remains undefined. The occurrence of Miocene strata on the Hebrides Terrace Seamount and adjacent areas on (Figure 75) remains unproven.

On the upper Hebrides Slope, a 2.85 m section of Middle to Upper Miocene (NN4–10) muddy, greenish-grey, bioturbated, glauconitic sandstone was recovered in BGS borehole BH88/07, BH88/07A, unconformably overlying upper Oligocene strata of the RPc megasequence (Stoker et al., 1994) (Figure 74). The Mid to Late Miocene age is well constrained by the association of the nannofossil Discoaster exilis, the planktonic foraminifera Orbulina universa, and the algal cyst Bolboforma metzmacheri (Stoker et al., 1994). Low-angle landward onlap implies that the Miocene strata represent a transgressive deposit; sporadic basinward onlap supports a bedload origin for the sediments, blanketing a slightly irregular, contemporary shelf. The textural and mineralogical maturity of the sand suggests a substantial period of exposure to moderate- to high-energy conditions before and during sediment transport (Leslie, 1992). The presence of dinoflagellate cysts, such as Tuberculodinium vancampoae, Homotryblium floripes, Polysphaeridium zoharyii, Operculodinium israelianum, Impagidinium strialatum and Nematosphaeropsis labyrinthus are indicative of warm climatic conditions with an oceanic influence (Stoker et al., 1994).

These sediments form part of an outer shelf–upper slope deposit, locally up to 50 m thick, which may have been more extensively developed at the time of deposition. BGS boreholes BH90/12, BH90/12A and BH90/13 penetrated respectively 57.36 m and 2.9 m of Upper Miocene (NN9–11) greensands preserved as outliers 15–20 km to the east of the main accumulation tested by borehole BH88/07, BH88/07A (Figure 75). Farther north, BGS borehole BH88/10 (Figure 75) penetrated a 41.4 m-thick section of reworked Middle Miocene (NN5–6) greensands preserved in a Quaternary channel about 50 km to the north-east of shelf-margin deposits. Comparison with the areas to the south and north of the report area would further support the idea of a previously more extensive cover of Miocene shelf deposits as rocks of similar facies and age have been sampled in wells 13/03-1 and 19/05-1 on the Malin Shelf (Figure 63), and from numerous BGS boreholes and commercial wells in the West Shetland region, including borehole BH90/03 (Stoker, 1999; Stoker and Varming, 2011) (Figure 77). The latter proved 29.78 m of green to dark green, muddy, glauconitic and bioclastic sandstone of Mid Miocene (NN4–6) age, which characterises the Muckle Ossa Sandstone that was deposited during a marine transgression of the West Shetland Shelf. Middle to Upper Miocene transgressive sandstones are also reported from the northern North Sea (Galloway, 2001; Faleide et al., 2002).

Pliocene to Holocene

The sediments deposited during the late Early Pliocene to Holocene interval comprise the RPa megasequence, which is widely distributed throughout the report area (Figure 76). It is best developed along the outer part of the Hebrides Shelf and eastern margin of the Rockall Basin, where the slope apron exceeds 600 ms TWTT in thickness on the Sula Sgeir and Barra–Donegal fans. It is commonly less than 100 ms TWTT thick in the western half of the Rockall Basin, being thin to absent along its western flank, as well as on rock platforms on the adjacent shelves and banks, and on the tops and flanks of the Rosemary Bank, Anton Dohrn and Hebrides Terrace seamounts, although these may be covered by a veneer of sediment. This pattern of sedimentation imparts a distinct east–west asymmetry to the geometry of the RPa megasequence, which reflects the build-out of the Hebridean margin in contrast to the relatively starved nature of the western part of the basin (Figure 61), though a small prograding sediment wedge is identified on the eastern flank of the Rockall High (Figure 76). In the southern Rockall Basin, the deep-water succession is more commonly between 200 and 400 ms TWTT in thickness.

The change in sedimentation style in the Early Pliocene represented a regional response to the differential uplift and subsidence of the Rockall Basin and its margins, instigated by epeirogenic movements (about 4 Ma) which changed the sediment supply to the Hebridean margin and, in part, modified the deep-water bottom-current system through submarine erosion and drift migration (Stoker et al., 2005a, b) (Figure 67). On seismic profiles, the C10 reflector marks the stratigraphical expression of this tectonic phase (Figure 80), (Figure 81), (Figure 82), (Figure 83). A further change is marked by the development of the Glacial Unconformity, which marks the onset of widespread glaciation of Britain and Ireland, and its offshore region, and is dated as early Mid Pleistocene (about 0.44 Ma) (Stoker et al., 1994). Glacial sediments contribute markedly to the upper part of the RPa megasequence above the Glacial Unconformity.

On the Hebridean margin, the abundance of stratigraphical units comprising the succession above the Glacial Unconformity (Figure 79) reflects a complexity that is largely a consequence of the diverse sediment–landform association inherent within the glacial depositional system. At the present time, this scheme has not been extended to the Rockall Basin succession. Although biostratigraphical markers, such as the last appearance datum (LAD) of the nannofossil Pseudoemiliania lacunosa, which coincides with the Glacial Unconformity, has been recognised in boreholes both on the Hebrides Slope and in the Rockall Basin, there is no high-resolution seismic-stratigraphical subdivision of the basinal succession beyond the limit of the Sula Sgeir and Barra–Donegal fans. On the Hebridean margin, the deposits below and above the Glacial Unconformity have recently been assigned to the Hebrides Margin Group and the Eilean Siar Glacigenic Group respectively. Reference to specific formations that comprise these groups will be made where appropriate for adding detail to the interpretation of megasequence RPa; however, for a fuller definition of the various units, the reader is referred to Stoker et al. (1993, 2010).

For ease of description, the Pliocene to Holocene succession is presented separately from the Hebridean margin and the Rockall Basin (the latter including the axial seamounts, and the shallow banks and highs flanking its western and northern margins) on account of the higher-resolution work that has been undertaken on this part of the Cenozoic record. As the emphasis throughout this section has been to summarise the Cenozoic succession to series level, ultra-high-resolution studies, such as those focused on latest Pleistocene to Holocene millennial-scale events associated with the deglaciation of the last British–Irish Ice Sheet, are beyond the scope of this report. For more information on the latter, with respect to the report area, the reader is directed towards the work of Peacock and Harkness (1990), Harland and Howe (1995), Austin and Kroon (1996), Peacock (1996), Howe et al. (1998), Kuijpers et al. (1998), Kroon et al. (2000), Knutz et al. (2001), Richter et al. (2001), Rasmussen et al. (2002), STRATAGEM Partners (2002), Wilson et al. (2002), Morri (2004), Knutz et al. (2007), Peck et al. (2007), Scourse et al. (2009), Hibbert et al. (2010) and references therein.

Hebridean Margin

The shelf succession preserved above the Glacial Unconformity displays an aggradational character that internally consists of stacked, sheet-like sequences, interpreted as composite glacial units containing ice-contact, ice-proximal and glacimarine deposits (Stoker et al., 1993). It is the glacial geomorphic shaping of these sequences, resulting in sub- to proglacial bedforms such as moraines and ice-contact fans (Stoker and Holmes, 1991), which has contributed to the higher-resolution stratigraphical complexity of the Eilean Siar Glacigenic Group (Figure 79).

The nature of the Hebridean margin succession has been tested in BGS boreholes, and a few commercial wells. In general terms, these data indicate that the Glacial Unconformity separates underlying sand-dominated strata of the Hebrides Margin Group from overlying mud-dominated sediments of the Eilean Siar Glacigenic Group (Figure 77). On the Hebrides Shelf, the Lower MacLeod Formation comprises the Hebrides Margin Group (Figure 79), and has been sampled by BGS boreholes BH85/05B, BH90/11 and BH90/13 (Figure 76). Collectively, these boreholes proved up to 43.2 m of shelly, bioturbated greenish grey sands and silts partly interbedded with dark grey, shelly, slightly pebbly, sands and clays. Dinoflagellate cysts, such as Amiculosphaera umbracula, Impagidinium multiplexum, Operculodinium israelianum, and Tectatodinium pellitum, together with the benthic foraminifera Cibicides lobatulus grossa suggest a Late Pliocene to Early Pleistocene age for these deposits.

On the upper Hebrides Slope, the equivalent section in BGS borehole BH88/07, BH88/07A proved 21.43 m of predominantly well-sorted sands, with interbedded bioturbated sandy muds and homogenous muds, overlying basal gravel (Figure 77). The latter is 0.25 m thick, forms a glauconite-rich lag deposit overlying the C10 unconformity, and contains a large number of reworked ‘clasts’ of lithified Miocene sands presumably derived from the underlying strata (Figure 74). Foraminifera date the gravel to the Early Pliocene Globorotalia puncticulata Zone, which, together with the occurrence of Uvigerina venusta saxonica, suggests a date of between 4.2 and 3.8 Ma (Stoker et al., 1994). Biostratigraphical and magnetostratigraphical data from the overlying sequence of interbedded sands and muds indicate that the bulk of the section is younger than 3 Ma. The top of the section, as marked by the Glacial Unconformity, represents the LAD of the nannofossil Pseudoemiliania lacunosa, which is dated throughout the north-east Atlantic at 0.44 Ma (Thierstein et al., 1977). Thus a Late Pliocene to early Mid-Pleistocene age has been assigned to the interval between 67.82 and 89.0 m in the borehole (Figure 74). This implies that there may be a hiatus between the Lower Pliocene lag gravel and the Upper Pliocene–Middle Pleistocene section. The age of the lag gravel correlates well with the vigorous pulse of bottom-current erosion associated with the C10 unconformity in the Rockall Basin, and confirms that the irregular erosion surface cutting into the Oligo-Miocene strata on the upper Hebrides Slope is an equivalent surface. Palaeontological analyses suggest that warm environmental conditions prevailed during deposition of most of the Pliocene section, though the appearance of ice-rafted debris in the form of scattered dropstones occurs at 86.0 m in the borehole, which coincides approximately with the Gauss–Matuyama polarity transition (2.48 Ma) measured at 85.8 m, implies that the climate was beginning to deteriorate from latest Pliocene to earliest Pleistocene time (Stoker et al., 1994). Direct coring of the major depocentres of the Sula Sgeir and Barra–Donegal fans has for the most part been restricted to the youngest (Eilean Siar Glacigenic Group) strata (see below). Although wells 164/25-1Z and 164/25-2 penetrated the entire Sula Sgeir Fan, the predominantly muddy succession indicated on the composite logs for the older part of the succession (Figure 77) is based on little direct observation of core. The Eilean Siar Glacigenic Group has been tested in many BGS boreholes along the length of the Hebrides Shelf, and is characterised by a variable assemblage of diamicton, gravel, sand and mud (Figure 76) and (Figure 77). Many of the stratigraphical units preserved on the outer shelf, such as the MacDonald, Wyville Thomson Ridge, Morrison, Banks, Conchar, Caoilte and Conan formations (Figure 79), are dominated by diamicton, a mixed lithology that consists of a poorly sorted admixture of mud, sand and gravel. The diamicton is commonly dark grey, firm to hard and very poorly sorted, varying from crudely stratified to massive and clast- to matrix-supported, to massive and matrix-supported (Stoker, 1988; Stoker et al., 1993). Clast composition is highly variable and includes Precambrian quartzite, psammite and gneiss, Torridonian and/or Devonian red-purple arkosic grit and sandstone, Dalradian and/or Devonian grey limestone, Permo-Triassic red-brown sandstones and mudstones, Cretaceous chalk and Palaeogene basalt, tuffaceous sandstone and lignite. All of these lithologies are typical of north-west Scotland and the adjacent shelf. Recovered clasts are up to 11 cm maximum dimension, vary from angular and faceted to rounded and smooth, and whilst predominantly fresh in appearance, encrustations of worm tubes and bryozoan nets are not uncommon. Shell debris is also common, though in many cases considerably reworked.

In borehole BH85/07, on the northern Hebrides Shelf (Figure 76), a diamicton recovered from the Morrison Formation displays an internal stratification depicted principally by flat-lying to inclined shell fragments preserved in beds 2 to 20 cm thick, though some clasts also conformed to the fabric (Stoker, 1988). More commonly, the diamictons recovered in most of the boreholes, including BH90/11, BH90/12, BH90/13, BH90/14 and BH90/16 (Banks, Conchar, Caoilte and Conan formations: southern Hebrides Shelf) and BH85/02, BH85/02A, BH85/02B and BH88/01 (MacDonald Formation: south-east Wyville Thomson Ridge to northern Hebrides Shelf), are of a massive structure with randomly orientated clasts and shell fragments (Selby, 1989; Stoker et al., 1993).

These diamicton-dominated sequences are likely to be constructed by a variety of ice-contact and ice-marginal processes active during the periodic growth and decay of the British–Irish Ice Sheet (Stoker and Holmes, 1991). The MacDonald and Conan formations include numerous mounded accumulations that form prominent ridges on the sea bed, which range from 10 to 30 m high, 2.5 to 4.5 km wide and can be traced for up to 70 km (Figure 81)a. These ridges are interpreted as submarine end moraines or morainal banks (Selby, 1989; Stoker et al., 1993). An ice-contact fan, 50 to 150 m thick, up to 10 km wide and at least 25 km long has also been reported in association within the MacDonald Formation, at the junction of the Hebrides Shelf and Wyville Thomson Ridge (Stoker and Holmes, 1991). These ice-contact bedforms are interpreted to mark former positions of the British–Irish Ice Sheet, which locally reached the edge of the Hebrides Shelf (Figure 84).

In older, buried sequences, such as the Banks and Conchar formations, any diagnostic landform morphology has been eroded prior to deposition of the overlying, younger sequences, resulting in their sheet-like geometry. Processes of lodgement, ice-marginal mass-flow, iceberg rafting and iceberg scouring have been (Figure 84) Glacial geomorphology of the Hebridean region, based on information derived from Selby (1989), Stoker and Holmes (1991), Stoker et al. (1993), Stoker and Bradwell (2005), and Bradwell et al. (2008). Chart shows the glaciation curve for the Hebridean margin based on the stratigraphical range of the sediments recovered in BGS borehole BH88/07, BH88/07A (Stoker et al. 1994), core MD95-2006 (Kroon et al., 2000; Knutz et al., 2001), and core MD04-2822 (Hibbert et al. (2010). British chronostratigraphical stages are based on Bowen (1999); the ages of marine isotope stage boundaries (MIS) are based on Martinsen et al. (1987) and Williams et al. (1988). Note the uncertainty in the timing of ‘Wolstonian’ events recognised in borehole BH88/07, BH88/07A. invoked for the origin of these diamicton-dominated sequences, in association with mass-flow sands, and glacifluvial sands and gravels (Stoker, 1988; Selby, 1989; Stoker and Holmes, 1991). The common occurrence of a restricted, low-diversity, yet indigenous microflora and microfauna, including dinoflagellate cysts such as Bitectatodinium tepikiense and Protoperidinium (round brown) cysts, and benthic foraminifera such as Elphidium clavatum, Cassidulina reniforme, Cibicides lobatulus and the left-coiling planktonic Neogloboquadrina pachyderma, supports an ice-marginal setting.

The MacDonald and Conan formations locally form basal diamictons in glacially overdeepened basins, such as the Sula Sgeir and St Kilda basins (Figure 84). On the flanks of these basins, the diamicton sequences commonly interdigitate with the basin-fill deposits, indicative of their contemporary development. The MacIver Formation and the Oisein and Fionn formations infill the Sula Sgeir and St Kilda basins respectively (Figure 79). Glacimarine dropstone muds, proved in BGS boreholes such as BH88/09 and BH88/10 (Figure 76) and (Figure 77) comprise these sequences, which were laid down during deglacial phases of the British–Irish Ice Sheet (Peacock et al., 1992; Stoker et al., 1993).

On the Hebrides Slope, the bulk of the Eilean Siar Glacigenic Group is assigned to the Upper MacLeod Formation, although the MacAulay and Gwaelo formations are locally differentiated in the upper part of the succession (Figure 79). At the edge of the Hebrides Shelf, the diamicton-dominated shelf sequences interdigitate with, or pass laterally into, the slope apron, which developed during phases of maximum shelf-edge glaciation (Stoker, 1990, 1995b; Stoker et al., 1994) (Figure 67) and (Figure 81). During these phases, vast amounts of glacial sediment were delivered directly to the Sula Sgeir and Barra–Donegal fans (Figure 81). Short sediment cores from the upper slope of the Sula Sgeir Fan recovered an interbedded succession of debris-flow diamictons, turbidite sands and muds from the acoustically structureless packages of hummocky mass-flow deposits above the Glacial Unconformity (Stoker, 1990). The higher-amplitude clinoforms separating the mass-flow units (Figure 81)a display a draped geometry, which implies intervals of reduced sediment supply to the slope, such as in an interstadial or interglacial phase (Stoker, 1995b). The MacAulay and Gwaelo formations represent the uppermost clinoform unit on the Sula Sgeir and Barra–Donegal fans, respectively, from which sediment cores confirm a drape of glacimarine/marine strata (Stoker et al., 1989, 1993; Knutz et al., 2001).

On the upper Hebrides Slope, between the Sula Sgeir and Barra–Donegal fans, borehole BH88/07, BH88/07A penetrated the entire preserved glacial succession, between sea bed and 67.82 m (Leslie, 1993; Stoker et al., 1994) (Figure 74). The base of the section, the Glacial Unconformity, is erosional, with evidence of iceberg scouring. The sediments of the overlying Upper MacLeod Formation consist almost entirely of grey, bioturbated, muds and sandy muds with subordinate thin, muddy sands. Abundant matrix-supported glacial dropstones occur throughout the succession, as do both upward fining- and upward-coarsening units 0.5 to 3.0 m thick. These sediments are of distal glacimarine origin, probably deposited by a combination of iceberg rafting of debris and the downslope and alongslope transport of fine-grained material as sediment plumes that generated the graded units.

The northerly flow pattern of the alongslope currents (Figure 84) suggests that the Barra–Donegal Fan may have been a major source of the glacimarine sediment deposited on the upper Hebrides Slope landward of the Geikie Escarpment (Leslie, 1993). Despite the lithological homogeneity of the glacial sediments, their acoustic signature is variable with weak reflections in the lower part of the post-Glacial Unconformity succession overlain by more variable, stronger, reflections (Figure 74). This variation has been attributed to differences in physical properties, such as the water content, which is reduced in the lower part of the succession compared to the upper part. This has been linked to the depositional history of the sediments, and especially differences in sediment accumulation rate as the lower, weakly layered sediments accumulated three times more rapidly than the strongly layered deposits (Stoker et al., 1994; Talbot et al., 1994).

This change in acoustic character broadly correlates with the Middle–Upper Pleistocene boundary as defined from biostratigraphical data (cf. Stoker et al., 1994). On the basis of the dominance of the nannofossil Gephyrocapsa caribbeanica, and the absence of P. lacunosa (LAD below the Glacial Unconformity), the section between 67.82 and 34.0 m was probably deposited between 0.44 and 0.38 Ma, most likely during marine oxygen isotope stages (MIS) 12 and lowermost 11, thus correlating with the Anglian glacial stage (Figure 84). Whilst the age of the sediments between 34.0 and 22.0 m remains uncertain, the near coincidence of the LAD and first appearance datum (FAD), respectively, of the nannofossils Gephyrocapsa ericsonii and Emiliania huxleyi implies that this interval was deposited between 0.38 and 0.27 Ma. This incorporates MIS 11 to 8, which broadly correlates with the Hoxnian to early ‘Wolstonian’ interval. However, no interglacial (neither Hoxnian (MIS 11) nor MIS 9) deposits were recovered, and the sediments cannot be assigned to any specific glacial stage(s), e.g. MIS 10, 8 or both. The absence of Gephyrocapsa ericsonii in the sediments above 22.0 m suggests that they are younger than 0.122 Ma, correlating with the Devensian glacial stage. This implies a major hiatus of up to 0.15 Ma at about 22.0 m, coinciding with the change in acoustic character, as well as intense iceberg scouring of the palaeo-sea bed at this level, upslope from the borehole site (Figure 74). The Devensian section can be subdivided at 13.32 m on the basis of the Emiliania huxleyi Acme Zone dated at about 0.06 Ma. The sediments between 22.0 and 13.32 m are tentatively assigned to MIS 5d to 4 (early and early mid-Devensian), whereas the section from 13.32 to 0.5 m is correlated with MIS 3 and 2. The section between 0.5 m and the sea bed is assigned to the early Holocene (pre-7.5 ka).

Borehole BH88/07, BH88/07A provides a temporal record of Mid to Late Pleistocene glaciation, and forms a key empirical basis for a tentative reconstruction of the British–Irish Ice Sheet on the Hebridean margin, linking the shelf sequences to the slope record during the last 0.5 Ma. A preliminary glaciation curve based on this borehole suggests that there may have been major expansion of the British–Irish Ice Sheet onto the margin during MIS 2, 4, 8/10 and 12 (Figure 84). Although the resolution of the borehole data precludes unambiguous delineation between MIS 12–8, it does indicate a significant input of sediment onto the Hebrides Slope during this interval (Anglian to Wolstonian), which implies shelf- edge glaciation probably associated with the lower parts of the Banks and Conchar formations. A well-defined network of east–west trending subglacial channels has been identified at the base of the Banks Formation, which confirms ice-sheet expansion onto the shelf leading out to the Barra–Donegal fan system (Stoker et al., 1993). The significance of the absence of sediments correlated to MIS 6 in borehole BH88/07, BH88/7A remains uncertain, especially in the light of a significant delivery of ice-rafted detritus to the Rockall Basin during this stage, as recorded in core MD04-2822 (Hibbert et al., 2010) (Figure 84). Peak glacial expansion on other parts of the UK margin has also been documented for this interval (Holmes, 1997).

Biostratigraphical evidence derived from basin-fill deposits (the MacIver, Oisein and Fionn formations) associated with the diamicton-dominated MacDonald and Conan formations suggests that the latter correlate, respectively, with early (MIS 4) and late (MIS 2) Devensian expansion of the British–Irish Ice Sheet. This implies that the ice sheet reached the edge of the Hebrides Shelf several times during the Devensian, though not necessarily reaching the shelf-edge along the entire margin at any one time. This is consistent with the record of ice-rafted detritus preserved in core MD04-2822 from the Rockall Basin, which suggests that the initial expansion of the Devensian ice sheet off north-west Britain occurred at about 70 ka BP (Hibbert et al., 2010) (Figure 84). Organic sediments, buried by till, in North Lewis have been dated as 37–31 ka BP (radiocarbon dates of 32–26 ka 14C BP), which implies an ice-free interval towards the end of MIS 3, with build-up of the last (Late Devensian) ice sheet occurring from 35–32 ka BP (Whittington and Hall, 2002).

Geomorphological evidence to the north and west of Lewis, including elongate subglacial bedforms and glacial moraines, indicates that the Hebrides Shelf was widely glaciated by Late Devensian ice (Stoker and Bradwell, 2005; Bradwell et al., 2008). Farther south, St Kilda, albeit subjected to valley glaciation, is reported to have not been inundated by Late Devensian ice (Sutherland et al., 1984). In contrast, the deep-water sedimentary record of the MIS 3/2 transition preserved on the Barra–Donegal Fan equates to an increase in sediment accumulation rate and the input of ice rafted detritus to increased sediment delivery from an expanded Late Devensian British–Irish Ice Sheet, its former presence indicated by the morainal banks of the Conan Formation (Knutz et al., 2001, 2002a, b; Wilson et al., 2002; Scourse et al., 2009). Thus, the

Late Devensian ice sheet most likely extended to, or close to, the shelf edge locally on both the southern and northern part of the Hebrides Shelf. This variation in the location of the ice margin may be related to the action of ice streams (Stoker and Bradwell, 2005; Bradwell et al., 2008). By incorporating these data with numerical modelling of the Late Devensian British–Irish ice sheet it has been suggested that the ice sheet attained a maximum extent to the shelf edge between about 29 and 19 ka (Bradwell et al., 2008; Greenwood and Clark, 2009; Hubbard et al., 2009; Scourse et al., 2009). However, it should be noted, that throughout this entire interval the ice sheet was probably subjected to rapid fluctuations in size, at centennial to millennial timescales, and experienced repeated but episodic and highly transient zones of fast-flowing warm-based ice (ice streaming). After 19 ka BP, the ice sheet underwent rapid thinning and retreat, and the ice margin was probably located at or near to the present-day coastline at about 16–15 ka BP (Graham et al., 1990; Stoker and Bradwell, 2005; Everest et al., 2006; Stone and Ballantyne, 2006). The complete demise of the last remnant of British glacier ice was achieved by about 10.5 ka BP, in the early Holocene (Hubbard et al., 2009; Golledge, 2010).

A widespread veneer of sand and gravel marks the rise of sea level across the Hebrides Shelf, during the ensuing Holocene interglacial period (Stoker et al., 1993). This veneer unconformably overlies the glacial diamictons, and largely records the effects of reworking during postglacial transgression of the shelf. In the Outer Hebrides, the general submergence of the islands is indicated by the absence of raised beaches, and the presence of several submerged peats (Peacock, 1984). Since this rise in sea level, the shelf has been essentially starved of sediment, whilst extensive blown sand, alluvium, and large areas of peat have developed in the Outer Hebrides (Peacock, 1984), with slopewash and soil overlying Late Devensian till on St Kilda (Sutherland et al., 1984).

Rockall Basin

Bottom-current activity prevailed throughout the deposition of the basinal sediments of the RPa megasequence. However, following the Early Pliocene reorganisation of bottom currents that created the C10 unconformity, the focus of sedimentation shifted. The sediment drifts migrated onto the eastern flank of the Rockall Basin and onto the Wyville Thomson Ridge (Figure 83), whereas the western and northern parts of the basin became areas of persistent sea-bed erosion (Figure 76) and (Figure 80). Erosion also prevailed around the bases of the seamounts, where moats exist at the present-day with a relief in excess of 200 m below the general level of the sea bed (Stoker, 1998; Howe et al., 2006). This pattern of sedimentation contrasts with the deeper-water southern Rockall Basin where the morphology of the Feni Ridge was generally maintained (Figure 82).

The depositional style established during the Miocene continued through the Pliocene–Holocene interval. Broad, flat-lying to gently domed, sheeted basinal drifts, and elongate mounded drifts, which onlap and migrate up the eastern flank of the Rockall Basin, characterise the RPa megasequence (Figure 81) and (Figure 83). The marginal, elongate mounded drifts vary from single to multicrested, and are locally associated with sediment waves (Howe et al., 1994; Stoker et al., 1998). The latter also occur as part of the basinal succession, passing laterally into broad sheeted drifts. The sediment waves illustrated in (Figure 83) form a sediment package up to 200 m thick, extend over an area of about 550 km², and are divisible into a basal set of climbing waves separated by a transitional zone from an upper set of sinusoidal waves, some of which have heights of 18 m and wavelengths of over 1 km. Howe et al. (1994) demonstrated an upcurrent/oblique current migration direction towards the Hebrides Slope, consistent with the migration direction of the broad sheeted and elongate mounded drifts. These sediment waves retain some sea-bed expression except where overlain by debris-flow deposits at the distal edge of the Sula Sgeir Fan. It should be noted that not all sediment waves in the report area are associated with alongslope currents; the development of sediment waves on the Barra–Donegal Fan has been attributed to the interaction of downslope (turbidity) and alongslope currents (Howe, 1996; Knutz et al., 2002a, b). Overlap or interdigitation between the basinal strata and the prograding slope is common along the Hebridean margin, recognised at a variety of scales from the large-scale seismic architecture of the margin, e.g. (Figure 81) and (Figure 83) to the small-scale interbedding of sediment facies in cores (see below) (Howe, 1996; Armishaw et al., 2000; Knutz et al., 2001).

The RPa megasequence pinches out on the western flank of the Rockall Basin, which is marked by an irregular, composite, erosion surface (Figure 71) and (Figure 80) that has been sculpted since the Early Pliocene reorganisation of bottom currents within the basin. A veneer of RPa is locally draped over the Miocene drifts (Figure 80)b, though the erosion surface is mostly mantled by a sandy sea-bed layer that includes gravel lag deposits (Howe et al., 2001). The latter implies prolonged exposure to phases of vigorous bottom-current activity.

In the southern Rockall Basin, the Feni Ridge displays high-amplitude, continuous, parallel to wavy-bedded reflections that represent the continued aggradation of this giant elongate sediment drift, following the C10 event (Figure 82). Sediment waves are common on the flanks of the Feni Ridge, and on the smaller drift adjacent to the Porcupine High. Wavelengths range from several hundreds of metres up to 2 km or more; wave heights can reach in excess of 30 m (de Haas et al., 2003). According to de Haas et al. (2003), waves in the subsurface appear to migrate towards the ridge crests, whereas the most recent sediment cover appears to be draped over the waves. Whilst the asymmetric profile of the crest of the Feni Ridge reflects a degree of drift migration, this has been locally enhanced due to slumping (Stoker et al., 2001).

The nature of the basinal sediments has been tested at several sites within and beyond the report area, including a large number of short sediment cores (Figure 77) and (Figure 78). The only continuously cored sections to penetrate (and largely recover) the entire succession in the Rockall Basin have been acquired south of the report area, at DSDP site 610 and ODP site 980/981 (Figure 63). In the southern Rockall Basin, DSDP site 610 cored the Feni Ridge and proved about 75 m of Lower to Upper Pliocene (NN15–16) nannofossil and diatom ooze overlain by 140 m of Upper Pliocene to Holocene (NN16/17–21) cyclic nannofossil ooze and calcareous muds, with abundant dropstones (Ruddiman et al., 1987). The sediments are interpreted as calcareous and siliceous biogenic contourites, with the upward increase in cyclicity related to glacial–interglacial oscillations (Kidd and Hill, 1987).

Farther north, a comparable sequence was recovered from an elongate mounded drift onlapping onto the western flank of the Rockall Basin at ODP sites 980 and 981. At these sites, a 113 m-thick section of Lower to Upper Pliocene (NN15–17) nannofossil ooze is overlain by a 158 m-thick cyclical sequence of Upper Pliocene to Holocene nannofossil oozes and clay (Jansen et al., 1996). The influx of ice rafted detritus is recorded from a sub-sea-bed depth of 138 m in this borehole, and is dated at 2.41 Ma. The LAD of Pseudoemiliania lacunosa, which marks the Glacial Unconformity on the Hebridean margin, is recorded at both these sites, at depths of 24 m (610) and 35 m (980/981) below sea bed, but has no resolvable seismic expression comparable with the shelf-margin unconformity.

West of the report area, DSDP site 116 and ODP site 982 proved a Pliocene–Holocene succession of nannofossil ooze, with a marked cyclic association of ooze and clay in the upper 60 m (Late Pliocene–Holocene) of the record (Laughton et al., 1972; Jansen et al., 1996). The influx of ice rafted detritus is dated at about 2.6 Ma at site 982.

In the North-east Rockall Basin, wells 164/25-1Z and 164/25-2 penetrated the basinal succession underlying the Sula Sgeir Fan, and reported a predominantly muddy succession with some indication of interbedded sands on the composite logs, though this is based on little direct observation of core (Figure 77). On the western flank of the Rockall Basin, BGS borehole 94/01 penetrated a 17 m-thick drape of the RPa megasequence that partially fills a moat adjacent to a relict Miocene elongate mounded drift (Figure 80)b, and recovered only traces of muddy, silty, and pebbly sand. An adjacent BGS rock core, 56-14/13 (Figure 76) and (Figure 78), showed the uppermost 2.1 m of the drape to consist of an interbedded sequence of massive to graded, bioturbated, thin- to medium-bedded sands and pebbly sands with rare very thin-bedded muds.

A number of cores have tested the upper part of the basinal succession (Figure 76), (Figure 77), (Figure 78). On the western flank of the Rockall Basin, BGS borehole BH94/04 indicates that the upper 25 m of the Upper Oligocene lowstand wedge of bioclastic sandstone (see above) has been reworked during the late Neogene and Quaternary, resulting in a cover of unconsolidated bioclastic sands. Adjacent BGS short sediment cores 57-13/53, 57-13/75 and 57-13/76 cored the uppermost 4.8 m of the wedge and confirm the reworked nature of the sediment, proving a sequence of massive to graded (upwards-fining and coarsening) contouritic sands, gravelly sands and sandy gravels, and massive, bioturbated, hemipelagic muds with ice-rafted dropstones (Howe et al., 2001). An age range of latest Pliocene to Holocene is estimated by biostratigraphical data, though the accuracy of the data may be affected during reworking of the sediments by bottom currents. To the south, short cores 57-14/47 and 48 sampled a 1.1 m Pleistocene veneer of massive, bioturbated mud with a sea-bed sandy layer, unconformable on the relict Miocene–Early Pliocene sediment drift illustrated in (Figure 80). A comparable sequence of Middle Pleistocene–Holocene sandy and gravelly contourites interbedded with hemipelagic muds has been proved elsewhere on the western flank of the Rockall Basin in short cores 57-13/57, 57-14/54 and 58-14/34 (Howe et al., 2001).

In the central part of the Rockall Basin, up to 3.54 m of massive, bioturbated hemipelagic muds with dropstones of Mid Pleistocene to Holocene age have been recovered in BGS short cores, including 57-13/26, 58-12/5, 58-12/9 and 58-11/2 (Figure 76) and (Figure 78). A coarser-grained 1.32 m-thick sequence of massive to graded contouritic sands, gravelly sands and gravels interbedded with hemipelagic muds, unconformable on Lower Pliocene ooze, was proved in core 57-12/33 near the base of the Anton Dohrn Seamount. Farther south, a giant piston core — MD04-2822 — sampled the distal margin of the Barra–Donegal Fan (Figure 76) and (Figure 84) to a depth of 37.7 m below sea bed. The core recovered predominantly clay to silty clay of a hemipelagic nature, with sporadic distal turbidite layers, small dropstones and rare bioturbation that spans the late Wolstonian to Holocene interval (MIS 6–1) (Hibbert et al., 2010). Ice-rafted detritus derived from the Hebridean margin has been identified in these sediments, and suggests an expansive shelf glaciation with marine calving margins for each of the last three main glacial stages (MIS 2, 4, 6). On the eastern flank of the Rockall Basin, a 30 m-long piston core (MD95-2006) recovered a Late Pleistocene to Holocene interbedded succession of hemipelagites and glacimarine muds, muddy and silty–muddy contourites, and thin sandy turbidites from the upper part of the Barra–Donegal Fan (Knutz et al., 2001, 2002a, b). This sediment accumulation has been influenced by frequent and extremely rapid shifts in terrigenous sediment supply (perhaps over time-spans of less than a century), reflecting changes in bottom-current flow strength, and fluctuations in glacimarine discharges from the British Ice Sheet, on the adjacent Hebrides Shelf, and downslope sedimentation, dating back to 45 ka BP (Kroon et al., 2000; Harland and Howe, 1995; Howe et al., 1998; Wilson et al., 2002). In particular, increased clay input and expanded sections (>20 m thick) characterise sedimentation during cold periods, such as MIS 2, whereas stronger current activity during warmer periods, e.g. the Holocene, has resulted in erosion and winnowing, and the development of condensed sections. Over most of the report area, the Holocene section is mostly less than 1m thick (Stoker et al., 1993).

At shallower depths, BGS boreholes BH94/02, BH94/03, BH94/05 and BH94/06 recovered a clastic succession of interbedded gravel, muddy sand, gravelly sandy mud and mud from the top of the Rockall High (Figure 77). Where sampled, this sediment cover is mostly preserved as a seismically unresolvable veneer. However, borehole BH94/05 penetrated a distinct bedform resembling a symmetrical shallow-marine, tidal, sand ridge, up to 50 m thick, and 3 km wide at a water depth between 200 and 300 m, close to the eastern edge of the high. It is clearly depicted in (Figure 76) as part of a unit that can be traced laterally for up to 50 km. Its location in deep water suggests that it is probably inactive, its formation most probably relating to a former lowstand of relative sea level. Similarly, the section tested in borehole BH94/03 is from a wave-cut platform or terrace (Figure 72), one of several that can be mapped at water depths of 200 to 400 m along the north-eastern edge of the Rockall High (see Physiography and sea-bed sediments section; (Figure 87)), and also indicative of former low sea level. The age of these features is poorly constrained; the biostratigraphical data from the boreholes give ranges from the Pliocene to the Holocene (Figure 77). Adjacent to Rockall islet, a beach conglomerate dredged from a depth of 173–183 m is reportedly of Holocene age (Roberts, 1975a). The prograding sediment wedge identified on the eastern flank of the Rockall High is interpreted to largely consist of Lower Pliocene to Middle Pleistocene strata (Stoker, 2002), as the high was not glaciated beyond the effect of iceberg scouring (Roberts, 1975b), and was thus a much reduced sediment source area. Although this wedge has not been sampled, borehole BH94/02 is located close to its seismically defined north-east margin (Figure 76).

On top of the George Bligh High, a stacked assemblage of massive-to-graded, thin- to thick-bedded, bioturbated, Pleistocene sands, up to 10.6 m thick, has been tested in BGS borehole BH94/07 and short sediment core 58-14/9 (Figure 76), (Figure 77), (Figure 78). The amalgamated nature of this sandy succession implies that it has been strongly influenced by persistent currents actively washing across the top of the dome. The deeper-water setting of the dome relative to the Rockall High suggests that this current activity was related to surface to intermediate-depth water masses.

On top of the Rosemary Bank Seamount, BGS borehole BH90/18 penetrated a 17.0 m-thick cover of upper Neogene to Quaternary strata. The top 4 m of the section was not recovered, however from 4.6 to 17.0 m the sequence consisted of Lower to Upper Pliocene (NN15–18) sands (Figure 76) and (Figure 77). However, a nearby short sediment core, 59-11/13, tested the uppermost 4.67 m and recovered a Middle Pleistocene to Holocene succession of massive-to-graded contourite sands and gravelly sands, with subordinate hemipelagic muds. There is no indication of an unconformity on seismic data (Howe et al., 2006). On the southern slope of the bank, sediment cores 59-11/16 and 59-11/17 sampled the upper 5.75 m of a discrete sediment drift and proved an interbedded sequence of sand, gravel and mud of Mid Pleistocene to Holocene age. A comparable sequence of interbedded sand and mud has been recovered from the top of the Anton Dohrn Seamount by BGS cores 57-11/67 and 57-12/19, which sampled up to 5.6 m of the upper Neogene to Quaternary drape (Figure 78). In common with Rosemary Bank, the age range is poorly constrained, ranging from Late Pliocene to Holocene (NN18–21). Both seamounts have been strongly affected by currents associated with surface and intermediate water masses, the cyclic pattern of sedimentation most likely reflecting the fluctuating glacial–interglacial climate conditions in late Neogene to Quaternary times (Morri, 2004). Thus, ongoing reworking and mixing of the sediment cover invoked by the current activity is most probably a contributory factor to the contradictory biostratigraphical age data. By way of contrast, an isolated occurrence of bioturbated, partially dolomitised, micritic limestone of latest Pliocene to Early Pleistocene age was sampled in core 57-12/41, beneath a veneer of Upper Pleistocene to Holocene sand (Figure 76) and (Figure 78).

Chapter 10 Economic geology

By Ken Hitchen and Martyn Quinn

Within the report area there has been no large-scale exploitation of mineral resources. The greatest potential resource is hydrocarbons (oil and gas) but exploration has so far been very limited and no oil or gas fields have yet been discovered. However the commercial ‘Benbecula’ gas discovery, made by well 154/01-1 in 2000, has been described as significant. Clathrates (frozen gas hydrates) may exist in the Rockall Basin but their presence has yet to be confirmed. Other potential resources include coal, sand and gravel, heavy minerals, carbonates and renewable energy such as wind and wave power.

Oil and gas

The first major seismic surveys across the Rockall Basin were conducted in the early to mid 1970s. These were speculative in nature and not related to a formal issue of exploration licences by the government. The first well to be drilled in the Rockall Basin was 163/06-1A in 1980. This is usually referred to as a ‘stratigraphical test well’ and was drilled by British National Oil Corporation on behalf of a consortium comprising the UK Department of Energy (now the Department for Energy and Climate Change) and 19 oil companies. It was not linked to a specific licence round and it was not a hydrocarbon exploration well. The well was intended to yield information about the deeper geology of the basin beneath the widespread lavas. Unfortunately the selected drilling location was on the north-west flank of the Darwin Volcanic Centre and the well was abandoned after drilling 1045 m of Paleocene volcanic rocks without penetrating the underlying Mesozoic succession (Morton et al., 1988).

Hydrocarbon exploration has been concentrated on the eastern side of the report area with most of the seismic data being acquired, and all of the wells being drilled, east of 10°W (Figure 2). The government has offered blocks for licensing here in the 5th (1977), 9th (1985), 11th (1989), 13th (1991) and 17th (1997) offshore licensing rounds. The 17th round licences resulted in six wells being drilled between 1997 and 2002. Following a Strategic Environmental Assessment of the Rockall Basin area, the government invited applications for exploration licences in the 25th offshore licensing round (2008) and 26th licensing round (2010). No applications were made for acreage in the Rockall Basin in either case.

The eleven hydrocarbon exploration wells drilled to date (Table 9) have targeted a variety of structures. Well 132/15-1 was drilled downdip of the crest of one of a series of rotated fault blocks. Immediately overlying the metamorphic basement was an interval interpreted as syn-rift and dated as Hauterivian to Cenomanian (Musgrove and Mitchener, 1996). This is one of the most direct lines of evidence for the age of rifting in the Rockall Basin although the actual lithology in the well was predominantly mudstone.

Wells 164/25-1ST and 164/25-2 targeted the West Lewis Basin and Ridge respectively (Isaksen et al., 2000, fig. 2). The former well proved that, beneath the Cenozoic, Cretaceous mudstones directly overlie Permo-Triassic sandstones at this location with the Lower Cretaceous being represented by only 2 m of Albian siltstones. In the second well, a 140 m-thick Upper Paleocene sandstone interval occurs between the upper and lower extrusive volcanic intervals. The sandstone was cored but contained no traces of hydrocarbons. The lower extrusive volcanic interval rests on metamorphic basement.

Well 164/07-1 targeted a Mesozoic structural four-way dip closure beneath the Upper Paleocene basalts in the North-east Rockall Basin. However at 3301.4 m TVDSS the well drilled into an intruded Albian to Coniacian mudstone succession where igneous material made up over 50 per cent of the interval (Archer et al., 2005). After drilling 1792.5 m of this succession the well was terminated at 5083.9 m TVDSS after penetrating over 70 sills but not a single potential hydrocarbon reservoir interval.

All the other hydrocarbon exploration wells were aimed at potential Paleocene or Eocene sandstone reservoir targets. In most cases the reservoir facies was encountered but the lithologies have proved very variable. A conglomeratic interval was proved in well 153/05-1. The sandstones in other wells commonly have a volcaniclastic or tuffaceous component.

Discoveries and shows

Including the stratigraphical test well, only twelve commercial deep wells have been drilled within the report area and no major hydrocarbon discoveries have been made to date. However well 154/01-1 is reported as a significant gas discovery (named ‘Benbecula’) with tests being conducted on a 94.2 m thick Upper Paleocene low-porosity sandstone resting directly on Upper Cretaceous (Maastrichtian) claystones. This sandstone is below the main extrusive volcanic interval in the well. A second well, drilled to appraise this sandstone interval further (well 154/01-2) was spudded in June 2006 a few kilometres to the north. In well 164/28-1A an Upper Paleocene (Thanetian) sandstone, more than 300 m thick yielded minor oil shows (fluorescence) mainly from the upper part of the interval. It has also been reported that that traces of hydrocarbons have been discovered in the volcanic rocks of well 163/06-1A (Morton et al., 1988). Although somewhat unusual this is not unique — hydrocarbon traces have also been described from the basalts of the Faroe Islands (Laier et al., 1997).

In Irish waters well 12/02-1 drilled the ‘Dooish’ prospect, 125 km from the Donegal coast, in 2002 (Figure 2). The well was terminated at 4116.9 m TVDSS after penetrating a substantial gas condensate column within a thick Permian sandstone reservoir (Shannon et al., 2005). The following year the well was re-entered, side-tracked (12/02-1Z) and deepened to 4434 m TVDSS. The two wells proved a 214 m hydrocarbon column in an underfilled structure. None of the petroleum samples analysed exhibited any significant correlation with potential source rocks penetrated by the wells. However the wells are significant as they prove that a working hydrocarbon system is present in the Rockall Basin.

The Corrib gasfield was originally discovered in 1996 by Irish exploration well 18/20-1 drilled 70 km off the coast of Co. Mayo, west Ireland. The subsequent appraisal well 18/20-2Z penetrated a 185 m gas column in a fair to good quality Triassic fluviatile sandstone reservoir. The trap is a north-east-trending, faulted anticline, contains approximately 1.2 TCF of gas in place and is presumed to be sourced from Westphalian Coal Measures (Dancer et al., 2005). The Corrib field is located in the Slyne Basin and illustrates the potential for hydrocarbon discoveries to be made in the basins marginal to the main Rockall Basin. Such basins exist in the UK and Ireland, and on both the east and west flanks of the Rockall Basin.

A small number of natural oil seeps has been recognised in the area of Rockall Bank using satellite synthetic aperture radar data (Hitchen, 2004) (Figure 5). These results are contained in a commercial study of the area west of 12°W. Lack of recognised seeps in the eastern part of the report area may be due to lack of data. The seeps are detected by the identification of thin oil slicks which have to meet specific criteria designed to omit false non-natural slicks caused by, for example, pollution events. Each slick is allocated a confidence factor that it is derived from the interpretation of the natural seep. Six slicks have been identified, all at the 60 per cent confidence level, on the eastern flank of Rockall High, possibly associated with the sub-basalt East Rockall High Basin, identified from gravity data (in UK waters) (Figure 5), or the Rónán Basin (in Irish waters). A seventh slick, identified with 90 per cent confidence, is located at the northern end of Rockall Bank where it is more difficult to explain geologically.

Potential source rocks

Potential hydrocarbon source rocks have been proved at many locations on the Atlantic Margin of the UK and Ireland. These include Carboniferous coals, mainly considered to be a source for gas, and Jurassic and Cretaceous marine and nonmarine sediments. From south to north the locations include the Porcupine Basin, the Slyne and Erris basins, the Inner Hebrides (notably Skye), the West Lewis and West Flannan basins and the Faroe–Shetland Basin (Butterworth et al., 1999; Hitchen and Stoker, 1993; Isaksen et al., 2000; Morton, 1993; Scotchman, 1998, 2001). Whether these potential source rocks are considered to be present in the deeper parts of the Rockall Basin depends on perceived crustal and palaeogeographical reconstructions and is a matter for debate. Cole and Peachey (1999) suggest that, owing to a pre-Cretaceous rifting event,‘Kimmeridge Clay equivalent’ source rocks might be present in the Rockall Basin. However none of the commercial wells in the report area has proved the presence of a hydrocarbon source rock although several BGS boreholes, drilled in the West Lewis and West Flannan Basins, proved Middle and Upper Jurassic and Lower Cretaceous potential source rocks (Hitchen and Stoker, 1993; Isaksen et al., 2000).

Carboniferous coals (or carbonaceous sediments) are the oldest likely hydrocarbon source rocks in the area. These are the presumed source for the Corrib gas field and may also have sourced the gas condensate in the Dooish discovery (Irish well 12/02-1). Although not proven in the report area, Westphalian sediments have been drilled by Irish wells 13/03-1 (Figure 2) and 19/05-1 in the Donegal Basin and Erris Trough respectively with the former well penetrating up to 50 coal horizons (Tate and Dobson, 1989).

Middle Jurassic (Bathonian) dark grey to black, organic-rich mudstones have been drilled close to sea bed, in an up-dip location, in the West Lewis Basin by BGS boreholes BH88/01 and BH90/02. The mudstones were deposited under freshwater, swamp, intertidal or lagoonal low-energy conditions and may be the time equivalent of the Cullaidh Shale Formation of the Hebrides Basin (e.g. on Skye). They possess total organic carbon (TOC) values ranging from 3.23 to 8.75 per cent (Table 10) and a low organic sulphur content (0.02–0.27 per cent). Although immature where drilled, they have the potential to produce paraffinic nonvolatile oil (Hitchen and Stoker, 1993; Isaksen et al., 2000).

Lower Cretaceous (Berriasian) mudstones have been drilled in both the West Lewis and West Flannan basins by BGS boreholes BH90/05 and BH90/09 respectively. Although deposited in an open marine environment the presence of lignite and plant fragments implies a relatively nearshore situation. Anoxic conditions may have existed at the sediment-water interface. The predominance of algal matter in these mudstones, and their age, suggest they are equivalent to the widespread latest Jurassic to earliest Cretaceous Kimmeridge Clay Formation which has sourced numerous oilfields in the northern North Sea and west of Shetland. The TOC values range from 1.40 to 14.50 per cent and the organic sulphur content is 0.76–2.07 per cent. The Berriasian mudstones have a mixed potential to generate gas and volatile oil and will expel their hydrocarbons at lower maturity levels than the Bathonian mudstones (Hitchen and Stoker, 1993; Isaksen et al., 2000).

Numerous other examples of potential Jurassic source rocks have been proved in offshore wells and boreholes on the Atlantic Margin of the UK (including the North Minch and Sea of Hebrides basins) and Ireland (including the Slyne, Erris and Porcupine basins). At various locations on the margin, Jurassic rocks of nearly every stage contain potential hydrocarbon source rocks (Butterworth et al., 1999; Scotchman et al., 1998; 2001). Whether any of these is present in the deeper parts of the Rockall Basin is one of the key questions affecting prospectivity of this basin. Several authors (including Roberts et al., 1999; Naylor and Shannon, 2005) have suggested that Jurassic rocks may be present here although they may be restricted to the margins as the basin comparmentalised during the main Cretaceous rifting phase. Furthermore, Jurassic shales were penetrated by Irish well 12/02-2 although their precise age and nature has not been reported.

Well 163/06-1A in the northern Rockall Basin proved 689 m of basalts directly overlying 356 m of dacites. The geochemistry of the dacites is unusual and the high Al, Ni and Cr contents, combined with their Sr and Pb isotopic compositions indicate they are not differentiates of a basaltic parent magma. It has been suggested that the dacites were derived by the melting of (?Cretaceous) black shales, and possibly also sandstones, in the Rockall Basin by being in contact with a high-temperature magma for a significant period of time (Morton et al., 1988). It requires further deep exploration drilling within the Rockall Basin to prove the presence or absence of such a potential source rock here.

Methane hydrates

In deep water, free-phase methane may occur in combination with water, in an ice-like form known as methane hydrate, which has the potential to store large volumes of methane and is considered a future hydrocarbon resource. It can also act as a cap for shallow gas accumulations. Hydrate formation requires moderate to high pressures typical of water depths of over 400 m. Assuming a gas source is present, the distribution of methane hydrate is dependent on pressure and temperature within the sediments and these can be used to model the extent and thickness of the methane hydrate stability zone (Figure 85). Hydrates are commonly recognised by the presence of a bottom simulating reflector, with a phase reversal, located at the base of the methane hydrate stability zone. No evidence for methane hydrate has been reported from the Rockall Basin. As well as a potential resource, hydrates are a potential hazard; when disturbed they can release large quantities of gas and destabilise sediments. Their potential presence needs to be considered in top-hole studies of deepwater wells.

Other potential resources

Although oil and gas remain the most potentially valuable future resources in the report area, other mineral or energy resources may also prove to be of value.

Coal

No coals (including lignites) have been encountered in commercial wells or BGS boreholes drilled within the report area. However some palaeogeographical reconstructions suggest that Upper Carboniferous sediments may originally have been deposited in the area, raising the possibility that coal-bearing successions may have been present. Even so, late Variscan uplift and erosion has reduced the likelihood of their preservation (for further discussion, see Cambrian to Carboniferous chapter).

The nearest coal successions that were formerly commercially exploited are located onshore at Ballycastle in Northern Ireland and at Machrihanish on the Mull of Kintyre. The coals that were mined at Ballycastle were exposed in sea cliffs and exploited from at least the early 18th century until the mines were finally abandoned in 1967 (Arthurs and Earls, 2004). The Machrihanish mine dates back to the fifteenth century and was worked intermittently for several centuries until finally closing in 1967. The mine extended offshore by about half a kilometre.

Adjacent to the report area, but east of the Outer Hebrides, lignitic coals have been drilled at depths of 64 m, 68 m and at 79 m below sea bed in BGS borehole BH80/14 (located just off the south-east coast of Harris). This borehole targeted a small basin in the hanging wall of the Minch Fault and proved Upper Oligocene terrestrial sediments. In the Canna Basin, BGS borehole BH78/01 drilled Upper Oligocene sediments comprising carbonaceous clays with plant remains and lignitic fragments (Evans et al., 1979).

Further afield, Oligocene, Lower Eocene, Upper Paleocene, Lower Cretaceous and Jurassic successions have been recorded. For instance, in borehole BH77/07 (drilled 100 km north of Cape Wrath), lignites of Oligocene age occur between the depths of 112.2 and 123.5 m (Evans et al. 1997). To the north-east, in the general west of Shetland area, several commercial wells have encountered coal-bearing intervals. Well 204/29a-2, drilled in the West Solan Basin, proved thin coals of Early Eocene to Late Paleocene age. Well 204/14-1, in the Judd Basin, is a typical example of the common occurrence of coals within the Early Eocene Balder formation. An Upper Paleocene coal, approximately 3 m thick, has been proved in commercial well 205/10-1A located on the Flett Ridge. Well 205/25-1, located in the West Shetland Basin, penetrated numerous thin coals within the Lower Cretaceous succession.

A thick coal-bearing Carboniferous succession has been proved south of the report area in Irish well 13/03-1 drilled in the Donegal Basin. This well drilled 949 m of Upper Carboniferous sediments containing up to 50 thin coal seams interbedded with sandstones, tuffaceous sandstones, siltstones and claystones (Stoker et al. 1993, fig. 29).

Sand and gravel

Dredging of marine aggregate makes a major contribution to the UK requirement for sand and gravel. It is used in the manufacture of concrete, as bulk fill in the construction industry, and for land reclamation. It also has more specialised uses as sand filters in the water industry, stabilisation of offshore oil and gas platforms and for beach replenishment (Ardus and Harrison, 1990). At present, exploitation of marine aggregates takes place only off the coasts of central and southern England. (Figure 91) summarises the distribution of sea-bed sediments in the report area. Amongst other factors, proximity to a ready market dictates where sand and gravel may be extracted offshore so (Figure 86) shows the eastern part of the report area (closer to Scotland) in more detail. Present technology limits the maximum water depth for extraction of sand and gravel by suction dredging to 45 m.

Heavy minerals

The onshore geology of the Outer Hebrides includes basic and ultrabasic intrusions. These may have provided a potential source for nearshore placer deposits of heavy minerals. However, there is no specific information on the existence or extent of such deposits in the vicinity of the Outer Hebrides.

Carbonates

Carbonates, in the form of shell sands from Scottish beaches have been used locally in agriculture as a fertiliser. In Iceland, offshore shell sands form the basis of a local cement industry. Other uses of lime include smelting, grit in chicken feed and road metal (Allen, 1983). Although an area around Orkney has been identified as the most promising area for commercial development, Allen (1983) notes that there may be deposits around the Outer Hebrides that may also have potential (Figure 86), and also (Figure 92) for carbonate percentage within sea-bed sediments of the whole report area). As with sand and gravel deposits, present technology limits the maximum water depth for extraction of shell sands by suction dredging to 45 m.

Renewable energy

Despite having a very high potential wind and wave resource, the report area currently has no existing offshore wind or wave energy generation facilities. Onshore windfarms have been proposed for the island of Lewis.

Chapter 11 Physiography and sea-bed sediments

By Carol Cotterill and Alick Leslie

Physiography

There are six distinct physiographical provinces within the area of this report, identified using the geographical names of Roberts et al. (1979):

• shelves that extend from subaerial islands to the shelfbreak, encompassing the Hebrides Shelf and Rockall Bank
• shelfbreaks
• the Hebrides and Rockall Bank slopes
• George Bligh Bank
• Rockall Trough
• the Hebrides Terrace, Anton Dohrn and Rosemary Bank seamounts.

Hebrides Shelf

The inner shelf, east of the Outer Hebrides, comprises Precambrian metamorphic basement highs, scoured by Pleistocene glacial activity, with elevated topography and lower-lying sedimentary basins. However, west and south-west of the Outer Hebrides, the shelf is dominated by the Outer Hebrides Platform (Figure 87). This is a gently sloping platform of predominantly Lewisian rocks, with a glaciated topography similar to that seen on land, and a total relief of 60 m or higher over the Stanton Banks (Pantin, 1990). The topographical highs are commonly formed from exposed rock, while lows are partially filled by Pleistocene and Holocene sediment.

The Flannan Ridge, including the Flannan Isles, is a northerly extension of the Outer Hebrides Platform. BGS cores taken during 2001 (e.g. 57-09/530 and 58-08/230) show a gravel lag overlying a metamorphic/alkali granitic basement rock of Lewisian age. The southern part of the Flannan Ridge has a rough topography showing pinnacles of weathered Lewisianoid rock. The St Kilda Massif at the south-western end of the Flannan Ridge comprises an exposed Palaeogene volcanic centre 25 km in diameter. On the platform surrounding the centre, distinct breaks of slope at water depths of 48 m, 100 m and 120 m have been identified (Jones et al., 1986).

The outer Hebrides Shelf, west of the Outer Hebrides, has less topographical expression than the inner shelf. The sea bed is commonly covered by a thick sequence of glacial deposits, which govern the sea-bed topography, including ridges with a relief of up to 50 m, overprinted by shallow troughs formed by the scouring of grounded icebergs. However, Cenozoic intrusive igneous rocks are exposed around the St Kilda archipelago.

A series of parallel ridges to the south-west of the St Kilda archipelago are interpreted to be terminal moraine complexes (Figure 87). These features range from less than 5 m to 50 m high, 1–8 km long, and can be mapped laterally for up to 70 km (Stoker, 1997; Bulat and Long, 2001). On a smaller scale, sediment bars, waves and ridges are common on the outer shelf, having their long axes parallel to the predominant north-easterly current direction (BGS, 1988b). These features are commonly composed of sand overlying a gravel substrate. On the outer margin of the shelf a number of channel-like features have also been identified, the most marked of which lies to the north-east of the Sula Sgeir Fan (Figure 87). This is interpreted to be a partially filled canyon related to the Sula Sgeir Fan depocentre on the slope.

Shelfbreaks

The Hebrides shelfbreak (Figure 88) represents a significant change in gradient of the sea bed (Vanney and Stanley, 1983). On the Hebrides margin the shelfbreak is between 60 and 100 km north-west of the Outer Hebrides and occurs at depths ranging from 150 to 250 m, although on the adjacent Wyville Thomson Ridge it is as deep as 500 m. The shelf itself has a gradient of between 0.07 and 0.2°.

Within the report area the morphology of the shelfbreak is variable, ranging from a distinct change of gradient to a gradual steepening of the sea floor. In the area of the Geikie Escarpment the shelfbreak is at its least distinct, with the most significant change in gradient, termed the slope gradient break, occurring on the escarpment itself (Figure 88), F–F’ and G–G’, with slope angles of more than 20° (Jones et al., 1986).

During the Plio-Pleistocene there was significant progradation of the shelfbreak on parts of the margin. In the area of the Sula Sgeir and Barra/Donegal fans supply of sediment to the slope has led to basinwards movement of up to 40 km. In between these depocentres, there appears to have been retreat of the shelfbreak in the last 400 Ka (Stoker, et al., 1993), possibly related to erosion by iceberg scouring.

The Rockall Bank shelfbreak has previously been described as occuring between the depths of 192 and 201 m (Roberts, 1975a). However, recent multibeam and echo sounder datasets show the shelfbreak to be located between 350 and 500 m. Jacobs (2006) identifies the shelfbreak as being at 350–400 m water depth along the south–south-easterly Rockall Bank, with a scarp slope rapidly dropping to 600 m water depth, and this interpretation is supported by BGS echo sounder data (92/01, line 23). The shelfbreak shoals northwards, rising to about 300 m water depth (BGS BH92/01 line 26; (Figure 88), C–C’). Below the shelfbreak, there are still steep scarp slopes of approximately 20° down to about 600 m water depth. West of 13° 50′W the slope angle decreases to about 2°, with the shelfbreak occurring at about 500 m water depth.

Hebrides Slope

The Hebrides Slope is an area of complex topography, reflecting the differences in Plio-Pleistocene sediment supply and pre-Neogene geology. The upper part of the slope, to a depth of approximately 500 m, is commonly marked by iceberg ploughmarks, below which slopes can be smooth, or disrupted by scarps, debris flow and contourite features.

In the southern part of the report area, the Hebrides Slope is dominated by the Barra and Donegal fans and the Peach Slide Complex. The Plio-Pleistocene depocentres contain sediment piles up to 1 km in thickness, covering an area of around 8400 km² (Roberts et al., 1979). Slope angles are greatest (5 to 7°) on the upper slope, falling to 1.5 to 3° on the mid slope and <1° on the lower slope. The Barra Fan surface is disturbed by at least five mass-wasting events, forming the Peach Slide Complex. Debris flows form a distinct lobate morphology at the sea bed. On the upper and mid slopes a number of scarp slopes record the head-wall positions of sediment failures (Figure 87).

North of the Barra Fan, the Hebrides Slope trends firstly north-west, then north then north-east, defining a bulge in the shelf margin. This bulge is related to the presence of the Geikie Volcanic Centre on the upper slope (Evans et al., 1989). In this area the upper slope has a very low gradient of around 0.6°, extending to a depth of 500–700 m, with no clear position for the shelfbreak (Figure 88), G–G’. Further north, the shelfbreak becomes defined at about 300 m water depth, with an additional slope-gradient break at about 700 m water depth (Figure 88), F–F’.

The mid slope is defined by the Geikie Escarpment (Figure 87), a steep scarp slope that appears to be related to a Late Oligocene erosive episode (Evans et al., 1989), which has since remained free of sediment. The Geikie Escarpment extends for 215 km around the bulge in the margin, and is 230 m in height north-west of St Kilda. Slope angles are commonly around 7° but can reach 26°. The base of the scarp lies at 910 m at its deepest, but shallows to around 700 m towards the north and south. A second deeper scarp underlies the Geikie Escarpment in the south of the bulge (Figure 87), termed the lower St Kilda scarp by Kenyon (1987), which may be related to the buried Hebridean Volcanic Escarpment. It does not curve to the north-east like the Geikie Escarpment and its expression diminishes north of 58° 30’N. At the base of both scarps, elongate ridges denote areas of contour current sedimentation.

North of the Geikie Escarpment the slope comprises part of another depocentre, the Sula Sgeir Fan (Figure 88), E–E’. Like the Barra Fan to the south, this is a Plio-Pleistocene sediment accumulation comprising stacked debris flows (Stoker, 1995b; Baltzer et al., 1998). A small number of slide scars are present on the upper slope of the southern Sula Sgeir Fan (Long et al., 2001). The Sula Sgeir Fan appears to have overlain, and possibly interacted with, a contourite ridge and moat complex, part of a sheeted and mounded drift in the North-east Rockall Trough (Howe et al., 2002; Masson et al., 2002) (Figure 87). The contourite ridge reaches a height of 150 m and is 20 km wide. The complex appears to be migrating upslope (Howe et al., 2002) and is interpreted to be active in the Holocene.

Rockall Bank

Rockall Bank forms an elongated plateau orientated south-west to north-east. Over most of its area it has a relatively planar, subdued topography, taking the form of a low gradient ramp. Above the shelfbreak much of the bank is covered by a thin veneer of biogenic carbonate sand and gravel (Scoffin et al., 1980). A broad topographical high at a water depth of around 230 m has been identified as a Pleistocene shallow marine sand bar (Figure 87).

Rock, either Lewisianoid metamorphic or Tertiary extrusive, exists on topographical highs including the subaerial Rockall Island. To the east of these highs the plateau gradient is around 0.1° (Roberts, 1975a). Echo sounder data on the shelf suggest that the bank surface is irregular with a relief of up to 5 m in places. This may be related to iceberg turbation of the sea bed during the late glacial period. However, other interpretations of this relief include sediment waves comprising carbonate sand and gravel.

To the east of Rockall Island a number of linear scarps/ridges have been identified at depths of between 140 and 500 m (Figure 87). On the southernmost of these ridges, BGS Borehole BH94/03 sampled a Pleistocene wave-cut terrace deposit, related to a lowstand (Stoker, 1995b). Three ridges south-west of Rockall Island are probably similar wave cut features related to variations in sea level.

Rockall Bank Slope

A scarp slope, at around 500 m depth, has been identified around most of Rockall Bank (Figure 88), C–C’. The height of the scarp diminishes to the north and has little expression west of 14°W. Underneath, and parallel to the scarp, Miocene sediments are exposed at sea bed. This Miocene package has been incised, forming several scarps on the lower slope (Figure 87). There has been little net deposition of sediment on the lower Rockall Slope during the Plio-Pleistocene although there is a thin drape of sediment, predominantly sand and gravel, over the Miocene drifts at the base of the Rockall Slope. A Plio-Pleistocene drift is present in the channel between Rockall and George Bligh banks.

Rockall Trough

The Rockall Trough is an area of deep water separating the UK and Irish Continental Shelf from the Rockall Plateau. In the report area it has a rough north–south trend, extending southwards into the Porcupine Abyssal Plain. Water depths range from less than 1000 m around the margins to over 2000 m in the south of the report area.

George Bligh Bank

George Bligh Bank reaches a minimum water depth of 450 m, with a small terrace forming above 500 m. Its eastern margin slopes relatively steeply to a depth of about 1000 m where a terrace and scarp are present (Figure 88), B–B’. Both the scarp and terrace cut into Miocene sediments (Figure 87). There is little deposition of Plio-Pleistocene sediments on the bank.

Wyville Thomson and Ymir ridges

The Wyville Thomson and Ymir ridges are west-north-west trending topographical highs that form a barrier between the Rockall Trough and the Faroe–Shetland Channel to the north-east (Figure 88), D–D’. The Wyville Thomson Ridge rises to a minimum depth of around 400 m to the north of the report area, whilst the Ymir Ridge rises to 600 m water depth. The topographical highs of both ridges are interpreted to be the result of Palaeogene inversion (e.g. Ritchie et al., 2008). Slope angles are in the region of 2–6°. Around the southern margin of the ridges, contourites form discontinuous mounds and waves.

Seamounts

Rosemary Bank

The Rosemary Bank Seamount summit reaches about 316 m water depth (Figure 88), A–A’, with a diameter of 60–70 km. However, pinnacles on the summit of this seamount, forming groups over 10 km across, have minimum depths of <400 m (Figure 89). It has an areal extent of about 5400 km², and is roughly circular, although a spur towards the south to south-west elongates it in plan view. A defined moat-drift complex surrounds the entire seamount, with sediment wave fields located to the east and west of the seamount (Howe et al., 2006). To the south-west, the moat to summit maximum vertical relief reaches about 2000 m, whilst to the north, with the sea floor being generally shallower, this reduces to about 1200 m. Scarp slopes form on the flanks of Rosemary Bank at water depths of 600–2300 m, reaching angles of 20–40° (Jacobs, 2006). These are most prominent on the south-west flank, and may be enhanced by slope failure.

South-west of Rosemary Bank, Eocene rocks are exposed at, or near to, the sea-bed, implying that there is an active erosional current regime. In other locations around Rosemary Bank, Neogene sediments infill topographical lows, indicating that erosion around this particular seamount has been ongoing since the Oligocene.

Anton Dohrn

The Anton Dohrn Seamount has a summit depth of about 650 m, with a diameter of about 45 km and a vertical relief of approximately 1600 m (Jacobs, 2006). It is roughly circular in plan view, with a shelfbreak occurring at approximately 850 m, where the domed summit falls away in steep scarp walls. These scarps are relatively symmetrical. There is a moat-like depression that surrounds the seamount, being slightly deeper around the north-west flank, reflecting the variation in sea-floor topography around the seamount. A slope apron rises up from this moat to about 1500 m water depth, where the sheer scarp walls begin.

To the south-west of Anton Dohrn is an area where Miocene rocks are exposed. The crest of a broad contourite drift can be seen to the south-west and west of the Anton Dohrn Seamount (Figure 87).

Hebrides Terrace

The Hebrides Terrace Seamount has a summit at approximately 1000 m water depth, with an overall vertical relief of 650–1000 m from the surrounding sea floor. It is elliptical in shape, with a diameter of 27 km by 40 km, and is being partially buried by sediments supplied from the Barra and Donegal fans (Figure 88), H–H’ to the east.

Oceanographic processes

The report area is influenced by the effects of tidal, wind driven and oceanic currents. As changes in modern sea-bed properties are commonly driven by tidal and wave generated shear stresses applied to the sea bed, persistent currents driven by regional changes in temperature and salinity gradients, or gravity driven bulk sediment movement, it is important to consider the locations and strengths of the dominant currents in this area (Figure 90).

On the Hebrides Shelf, wind driven, wave-induced currents affect the sea bed to a depth of over 200 m, with 50 year storms creating currents of over 1 m/s at the shelf break and over 4 m/s on parts of the inner shelf. The presence of asymmetrical mobile bedforms along the shelf indicates that the currents mobilise sediments in a north-westerly direction parallel to the coastline, occasionally smothering and abrading the edge of the Hebrides Shelf. The currents scour sand from adjacent sandy gravels, forming sandy ridges with gravel-rich troughs.

Interaction between long-swell waves and the sea bed, and the resulting friction, causes orbital wave currents to form, particularly over the outer and mid Hebrides Shelf where the sea bed is above the wave base. Maximum wave-orbital current speeds at the sea bed in about 150 m water depth are estimated to exceed 4 m/s on the Hebrides Shelf (Pantin, 1990), whilst in 200 m water depth this decreases to about 2.5 m/s. These figures are an order of magnitude larger than the mean peak spring-tide generated currents at the same locations, and so these orbital currents are capable of mobilising sea-bed sediments at times when the tidal or residual currents are too weak. The orbital motion of these currents tends to flatten and erode the bedforms produced by tidal or residual currents, particularly during storm events.

In shallow water depths and nearshore environments, and on the open ramps of the Hebrides Shelf that face long ocean fetches, there is a positive correlation between coarsening grain sizes, the amount of suspended mud and sand and the orbital wave velocities recorded. However, as most sea-bed sample records are taken during ‘fair weather’, it is not known how much the distribution of sediments and erosion/transport of bedforms changes during poor operational weather periods when storm-generated velocities may be dominant.

On the outer shelf and upper slope an oceanic, north-easterly flowing current, European Slope Current (ESC), comprising part of the North Atlantic Water (NAW), transports relatively warm water from the Atlantic into the Norwegian Sea, whilst the Scottish Coastal Water Current (SCWC) acts on the inner shelf (Figure 90). They are partly wind driven and show some element of diurnal and seasonal variation, although the predominant current flow appears to be north-east with a mean velocity of 0.2 m/s (Souza et al., 2001). On the Barra Fan some current data suggest that maximum current velocities involve downslope movement of water of about 0.25 m/s, not related to NAW flow.

Within the Rockall Trough three water masses are active. The ESC forms the upper part of the water column to a depth of less than 1000 m with flow velocities greatest at a depth of 200 m several km west off the shelfbreak (Souza et al., 2001). Shallow ESC continues north-east across the Wyville Thomson Ridge into the Faroe–Shetland Channel. At depths greater than 500 m the ESC is deflected by the ridge and flows west around the northern margin of the Rockall Trough. A deep northerly flowing current, comprising North Atlantic Deep Water and possibly some Mediterranean deep water, flows along the base of the Hebrides Slope and is then also deflected to the west around the Wyville Thomson and Ymir ridges. Maximum current velocities are in the order of 0.3 m/s. Sheet-like sand drifts at water depths of at least 1000 m indicate that these deep-water residual oceanic currents are strong enough to mobilise sediments without the need for the additional effects of orbital or tidal currents.

From the north, Norwegian Sea Deep Water crosses the Wyville Thomson Ridge and passes west then south down the eastern margin of Rockall Bank. This overflow generates maximum current velocities in excess of 1 m/s in the overflow channel west of the Ymir Ridge, diminishing towards the west and south. It is possible that the deflection of deep currents eastwards south of Rosemary Bank and westwards at the Wyville Thomson Ridge form anti-clockwise gyres within the Rockall Trough.

Over the Rockall Bank tidal currents are less intensive, however wind-driven currents produce high velocities in shallow water depths. Currents coming from the Atlantic flow to the north and north-east with maximum velocities of less than 0.5 m/s. Temperature variation between the shallow waters over the bank and warmer water at equivalent depth over the basin suggest upwelling (Scoffin et al., 1980) but this has not been proven. There is an indication of clockwise flow around Rockall Bank by intermediate (600 m water depth) currents (Figure 90).

Sediment distribution

The distribution patterns of sea-bed gravels, sandy gravels, gravelly sands and carbonate are positively correlated with stress on the sea bed. Currents, enhanced by topographical constrictions including islands, bedrock headlands and elevated submarine platforms, are the primary factor governing the distribution of coarser-grained sediments as opposed to water depth. Whilst much is known about the Hebrides Shelf and eastern margin of the Rockall Bank, there are few data points in the remainder of the study area. This has resulted in a more simplified sediment distribution pattern in the Rockall Basin (Figure 91).

Overview

In general, sand and gravel are common on the shelves, with mud common below 500 m water depth (Figure 91). The submarine shelves and slopes within the report area are being enriched in biogenic carbonate during the present interglacial. The highest values of carbonate are centred around the rock pinnacles, skerries, islands, rock platforms and rock escarpments (e.g. Ferentinos, 1976; Farrow et al., 1984), as a result of both high production rates in these areas and strong tidal and wave-induced longshore drift (Figure 92).

Following the end of the last major glacial period, global sea-level rise of at least 120 m, and subsequent marine inundation of the continental shelves, formed a regional unconformity across the Hebrides Shelf, creating an erosional boundary between underlying glacial sediments and overlying interglacial surficial and sea-bed sediments.

The glacigenic and cold-water late glacial non-glacigenic sediments have been dated to approximately 11 300 to 9560 radiocarbon years old (Jones et al., 1986). They are unconsolidated and typically poorly sorted. Grain sizes range from clay and mud (containing a high proportion of rock flour) to boulder size (based on the Wentworth grain size scale), with the bulk of the sediments being muds, with smaller proportions of gravel and sand. The glacigenic sediments are themselves underlain by sedimentary, metamorphic and igneous consolidated bedrock units.

The present cover of sea-bed sediments forms a thin veneer that is commonly less than 1 m in thickness over the pre-Holocene surface, although in places up to 5 m of Holocene sediments have accumulated in sand waves. The sediments comprise a mixture of biogenic carbonate and siliciclastic material (Scoffin, 1988), encompassing a wide range of sediment grain sizes. The siliciclastic sediments are mostly derived from reworking of glacial materials during the Holocene transgression. Carbonate production has been continuous throughout the Holocene with variations in water depth, substrate and current velocity creating several zones where particular organisms dominate (Scoffin, 1988).

On the Hebrides Slope there is a progressive fining with depth. In depths from 200 to 1000 m, persistent currents, and shelfbreak spillover, deposit sand and muddy sand sea-bed sediments onto muds containing glacial dropstones (Masson et al., 2002). The surficial post-glacial deposits are usually thin (less than 0.5 m) and well sorted. However, thicknesses of up to 4 m have accumulated in locations associated with slope failure, where depressions have been infilled. The sand sheet formed in the south of the report area through this shelfbeak spillover process is called the Barra contourite sand sheet (Stow et al., 2002).

In water depths over 1000 m, the transition between late glacial and modern interglacial sediments is not readily identifiable from physical properties and changes in the sea bed. Instead, the change is identified through climate change indicators contained within fossil assemblages. In general, Holocene sediments form a thin (less than 0.2 m) continuous drape of fine-grained mud-prone sediments. However, strong currents related to the overflow of Norwegian Sea Bottom Water via the Faroe–Shetland Channel have formed a zone of coarse-grained sediments in the north and west Rockall Basin, with little active deposition occurring for much of the Neogene.

In the eastern Rockall Trough, the basin floor is dominated by sheeted contourite drifts (Howe et al., 2002; Masson et al., 2002). A broad north-west-trending drift and moat complex is found in the north-eastern Rockall Trough (refer to Figure 87), with the moat reaching up to 100 m in depth (Masson et al., 2002), and relatively steep (up to 5°) flanks. At the base of the Hebrides Slope, this complex is overlain by the Sula Sgeir Fan topography. To the south-east of the report area, geostrophic currents have resulted in the deposition of an extensive sediment drift called the Feni Drift (Figure 63), containing Neogene sediment waves (Masson and Kidd, 1986).

Mud

On the Hebrides Shelf mud-prone sediments are generally uncommon. However, muddy sands are found within the St Kilda Basin and in smaller depocentres south-west of Barra. Mud becomes a significant component of the sea-bed sediment on the Hebrides Slope to the south of the report area, at a water depth of approximately 500 m, where it underlies remobilised sand sheets. The majority of the muds on the floor of the Rockall Trough are indistinguishable from the underlying glacial sediments. It is probable that the rate of deposition in these basinal areas, away from deposition by slides and mass failures, is low at present. On both Rockall and George Bligh banks' mean sediment grain size decreases with increasing depth, with mud being a minor component of the sediments on the flanks of both banks. Above 500 m gravelly muddy sand is common, the majority of which is biogenic in origin (e.g. BGS core 58-14/28). On Rockall Bank itself there is no significant mud in the sea-bed sediments. Recordings of low current velocities from the western side of the bank suggest that muddy sand might be a component of the sediment on the margin of the Hatton–Rockall Basin (e.g. BGS core 57-16/17).

Sand

Sand-prone sediment is common across much of the Hebrides Shelf and is the dominant sediment type north of Lewis, west of the Uists and south-west of Barra (Figure 91). To the south-west of Barra, the accumulations of sand are interpreted to have been reworked from the raised outer shelf (Ferentinos, 1976). The St Kilda Basin, south of the St Kilda archipelago, contains an accumulation of sand and, in the basin centre, muddy sand. This sand accumulation is commonly under 0.5 m in thickness and overlies a carbonate gravel lag deposit that has given radiocarbon ages of 11.3 to 9.56 ka BP (Jones et al., 1986). In other lows on either side of the Flannan Ridge and west of Sula Sgeir the sediment distribution shows zonation, with increasingly well-sorted sands in the centre of the basins.

On the Hebrides Shelf sand occurs as isolated patches and streaks overlying gravel-rich lag deposits (Figure 91) and (Figure 93), and infilling depressions such as iceberg ploughmarks. In some instances the sands can be seen to form ridges aligned parallel to the oceanic currents with gravel lags exposed between ridges. This process of reworking sands on the outer shelf has been termed ‘sea-floor polishing’ (Viana et al., 1994, 1998; Stow et al., 2002).

Sand is a common component of the Hebrides Slope, in part being supplied by shelf spillover (Stow and Mayall, 2000). However, oceanic bottom currents that flow to the north are also important factors in the remobilising of sandy sediments along the Hebrides Shelf and Slope. To the south of the report area lies the Barra contourite sand sheet (Stow et al., 2002), extending from a depth of between 170 and 300 m on the outer shelf and upper slope to between 1200 and 1500 m near the base of slope, and overlying muddy sediments. This has formed over the mid slope region of the Barra Fan, as bottom currents smooth the irregular slope topography created by previous episodes of debris flows and slumping (Armishaw et al., 2000). The thickness of the layer appears to be in part controlled by underlying topography, with sand infilling depressions caused by earlier sediment failure. The sediment distribution data suggest that this sand sheet extends to the north, being present on the mid to lower slope in the north-east Rockall Trough. There are no data in the northern Rockall Trough to further constrain the extent of this sand sheet.

In the north-eastern Rockall Trough, muddy sands are found in water depths over 1000 m. This zone of muddy sand is interpreted to continue westwards across the northern Rockall Trough. There is also a sheet of sandy sediment that extends eastwards from the base of the Rockall Bank, into the Rockall Trough towards Anton Dohrn Seamount, possibly suggesting some element of an anticlockwise gyre in the northern Rockall Trough.

Gravelly sands are present on the summits of seamounts within the Rockall Trough, and on the crests of the Wyville Thomson and Ymir ridges. They are assumed to be a thin layer reworked by storm and deep oceanic currents. However, this is based on limited sample data. Sand-prone sediments have also been recovered from the south-western margins of both Anton Dohrn and Rosemary Bank seamounts, near the base of slope. This might reflect some form of current activity and winnowing related to current movements around the topographical highs, (the ‘moats’ south-west of these seamounts also indicate some element of scouring), but the lack of data points make this interpretation speculative.

Gravel

On the Hebrides Shelf, large areas of sea-bed sediments north-west of Lewis and west of South Uist and Barra contain more than 30 per cent gravel (Figure 91). The gravel is most common on topographical highs and represents sediment that has not been redistributed by currents. These gravels contain both lithic clasts derived from the local bedrock, encompassing igneous, metamorphic and sedimentary rock types, and bioclastic fragments. To the west of Barra, lithic gravels occur on the summits and leeward margins of banks on the outer shelf, and on shoreward facing slopes on the inner shelf (Ferentinos, 1976). South-west of Barra, compacted cobble beds and cobble ridges up to 0.2 m in height have formed (Figure 94)a, possibly in response to the early Holocene transgression of the shelf and winnowing of less consolidated sediments.

On the Hebrides Slope gravel becomes less abundant with depth, but remains a component of the sea-bed sediment to depths of about 1000 m. It is known that gravel-prone sediments are uncommon on the basin floor, with muds and silts dominating (e.g. BGS cores 58-12/5, 58-11/3 and 58-13/19). However, on the western margin of the Rockall Trough, gravel-prone sediment is common at the base of the eastern slope of the Rockall Bank (e.g. BGS core 57-13/53), associated with the deep current moving southwards (Figure 90). Current velocity data suggest that this area of coarse-grained sediment could extend across the north-western Rockall Trough margin south of Lousy and Bill Bailey’s banks. However, samples from this area suggest a complex pattern of sea-bed sediments including muds, sands, lithic fragments and cobbles in addition to gravel. It is possible that the gravel component of the sea-bed sediments represents a relict lag deposit over which an active sand and mud layer is being mobilised by the current.

On Rockall Bank coarse-grained sediment is most common in shallow water (Roberts, 1975b). Around Rockall Island the sediment is sandy gravel and comprises almost entirely biogenic carbonate. Limited sample data indicates that slightly gravelly sediment is present on the plateaux of the seamounts. Scattered sample sites showing gravel as a component of basinal sea-bed sediments are probably sampling pre-Holocene glacigenic sediments containing ice-rafted debris, in areas where present-day deposition rates are low.

Bedrock

A significant part of the Inner Hebrides Shelf comprises uncovered bedrock (Figure 91), most of which is Precambrian Lewisianoid metamorphic basement (e.g. BGS rock drill cores 57-09/537 and 56-08/920). Basement rocks are also exposed at the sea bed around the Flannan Isles, Sula Sgeir and west of the Outer Hebrides. In the surrounding sedimentary basins the softer rocks have been eroded, with the topographical lows being later partially infilled by glacial deposits. The basement and basin topography observed on the present-day sea bed therefore reflects the underlying structural trends.

Cenozoic extrusive igneous rocks are exposed around the islands of St Kilda and Rockall Bank (e.g. BGS core 57-13/54). Late Cretaceous to Eocene igneous rocks, predominantly transitional to alkali basalts, and sedimentary rocks containing igneous clasts, have been recovered from the flanks of the three seamounts in this report area (O’Connor et al., 2000). The top of Anton Dohrn Seamount shows a poorly sorted conglomerate composed of igneous rocks and carbonate clasts overlain by a terrigenous/carbonate sandy mud (BGS core 57-12/18), whilst core 59-11/12, taken from the top of Rosemary Bank, shows five separate extrusive lava flows, with thin bioclastic limestone interbeds.

To the north of the study area, Eocene tuffs were recovered from the Ymir Ridge (Jones and Ramsay, 1982), and it is believed that Palaeogene igneous rocks are exposed on both the Ymir and Wyville Thomson ridges. In the east Rockall Basin, Lower Oligocene chalk has been sampled from water depths of around 700 m on the Geike Escarpment (Jones et al., 1986). In the north-west Rockall Trough, the margin of the Plio-Pleistocene drift occurs in deep water east of the Rockall and George Bligh Banks (Figure 87). To the west of this margin, vigorous current activity means that Miocene rocks are near to the surface, with only a veneer of sand and gravel (Howe et al., 2001; BGS core 58-14/54), or completely exposed.

Carbonate

From the start of the present interglacial the main sedimentary source has come from biogenic carbonate, as the Hebrides Shelf has become increasingly starved of terrestrial input. Regional investigations have shown that the large areas of bedrock exposure and stoney reefs with gravel are the major sources of the pure biogenic carbonate supplied to the Hebrides Shelf. The largest such area is a platform of mixed bedrock and gravel to the west of the Outer Hebrides (Figure 92). The Hebrides Shelf can therefore be termed a temperate carbonate shelf.

Mollusc fragments, predominantly from bivalves, comprise 30–55 per cent of the total carbonate in sea-bed sediments on the open shelves (Wilson, 1979a). Over the remainder of the Hebrides Shelf, carbonate grains are derived from foraminifera, calcareous algae, echinoderms, bryozoans, serpulids and scaphopods (Wilson 1979a; Scoffin, 1988). Radiocarbon ages derived from bulk samples give ages between 8000 and 2800 years, suggesting an active, constant generation and gradual build up of carbonate on the shelf during the Holocene.

The sea-bed sediments on the Rockall Bank show more extensive carbonate concentrations, and greater biological species diversity than environments on the Hebrides Shelf and slope at similar water depths (Narayanaswamy et al., 2005). The faunal assemblage comprise molluscs, bryozoans and serpulids, most commonly in shallower water, and benthic foraminifera are common below 120 m (Scoffin et al., 1980). Patches of the coral Lophelia pertusa up to 50 m in diameter are also common (Wilson, 1979b). It is suggested that this pattern occurs as a result of reduced smothering from (glacigenic) terrigenous silts and sands on the Rockall Bank. Over the shallow areas of the bank, the carbonate percentage is over 80 per cent, showing biogenic zonation dependant on water depth and substrate (Scoffin et al., 1980).

The Rockall Escarpment, situated on the east flank of the Rockall Bank, is part of a continuous rock-reef, carbonate-reef, moat- and carbonate-rich elongated drift that fringes the Rockall Bank, lying in water depths of between 300 and 600 m. This zone is beyond the modern photic zone, but above the wave base during periods of sea-level fall in glacial times. What is unusual about this fringe is that it is physically decoupled from the carbonate-rich sediments centred in shallower waters on the Rockall Bank (Figure 92). However, the continuity of this fringing carbonate reef system around the Rockall Bank is unknown. A less extensive bedrock and carbonate reef occurs beneath about 1000 m water depth. However, BGS borehole BH94/04 shows the sediments in this area still have up to 98 per cent carbonate composition in the bioclastic limestone retrieved.

Much of the carbonate within the sand fraction on the Anton Dohrn and Rosemary Bank seamounts is composed of planktonic foraminiferal tests, whilst calcareous foraminifera provide a significant proportion of the biogenic carbonate found in the deeper waters on the outer Hebrides shelf and slope. An area of slope sediments with more than 60 per cent carbonate lies in the sea bight at the junction of the Hebrides Shelf with the Wyville Thomson Ridge. It is suggested that the high values of biogenic carbonate in this region originate from a mixed carbonate reef and bedrock reef system (Figure 94) that has been protected from intensive fishing by the rugged sea-bed topography. Its elevated position has also prevented the reefs from being smothered by mobile sediments.

Sedimentary bedforms

There is a wide range of sedimentary bedform types within the report area (Figure 95). Mobile bedforms, such as ripples and sandwave fields, correspond to locations where sediment transport and re-working is enhanced by currents and topographical constraints. Gravity-driven bedforms, by their nature, are intrinsically linked into sea-bed failures on moderate to steep slopes, and are commonly located upslope from elongated drifts. These include the large areas of scarps formed around the seamounts, the Wyville Thomson Ridge and the deep water slopes on the east margin of the Rockall Bank. Unstable areas also exist on the Hebrides Shelf where bedrock was shaped into scarps and basins by glacial movement. This process can transport modern sea-bed sediments, characteristic of shallow-water environments, into deeper waters, forming bed forms that appear unusual in their location and composition. Static bed forms, such as pockmarks, iceberg ploughmarks and glacigenic troughs, underpin the variations in sea-bed geomorphology and grain size, enabling biological communities to establish themselves.

Ripples, sandwaves and drift deposits

Ripples are a common form of mobile bed form in the survey area, being present on the shelves, banks and margins of the Rockall Trough and Hebrides Shelf (Figure 93). On the outer shelf the sand bodies show a range of ripple geometries, reflecting the interaction of several current directions. The predominant orientation of sand bodies and ribbons parallel the bathymetric contours. North–south oriented linear sand features have amplitudes of up to 0.2 m and can be seen to either overlie a gravel substrate, or have gravel-rich troughs (Figure 94)b. Current velocities are sufficient to form scoured depressions around cobbles at the sea bed.

Larger-scale sand waves observed near the Outer Hebrides (Figure 95) commonly overlie bedrock, and lie in close proximity to significant geomorphological barriers. The crestlines of the bed forms are aligned transverse to the direction of travel of the near-bed currents (Figure 90), with the steepest slopes facing the direction of bedload transport (Figure 93).

Sand streaks, ribbons and ridges are common on the Hebrides Shelf, ranging from small features several metres in width to large wave forms tens of metres across. Most long axes are oriented parallel both to the inshore tidal currents and to oceanic currents on the outer shelf (Figure 93). Offshore St Kilda, however, sand ribbons form a concentric pattern around the archipelago (Perry, 1987), whilst on the slope north-west of St Kilda, the sands form ribbons up to 1500 m in length, that have an anastomosing pattern (Kenyon, 1986), orientated parallel to the dominant tidal currents that flow north to north-easterly across the shelf (Kenyon, 1986) (Figure 90).

A lack of significant modern sedimentary inputs from the mainland and islands onto the Hebrides Shelf prevents the formation of more widespread large-scale bedforms on the shelf. What input there is, is either deposited in the sinks provided by the fjords and glacigenic troughs, or is swept out of the North Channel into the Irish Sea (Holmes and Tappin, 2005). In addition, geomorphological barriers prevent significant along-shelf transport, whilst orbital wave currents prevent the growth of large bedforms on the open shelf. Sediment waves on the Barra Fan have formed through a combination of diffuse persistent currents and weak turbidity currents. Several areas of broad sediment waves have also been observed in the north-eastern Rockall Trough, adjacent to the Sula Sgeir Fan, formed through the deposition of suspended muds in hemipelagic environments (Figure 95). They have wavelengths of up to 2 km and a maximum amplitude of 20 m and are covered by a sand sheet. It is estimated that they laterally migrate at 0.4–0.9 m per thousand years (Masson et al., 2002).

In the basinal parts of the Rockall Trough several other areas of broad bedform features have been identified (Figure 95), showing symmetrical waves with amplitudes of metres and wavelengths of up to 1 km (Long et al., 2001). Some exhibit up-slope migration, whilst north-west of Anton Dohrn, an irregular sea-bed topography with poorly developed waveforms suggests modern draping of an earlier buried wavefield. It is uncertain whether the bedforms are presently active or relict features, although high resolution seismic profiles suggest that there has been some Holocene development of waves in the eastern Rockall Trough (Howe, 1996).

On the Hebrides Slope contourite sands are also a common component of the sea bed, showing elongation parallel to the bathymetric contours. The contourite sediments have a broad ridge and moat topography in the north-eastern Rockall Trough (Figure 96) whilst elongate sediment waves with wavelengths up to 2 km parallel the slope contours downslope from the Geikie Escarpment (Figure 95). These sands are also rippled with complex patterns of lunate ripples similar to those seen on shelf sand bodies (Figure 94)c. Sand waves with an amplitude of 1–2 m and wavelengths up to 8 m have also been reported from the upper Hebrides Slope (Eden et al., 1971).

Parallel-bedded, flat-lying to undulatory sediment-drift deposits dominate the Rockall Trough. These vary from broad-domed, sheeted drifts in the basin, to single- and multicrested elongate mounded drifts on the basin flanks (Stoker et al., 1998; 2001). Much of the basin fill comprises seismically parallel laminated, fine-grained sediments that are termed a sheeted contourite drift (Howe et al., 1994, 2002). The grain-size composition of these drifts reflects the sedimentology of the surrounding area. Thus the drifts near the Rockall escarpment are rich in biogenic carbonate whilst those against the Hebrides Slope are predominantly re-worked glacigenic sediments.

Sediment data suggest that the crests and flanks of the seamounts are areas of increased current velocity. However, recent high-resolution multibeam data shows relatively smooth summits, with evidence of dropstones and trawl-marks (Anton Dohrn), but few sedimentary bedforms (Jacobs, 2006).

The depressions around Rosemary Bank and exposure of Eocene rock on the flanks of all the seamounts implies an active, erosional current regime. On the slope of Rockall and George Bligh banks the seabed sediment distribution suggests that there is active sediment movement, with strong geostrophic currents forming erosive features (Jacobs, 2006), preventing the formation of sedimentary bedforms. Sediment waves with crests parallel to contours underlie the steep flank on the eastern margin of Rockall Bank. Condensed, muddy, sandy and gravel-lag contourites occur on the slopes of Rockall and George Bligh banks (Howe et al., 2001). Coarse sands, gravel to boulder sized clasts and broken shell fragments cap the seamounts (Jacobs, 2006; Stoker et al., 1993), indicative of a high energy environment with few bed form features.

Debris flows and mass wasting

The shelfbreak on the Hebrides Shelf is sharply defined, occurring in water depths of between 150 and 250 m. The debris flows spilling over the shelfbreak during the last 4.7 Ma have formed the Sula Sgeir and Barra Fans (Figure 95), burying the bedrock and partly obliterating modern sea-bed bed forms forming through persistent currents. Therefore modern sediments deposited on the fans and shelf slopes are commonly unregulated by the underlying bedrock geology, with the fans themselves now forming large static bed forms that interact with the current regime.

The potential for sea-bed failures is correlated with slope steepness and sea-bed/sub-sea-bed geotechnical properties of the sediments. Ground acceleration from earthquakes can destabilise unconsolidated sediments, causing unpredictable (timing) sea-bed failures. Large-scale variations in the sea-bed topography of these landslides occur between their head-wall scarp and their toe. Bed forms in the compression zone at the toe of a slide range from small ridges to large outward facing scarps that can be over 10 m higher than the surrounding sea bed and laterally extend up to 20 km. Where toe scarps are large enough to interact with persistent currents, moats and elongated sediment waves can also be found (Knutz et al., 2002a, b). On the Sula Sgeir Fan, translated sediments forming mid-slope fans have been cut by a more recent slide head-wall scar that extends laterally for over 20 km (Evans et al., 2005).

The Barra Fan deposits have been cut by at least five erosive events, which indicate several large slides have occurred, displacing 1830 km3 of material (Holmes et al., 1998). These have been termed the Peach Slide Complex. The Sula Sgeir Fan also shows evidence of three significant bottleneck slides (Kenyon, 1987), where the maximum slide scar width is upslope, narrowing downslope.

Glacigenic troughs

Periodic glaciations have shaped the complex geomorphology of the western coast of Scotland. Large enclosed glacigenic troughs remain on the Hebrides Shelf, aligned with the younger and softer bedrock (Figure 95). The pinnacles, skerries and platforms standing out from the surrounding sea bed are igneous and/or metamorphic bedrock that was more resistant to glacigenic erosion.

Iceberg ploughmarks

These have been documented on the outer Hebrides Shelf and upper slope, prevalent in 200–500 m water depth, and absent in over 700 m water depth (Figure 95). The ploughmarks can reach 2 km in length, 200 m in width and can be up to 25 m in depth. The density of ploughmarks decreases downslope to the west, from total coverage and merging of ploughmarks on the eastern shelf, to individual ploughmarks being separated by up to 500 m at the western limit of these features. They commonly display a downslope turn or pit, created when the icebergs pivoted around their anchor points. Near-bed Holocene currents have winnowed the berms formed by the iceberg ploughmarks, resulting in complex patterns of gravel and exposed boulders along ridges. Sediment was pushed aside by the movement of the icebergs, whilst fine-grained sands and muds accumulated within the troughs.

On Rockall Bank, iceberg ploughmarks occur in water depths between about 100 and 500 m (Holmes et al., 2006; Jacobs, 2006). They are also present on the summit of George Bligh Bank to a water depth of 500 m (Jacobs, 2006). Iceberg scour is the only glacigenic static bedform common to both the Rockall Bank, George Bligh Bank and the Hebrides Shelf.

Turbidite channels

Turbidite channels extend downslope from the shelfbreak of the Hebrides Shelf (Figure 95). They appear to have formed in areas where the slopes were not covered by glacigenic debris flows, due to the landward presence of shelf moraines trapping material. They formed during the last glacial period, transporting debris downslope through gravity driven turbidity currents (Knutz, 1999). However, they became more or less inactive at the start of the modern interglacial period.

In the uppermost reaches, the morphology of these channels is partially obscured by iceberg ploughmarks. In the middle reaches, they are over 50 m deep from the shoulder to axis. On the lower slopes, the channels have interacted with large landslide deposits, forming large sediment waves, composed of turbidites, drifted and hemipelagic sediments (Knutz, 1999). Levees have formed as a result of overspill from the channel axes. Below about 700 m water depth, the levees become asymmetrical, becoming larger on one side of the channel, forming subtle sediment waves that run subparallel to the channel axis.

Pockmarks

Pockmarks have been observed south of the Wyville Thomson Ridge, in the north-east Rockall Trough (Figure 95). They occur in muddy sands and sandy mud, and are between 0.5 and 2 m in depth from the shoulder to the centre, and have been documented occurring in fields where densities range from 10 to 50 per km² (UKOOA, 2000; Fugro-Geoteam, 1997). It is not known if this density range is resolution or location dependant. They are usually 50 m or less in diameter, commonly marked by hyperbolae, and occur in approximately 1000–1100 m water depth, bounded at the upper limit by the Darwin Mounds. The centre of some of these features contains a boulder over 0.3 m diameter (Tim Smith, Conoco, pers comm.). Photographic evidence indicates these are dropstones originating from icebergs (Tappin et al., 2001). Sidescan survey records (BGS survey 1985/7 lines11–15) indicate that the majority of the pockmarks are currently inactive, showing no evidence of fluid seepage (Long, 1992). Therefore it is surmised that current sea-bed scour around the boulders is causing the pockmarks to remain open and not infill.

Isolated pockmarks have been identified on 3D seismic images at other locations in the north-east Rockall Trough, and it is thought possible that other pockmark fields may exist in slope and basin areas where high resolution surveys have yet to be undertaken.

Darwin mounds

The Darwin mounds are a series of mounds occurring in about 900 to 1000 m water depth, subparallel to the trend of the Wyville Thomson Ridge. They have been identified as biological communities containing cold water corals, with surficial sediments being silica sand with minor proportions of coral carbonate (Holmes et al., 1998). It has been suggested that expulsion of sand, entrained in buoyant fluids or gases may have initially created these features (Masson et al., 2003). The close proximity of these mounds to the pockmarks immediately downslope, combined with evidence of shallow gas bearing facies from sub-sea-bed acoustic returns (Baltzer et al., 1998), support this hypothesis.

Chapter 12 Geohazards

By Dave Long

A geohazard is a geological feature or event that involves a degree of risk of detriment to life, property or the environment. For the purpose of this chapter they are primarily located at, or close to, sea bed and would need to be evaluated by anyone proposing to use the sea bed for any activity. The details presented result from a wide range of studies, both regional and site specific. The distribution of data is not even but is focused primarily in areas of oil industry interest and areas considered for habitat protection measures. A range of industry- and government-funded surveys have been undertaken and their integration allows better understanding of the features identified. For the purpose of discussion, geohazards within the report area are discussed in terms of shelf, slope or basinal settings and also in relation to seismicity.

Shelf

Carbonate sand waves

The sea-floor sediments on the shelf west of the Hebrides are carbonate rich (Allen, 1983). They commonly occur as migrating sand waves and therefore are a hazard to constructions on the sea bed through burial and scour of surface and buried infrastructure. The bed forms migrate in a clockwise direction on the UK shelf moving from west of Ireland, where much carbonate is produced, northwards and eastwards to the Orkneys and exiting into the North Sea. The rate of migration is poorly known. The largest sediment bodies are near the shelfbreak west of Barra but the potentially most active are located where near sea-bed currents are strongest.

Bare rock platforms

Exposed rock creates constraints to pipeline and cable routing, as such infrastructure cannot be buried. Rock platforms also create problems for anchoring and ‘spudding-in’. Pipelines may require to be covered with sediments to protect them from trawling activity. Sharp changes in topography also create spanning and flexing points for pipelines and cables, with abrasion a concern, particularly for the latter. Spans also create vulnerable locations for snagging by trawls.

Extensive bare rock platforms have been mapped west of the Western Isles extending to about 120 m water depth. They also occur around the Flannan Isles but are less extensive around St Kilda (Figure 97). The platforms west of the Hebrides and the Flannan Isles are typically crossed by deep crevices where jointing in the Lewisian bedrock has been eroded. The crevices commonly contain boulders and produce difficult terrain for the laying of pipelines or cables. Around St Kilda, the topography is uneven and there is much bare rock. A small area of uneven rock exposure occurs in the south-east corner of the report area as part of Stanton Bank. Although there is a small area of rock exposure on Rockall Bank around Rockall islet, gravels predominate over the bank.

Morainic deposits

During the glacial periods, ice-sheets flowed across the shelf and locally extended beyond the shelfbreak, most notably at the Barra and Sula Sgeir fans (see Cenozoic sedimentary rocks chapter). They modified the landscape and deposited glacial material. In some instances very well-compacted glacial material is close to sea bed, commonly in the form of moraines or drumlins. Both features can create difficult foundation conditions. The variable and commonly coarse-grained lithology of these mounds, together with their uneven surface and occasional buried boulders, can cause problems for drilling and pile installation, and make estimates of shear strength profiles uncertain.

Moraines are most notable on the outer Hebrides shelf where prominent ridges have been mapped parallel to the shelfbreak west of St Kilda (Figure 97). On acoustic profiles, the ridges appear structureless to chaotic with a few hyperbolic reflectors typical of buried boulders. Limited geotechnical data exists (Selby, 1989) but the diamictons are described as firm to very stiff, and the cores contain pebbles and reworked shells. In contrast, commonly between these ridges on the outer shelf, are small basins infilled with very soft sediments overlying the morainic material.

Shallow gas

The St Kilda Basin (Figure 84) is a bathymetric depression (about 150 m water depth) south of St Kilda partially infilled with very soft sediments of late postglacial age (Peacock et al., 1992). High-resolution seismic records sometimes show gas blanking. It is thought that the gas is derived in situ and assumed to be biogenic methane.

Slope features

Iceberg ploughmarks

Iceberg ploughmarks have been extensively mapped on the shelf edge. They also extend downslope to about 500 m water depth. Although appearing chaotic there is a tendency for them to be aligned along bathymetric contours. They are usually 0.5–1.5 m deep and vary between 20 and 100 m wide. There are occasional pits where a berg sat or turned creating a wide area of disturbance. Either side of the ploughmarks are ridges of excavated material, 3–4 m high, known as berms. These have usually been winnowed exposing boulders on their crests; the winnowed material forms the soft clays infilling the hollows thereby creating an area of contrasting physical properties across a ploughmark.

In addition as the iceberg sat on the sea bed the pre-existing sediments in some places were heavily loaded causing them to become overconsolidated.

Submarine landslides

At the southern end of the Hebrides margin there is extensive evidence for large scale translational events within the Barra Fan, a site of extensive sedimentation throughout the Pleistocene (Armishaw et al., 1998, 2000). For the slope area north of the Hebrides Terrace Seamount, there is evidence of up to five slide events, which together are known as the Peach Slide Complex (Figure 97) (Holmes et al., 1998). South of the Hebrides Terrace Seamount the slope is known as the Donegal Fan, most of which is south of the UK/Ireland median line and contains evidence of many slides including small events thought to be recent (Holmes et al., 2001). This has also been referred to as the Foreland Slide Complex (Figure 76) (Holmes, 2003).

On the Barra Fan, the oldest large-scale submarine landslide in the multistage Peach Slide Complex is termed Event 1. This eroded deeply into the underlying sediments and locally into the early Pliocene (C10) unconformity, but is otherwise undated (Holmes et al., 1998). Event 2 is also undated. Event 3 has been dated at approximately 21 to 18 ka 14C BP, thus demonstrating that at least one large-scale submarine landslide occurred during a period of rapid glacigenic sedimentation (Knutz et al., 2002a). Event 4 truncates glacigenic debris-flow deposits but is overlain by channel and levee systems that connected to a turbidite system, which effectively closed down after approximately 14 ka ¹⁴C BP (Knutz et al., 2002b). As Event 4 is also covered by a thick sequence of hemipelagic sediments, it probably occurred during the retreat stages of the Devensian glaciation. The fifth event is very small and considered to be late glacial in age.

The erosion hollow left behind after the slide events becomes a site of preferential deposition of contourite deposits (Armishaw et al., 2000). Geotechnical analysis of samples within the failure area indicates that the shallowest samples still have factors of safety just above (Hobbs and Holmes, 2002).

A smaller area with strong evidence for slope failure occurs in the north-east Rockall Trough as the Sula Sgeir Fan. There is clear morphological evidence for sliding on both its south-western and north-eastern flanks (Baltzer et al., 1998). Debris-flow translation chutes extend from just below the shelfbreak to almost 1000 m water depth. These are the so-called bottleneck slides of Kenyon (1987) that both pre- and post-date near sea-bed packages of glacigenic debris-flow deposits. The translated sediments from the bottleneck slides form midslope fans, part of which are truncated by the head-wall scar of a larger slide, over 20 km wide, which itself is partly draped by glacigenic debris-flow deposits. Therefore, a pre- or early Devensian age is proposed for the event (Evans et al., 2005). The immediate area shows sub-seabed acoustic scatter attributable to fluid ascent, possibly gas (Baltzer et al., 1998) and the areas around the slides include sea-bed pockmarks (Masson, 2001).

Buried slides are also evident on the Hebridean margin. The most notable feature is associated with the Geikie Escarpment. Strachan and Evans (1991) described a buried sediment failure to the south of the Sula Sgeir Fan extending 4.1 km above the main Geikie Escarpment. The failure is of uncertain lateral extent (but probably less than 10 km width) as it has only been identified on a few seismic profiles. It occurs at a point where the slope is about 1.58°, which is shallow relative to other slopes of the escarpment zone. The failed sediment, up to 43 m thick, appears featureless with depressions on its upper surface that may represent dewatering features. It is unclear how much sediment has moved or has densified in situ. However the vertical reduction of about 15 m suggests at least some sediment is likely to have moved downslope, probably to be deposited at the foot of the Geikie Escarpment.

The failure is covered by up to 40 m of undisturbed, acoustically well-bedded sediment, and the glide plane is formed of acoustically similar sediments (Strachan and Evans, 1991, fig. 1). Although direct seismic correlation with the dated section in the nearby BGS borehole BH88/07A (Stoker et al., 1994) is not possible due to attenuation of the sediments at the escarpment, a tentative early Devensian (Weichselian) age can be suggested by correlation of the reflector characteristics of the sediments.

This range of downslope movements on the Hebridean slope provides an explanation for the variation in strength of the sediments. Shallow cores reveal underconsolidated to overconsolidated sediments (Paul et al., 1998).

The western margin of the Rockall Trough also shows a wide range of evidence for slope failures. One of the earliest studies (Flood et al., 1979) suggested a large scale event at the south-eastern end of the UK part of Rockall Bank. This slide measures 100 km wide and extends 160 km into Irish waters to water depths in excess of 2000 m where it was recognised initially on GLORIA sidescan data (Unnithan et al., 2001) and subsequently on the multibeam data of the Irish National Seabed Survey (Elliott et al., 2009). The more recent studies show that this feature extends over an area of more than 6000 km² within Irish waters. This has been termed the Rockall Bank Mass Failure (Elliott et al., 2009). It comprises a series of failures represented by scarps and debris lobes. Individual lobes extend up to 130 km on to the floor of the Rockall Trough. Flood et al. (1979) dated the slumping as occurring at approximately 15–16 ka, which is in broad agreement with the 14C dates of 10 to 21 ka BP recorded by Øvrebø et al. (2005) and coincides with the end of the last glacial maximum. Elliott et al. (2010) consider the instability of the eastern Rockall Bank margin to be related to fluctuations of current flow along the margin associated with development of the Feni Drift, both through the deposition and erosion of the sediment pile. Recent detailed surveys suggest a more complex situation along the eastern margin of Rockall Bank with many small events below 600 m water depth (Jacobs, 2006). The most detailed surveys show slides ranging from a few hundred metres across to several kilometres. Some show a series of steps about 10 m high consistent with retrogressive failure (Long et al., 2010). The age of these events is unclear except where they intercept the sediment drift implying an age within the last million years.

The seamounts within the Rockall Trough are also sites of slope failure. Detailed mapping of Rosemary Bank has revealed three slide scars at the south-west corner, each about 5 km across (Howe et al., 2006), and where a debris flow up to 100 ms thick has been recognised within the adjacent Mid Miocene to Early Pliocene sequence extending at least 15 km from the slide scar. Multibeam imaging of Anton Dohrn shows a slope failure interpreted as a rockfall on its north-west flank. It is considered a rockfall because of the steep slope of its source area and the hummocky terrain of the deposit on the lower slope, confirmed by video (Stewart et al., 2009).

Rock outcrop

Recent video surveys have confirmed that there is abundant rock exposure on the slopes of the seamounts and the eastern flank of Rockall Bank (Figure 97). Here slopes up to 60° have been measured and comprise mainly igneous rock types (e.g. Howe et al., 2006), but Cenozoic sediments, probably Eocene, also crop out (Long et al., 2010). It has been suggested that Cenozoic sediments also crop out along the Geikie Escarpment on the Hebridean margin (Jones et al., 1986).

Basin-floor features

Pockmarks

An extensive area in excess of 1000 km² in the north-east Rockall Trough, towards the base of the south-eastern flank of the Ymir Ridge, has a muddy sea bed and shows circular to ellipsoidal sea-bed craters up to 300 m or more diameter and up to 1 m or more deep below the surrounding sea bed. These features have been mapped from TOBI sidescan sonar data (Masson et al., 2003) and from the sea-bed pick from 3D seismic volumes (Long et al., 2001; Leslie et al., 2002), and are interpreted as pockmarks (Figure 97). The spots of high sonar backscatter recorded from the centre of many of the pockmarks are proven as boulders (about 0.5 m) and other gravel (Holmes et al., 2001). The pockmarks probably originate from fluid expulsion and sea-bed sediment excavation, thus the boulders and sea-bed gravel observed in the pockmarks are probably relict from excavation of glacigenic sea-bed and shallow sediments by escaping fluids. The pockmarks at sea bed extend to the positive topography formed at the periphery of the Sula Sgeir Fan and to the structural trend of the Ymir Ridge cropping at or near sea bed. The possibility is that some of the pockmarks may originate from the ascent of gas and other light hydrocarbons. Alternatively, the juxtaposition of the pockmarks with the positive topography is consistent with their having a possible origin from pore fluids. These may be presumed to have been expelled at the sea bed as a result of formation compaction over the basement high formed by the Ymir Ridge, neotectonics associated with the basement or sub-sea-bed faulting, or by sediment compaction associated with the Sula Sgeir Fan.

Darwin mounds

Downslope of the pockmarks on the floor of the north-east Rockall Trough are small mounds of well-sorted sand (Figure 95). The elevated position atop of these mounds is exploited by the cold water coral Lophelia pertusa. These mounds have been attributed to the fluid escape in a similar manner to the adjacent pockmarks (Masson et al., 2003). It would seem that the fluid escape is not a controlling factor in the distribution of the coral rather the elevated position afforded by the mounds. The distribution of mounds and pockmarks suggests a gradual transition from mounds in the north to pockmarks in the south. This, combined with the lack of bioclastic material in the mound sediments, suggests that both mounds and pockmarks are created by fluid escape from below the sea floor. Mounds occur where fluids carry subsurface sand to the surface, where it forms mounds because bottom currents are not strong enough to disperse it. Pockmarks form where muddy material is eroded by fluid escape but dispersed by bottom currents. These mounds are now protected by a Special Area of Conservation (SAC) order but previously the coral growths had been badly damaged by trawling activity (Roberts et al., 2006).

Methane hydrates

Methane hydrates are a type of clathrate in which a three-dimensional framework of water molecules is stabilised by molecules of methane held in the centre of molecular cages. Methane hydrates are only stable at low temperatures and/or high pressures. Assuming a gas source is present, the vertical and lateral extent of any particular accumulation is dependent on the interaction between the controlling factors of bottom water temperature, pressure and geothermal gradient. The base of accumulated hydrate may be recognised on seismic profiles as a reflector that tends to mimic the sea bed topography. This is known as a bottom simulating reflector (BSR) and it may cross-cut stratigraphical reflectors. However, BSRs only occur where there is free gas trapped beneath the hydrate. It should be noted that hydrate can occur within sediments without any BSR being evident on seismic profiles.

Accumulations of hydrate may cover hundreds of square kilometres, and be several hundred metres thick. Hydrates could become an important resource in themselves, or may act as a cap rock for gaseous hydrocarbons trapped underneath. However they are a potential hazard as any disturbance of a hydrate layer (such as drilling induced heating) may cause the sudden release of large volumes of methane both from melting the hydrate as well as releasing any free gas trapped beneath the hydrate level. Similarly, free gas released from deep below the sea bed could rise and form hydrates at, or near, the sea bed freezing within drilling equipment as it enters the hydrate stability zone.

Hydrate formation is possible with the bottom water temperatures and pressures found in the Rockall Basin, in water depths of over 400 m (Long, 2000; Camps et al., 2009). Although no direct evidence has been noted for their actual occurrence in the Rockall Basin, modelling using a geothermal gradient of 30°C/km has delineated extensive areas where the methane hydrate stability zone occurs beneath the sea bed, reaching 400 m thick to the south of Anton Dohrn, but with hydrate free areas on the tops of some of the banks, notably Rockall Bank, George Bligh Bank, Anton Dohrn Seamount, Rosemary Bank, the Ymir Ridge and the Wyville Thomson Ridge but not the Hebrides Terrace Seamount. The presence of this zone needs to been borne in mind as drilling through it, may allow gas to migrate upwards.

A cross-cutting reflector has occasionally been observed within the Oligocene to Miocene succession within the Rockall Basin and has even been considered a BSR but is now considered to relate to silica digenesis, the Opal A–Opal CT transition formed in Mid to Late Miocene times (Davies and Cartwright, 2002). This silica rich porcellanite was encountered in BGS borehole BH87/09 where it obstructed drilling.

Sedimentary bed forms

In the deep water areas, multibeam morphological mapping has shown the presence of a sediment drift along the base of the eastern flank of Rockall Bank, along the base of the Ymir Ridge, large sediment waves on the floor of the Rockall Trough and a wide hollow around the Rosemary Bank and Anton Dohrn seamounts (Figure 95) (Masson et al., 2002). However sediment movement associated with these features is unlikely to be rapid enough to pose a problem to sea-bed infrastructure.

Seismology

The low number of seismic events recorded in the area reflects the low activity, the poor historical documentation for the area and the limited seismicity recording for the west of Scotland. A study of historical records revealed an event in 1686 in the vicinity of St Kilda (Musson, 1998). One of the first recorded events was a series of events in the vicinity of the Hebrides Terrace Seamount. There were two events in the 1980s, these were magnitude 3.4 to 4 (Jacob et al., 1983).

Two smaller events occurred in the same area in 2007 (Galloway, 2008).

Earthquakes equal to, or greater than, 4 ML (Richter local magnitude) are taken into account as part of the structural design criteria for developments tied to the sea bed (Musson et al., 1997). The largest earthquakes have occurred nearshore in the areas of fjords and adjacent deeply glaciated valleys. The largest event in historic times was the 5.2 ML Argyll earthquake in 1880. The largest instrumentally recorded earthquake was the 4.4 ML Kintail earthquake in 1974. A further ten historic earthquakes and four instrumental earthquakes of 4.0 ML or more have occurred on the nearshore continental shelf.

In the deep-water oceanic environments, two earthquakes of magnitude 2.9 ML and 3.3 ML occurred adjacent to the Hebrides Terrace Seamount, a former volcanic centre separating the Donegal Fan from the Barra Fan. Also, an earthquake of 3.5 ML was located adjacent to the Mammal Volcanic Centre (in the Hatton Basin, west of Rockall) and a magnitude 2.6 ML event occurred in the north-east Rockall Trough adjacent to the Sula Sgeir Fan. Earthquakes with magnitude greater than 4.0 ML have not been recorded outside of the nearshore areas since 1970, so that there is a low risk that earthquakes equal to, or larger, than 4 ML will occur in these areas in the future.

References

Abraham, D A, and Ritchie, J D. 1991. The Darwin Complex, a Tertiary igneous centre in the Northern Rockall Trough. Scottish Journal of Geology, Vol. 27, 113–125.

Abreu, V S, and Anderson, J B. 1998. Glacial eustasy during the Cenozoic: sequence stratigraphic implications. American Association of Petroleum Geologists, Vol. 82, 1385–1400.

Åhäll, K-I, Connelly, J N, and Brewer, T S. 2000. Episodic Rapakivi magmatism due to distal orogenesis?: Correlation of 1.69–1.50 Ga orogenic and inboard, ‘anorogenic’ events in the Baltic Shield. Geology, Vol. 28, 823–826.

Allen, N H. 1983. Recent temperate carbonate deposits on the continental shelf north and west of Scotland: distribution, sedimentology and reserves. Unpublished PhD thesis, University of Strathclyde.

Allen, P A, and Mange-Rajetzky, M A. 1992. Devonian–Carboniferous sedimentary evolution of the Clair area, offshore north-western UK: impact of changing provenence. Marine and Petroleum Geology, Vol. 9, 29–52.

Andersen, M S. 1988. Late Cretaceous and early Tertiary extension and volcanism around the Faeroe Islands. 115–122 in Early Tertiary Volcanism and the Opening of the NE Atlantic. Morton, A C, and Parson, L M (editors). Geological Society of London Special Publication, No. 39.

Anderton, R, Bridges, P H, Leeder, M R, and Sellwood, B N. 1979. A dynamic stratigraphy of the British Isles: a study in crustal evolution. (London: George Allen & Unwin.)

Archer, S G, Bergman, S C, Iliffe, J, Murphy, C M, and Thornton, M. 2005. Palaeogene igneous rocks reveal new insights into the geodynamic evolution and petroleum potential of the Rockall Trough, NE Atlantic Margin. Basin Research, Vol. 17, 171–201.

Ardus, D A, and Harrison, D J. 1990. The assessment of aggregate resources from the UK Continental Shelf. 113–128 in Assessment and Utilisation. Ardus D A, and Champ, M A (editors). Ocean Resources, Vol. 1.

Armishaw, J E, Holmes, R, and Stow, D A V. 1998. Morphology and sedimentation on the Hebrides slope and Barra Fan, NW UK continental margin. 81–104 in Geological processes on continental margins: sedimentation, mass-wasting and stability. Stoker, M S, Evans, D, and Cramp, A (editors). Geological Society of London Special Publication, No. 129.

Armishaw, J E, Holmes, R, and Stow, D A V. 2000. The Barra Fan: a bottom-current reworked, glacially-fed submarine fan system. Marine and Petroleum Geology, Vol. 17, 219–238.

Arthurs, J W, and Earls, G. 2004. Minerals. 255–272 in The Geology of Northern Ireland — Our Natural Foundation. (Second Edition). Mitchell, W I (editor). (Belfast: Geological Survey of Northern Ireland.)

Austin, W E N, and Kroon, D. 1996. Late glacial sedimentology, foraminifera and stable isotope stratigraphy of the Hebridean continental shelf, north-west Scotland. 187–213 in Late Quaternary Palaeoceanography of the North Atlantic Margins. Andrews, J T, Austin, W E N, Bergsten, H, and Jennings, A E (editors). Geological Society of London Special Publication, No. 111.

Baba, S. 1999. Evolution of the Lewisian Complex in South Harris, North-west Scotland, and its Relationship to the North Atlantic Craton in the Palaeoproterozoic (2.0 Ga). Journal of Geosciences, Osaka University, Vol. 42, 115–125.

Baltzer, A, Holmes, R, and Evans, D. 1998. Debris flows on the Sula Sgeir Fan, NW of Scotland. 105–115 in Geological processes on continental margins: sedimentation, mass-wasting and stability. Stoker, M S, Evans, D, and Cramp, A (editors). Geological Society of London Special Publication, No. 129.

Bannister, K. 1995. A palaeomagnetic polarity study of 7 cores from in and around the Rockall Trough. Unpublished undergraduate thesis, Department of Geology and Geophysics, University of Edinburgh.

Barton, A J, and White, R S. 1997. Crustal structure of the Edoras Bank continental margin and mantle thermal anomalies in the North Atlantic. Journal of Geophysical Research, Vol. 102, B2, 3109–3129.

Barton, P J. 1992. LISPB revisited: a new look under the Caledonides of northern Britain. Geophysical Journal International, Vol. 110, 371–391.

Bassi, G. 1995. Relative importance of strain rate and theology for the mode of continental extension. Geophysical Journal International, Vol. 122, 195–210.

Bell, B R, and Butcher, H. 2002. On the emplacement of sill complexes: evidence from the Faroe–Shetland Basin. 307–329 in The North Atlantic Igneous Province: stratigraphy, tectonic, volcanic and magmatic processes. Jolley, D W, and Bell, B (editors). Geological Society of London Special Publication, No. 197.

Bentley, P A D, and Scrutton, R A. 1987. Seismic investigations into the basement structure of Southern Rockall Trough. 667–675 in Petroleum Geology of north-west Europe. Proceedings of the 3rd Conference. Brooks, J, and Glennie, K W (editors). (London: Graham and Trotman.)

Berggren, W A, and Schnitker, D. 1983. Cenozoic marine environments in the North Atlantic and Norwegian–Greenland Sea. 495–548 in Structure and development of the Greenland–Scotland Ridge: new methods and concepts. Bott, M H P, Saxov, S, Talwani, M, and Thiede, J (editors). (New York: Plenum.)

Berggren, W A, Kent, D V, Swisher, III, C C, and Aubry, M-P. 1995a. A revised Cenozoic geochronology and chronostratigraphy. 129–212 in Geochronology, time scales and stratigraphic correlation: framework for an historical geology.

Berggren, W A, Kent, D V, Aubry, M-P, and Hardenbol, J (editors). Society for Economic Palaeontologists and Mineralogists Special Publication, No. 54.

Berggren, W A, Hilgen, F J, Lengereis, C G, Kent, D V, Obradovich, J D, Raffi, I, Raymo, M E, and Shackleton, N J. 1995b. Late Neogene chronology: new perspectives in high-resolution stratigraphy. Geological Society of America Bulletin, Vol. 107, 1272–1287.

Bernstein, S, Keleman, P B, and Brooks, C K. 1998. Depleted spinel harzburgite xenoliths in Tertiary dykes from East Greenland: restites from high degree melting. Earth and Planetary Science Letters, Vol. 154, 221–235.

Blackbourn, G A. 1987. Sedimentary environments and stratigraphy of the Late Devonian–Early Carboniferous Clair Basin, west of Shetland. 75–91 in European Dinantian Environments. Miller, J, Adams, A E, and Wright, V P (editors). (Wiley).

Boldreel, L O, and Andersen, M S. 1993. Late Paleocene to Miocene compression in the Faroe–Rockall area. 1025–1034 in Petroleum geology of north-west Europe. Proceedings of the 4th conference. Parker, J R (editor). (London: The Geological Society.)

Boldreel, L O, and Andersen, M S. 1994. Tertiary development of the Faroe–Rockall Plateau based on reflection seismic data. Bulletin of the Geological Survey of Denmark, Vol. 41, 162–180.

Boldreel, L O, and Andersen, M S. 1995. The relationship between the distribution of Tertiary sediments, tectonic processes and deep-water circulation around the Faeroe Islands. 145–158 in The tectonics, sedimentation and palaeoceanography of the North Atlantic Region. Scrutton, R A, Shimmield, G B, Stoker, M S, and Tudhope, A W (editors). Geological Society of London Special Publication, No. 90.

Boldreel, L O, and Andersen, M S. 1998. Tertiary compressional structures on the Faroe–Rockall Plateau in relation to north-east Atlantic ridge-push and Alpine foreland stresses. Tectonophysics, Vol. 300, 13–28.

Borg, S G, De Paolo, D J, and Smith, B M. 1990. Isotopic structure and tectonics of the central Trans-antarctic Mountains. Journal of Geophysical Research, Vol. 95, 6647–6667.

Bott, M H P. 1978. The origin and development of the continental margin between the British Isles and SE Greenland. Geological Journal, Vol. 10, 377–392.

Bott, M H P, and Browitt, C W A. 1975. Interpretation of geophysical observations between the Orkney and Shetland Islands. Journal of the Geological Society of London, Vol. 131, 353–371.

Bott, M H P, and Watts, A B. 1971. Deep structure of the continental margin adjacent to the British Isles. 89–109 in The geology of the East Atlantic continental margin 2. Europe. Delaney, F M X (editor). Report of the Institute of Geological Sciences, No. 70/14.

Bott, M H P, Armour, A R, Himsworth, E M, Murphy, T, and Wylie, G. 1979. An explosion seismology investigation of the continental margin west of the Hebrides, Scotland, at 58ºN. Tectonophysics, Vol. 59, 217–231.

Bowen, D Q. 1999. On the correlation and classification of Quaternary deposits and land–sea correlations. 1–9 in A revised correlation of quaternary deposits in the British Isles. Bowen, D Q (editor). Geological Society of London Special Report, No. 23.

Bown, J W, and White, R S. 1995. Finite duration rifting, melting and subsidence at continental margins. 31–54 in Rifted continent–ocean boundaries. Banda, E, Torné, M, and Talwani, M (editors). (Dordrecht, The Netherlands: Kluwer Academic Publishers.)

Bradwell, T, Stoker, M S, Golledge, N R, Wilson, C K, Merritt, J W, and seven others. 2008. The northern sector of the last British Ice Sheet: Maximum extent and demise. Earth Science Reviews, Vol. 88, 207–226.

Brekke, H. 2000. The tectonic evolution of the Norwegian Sea continental margin with emphasis on the Vøring and Møre basins. 327–378 in Dynamics of the Norwegian margin. Nøttvedt, A (editor). Geological Society of London Special Publication, No. 167.

Brewer, J A, and Smythe, D K. 1984. MOIST and the continuity of crustal reflector geometry along the Caledonian–Appalachian orogen. Journal of the Geological Society of London, Vol. 141, 105–120.

Brewer, J A, and Smythe, D K. 1986. Deep structure of the foreland to the Caledonian Orogen, North-west Scotland: results of the BIRPS WINCH profile. Tectonics, Vol. 5, 171–194.

Brewer, J A, Matthews, D H, Warner, M R, Hall, J, Smythe, D K, and Whittington, R. 1983. BIRPS deep seismic reflection studies of the British Caledonides. Nature, Vol. 305, 206–210.

Bridgwater, D, Austrheim, H, Hansen, B T, Mengel, F, Pedersen, S, and Winter, J. 1990. The Proterozoic Nagssugtoquidian mobile belt of south-east Greenland: a link between the eastern Canadian and Baltic shields. Geoscience Canada, Vol. 17, 305–310.

British Geological Survey. 1983. Malin. Sheet 55°N–08°W. Solid geology. 1:250 000. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey. 1986a. Little Minch. Sheet 50°N–08°W. Solid geology. 1:250 000. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey. 1986b. Tiree. Sheet 56°N–08°W. Solid geology. 1:250 000. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey. 1986c. Rona. Sheet 59°N–06°W. Solid geology. 1:250 000. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey. 1988a. Sula Sgeir. Sheet 59°N–08°W. Solid geology. 1:250 000. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey. 1988b. Little Minch. Sheet 57°N–08°W. Sea bed sediments and Quaternary Geology. 1:250 000. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey. 1989a. Geikie. Solid geology. 1:250 000. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey. 1989b. Sutherland. Sheet 58°N–06°W. Solid geology. 1:250 000. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey. 1990. Lewis. Sheet 58°N–08°W. Solid geology. 1:250 000. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey. 1991. St Kilda. Sheet 57°N–10°W. Solid geology. 1:250 000. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey. 1992. Peach. Sheet 56°N–10°W. Solid geology. 1:250 000. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey and Petroleum Affairs Division. 2002. Central Rockall Basin. 56°–58°N, 08°–15°W. 1:500 000. Bedrock geology. (Southampton: Ordnance Survey for British Geological Survey.)

British Geological Survey. 2007. North Rockall Basin. 58°–60°N, 06°–13°W. 1:500 000. Bedrock geology. (Southampton: Ordnance Survey for British Geological Survey.)

Brook, J, and Glennie, K W (editors). 1987. Petroleum geology of north-west Europe. (London: Graham and Trotman.)

Brook, M. 1984. The age of the Conachair Granite. 40–41 in St Kilda: an illustrated account of the geology. Harding, R R, Merriman, R J, and Nancarrow, P H A (editors). Report of the British Geological Survey, 16/7. HMSO, London.

Bruce, D, and Stemmerik, L. 2003. Carboniferous. 83–89 in The millennium atlas: petroleum geology of the central and northern North Sea. Evans, D, Graham, C, Armour, A, and Bathurst, P (editors and co-ordinators). (London: The Geological Society.)

Buchan, K L, Mertanen, S, Park, R G, Pesonen, L J, Elming, S A, Abrahamsen, N, and Bylund, G. 2000. The drift of Laurentia and Baltica in the Proterozoic: a comparison based on key palaeomagnetic poles. Tectonophysics, Vol. 319, 167–198.

Buckley, J S, and Bailey, R J. 1975. Geophysical evidence on the nature of the Hebrides Terrace Seamount. Scottish Journal of Geology, Vol. 11, 37–45.

Bulat, J, and Long, D. 2001. Images of the sea bed in the Faroe–Shetland Channel from commercial 3D seismic data. Marine Geophysical Researches, Vol. 22, 345–367.

Bull, J M, and Masson, D G. 1996. The southern margin of the Rockall Plateau: stratigraphy, Tertiary volcanism and plate tectonic evolution. Journal of the Geological Society of London, Vol. 147, 495–506.

Bunch, A W H. 1979. A detailed seismic structure of Rockall Bank (55ºN, 15ºW) — a synthetic seismogram analysis. Earth and Planetary Science Letters, Vol. 45, 453–463.

Butler, C A, Holdsworth, R E, and Strachan, R A. 1995. Evidence for Caledonian sinsitral strike-slip motion and associated fault zone weakening, Outer Hebrides Fault Zone, NW Scotland. Journal of the Geological Society of London, Vol. 152, 743–746.

Butterworth, P, Holba, A, Hertig, S, Hughes, W, and Atkinson, C. 1999. Jurassic nonmarine source rocks and oils of the Porcupine Basin and other North Atlantic margin basins. 471–486 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Cameron, I B, and Stephenson, D. 1985. British regional geology: the Midland Valley of Scotland (3rd edition). (London: HMSO for the British Geological Survey.)

Camps, A P, Long, D, Rochelle, C A, and Lovell, M A. 2009. Mapping hydrate stability zones offshore Scotland. 81–91 in Sediment-hosted gaas hydrates: new insights on natural and syntheitc systems. Long, D, Lovell, M A, Rees, J G, and Rochelle, C A (editors). Geological Society of London Special Publication, No. 319.

Carter, S. 1994. Palaeomagnetic polarity study of Hebridean offshore lavas. Unpublished undergraduate thesis, Department of Geology and Geophysics, University of Edinburgh.

Chadwick, R A, and Evans, D J. 1995. The timing and direction of Permo-Triassic extension in southern Britain. 161–192 in Permian and Triassic rifting in Northwest Europe. Boldy, S A R (editor). Geological Society of London Special Publication, No. 91.

Chadwick, R A, and Pharaoh, T C. 1998. The seismic reflection Moho beneath the United Kingdom and adjacent areas. Tectonophysics, Vol. 299, 255–279.

Chadwick, R A, and 16 others. 2001. Geology of the Isle of Man and its offshore area. British Geological Survey Research Report, RR/01/06.

Chalmers, J A. 1997. The continental margin off southern Greenland: along-strike transition from an amagmatic to a volcanic margin. Journal of the Geological Society of London, Vol. 154, 571–576.

Chalmers, J A, Pulvertaft, T C R, Christiansen, F G, Larsen, H C, and Ottersen, T G. 1993. The southern West Greenland continental margin: rifting history, basin development and petroleum potential. 915–931 in Petroleum geology of north-west Europe. Proceedings of the 4th conference. Parker, J R (editor). (London: The Geological Society.)

Chambers, L M. 2000. Age and duration of the British Tertiary Igneous Province: implications for the development of the ancestral Iceland plume. Unpublished PhD thesis, University of Edinburgh.

Chambers, L M, Darbyshire, F, Noble, S, and Ritchie, J D. 2005. NW UK continental margin; chronology and isotope geochemistry. British Geological Survey Commissioned Report, CR/05/095.

Chapman, T J, Broks, T M, Corcoran, D V, Duncan, L A, and Dancer, P N. 1999. The structural evolution of the Erris Trough, offshore north-west Ireland, and implications for hydrocarbon generation. 455–469 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Chappell, A R, and Kusznir, N J. 2005. Crustal thickness mapping using satellite gravity data: implications for the formation of the southern Rockall Trough. Geophysical Research Abstracts, Vol. 7, 07953.

Chesher, J C, Smythe, D K, and Bishop, P. 1983. The geology of the Minches, Inner Sound and Sound of Raasay. Report of the Institute of Geological Sciences, No. 83/6.

Cliff, R A, Gray, C, and Huhma, H. 1983. A Sm-Nd isotopic study of the South Harris Igneous Complex, the Outer Hebrides. Contributions to Mineralogy and Petrology, Vol. 82, 91–98.

Cliff, R A, Rex, D C, and Guise, P G. 1998. Geochronological studies of Proterozoic crustal evolution in the northern Outer Hebrides. Precambrian Research, Vol. 91, 401–418.

Clift, P D, and Turner, J. 1998. Palaeogene igneous underplating and subsidence anomalies in the Rockall–Faeroe–Shetland area. Marine and Petroleum Geology, Vol. 15, 223–243.

Climap Members. 1981. Seasonal reconstruction of the Earth’s surface at the last glacial maximum. Map and Chart Series MC-36. (Boulder, Colorado: Geological Society of America.)

Cloetingh, S, Gradstein, F M, Kooi, H, Grant, A C, and Kaminski, M. 1990. Plate reorganisation: a cause of rapid late Neogene subsidence and sedimentation around the North Atlantic. Journal of the Geological Society of London, Vol. 147, 495–506.

Cockburn, A M. 1935. The geology of St Kilda. Transactions of the Royal Society of Edinburgh, Vol. 58, 511–547.

Coffin, M F, and Eldholm, O. 1992. Volcanism and continental breakup: a global compilation of a large igneous provinces. 17–30 in Magmatism and the causes of continental break-up. Storey, B C, Alabaster, T, and Pankhurst, R J (editors). Geological Society of London Special Publication, No. 68.

Cole, J E, and Peachey, J. 1999. Evidence for pre-Cretaceous rifting in the Rockall Trough: an analysis using quantitative plate tectonic modelling. 359–370 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Corcoran, D V, and Mecklenburgh, R. 2005. Exhumation of the Corrib gas field, Slyne Basin, offshore Ireland. Petroleum Geoscience, Vol. 11, 239–256.

Corfield, S, Murphy, N, and Parker, S. 1999. The structural and stratigraphic framework of the Irish Rockall Trough. 407–420 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Coward, M P, and Enfield, M A. 1987. The structure of the West Orkney and adjacent basins. 687–696 in Petroleum geology of north-west Europe. Brooks, J, and Glennie, K W (editors). (London: Graham and Trotman.)

Coward, M P, Enfield, M A, and Fischer, M W. 1989. Devonian basins of northern Scotland: extension and inversion related to Late Caledonian–Variscan tectonics. 275–308 in Inversion tectonics. Cooper, M A, and Williams, G D (editors). Geological Society of London Special Publication, No. 44.

Coward, M P, Dewey, J F, Hempton, M, and Holroyd, J. 2003. Tectonic evolution. 17–33 in The millennium atlas: petroleum geology of the central and northern North Sea. Evans, D, Graham, C, Armour, A, and Bathurst, P (editors and co-ordinators). (London: The Geological Society.)

Croker, P F. 1995. The Clare Basin: a geological and geophysical outline. 327–339 in The petroleum geology of Ireland’s offshore basins. Croker, P F, and Shannon, P M (editors). Geological Society of London Special Publication, No. 93.

Croker, P F, and Shannon, P M. 1987. The evolution and hydrocarbon prospectivity of the Porcupine Basin, Offshore Ireland. 633–642 in Petroleum geology of north-west Europe. Proceedings of the 3rd conference. Brooks, J, and Glennie, K W (editors). (London: Graham and Trotman.)

Daly, J S. 2001. Precambrian. 7–45 in The Geology of Ireland. Holland, C H (editor). (Edinburgh: Dunedin Academic Press.)

Daly, J S, Muir, R J, and Cliff, R A. 1991. A precise U-Pb zircon age for the Inishtrahull syenitic gneiss, County Donegal, Ireland. Journal of the Geological Society of London, Vol. 148, 639–642.

Daly, J S, Heaman, L M, Fitzgerald, R C, Menuge, J F, Brewer, T S, and Morton, A C. 1995. Age and crustal evolution of crystalline basement in western Ireland and Rockall. 433–434 in The petroleum geology of Ireland’s offshore basins. Croker, P F, and Shannon, P M (editors). Geological Society of London Special Publication, Vol. 93.

Dancer, P N, Algar, S T, and Wilson, I R. 1999. Structural evolution of the Slyne Trough. 445–453 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: Graham and Trotman.)

Dancer, P N, Kenyon-Roberts, S M, Downey, J W, Baillie, J M, Meadows, N S, and Maguire, K. 2005. The Corrib gas field, offshore west of Ireland. 1035–1046 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: Graham and Trotman.)

Davies, R J, and Cartwright, J. 2002. A fossilized Opal A to Opal C/T transformation on the north-east Atlantic margin: support for a significantly elevated palaeogeothermal gradient during the Neogene? Basin Research, Vol.14, 467–486.

Davies, R J, Cartwright, J, Pike, J, and Line, C. 2001. Early Oligocene initiation of North Atlantic Deep Water formation. Nature, Vol. 410, 917–920.

De Haas, H, van Weering, T C E, and Stoker, M S. 2003. Development of a sediment drift: Feni Drift, north-east Atlantic Margin. 197–201 in European margin sediment dynamics: side-scan sonar and seismic images. Mienert, J, and Weaver, P (editors). (Berlin: Springer.)

Dean, K, McLachlan, K, and Chambers, A. 1999. Rifting and the development of the Faeroe–Shetland Basin. 533–544 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Dearnley, R. 1962. An outline of the Lewisian complex of the Outer Hebrides in relation to that of the Scottish mainland. Quarterly Journal of the Geological Society of London, Vol. 118, 143–176.

Dearnley, R. 1963. The Lewisian Complex of South Harris. Quarterly Journal of the Geological Society of London, Vol. 119, 243–307.

Dearnley, R, and Dunning, F W. 1968. Metamorphosed and deformed pegmatites and basic dykes in the Lewisian complex of the Outer Hebrides and their geological significance. Quarterly Journal of the Geological Society of London, Vol. 123, 353–378.

De Paolo, D J, Linn, A M, and Schubert, G. 1991. The continental crustal age distribution: methods of determining mantle separation ages from Sm-Nd isotopic data and application to the south-western United States. Journal of Geophysical Research, Vol. 96, 2071–2088.

De Silva, N R. 1999. Sedimentary basins and petroleum systems offshore Newfoundland and Labrador. 41–61 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Dickin, A P. 1992. Evidence for an Early Proterozoic crustal province in the North Atlantic region. Journal of the Geological Society of London, Vol. 149, 483–486.

Dickin, A P, and Bowes, D R. 1991. Isotopic evidence for the extent of early Proterozoic basement in Scotland and north-west Ireland. Geological Magazine, Vol. 128, 385–388.

Dickin, A P, and Durrant, G P. 2002. The Blackstones Bank igneous complex: geochemistry and crustal context of a submerged Tertiary igneous centre in the Scottish Hebrides. Geological Magazine, Vol. 139, 199–207.

Dietrich, V J, and Jones, E J W. 1980. Volcanic rocks from Rosemary Bank (Rockall Trough, NE Atlantic). Marine Geology, Vol. 35, 287–297.

Dobson, M R, and Whittington, R J. 1992. Aspects of the geology of the Malin Sea area. 291–311 in Basins on the Atlantic seaboard: petroleum geology, sedimentology and basin evolution. Parnell, J (editor). Geological Society of London Special Publication, No. 62.

Dolan, J F. 1986. The relationship between the ‘R2’ seismic reflector and a zone of abundant detrital and authigenic smectites, Deep Sea Drilling Project Hole 610, Rockall Plateau region. 1105–1109 in Initial Reports of the Deep Sea Drilling Project, Vol. 94. Ruddiman, W F, Kidd, R B et al. (editors). (Washington, DC: US Government Printing Office.)

Doré, A G, and Lundin, E R. 1996. Cenozoic compressional structures on the north-east Atlantic margin; nature, origin and potential significance for hydrocarbon exploration. Petroleum Geoscience, Vol. 2, 299–311.

Doré, A G, and Vining, B A (editors). 2005. Petroleum geology: north-west Europe and global perspectives. Proceedingsof the 6th conference. (London: The Geological Society.)

Doré, A G, Lundin, E R, Jensen, L N, Birkeland, Ø, Eliassen, P E, and Fichler, C. 1999. Principal tectonic events in the evolution of the north-west European Atlantic margin. 41–61 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Driscoll, N W, Hogg, J R, Christie-Blick, N, and Karner, G D. 1995. Extensional tectonics in the Jeanne d’Arc Basin, offshore Newfoundland: implications for the timing of break-up between Grand Banks and Iberia. 1–28 in The tectonics, sedimentation and palaeoceanography of the North Atlantic region. Scrutton, R A, Stoker, M S, Shimmield, G B, and Tudhope, A W (editors). Geological Society of London Special Publication, No. 90.

Duindam, P, and van Hoorn, B. 1987. Structural evolution of the West Shetland continental margin. 765–773 in Petroleum geology of north-west Europe. Brooks, J, and Glennie, K W (editors). (London: Graham and Trotman.)

Earle, M M, Jankowski, E J, and Vann, I R. 1989. Structural and stratigraphic evolution of the Faroe–Shetland Channel and Northern Rockall Trough. 461–469 in Extensional tectonics and stratigraphy of the North Atlantic margins. Tankard, A J, and Balkwill, H R (editors). Memoir of the American Association of Petroleum Geologists, No. 46.

Eden, R A, Ardus, D A, Binns, P F, McQuillin, R, and Wilson, J B. 1971. Geological investigations with a manned submersible off the west coast of Scotland 1969–1970. Report of the Institute of Geological Sciences, No. 71/16.

Egerton, P D. 1998. Seismic characterisation of Palaeogene depositional sequences: north-eastern Rockall Trough. 217–228 in Geological processes on continental margins: sedimentation, mass-wasting and stability. Stoker, M S, Evans, D, and Cramp, A (editors). Geological Society of London Special Publication, No. 129.

Eiríksson, J. 1981. Lithostratigraphy of the upper Tjörnes sequence, North Iceland: the Breidavík Group. Acta Naturalia Islandica, Vol. 29, 1–37.

Eldholm, O, Tsikalas, F, and Faleide, J I. 2002. Continental margin off Norway 62–75ºN: Palaeogene tectono-magmatic segmentation and sedimentation. 39–68 in The North Atlantic igneous province: stratigraphy, tectonic, volcanic and magmatic processes. Jolley, D W, and Bell, B R (editors). Geological Society of London Special Publication, No. 197.

Ellett, D J, and Roberts, D G. 1973. The overflow of Norwegian Sea deep water across the Wyville Thomson Ridge. Deep Sea Research, Vol. 20, 819.

Elliott, G M, Shannon, P M, Haughton, P D W, and Øvrebø, L K. 2010. The Rockall Bank Mass Flow: collapse of a moated contourite drift onlapping the eastern flank of Rockall Bank, west of Ireland. Marine and Petroleum Geology, Vol 27, 92–107.

Emeleus, C H, and Goyopari, M C. 1992. British Tertiary Volcanic Province. Geological Conservation Review Series, No. 4. (London: Chapman and Hall.)

Enfield, M A, and Coward, M P. 1987. The structure of the West Orkney Basin, northern Scotland. Journal of the Geological Society of London. Vol. 144, 871–884.

England, R W, and Hobbs, R W. 1997. The structure of the Rockall Trough imaged by deep seismic reflection profiling. Journal of the Geological Society of London, Vol. 154, 497–502.

Evans, C R. 1965. Geochronology of the Lewisian basement near Lochinver, Sutherland. Nature, Vol. 207, 54–56.

Evans, D, Wilkinson, I L, and Craig, D L. 1979. The Tertiary sediments of the Canna Basin, Sea of the Hebrides. Scottish Journal of Geology, Vol. 15, 329–332.

Evans, D, Abraham, D A, and Hitchen, K. 1989. The Geikie igneous centre, west of Lewis: its structure and influence on Tertiary geology. Scottish Journal of Geology, Vol. 25, 339–352.

Evans, D, Hallsworth, C, Jolley, D W, and Morton, A C. 1991. Late Oligocene terrestrial sediments from a small basin in the Little Minch. Scottish Journal of Geology, Vol. 27, 33–40.

Evans, D, Morton, A C, Wilson, S, Jolley, D, and Barreiro, B A. 1997. Palaeoenvironmental significance of marine and terrestrial Tertiary sediments on the north-west Scottish shelf in BGS borehole BH77/7. Scottish Journal of Geology, Vol. 33, 31–42.

Evans, D, Harrison, Z, Shannon, P M, Laberg, J S, Nielson, T, Ayers, S, Holmes, R, Hoolt, R J, Lindberg, B, Haflidson, H, Long, D, Kuijpes, A, Anderson, E S, and Bryn, P. 2005. Palaeoslides and other mass failures of the Pliocene to Pleistocene age along the Atlantic continental margin of north-west Europe. Marine and Petroleum Geology, Vol. 22, 1131–1148.

Everest, J D, Bradwell, T, Fogwill, C J, and Kubik, P W. 2006. Cosmogenic ¹⁰Be age constraints for the Wester Ross readvance moraine: insights into British ice-sheet behaviour. Geografiska Annaler, Vol. 88A, 9–17.

Faleide, J I, Kyrkjebø, R, Kjennerud, T, Gabrielsen, R H, Jordt, H, Fanavoll, S, and Bjerke, M D. 2002. Tectonic impact on sedimentary processes during Cenozoic evolution of the northern North Sea and surrounding areas. 235–269 in Exhumation of the North Atlantic Margin: timing, mechanisms and implications for petroleum exploration. Doré, A G, Cartwright, J A, Stoker, M S, Turner, J P, and White, N (editors). Geological Society of London Special Publication, No. 196.

Farrow, G E, Allen, N H, and Akpan, E T. 1984. Bioclastic carbonate sedimentation on a high-latitude, tide-dominated shelf: north-east Orkney Islands, Scotland. Journal of Sedimentology and Petrology, Vol. 54, 373–393.

Ferentinos, G K. 1976. Sedimentary distribution and transport processes on the outer continental shelf of the Hebridean Sea. Marine Geology, Vol. 20, 41–56.

Ferragne, A, Latouche, C, and Parra, M. 1984. Lower Eocene red clay sedimentation on the George Bligh/Rockall Channel. Geo-Marine Letters, Vol. 4, 77–82.

Fettes, D J, and Mendum, J R. 1987. The evolution of the igneous complex in the Outer Isles. 27–44 in Evolution of the Lewisian and comparable Precambrian high grade terrains. Park, R G, and Tarney, J (editors). Geological Society of London Special Publication, No. 27.

Fettes, D J, Mendum, J, Smith, D I, and Watson, J V. 1992. Geology of the Outer Hebrides. Memoir of the British Geological Survey, Sheet 558–59, 568–69, 578–79, 588–59, 598–99, S104–105, S111–112 (Scotland).

Fisher, J. 1956. Rockall. (London: Geoffrey Bles Limited.)

Flack, C, and Warner, M. 1990. Three dimensional mapping of seismic reflections from the crust and upper mantle, north-west of Scotland. Tectonophysics, Vol. 173, 469–481.

Fleet, A J, and Boldy, S A R (editors). 1999. Petroleum geology of north-west Europe. Proceedings of the 5th conference. (London: The Geological Society.)

Flood, R D, Hollister C D, and Lonsdale, P. 1979. Disruption of the Feni sediment drift by debris flows from the Rockall Bank. Marine Geology, Vol. 32, 311–334.

Flower, B P. 1999. Data report: planktonic foraminifers from the subpolar North Atlantic and Nordic seas: sites 980–987 and 907. 19–34 in Proceedings of the Ocean Drilling Program, Scientific Results, Vol. 162. Jansen, E, Raymo, M E, Blum, P and Herbert, T D (editors). (College Station, Texas: Ocean Drilling Programme.)

Foulger, G. 2002. Plumes, or plate tectonic processes? Astronomy and Geophysics, Vol. 43, 6.19–6.23

Foulger, G R, and Anderson, D L. 2005. A cool model for the Iceland hotspot. Journal of Volcanism and Geothermal Research, Vol. 141, 1–22.

Fowler, S R, White, R S, Spence, G D, and Westbrook, G K. 1989. The Hatton Bank continental margin-II. Deep structure from two-ship expanding spread seismic profiles. Geophysical Journal, Vol. 96, 295–309.

Francis, E H. 1983. Carboniferous. 253–296 in Geology of Scotland (2nd edition). Craig, G Y (editor). (Edinburgh: Scottish Academic Press.)

Fraser, S I, Robinson, A M, Johnson, H D, Underhill, J R, Kadolsky, D G A, Connell, R, Johannessen, P, and Ravnas, R. 2003. Upper Jurassic. 157–189 in The millennium atlas: petroleum geology of the central and northern North Sea. Evans, D, Graham, C, Armour, A, and Bathurst, P (editors and coordinators). (London: The Geological Society.)

Friend, C R L, and Kinny, P D. 2001. A reappraisal of the Lewisian Gneiss Complex: geochronological evidence for its tectonic assembly from disparate terranes in the Proterozoic. Contributions to Mineralogy and Petrology, Vol. 142, 198–218.

Fugro-Geoteam. 1997. Rig Site Survey Report Location UKCS 164/7-A 27/04/097-17/05/97 Report No. M1024 (3 volumes). Report to Conoco.

Fyfe, J A, Long, D, and Evans, D. 1993. United Kingdom offshore regional report: the geology of the Malin–Hebrides sea area. (London: HMSO for the British Geological Survey).

Galloway, D D. 2008. Bulletin of British earth-quakes. 2007. British Geological Survey Open Report, OR/08/048.

Galloway, W E. 2001. Seismic expressions of deep-shelf depositional and erosional morphologies, Miocene Utsira Formation, North Sea Basin. Marine Geophysical Researches, Vol. 22, 309–321.

Garde, A A, Hamilton, M A, Chadwick, B, Grocott, J, and McCaffrey, K J W. 2002. The Ketalidian Orogen of South Greenland: geochronology, tectonics, magmatism, and forearc accretion during Palaeoproterozoic oblique convergence. Canadian Journal of Earth Sciences, Vol. 39, 765–793.

Gibbard, P L, Head, M J, Walker, M J C, and The Subcommission on Quaternary Stratigraphy. 2010. Formal ratification of the Quaternary System/Period and the Pleistocene Series/Epoch with a base at 2.58 Ma. Journal of Quaternary Science, Vol. 25, 96–102.

Gillespie, M R, Stephenson, D, and Millward, D. 2008. BGS classification of lithodemic units: proposals for classifying units of intrusive rock. British Geological Survey Research Report, RR/08/05.

Glennie, K W. 2000. Invited review: Contribution of the Scottish Offshore to the advancement of geology. Scottish Journal of Geology, Vol. 36, 17–31.

Glennie, K W. 2002. Permian and Triassic. 301–322 in The geology of Scotland (4th edition). Trewin, N H (editor). (London: The Geological Society.)

Goes, S, Govers, R, and Vacher, P. 2000. Shallow mantle temperatures under Europe from P and S wave tomography. Journal of Geophysical Research, Vol. 105, 11153–11169.

Golledge, N R. 2010. Glaciation of Scotland during the Younger Dryas: a review. Journal of Quaternary Science, Vol. 25, 550–566. DOI10.1002/jpgs.1319.

Gower, C F, Schärer, U, and Heaman, L M. 1992. The Labradorian orogeny in the Grenville Province, eastern Labrador, Canada. Canadian Journal of Earth Sciences, Vol. 29, 1944–1957.

Gradstein, F M, and Berggren, W A. 1981. Flysch-type agglutinating foraminifera and the Maastrichtian to Palaeogene history of the Labrador and North Seas. Marine Micropalaeontology, Vol. 6, 211–268.

Gradstein, F, Ogg, J G, and Smith, A G. 2004. A Geologic Time Scale 2004. (Cambridge: Cambridge University Press.)

Graham, D K, Harland, R, Gregory, D M, Long, D, and Morton, A C. 1990. The biostratigraphy and chronostratigraphy of BGS borehole 78/4, North Minch. Scottish Journal of Geology, Vol. 26, 65–75.

Graham, J R. 2001. Devonian. 201–239 in The Geology of Ireland. Holland, C H (editor). (Edinburgh: Dunedin Academic Press.)

Green, P F, Duddy, I R, and Hegarty, K A. 2002. Quantifying exhumation in sedimentary basins of the UK from apatite fission-track analysis and vitrinite reflectance data: precision, accuracy and latest results. 331–354 in Exhumation of the North Atlantic margin: timing, mechanisms and implications for petroleum exploration. Doré, A G, Cartwright, J A, Stoker, M S, Turner, J P, and White, N (editors). Geological Society of London Special Publication, No. 196.

Greenwood, S L, and Clark, C D. 2009. Reconstructing the last Irish Ice Sheet 2: a geomorphologically-driven model of ice sheet growth, retreat and dynamics. Quaternary Science Reviews, Vol. 28, 3101–3123.

Hall, J, Brewer, J A, Matthews, D H, and Warner, M R. 1984. Crustal structure across the Caledonides from the WINCH seismic reflection profile: influences on the evolution of the Midland Valley of Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences, Vol. 75, 97–109.

Hanisch, J. 1984. The Cretaceous opening of the north-east Atlantic. Tectonophysics, Vol. 101, 1–23.

Harding, R R, Merriman, R J, and Nancarrow, P H A. 1984. St Kilda: an illustrated account of the geology. Report of the British Geological Survey, Vol. 16, No. 7.

Harland, R, and Howe, J A. 1995. Dinoflagellate cysts and Holocene oceanography of the north-eastern Atlantic Ocean. The Holocene, Vol. 5, 220–228.

Harrison, R K (editor). 1975. Expeditions to Rockall 1971–72. Report of the Institute of Geological Sciences, No. 75/1.

Haszeldine, R S, and Russell, M J. 1987. The Late Carboniferous northern North Atlantic Ocean: implications for hydrocarbon exploration from Britain to the Arctic. 1163–1175 in Petroleum geology of north-west Europe. Proceedings of the 3rd conference. Brooks, J, and Glennie, K W (editors). (London: Graham and Trotman.)

Haug, G H, and Tiedemann, R. 1998. Effect of the formation of the Isthmus of Panama on Atlantic Ocean thermohaline circulation. Nature, Vol. 393, 673–675.

Haughton, P, Praeg, D, Shannon, P M, Lawrence, A, Tyrrell, S, and Morrissey, T. 2005. First results from shallow stratigraphic boreholes on the eastern flank of the Rockall basin, offshore western Ireland. 1077–1094 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Hauser, F, O’Reilly, B M, Jacob, A W B, Shannon, P M, Makris, J, and Vogt, U. 1995. The crustal structure of the Rockall Trough: differential stretching without underplating. Journal of Geophysical Research, Vol. 100, 4097–4116.

Hawkes, J R, Merriman, R J, Harding, R, and Darbyshire, D P F. 1975. Rockall Island: new geological, petrological, chemical and Rb-Sr age data. 11–51 in Expeditions to Rockall 1971–72. Harrison, R K (editor). Report of the Institute of Geological Sciences, 75/1.

Hibbert, F D, Austin, W E N, Leng, M J, and Gatliff, R W. 2010. British ice sheet dynamics inferred from North Atlantic ice-rafted debris records spanning the last 175 000 years. Journal of Quaternary Science, Vol. 25, 461–482.

Hill, P R. 1987. Characteristics of sediments from Feni and Gardar drifts, sites 610 and 611, Deep Sea Drilling Project Leg 96. 1075–1082 in Initial Reports of the Deep Sea Drilling Project, Vol. 94. Ruddiman, W F, Kidd, R B et al. (editors). (Washington DC: US Government Printing Office.)

Hillis, R R, Holford, S P, Green, P F, Doré, A G, Gatliff, R W, Stoker, M S, Thomson, K, Turner, J P, Underhill, J R, and Williams, G A. 2008. Cenozoic exhumation of the southern British Isles. Geology, Vol. 36, 371–374.

Himsworth, E M. 1973. Marine geophysical studies between north-west Scotland and the Faeroe Plateau. Unpublished PhD thesis, University of Durham.

Hitchen, K. 2004. The geology of the UK Hatton–Rockall margin. Marine and Petroleum Geology, Vol. 21, 993–1012.

Hitchen, K, and Ritchie, J D. 1987. Geological review of the West Shetland area. 737–749 in Petroleum geology of north-west Europe. Proceedings of the 3rd conference. Brooks, J, and Glennie, K W (editors). (London: Graham and Trotman.)

Hitchen, K, and Ritchie, J D. 1993. New ages, and a provisional chronogy for, the offshore part of the British Tertiary Igneous Province. Scottish Journal of Geology, Vol. 29, 73–85.

Hitchen, K, and Stoker, M S. 1993. Mesozoic rocks from the Hebrides Shelf and the implications for hydrocarbon prospectivity in the northern Rockall Trough. Marine and Petroleum Geology, Vol. 10, 246–254.

Hitchen, K, Stoker, M S, Evans, D, and Beddoe-Stephens, B. 1995a. Permo-Triassic sedimentary and volcanic rocks in basins to the north and west of Scotland. 87–102 in Permian and Triassic rifting in north-west Europe. Boldy, S A R (editor). Geological Society of London Special Publication, No. 91.

Hitchen, K, Jolley, D W, Harland, R, and Jones, J M. 1995b. Regional implications of micropalaeontological and palaeo-environmental investigations of the Early Eocene interval in BGS borehole BH90/3, offshore north-west Scotland. Scottish Journal of Geology, Vol. 31, 117–124.

Hitchen, K, Morton, A C, Mearns, E W, Whitehouse, M, and Stoker, M S. 1997. Geological implications from geochemical and isotopic studies of Upper Cretaceous and lower Tertiary igneous rocks around the northern Rockall Trough. Journal of the Geological Society of London, Vol. 154, 517–521.

Hitchen, K, Gillespie, E J, and Leslie, A. 2002. Preliminary geological results of sea-bed sampling in the Hebrides–Rockall area from the RRS James Clark Ross in 2001. British Geological Survey Internal Report, IR/02/049.

Hobbs, P R N, and Holmes, R. 2002. Slope stability assessment based on vibrocore geotechnical data from the Barra and Donegal fans, eastern Rockall Trough. British Geological Survey Commissioned Report, CR/02/171.

Holford, S P, Turner, J P, and Green, P F. 2005. Reconstructing the Mesozoic–Cenozoic exhumation history of the Irish Sea basin system using apatite fission-track analysis and vitrinite reflectance. 1095–1107 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Holford, S P, Green, P F, Duddy, I R, Turner, J P, Hillis, R R, and Stoker, M S. 2009. Regional intraplate exhumation episodes related to plate-boundary deformation. Geological Society of America Bulletin, Vol. 121, 1611–1628.

Holford, S P, Green, P F, Hillis, R R, Underhill, J R, Stoker, M S, and Duddy, I R. 2010. Multiple post-Caledonian exhumation episodes across north-west Scotland revealed by apatite fission-track analysis. Journal of the Geological Society of London, Vol. 167, 675–694.

Holmes, R. 1997. Quaternary stratigraphy: the offshore record. 72–94 in Reflections on the Ice Age in Scotland. Gordon, J E (editor). (Glasgow: Scottish Natural Heritage).

Holmes, R. 2003. Holocene shelf-margin submarine landslides, Donegal Fan, eastern Rockall Trough. 179–182 in European margin sediment dynamics: side-scan sonar and seismic images. Mienert, J, Weaver, P (editors). (Berlin: Springer.)

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

Holmes, R, Long, D, and Dodd, L R. 1998. Large-scale debrites and submarine landslides on the Barra Fan, west of Britain. 67–79 in Geological processes on continental margins: sedimentation, mass-wasting and stability. Stoker, M S, Evans, D, and Cramp, A (editors). Geological Society of London Special Publication, No. 129.

Holmes, R, Hobbs, P R N, Bulat, J, and Jones, S. 2001. Shelf edge and slope stability west of Hebrides. British Geological Survey Commissioned Report, CR/01/20.

Holmes, R, Hitchen, K, and Ottemoller, L. 2006. Strategic Environment Assessment Area 7. Hydrocarbon prospectivity, earthquakes, continental shelves and Rockall Trough surficial and sea-bed geology and sea-bed processes. British Geological Survey Commissioned Report, CR/06/063.

House, M R, Richardson, J B, Chaloner, W G, Allen, J R L, Holland, C H, and Westoll, T S. 1977. A correlation of Devonian rocks in the British Isles. Geological Society of London Special Report, No. 7.

Howe, J A. 1996. Turbidite and contourite sediment waves in the northern Rockall Trough, North Atlantic Ocean. Sedimentology, Vol. 43, 219–234.

Howe, J A, Stoker, M S, and Stow, D A V. 1994. Late Cenozoic sediment drift complex, north-east Rockall Trough, North Atlantic. Paleoceanography, Vol. 9, 989–999.

Howe, J A, Harland, R, Hine, N M, and Austin, W E N. 1998. Late Quaternary and palae-oceanographic change in the northern Rockall Trough, North Atlantic Ocean. 269–286 in Geological processes on continental margins: sedimentation, mass-wasting and stability. Stoker, M S, Evans, D, and Cramp, A (editors). Geological Society of London Special Publication, No. 129.

Howe, J A, Stoker, M S, and Woolfe, K J. 2001. Deep-marine sea-bed erosion and gravel lags in the north-western Rockall Trough, North Atlantic Ocean. Journal of the Geological Society of London, Vol. 158, 427–438.

Howe, J A, Stoker, M S, Stow, D A V, and Akhurst, M A. 2002. Sediment drifts and contourite sedimentation in the north-eastern Rockall Trough and Faroe–Shetland Channel, North Atlantic Ocean. 65–72 in Deep-water contourite systems: modern drifts and ancient series, seismic and sedimentary characteristics. Stow, D A V, Pudsey, C J, Howe, J A, Faugeres, J–C, and Viana, A R (editors). Geological Society of London Memoir, No. 22.

Howe, J A, Stoker, M S, Masson, D G, Pudsey, C J, Morris, P, Larter, R D, and Bulat, J. 2006. Sea-bed morphology and the bottom-current pathways around Rosemary Bank Seamount, northern Rockall Trough, North Atlantic. Marine and Petroleum Geology, Vol. 23, 165–181.

Hubbard, A, Bradwell, T, Golledge, N, Hall, A, Patton, H, Sugden, D, Cooper, R, and Stoker, M S. 2009. Dynamic cycles, ice streams and their impact on the extent, chronology and deglaciation of the British–Irish ice sheet. Quaternary Science Reviews, Vol. 28, 758–776.

Hudson, J D. 1963. The recognition of salinity-controlled mollusc assemblages in the Great Estuarine Series of the Inner Hebrides. Palaeontology, Vol. 6, 318–326.

Hudson, J D, and Trewin, N H. 2002. Jurassic. 323–350 in The geology of Scotland (4th edition). Trewin, N H (editor). (London: The Geological Society.)

Huuse, M. 2002. Cenozoic uplift and denudation of southern Norway: insights from the North Sea Basin. 209–233 in Exhumation of the North Atlantic margin: timing, mechanisms and implications for petroleum exploration. Doré, A G, Cartwright, J A, Stoker, M S, Turner, J P, and White, N (editors). Geological Society of London Special Publication, No. 196.

Isaksen, G H, Wilkinson, D R, and Hitchen, K. 2000. Geochemistry of organic-rich Cretaceous and Jurassic mudstones in the West Lewis and West Flannan basins, offshore north-west Scotland: implications for source rock presence in the north-east Rockall Trough. Marine and Petroleum Geology, Vol. 17, 27–42.

Irving, E. 1964. Palaeomagnetism and its application to geological and geophysical problems. (New York: John Wiley and Sons).

Jacob, A W B, Neilson, G, and Ward, V. 1983. A seismic event near the Hebrides Terrace Seamount. Scottish Journal of Geology, Vol. 19, 287–296.

Jacob, A W B, Shannon, P M, Makris, J, Hauser, F, Vogt, U, and O’Reilly, B M. 1995. An overview of the results of the RAPIDS seismic project, North Atlantic. 429–431 in The petroleum exploration of Ireland’s offshore basins. Shannon, P M, Haughton, P D W, and Corcoran, D V (editors). Geological Society of London Special Publication, No. 188.

Jacobs, C L. 2006. An appraisal of the surface geology and sedimentary processes with SEA7, the UK Continental Shelf. Research and Consultancy Report, No.18, National Oceanography Centre, Southampton.

Jansen, E, and Raymo, M E. 1996. 1. Leg 162: New frontiers on past climates. 5–29 in Proceedings of the Ocean Drilling Program, Initial Reports, Vol. 162. Jansen, E, Raymo, M E, Blum, P et al. (editors) (College Station, Texas: Ocean Drilling Program).

Jansen, E, Raymo, M E, Blum, P et al. (editors). 1996. Proceedings of the Ocean Drilling Program, Initial Reports, Vol. 162. (College Station, Texas: Ocean Drilling Program.)

Japsen, P, and Chalmers, J A. 2000. Neogene uplift and tectonics around the North Atlantic: overview. Global and Planetary Changes, Vol. 24, 165–173.

Jehu, T J, and Craig, R M. 1934. Geology of the Outer Hebrides. Part 4. North Harris and Lewis. Transactions of the Royal Society of Edinburgh, Vol. 57, 839–874.

Johnson, H, Ritchie, J D, Gatliff, R W, Williamson, J P, Cavill, J E, and Bulat, J. 2001. Aspects of the structure of the Porcupine and Porcupine Seabight basins as revealed from gravity modelling of regional seismic transects. 265–274 in The petroleum exploration of Ireland’s offshore basins. Shannon, P M, Haughton, P D W, and Corcoran, D V, (editors). Geological Society of London Special Publication, No. 188.

Johnson, H, Ritchie, J D, Hitchen, K, McInroy, D B, and Kimbell, G S. 2005. Aspects of the Cenozoic deformational history of the north-east Faroe–Shetland Basin, Wyville–Thomson Ridge and Hatton Bank areas. 993–1007 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Johnstone, G S, and Mykura, W. 1989. British regional geology: the Northern Highlands of Scotland (4th edition). (London: HMSO for the British Geological Survey).

Jolley, D W, and Whitham, A G. 2004. A stratigraphical and palaeoenvironmental analysis of the sub-basaltic Palaeogene sediments of East Greenland. Petroleum Geoscience, Vol. 10, 53–60.

Jolley, D W, Morton, A C, and Prince, I. 2005. Volcanogenic impact on phytogeograpahy and sediment dispersal patterns in the north-east Atlantic. 969–975 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Jones, E J W. 1978. Seismic evidence for sedimentary troughs of Mesozoic age on the Hebridean continental margin. Nature, Vol. 272, 789–792.

Jones, E J W. 1981. Seismic refraction shooting on the continental margin west of the Hebrides, north-west Scotland. Journal of Geophysical Research, Vol. 86, 11553–11574.

Jones, E J W, and Ramsay, A T S. 1982. Volcanic ash deposits of early Eocene age from the Rockall Trough. Nature, Vol. 299, 342–344.

Jones, E J W, Mitchell, J G, Shido, F, and Phillips, J D. 1972. Igneous rocks dredged from the Rockall Plateau. Nature, Vol. 237, 118–120.

Jones, E J W, Ramsay, A T S, Preston, N J, and Smith, A C S. 1974. A Cretaceous guyot in the Rockall Trough. Nature, Vol. 251, 129–131.

Jones, E J W, White, R S, Hughes, V J, Matthews, D H, and Clayton, B R. 1984. Crustal structure of the continental shelf off north-west Britain from two-ship seismic experiments. Geophysics, 49, 1605–1621.

Jones, E J W, Perry, R G, and Wild, J L. 1986. Geology of the Hebridean margin of the Rockall Trough. Proceedings of the Royal Society of Edinburgh, Vol. 88B, 27–51.

Jones, E J W, Siddall, R, Thirlwall, M F, Chroston, P N, and Lloyd, A J. 1994. Anton Dohrn Seamount and the evolution of the Rockall Trough. Oceanologica Acta, Vol. 17, 237–247.

Jones, S M, White, N, Clarke, B J, Rowley, E, and Gallagher, K. 2002. Present and past influence of the Iceland Plume on sedimentation. 13–25 in Exhumation of the North Atlantic margin: timing, mechanisms and implications for petroleum exploration. Doré, A G, Cartwright, J A, Stoker, M S, Turner, J P, and White, N (editors). Geological Society of London Special Publication, No. 196.

Joppen, M, and White, R S. 1990. The structure and subsidence of Rockall Trough from two-ship seismic experiments. Journal of Geophysical Research, Vol. 95, 19821–19837.

Kalsbeek, F, Nutnam, A P, and Taylor, P N. 1993. Palaeoproterozic basement province in the Caledonian fold belt of north-east Greenland. Precambrian Research, Vol. 63, 163–178.

Kelley, S P, Reddy, S M, and Maddock, R. 1994. Laser-probe 40Ar/39Ar investigation of a pseudo-tachylite and its host rock from the Outer Hebrides thrust, Scotland. Geology, Vol. 22, 443–446.

Kenyon, N H. 1986. Evidence from bedforms for a strong poleward current along the upper continental slope of north-west Europe. Marine Geology, Vol. 72, 187–198.

Kenyon, N H. 1987. Mass-wasting features on the continental slope of north-west Europe. Marine Geology, Vol. 74, 57–77.

Keser Neish, J. 1993. Seismic structure of the Hatton–Rockall area: an integrated seismic/modelling study from composite datasets, 1047–1056 in Petroleum geology of north-west Europe. Proceedings of the 4th conference. Parker, J R (editor). (London: The Geological Society.)

Keser Neish, J. 2003. The Faroese region: a standard structural nomenclature system. Faroese Geological Survey, Tórshavn, Faroe Islands.

Keser Neish, J. 2005. Faroese area; structural interpretation of seismic data in a basalt environment. 131–145 in Faroe Islands Exploration Conference: Proceedings of the 1st Conference. Ziska, H, Varming, T, and Bloch, D (editors). Annales Societatis Scientarum Faeroensis, Supplementum, Vol. 43.

Keser Neish, J, and Ziska, H. 2005. Structure of the Faroe-Bank Channel. 873–885 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Kidd, R B, and Hill, P R. 1987. Sedimentation on Feni and Gardar sediment drifts. 1217–1244 in Initial reports of the Deep Sea Drilling Project, Vol. 94. Ruddiman, W F, Kidd, R Bet al. (editors). (Washington DC: US Government Printing Office.)

Kimbell, G S, Gatliff, R W, Ritchie, J D, Walker, A S D, and Williamson, J P. 2004. Regional three-dimensional modelling of the north-east Atlantic margin. Basin Research, Vol. 16, 259–278.

Kimbell, G S, Ritchie, J D, Johnson, H, and Gatliff, R W. 2005. Controls on the structure and evolution of the north-east Atlantic margin revealed by regional potential field imaging and 3D modelling. 933–945 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Kinny, P D, Friend, C R L, and Love, G J. 2005. Proposal for a terrane-based nomenclature for the Lewisian Gneiss Complex of north-west Scotland. Journal of the Geological Society of London, Vol. 162, 175–186.

Kirton, S R, and Hitchen, K. 1987. Timing and style of crustal extension north of the Scottish mainland. 501–510 in Continental Extensional Tectonics. Coward, M P, Dewey, J F, and Hancock, P L (editors). Geological Society of London Special Publication, No. 28.

Klemperer, S L, and Hobbs, R. 1991. The BIRPS Atlas. Deep seismic reflection profiles around the British Isles. (Cambridge: Cambridge University Press.)

Klemperer, S L, Ryan, P D, and Snyder, D B. 1991. A deep seismic reflection transect across the Irish Caledonides. Journal of the Geological Society of London, Vol. 148, 149–164.

Klingelhöfer, F, Edwards, R A, Hobbs, R W, and England, R W. 2005. Crustal structure of the north-east Rockall Trough from wide-angle seismic data modelling. Journal of Geophysical Research, Vol. 110, B11105, 1–25.

Knott, S D, Burchell, M T, Jolley, E J, and Fraser, A J. 1993. Mesozoic to Cenozoic plate reconstructions of the North Atlantic and hydrocarbon plays of the Atlantic margins. 953–974 in Petroleum geology of north-west Europe. Proceedings of the 4th conference. Parker, J R (editor). (London: The Geological Society.)

Knutz, P C. 1999. Late Pleistocene mass flow events on the northern Barra Fan: relationships between seismic facies and lithology of the North East Atlantic. Slope processes: multidisciplinary approaches, Abstract Book. Conference 24–27th January 1999, Southampton Oceanography Centre, Southampton, UK, 47pp.

Knutz, P C, Austin, W E N, and Jones, E J W. 2001. Millennial-scale depositional cycles related to British Ice Sheet variability and North Atlantic paleocirculation since 45 kyr BP, Barra Fan, UK margin. Paleoceanography, Vol. 16, 53–64.

Knutz, P C, Jones, E J W, Austin, W E N, and Van Weering, T C E. 2002a. Glacimarine slope sedimentation, contourite drifts and bottom current pathways on the Barra Fan, UK North Atlantic margin. Marine Geology, Vol.188, 129–146.

Knutz, P C, Jones, E J W, Howe, J A, van Weering, T C E, and Stow, D A V. 2002b. Wave-form sheeted contourite drift on the Barra Fan, north-west UK continental margin. 85–97 in Deep-water contourite systems: modern drifts and ancient series. Stow, D A V, Pudsey, C J, Howe, J A, Faugères, J-C, and Viana, A R (editors). Geological Society of London Memoir, No. 22.

Knutz, P C, Zahn, R, and Hall, I R. 2007. Centennial-scale variability of the British Ice Sheet: implications for climate forcing and Atlantic meridional overturning circulation during the last deglaciation. Paleoceanography, Vol. 12, PA1207.

Krabbendam, M, Prave, T, and Cheer, D. 2008. A fluvial origin for the Neoproterozoic Morar Group, north-west Scotland; implications for Torridon–Morar Group correlation and the Grenville Orogen foreland basin. Journal of the Geological Society of London, Vol. 165, 379–394.

Kristofferson, Y. 1978. Sea-floor spreading and the early opening of the North Atlantic. Earth and Planetary Science Letters, Vol. 38, 273–290.

Kroon, D, Shimmield, G, Austin, W E N, Derrick, S, Knutz, P, and Shimmield, T. 2000. Century- to millennial-scale sedimentological–geochemical records of glacial–Holocene sediment variations from the Barra Fan (north-east Atlantic). Journal of the Geological Society of London, Vol. 157, 643–653.

Kuijpers, A, Andersen, M S, Kenyon, N H, Kunzendorf, H, and van Weering, T C E. 1998. Quaternary sedimentation and Norwegian Sea overflow pathways around Bill Bailey Bank, north-eastern Atlantic. Marine Geology, Vol. 152, 101–127.

Laberg, J S, Stoker, M S, Dahlgren, K I T, de Haas, H, Haflidason, H, Hjelstuen, B O, Nielsen, T, Shannon, P M, Vorren, T O, van Weering, T C, and Ceramicola, S. 2005. Cenozoic alongslope processes and sedimentation on the north-west European Atlantic margin. Marine and Petroleum Geology, Vol. 22, 1069–1088.

Laier, T, Nytoft, H P, Jorgensen, O, and Isaksen, G H. 1997. Hydrocarbon traces in the Tertiary basalts of the Faeroe Islands. Marine and Petroleum Geology, Vol. 14, 257–266.

Lailey, M, Stein, A M, and Reston, T J. 1989. The Outer Hebrides fault, Scotland; a major Proterozoic structure in north-west Britain. Journal of the Geological Society of London, Vol. 146, 253–260.

Laughton, A S, Berggren, W A, and 8 others. 1972. Initial Reports of the Deep Sea Drilling Project, Vol. 12. (Washington, DC: US Government Printing Office.)

Lavier, L, and Manatschal, G. 2006. A mechanism to thin the continental lithosphere at magma-poor margins. Nature, 440, 324–328.

Lawver, L A, Müller, R D, Srivastava, S P, and Roest, W. 1990. The opening of the Arctic Ocean. 29–62 in Geologic history of the polar oceans: Arctic versus Antarctic. Bleil, U, and Thiede, J (editors). (Dordrecht, Netherlands: Kluwer).

Lear, C H, Rosenthal, Y, and Wright, J D. 2003. The closing of a seaway: ocean water masses and global climate change. Earth and Planetary Science Letters, Vol. 210, 425–436.

Leleu, S, and Hartley, A J. 2010. Controls on the stratigraphic development of the Triassic Fundy Basin, Nova Scotia: implications for the tectonostratigraphic evolution of Triassic Atlantic basins. Journal of the Geological Society of London, Vol. 167, 437–454.

Leslie, A B. 1992. A sedimentological study of Tertiary and Quaternary sediments in borehole BH88/7, BH88/7A, Hebrides Slope, northern Rockall Trough region. British Geological Survey Technical Report, WB/92/16.

Leslie, A B. 1993. Shallow Plio-Pleistocene contourites on the Hebrides Slope, north-west UK continental margin. Sedimentary Geology, Vol. 82, 61–78.

Leslie, A B, Bulat, J, Long, D, and Jones, S M. 2002. Sea-bed Image Atlas. British Geological Survey Commissioned Report, CR/02/301.

Lewis, A G, Beckmann, G E J, and Merriman, R J. 1975. Notes on geomagnetic work and palaeomagnetic data. 50–51 in Harrison, R K (editor). Expeditions to Rockall 1971–72. Report of the Institute of Geological Sciences, 75/1.

Long, D. 1992. Devensian late-glacial gas escape in the central North Sea. Continental Shelf Research, Vol. 12, 1097–1110.

Long, D. 2000. The methane hydrate stability zone north and west of Scotland. British Geological Survey Commissioned Report, CR/00/104.

Long, D, Bulat, J, and Leslie, A B. 2001. Seabed imagery of the north-east Rockall Trough derived from 3D exploration seismic. British Geological Survey Commissioned Report, CR/01/161.

Long, D, Howell, K L, Davies, J, and Stewart, H. 2010. JNCC Offshore nature survey of Anton Dohrn Seamount and East Rockall Bank Areas of Search. JNCC Report Series, No. 437.

Love, G J. 2004. The origins and accretionary development of the Lewisian Gneiss Complex of north-west Scotland: constraints from in situ U-Pb and Hf isotopic analysis of accessory minerals. Unpublished PhD thesis, Curtin University of Technology, Perth.

Love, G J, Kinny, P D, and Friend, C R L. 2004. Timing of magmatism and metamorphism in the Gruinard Bay area of the Lewisian Gneiss Complex: comparisons with the Assynt Terrane and implications for terrane accretion. Contributions to Mineralogy and Petrology, Vol. 146, 620–636.

Lovell, J P B. 1991. Permian and Triassic. 421–438 in Geology of Scotland (3rd edition). Craig, G Y (editor). (London: The Geological Society.)

Lowe, C, and Jacob, A W B. 1989. A north–south seismic profile across the Caledonian Suture zone in Ireland. Tectonophysics, Vol. 168, 297–381.

Lundin, E R, and Doré, A G. 2002. Mid-Cenozoic post-breakup deformation in the ‘passive’ margins bordering the Norwegian–Greenland Sea. Marine and Petroleum Geology, Vol. 19, 79–93.

Lundin, E R, and Doré, A G. 2005a. North-east Atlantic break-up: a re-examination of the Iceland mantle plume model and the Atlantic–Arctic linkage. 739–754 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Lundin, E R, and Doré, A G. 2005b. Fixity of the Iceland ‘hotspot’ on the Mid-Atlantic Ridge: observational evidence, mechanisms, and implications for Atlantic volcanic margins. 627–651 in Plates, plumes, and paradigms. Foulger, G R, Natland, J H, Presnall, D C, and Anderson, D L (editors). Geological Society of America Special Paper, No. 388.

MacDonald, H, Allan, P M, and Lovell, J P B. 1987. Geology of the oil accumulation in Block 26/28, Porcupine Basin, offshore Ireland. 643–651 in Petroleum geology of north-west Europe. Proceedings of the 3rd conference. Brooks, J, and Glennie, K W (editors). (London: Graham and Trotman.)

MacKenzie, G D, Shannon, P M, Jacob, A W B, Morewood, N C, Makris, L, Gaye, M, and Egloff, F. 2002. The velocity of the sediments in the southern Rockall Basin: results from new wide-angle seismic modelling. Marine and Petroleum Geology, Vol. 19, 989–1003.

Maclennan, J, and Lovell, J P B. 2002. Control of regional sea level by regional uplift and subsidence caused by magmatic underplating of Earth’s crust. Geology, 30, 675–678.

Makris, J, Ginzburg, A, Shannon, P M, Jacob, A W B, Bean, C J, and Vogt, U. 1991. A new look at the Rockall region, offshore Ireland. Marine and Petroleum Geology, Vol. 8, 410–416.

Marcantonio, F, Dickin, A P, McNutt, R H, and Heaman, L. 1988. A 1800-million-year-old Proterozoic gneiss terrane in Islay with implications for crustal structure and evolution of Britian. Nature, Vol. 335, 62–64.

Marshall, J E A. 1991. Palynology of the Stonehaven Group, Scotland; evidence for a Mid-Silurian age and its geological implications. Geological Magazine, Vol. 128, 283–286.

Marshall, J E A, and Hewett, T. 2003. Devonian. 65–81 in The millennium atlas: petroleum geology of the central and northern North Sea. Evans, D, Graham, C, Armour, A, and Bathurst, P (editors and co-ordinators). (London: The Geological Society.)

Martini, E. 1971. Standard Tertiary and Quaternary calcareous nannoplancton zonation. 739–785 in Proceedings of the II Planktonic Conference Roma, 1970. Farinacci, A (editor). (Rome: Tecnoscienza).

Martinsen, D G, Pisias, N G, Hays, J D, Imbrie, J, Moore, T C, and Shackleton, N J. 1987. Age dating and the orbital theory of the ice ages; development of a high-resolution 0 to 300 000-year chronostratigraphy. Quaternary Research, Vol. 27, 1–29.

Mason, A J, Parrish, R R, and Brewer, T S. 2004a. U-Pb geochronology of Lewisian orthogneisses in the Outer Hebrides, Scotland: implications for the tectonic setting and correlation of the South Harris Complex. Journal of the Geological Society of London, Vol. 161, 45–54.

Mason, A J, Temperley, S, and Parrish, R R. 2004b. New light on the construction, evolution and correlation of the Langavat Belt (Lewisian Complex), Outer Hebrides, Scotland: field, petrographic and geochronological evidence for an early Proterozoic imbricate zone. Journal of the Geological Society of London, Vol. 161, 837–848.

Masson, D G. 2001. Sedimentary processes shaping the eastern slope of the Faroe–Shetland Channel. Continental Shelf Research, Vol.21, 825–857.

Masson, D G, and Kidd, R B. 1986. Revised Tertiary seismic stratigraphy of the southern Rockall Trough. 1117–1126 in Initial Reports of the Deep Sea Drilling Project, Vol. 94. Ruddiman, W F, Kidd, R Bet al. (editors). (Washington, DC: US Government Printing Office.)

Masson, D G, Howe, J A, and Stoker, M S. 2002. Bottom-current sediment waves, sediment drifts and contourites in the northern Rockall Trough. Marine Geology, Vol. 192, 215–237.

Masson, D G, Bett, B J, Billett, D S M, Jacobs, C L, Wheeler, A J, and Wynn, R B. 2003. The origin of deep-water, coral-topped mounds in the northern Rockall Trough, north-east Atlantic. Marine Geology, Vol. 194, 159–180.

McBride, J H, Snyder, D B, Tate, M P, England, R W, and Hobbs, R W. 1995. Upper mantle reflector structure and origin beneath the Scottish Caledonides. Tectonics, Vol. 14, 1351–1367.

McClay, K R, Norton, M G, Coney, P, and Davis, G H. 1986. Collapse of the Caledonide Orogen and the Old Red Sandstone. Nature, Vol. 525, 147.

McDonnell, A, and Shannon, P M. 2001. Comparative Tertiary stratigraphic evolution of the Porcupine and Rockall basins. 323–344 in The petroleum exploration of Ireland’s offshore basins. Shannon, P M, Haughton, P D W, and Corcoran, D V (editors). Geological Society of London Special Publication, No. 188.

McGeary, S E, Cheadle, M J, Warner, M R, and Blundell, D J. 1987. Crustal structure of the continental shelf around Britain derived from BIRPS deep seismic profiling. 33–41 in Petroleum geology of north-west Europe. Proceedings of the 3rd conference. Brooks, J, and Glennie, K W (editors). (London: Graham and Trotman.)

McInroy, D B, Hitchen, K, and Stoker, M S. 2006. Potential Eocene and Oligocene stratigraphic traps of the Rockall Plateau, north-east Atlantic Margin. 247–266 in The deliberate search for the stratigraphic trap. Allen, M, Morgan, R, Goffey, G, and Walker, I (editors). Geological Society of London Special Publication, No. 254.

McKenzie, D, and Bickle, M J. 1988. The volume and composition of melt generated by extension of the lithosphere. Journal of Petrology, Vol. 29, 625–679.

McQuillin, R, and Watson, J V. 1973. Large-scale basement structures of the Outer Hebrides in the light of geophysical evidence. Nature Physical Science, Vol. 245, 1–3.

Meadows, N S, Macchi, L, Cubitt, J M, Bailey, N J L, and Johnson, B. 1987. Sedimentology and reservoir potential in the west of Shetland, UK exploration area. 723–736 in Petroleum geology of north-west Europe. Proceedings of the 3rd conference. Brooks, J, and Glennie, K W (editors). (London: Graham and Trotman.)

Megson, J B. 1987. The evolution of the Rockall Trough and implications for the Faroe–Shetland Trough. 653–665 in Petroleum geology of north-west Europe. Proceedings of the 3rd conference. Brooks, J, and Glennie, K (editors). (London: Graham and Trotman.)

Menzies, M A, Halliday, A N, Palacz, Z, Hunter, R H, Upton, B G J, Aspen, P, and Hawkesworth, C J. 1987. Evidence from mantle xenoliths for an enriched lithospheric keel under the Outer Hebrides. Nature, Vol. 325, 44–47.

Millennia Limited. 2004. Biostratigraphic analysis of the Enterprise Wildcat Exploration Well 12/2-1Z (interval: 3841 m to 4471 m) Rockall Trough, Licence 2/94, West of Ireland.

Miles, P R, and Roberts, D G. 1981. The magnetisation of Rosemary Bank Seamount, Rockall Trough, North-east Atlantic. Earth and Planetary Science Letters, Vol. 54, 442–448.

Miller, J A, and Mohr, P A. 1965. Potassium-argon age determination on rocks from St Kilda and Rockall. Scottish Journal of Geology, Vol. 1, 93–99.

Miller, J A, Matthews, D H, and Roberts, D G. 1973. Rocks of Grenville age from Rockall Bank. Nature Physical Science, Vol. 246, 61.

Miller, K G, and Tucholke, B E. 1983. Development of Cenozoic abyssal circulation south of the Greenland–Scotland Ridge. 549–589 in Structure and development of the Greenland–Scotland Ridge. Bott, M H P, Saxov, S, Talwani, M, and Thiede, J (editors). (New York: Plenum Publishing Corporation.)

Morewood, N C, Shannon, P M, and Mackenzie, G D. 2004. Seismic stratigraphy of the southern Rockall Basin: a comparison between wide-angle seismic and normal incidence reflection data. Marine and Petroleum Geology, Vol. 21, 1149–1163.

Morewood, N C, McKenzie, G D, Shannon, P M, O’Reilly, B M, Readman, P W, and Makris, J. 2005. The crustal structure and regional development of the Irish Atlantic margin region. 1023–1033 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Morgan, G E. 1984. Palaeomagnetism. 38–39 in St Kilda: an illustrated account of the geology. Harding, R R, Merriman, R J, and Nancarrow, P H A (editors). British Geological Survey Report, No. 16/7.

Morgan, J V, Barton, P J, and White, R S. 1989. The Hatton Bank continental margin–III. Structure from wide-angle OBS and multichannel seismic refraction profiles. Geophysical Journal International, Vol. 98, 367–384.

Morgan, J V, Hadwin, M, Warner, M, Barton, P, and Morgan, R P Ll. 1994. The polarity of deep seismic reflections from the lithospheric mantle: evidence for a relic subduction zone. Tectonophysics, Vol. 232, 319–328.

Morgan, R P Ll, Barton, P J, Warner, M, Morgan, J V, Price, C, and Jones, K. 2000. Lithospheric structure north of Scotland – I. P-wave modelling, deep reflection profiles and gravity. Geophysical Journal International, Vol. 142, 716–736.

Morgan, W J. 1971. Convection plumes in the lower mantle. Nature, Vol. 230, 42–43.

Morri, C M. 2004. The last glacial–interglacial cycle: palaeocurrent reconstruction and sediment response to climate change in the North Atlantic. Unpublished PhD thesis, University of the Highlands and Islands and the Scottish Association for Marine Science.

Morton, A C, and Taylor, P N. 1991. Geochemical and isotopic constraints on the nature and age of the basement rocks from Rockall Bank, north-east Atlantic. Journal of the Geological Society of London, Vol. 148, 631–634.

Morton, A C, Dixon, J E, Fitton, J G, Macintyre, R M, Smythe, D K, and Taylor, P N. 1988. Early Tertiary volcanic rocks in well 163/6-1A, Rockall Trough. 293–308 in Early Tertiary volcanism and the opening of the north-east Atlantic. Morton, A C, and Parson, L M (editors). Geological Society of London Special Publication, No. 39.

Morton, A C, Hitchen, K, Ritchie, J D, Hine, N M, Whitehouse, M, and Carter, S G. 1995. Late Cretaceous basalts from Rosemary Bank, northern Rockall Trough. Journal of the Geological Society of London, Vol. 152, 947–952.

Morton, A C, Hitchen, K, Fanning, C M, Yaxley, G, Johnson, H, and Ritchie, J D. 2009. Detrital zircon age constraints on the provenance of sandstones on Hatton Bank and Edoras Bank, north-east Atlantic. Journal of the Geological Society of London, Vol. 166, 137–146.

Morton, N. 1989. Jurassic sequence stratigraphy in the Hebrides Basin, north-west Scotland. Marine and Petroleum Geology, Vol. 6, 243–260.

Morton, N. 1993. Potential reservoir and source rocks in relation to Upper Triassic to Middle Jurassic sequence stratigraphy, Atlantic margin basins of the British Isles. 285–297 in Petroleum geology of north-west Europe. Proceedings of the 4th conference. Parker, J R. (editor). (London: The Geological Society.)

Mosar, J, Lewis, G, and Torsvik, T H. 2002. North Atlantic sea-floor spreading rates: implications for the Tertiary development of inversion structures of the Norwegian–Greenland Sea. Journal of the Geological Society of London, Vol. 159, 503–515.

Muir, R J, Fitches, W R, and Maltman, A J. 1989. An early Proterozoic link between Greenland and Scandinavia in the Inner Hebrides of Scotland. Terra Abstracts, Vol. 1, 5.

Muir, R J, Fitches, W R, and Maltman, A J. 1992. Rhinns Complex: A missing link in the Proterozoic basement of the North Atlantic region. Geology, Vol. 20, 1043–1046.

Muir, R J, Fitches, W R, and Maltman, A J. 1994. The Rhinns Complex: Proterozoic basement on Islay and Colonsay, Inner Hebrides, Scotland, and on Inishtrahull, north-west Ireland. Transactions of the Royal Society of Edinburgh: Earth Sciences, Vol. 85, 77–90.

Murphy, N J, and Croker, P F. 1992. Many play concepts seen over wide areas in Erris, Slyne troughs off Ireland. Oil and Gas Journal, Vol. 90 (37), 14 September 1992, 92–97.

Musgrove, F W, and Mitchener, B. 1996. Analysis of the pre-Tertiary rifting history of the Rockall Trough. Petroleum Geoscience, Vol. 2, 353–360.

Mussett, A E, Dagley, P, and Skelhorn, R R. 1988. Time and duration of activity in the British Tertiary Igneous Province. 337–348 in Early Tertiary volcanism and the opening of the north-east Atlantic. Morton, A C, and Parson, L M (editors). Geological Society of London Special Publication, No. 39.

Musson, R M W. 1998. Historical seismicity of the Western Frontiers Area. British Geological Survey Technical Report, WL/98/26.

Musson, R M W, Long, D, Pappin, J W, Lubkowski, Z A, and Booth, E. 1997. UK continental shelf seismic hazard. Health and Safety Executive, Offshore Technology Report. OTH 93416. ISBN 0 7176 13194.

Nadin, P A, Houchen, M A, and Kuznir, N J. 1999. Evidence for pre-Cretaceous rifting in the Rockall Trough: an analysis using quantitative 2D structural/stratigraphic modelling. 371–378 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R. (editors). (London: The Geological Society.)

Narayanaswamy, B E, Howell, K L, Hughes, D J, Davies, J S, Roberts, J M, and Black, D D. 2005. DTI Strategic Environmental Assessment Area 7: Photographic Analysis. 177 pp.

Naylor, D, and Shannon, P M. 2005. The structural framework of the Irish Atlantic Margin. 1009–1021 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Naylor, D, Shannon, P M, and Murphy, N. 1999. Irish Rockall Basin region — a standard structural nomenclature system. Petroleum Affairs Division, Dublin, Special Publication, No. 1/99.

Naylor, D, Shannon, P, and Murphy, N. 2002. Porcupine–Goban region — a standard structural nomenclature system. Petroleum Affairs Division, Dublin, Special Publication, No. 1/02.

Neves, R, and five others. 1973. Palynological correlation within the Lower Carboniferous of Scotland and northern England. Transactions of the Royal Society of Edinburgh, Vol. 69, 4–67.

Nicholas, C J. 1994. New stratigraphical constraints on the Durness Group of north-west Scotland. Scottish Journal of Geology, Vol. 30, 73–85.

Nichols, G J. 2005. Sedimentary evolution of the Lower Clair Group, Devonian, West of Shetland: climate and sediment supply controls on fluvial, aeolian and lacustrine deposition. 957–967 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Nisbet, H C, and Bowes, D R. 1961. The geology of north Rona. Transactions of the Geological Society of Glasgow, Vol. 24, 169–189.

Norris, R D, Klaus, A, and Kroon, D. 2001. Mid-Eocene deep water, the Late Palaeocene thermal maximum and continental slope mass wasting during the Cretaceous–Palaeogene impact. 23–48 in Western North Atlantic Palaeogene and Cretaceous palaeoceanography. Kroon, D, Norris, R D, and Klaus, A (editors). Geological Society of London Special Publication, No. 183.

Nunns, A G. 1983. Plate tectonic evolution of the Greenland–Scotland Ridge and surrounding regions. 91–108 in Structure and development of the Greenland–Scotland Ridge: new methods and concepts. Bott, M H P, Saxov, S, Talwani, M, and Thiede, J (editors). (New York: Plenum.)

O’Connor, J M, Stoffers, P, Wijbrans, J R, Shannon, P M, and Morrissey, T. 2000. Evidence from episodic seamount volcanism for pulsing of the Iceland plume in the past 70 Myr. Nature, Vol. 408, 954–958.

Ofoegbu, C O, and Bott, M H P. 1985. Interpretation of the Minch linear magnetic anomaly and of a similar feature on the shelf north of Lewis by non-linear optimisation. Journal of Geophysical Sciences, Vol. 142, 1077–1087.

Oliver, G J H. 2002. Chronology and terrane assemblies, new and old controversies. 201–212 in The geology of Scotland. (4th edition). Trewin, N H (editor). (London: The Geological Society.)

Omran, M A. 1990. Geophysical studies in the Hebrides Terrace Seamount area. Unpublished PhD Thesis, University of Wales.

O’Reilly, B M, Hauser, F, Jacob, A W B, Shannon, P M, Makris, J, and Vogt, U. 1995. The transition between the Erris and the Rockall Basins: new evidence from wide-angle seismic data. Tectonophysics, Vol. 241, 143–163.

O’Reilly, B M, Hauser, F, Jacob, A W B, and Shannon, P M. 1996. The lithosphere below the Rockall Trough: wide-angle seismic evidence for extensive serpentinisation. Tectonophysics, Vol. 255, 1–23.

Øvrebø, L K, Haughton, P D W, Shannon, P M. 2005. Temporal and spatial variations in late Quaternary slope sedimentation along the under-supplied margins of the Rockall Trough, offshore west Ireland. Norwegian Journal of Geology, Vol. 85, 279–294.

Owens, B, and Marshall, J (compilers). 1978. Micropalaeontological biostratigraphy of samples from around the coast of Scotland. Report of the Institute of Geological Sciences, No. 78/20.

Palmer, T J, McKerrow, W S, and Cowie, J W. 1980. Sedimentation evidence for a stratigraphical break in the Durness Group. Nature, Vol. 287, 720–722.

Pankhurst, R J. 1982. Geochronological tables for British igneous rocks. 575–580 in Igneous rocks of the British Isles. Sutherland, D S (editor). (Chichester, England: John Wiley and sons.)

Pantin, H M. 1990. The sea-bed sediments around the United Kingdom: their bathymetric and physical environment, mineral composition and associated bedforms. British Geological Survey Research Report, SB/90/1.

Park, R G. 1995. Palaeoproterozoic Laurentia–Baltica relationships: a view from the Lewisian. 211–224 in Early Precambrian Processes. Coward, M P, and Reis, A C (editors). Geological Society of London Special Publication, Vol. 95.

Park, R G. 2005. The Lewisian terrane model: a review. Scottish Journal of Geology, Vol. 41, 105–118.

Park, R G, Cliff, R A, Fettes, D G, and Stewart, A D. 1994. Precambrian rocks in north-west Scotland west of the Moine Thrust: the Lewisian Complex and the Torridonian. 6–22 in A Revised Correlation of Precambrian rocks in the British Isles. Gibbons, F C, and Harris, A L (editors). Geological Society of London Special Report, Vol. 22.

Park, R G, Stewart, A D, and Wright, D T. 2002. The Hebridean terrane. 45–80 in The geology of Scotland (4th edition). Trewin, N H (editor). (London: The Geological Society.)

Park, R G, Kinny, P D, Friend, C R L, and Love, G J. 2005. Discussion on a terrane-based nomenclature for the Lewisian Gneiss Complex of north-west Scotland. Journal of the Geological Society of London, Vol. 162, 893–895.

Parker, J R (editor). 1993. Petroleum geology of north-west Europe. Proceedings of the 4th conference. (London: The Geological Society.)

Paul, M A, Talbot, L A, and Stoker, M S. 1998. Shallow geotechnical profiles, acoustic character and depositional history in glacially influenced sediments from the Hebrides and West Shetland Slopes. 117–131 in Geological processes on continental margins: sedimentation, mass-wasting, and stability. Stoker, M S, Evans, D, and Cramp, A (editors). Geological Society of London Special Publication, No. 129.

Peacock, J D. 1984. Quaternary geology of the Outer Hebrides. British Geological Survey Report, Vol. 16, No. 2.

Peacock, J D. 1996. Marine molluscan proxy data applied to Scottish late glacial and Flandrian sites: strengths and limitations. 215–228 in Late Quaternary palaeoceanography of the North Atlantic margins. Andrews, J T, Austin, W E N, Bergsten, H, and Jennings, A E (editors). Geological Society of London Special Publication, No. 111.

Peacock, J D, and Harkness, D D. 1990. Radiocarbon ages and the full-glacial to Holocene transition in seas adjacent to Scotland and southern Scandinavia: a review. Transactions of the Royal Society of Edinburgh, Vol. 81, 385–396.

Peacock, J D, Austin, W E N, Selby, I, Graham, D K, Harland, R, and Wilkinson, I P. 1992. Late Devensian and Flandrian palaeoenvironmental changes on the Scottish continental shelf west of the Outer Hebrides. Journal of Quaternary Science, Vol. 7, 145–161.

Peck, V L, Hall, I R, Zahn, R, Grousset, F, Hemming, S R, and Scourse, J D. 2007. The relationship of Heinrich events and their European precursors over the past 60 ka BP: a multi-proxy ice-rafted debris provenance study in the North East Atlantic. Quaternary Science Reviews, Vol. 26, 862–875.

Penn, I E, Holliday, D W, Kirby, G A, Kubala, M, Sobey, R A, Mitchell, W I, Harrison, R K, and Beckinsale, R D. 1983. The Larne No.2 Borehole: discovery of a new Permian volcanic centre. Scottish Journal of Geology, Vol. 19, 333–346.

Pérez-Gussinyé, M, and Reston, T J. 2001. Rheological evolution during extension at non-volcanic rifted margins: onset of serpentinization and development of detachments leading to continental breakup. Journal of Geophysical Research, Vol. 106, 3961–3975.

Pérez-Gussinyé, M, Reston, T J, and Phipps-Morgan, J. 2001. Serpentinization and magmatism during extension at non-volcanic margins: the effect of initial lithospheric structure. 551–576 in Non-volcanic rifting of continental margins: a comparison of evidence from land and sea. Wilson, R C L, Whitmarsh, R B, Taylor, B, and Froitzheim, N (editors). Geological Society of London Special Publication, No. 187.

Péron-Pinvidic, G, and Manatschal, G. 2010. From microcontinents to extensional allochthons: witnesses of how continents rift and break apart? Petroleum Geoscience, Vol. 16, 189–197.

Perry, R G. 1987. A geological and geophysical investigation of the Hebridean continental margin. Unpublished PhD Thesis, University of London.

Pharaoh, T C. 1999. Palaeozoic terranes and their lithospheric boundaries within the Trans-European Suture Zone (TESZ); a review. 17–41 in Geophysical investigations of the Trans-European Suture zone. Thybo, H, Pharaoh, T, and Guterch, A (editors). Special Issue of Tectonophysics, Vol. 314.

Pidgeon, R T, and Aftalion, M. 1972. The geochronological significance of discordant U-Pb ages of oval-shaped zircons from a Lewisian gneiss from Harris, Outer Hebrides. Earth and Planetary Science Letters, Vol. 17, 268–274.

Pitcher, W S, Elwell, R W D, Tozer, C F, and Cambray, F W. 1964. The Leannan Fault. Quarterly Journal of the Geological Society of London, Vol. 120, 241–273.

Planke, S, Symonds, P A, Alvesstad, E, and Skogseid, J. 2000. Seismic volcanostratigraphy of large-volume basaltic extrusive complexes on rifted margins. Journal of Geophysical Research, Vol. 105, B8, 19 335–19 351.

Poudjom Djomani, Y H, O’Reilly, S Y, Griffin, W L, and Morgan, P. 2001. The density structure of subcontinental lithosphere through time. Earth and Planetary Science Letters, Vol. 184, 605–621.

Powell, C M R, and Sinha, M C. 1987. The PUMA experiment west of Lewis, UK. Geophysical Journal of the Royal Astronomical Society, Vol. 89, 259–264.

Praeg, D, Stoker, M S, Shannon, P M, Ceramicola, S, Hjelstuen, B O, and Mathiesen, A. 2005. Episodic Cenozoic tectonism and the development of the north-west European ‘passive’ continental margin. Marine and Petroleum Geology, Vol. 22, 1007–1030.

Price, C, and Morgan, J. 2000. Lithospheric structure north of Scotland — II. Poisson’s ratios and waveform modelling. Geophysical Journal International, Vol. 142, 737–754.

Pudsey, C J, Morris, P, and Larter, R D. 2004. Rosemary: visually stunning. Marine Scientist, Vol. 6, 16–17.

Rainbird, R H, Hamilton, M A, and Young, G M.2001. Detrital zircon geochronology and provenance of the Torridonian, north-west Scotland. Journal of the Geological Society of London, Vol. 158, 15–27.

Ramsay, A T S, Smart, C W, and Zachos, J C. 1998. A model of early to middle Miocene deep ocean circulation for the Atlantic and Indian Oceans. 55–70 in Geological evolution of ocean basins: Results from the Ocean Drilling Program. Cramp, A, MacLeod, C J, Lee, S V, and Jones, E J W (editors). Geological Society of London Special Publication, No. 131.

Rasmussen, T L, Bäckström, D, Heinemeier, J, Klitgaard-Kristensen, D, Knutz, P C, Knutz, Kuijpers, A, Lassen, S, Thomsen, E, Troelstra, S R, and Van Weering, T C E. 2002. The Faroe–Shetland Gateway: Late Quaternary water mass exchange between the Nordic seas and the north-eastern Atlantic. Marine Geology, Vol. 188, 165–192.

Rattey, R P, and Hayward, A W. 1993. Sequence stratigraphy of a failed rift system; the Middle Jurassic to Early Cretaceous basin evolution of the Central and Northern North Sea. 215–249 in Petroleum geology of north-west Europe. Proceedings of the 4th conference. Parker, J R (editor). (London: The Geological Society.)

Read, W A, Browne, M A E, Stephenson, D, and Upton, B J G. 2002. Carboniferous. 251–299 in The geology of Scotland (4th edition). Trewin, N H (editor). (London: The Geological Society.)

Redfern, R. 2000. Origins: the evolution of continents, ocean and life. (London: Cassell & Co.)

Reston, T J. 2007. Extension discrepancy at North Atlantic non-volcanic rifted margins: depth-dependent stretching or unrecognised faulting? Geology, Vol. 35, 367–370.

Reston, T J. 2009a. The structure, evolution and symmetry of the magma-poor rifted margins of the North and Central Atlantic: a synthesis. Tectonophysics, Vol. 468, 6–27.

Reston, T J. 2009b. The extension discrepancy and syn-rift subsidence deficit at rifted margins. Petroleum Geoscience, Vol. 15, 217–237.

Richardson, K R, Smallwood, J R, White, R S, Snyder, D B, and Maguire, P K H. 1998. Crustal structure beneath the Faroe Islands and the Faroe–Iceland Ridge. Tectonophysics, Vol. 300, 159–180.

Richey, J E. 1961. British regional geology: the Tertiary volcanic districts (3rd edition). (London: HMSO for Institute of Geological Sciences.)

Richter, T O, Lassen, S, Van Weering, T C E, and De Haas, H. 2001. Magnetic susceptibility patterns and provenance of ice-rafted material at Feni Drift, Rockall Trough: implications for the history of the British–Irish ice sheet. Marine Geology, Vol. 173, 37–54.

Rise, L, Ottesen, D, Larsen, E, Lundin, E, Olsen, L, and Thorsnes, T. 2002. Large-scale development of the mid-Norwegian shelf and margin with emphasis on the last 3 million years. (Trondheim: Norwegian Geological Survey.)

Ritchie, J D, and Gatliff, R W. 1996. Rockall Trough: the last frontier. Geoscientist, Vol. 6, 11–14.

Ritchie, J D, and Hitchen, K. 1996. Early Palaeogene offshore igneous activity to the north-west of the UK and its relationship to the North Atlantic igneous province. 63–78 in Correlation of the early Paleogene in north-west Europe. Knox, R W O’B, Corfield, R M, and Dunay, R E (editors). Geological Society of London Special Publication, No. 101.

Ritchie, J D, Hitchen, K, and Edwards, J W F. 1997. The Sigmundur Complex, a ?Tertiary igneous centre in the northern Rockall Trough. Scottish Journal of Geology, Vol. 33, 97–103.

Ritchie, J D, Gatliff, R W, and Richards, P C. 1999. Early Tertiary magmatism in the offshore north-west UK margin and surrounds. 573–584 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Ritchie, J D, Johnson, H, and Kimbell, G S. 2003a. The nature and age of Cenozoic contractional deformation within the north-east Faroe–Shetland Basin. Marine and Petroleum Geology, Vol. 20, 399–409.

Ritchie, J D, Johnson, H, Browne, M A E, and Monaghan, A A. 2003b. Late Devonian–Carboniferous tectonic evolution within the Firth of Forth, Midland Valley; as revealed from 2D seismic reflection data. Scottish Journal of Geology, Vol. 39, 121–134.

Ritchie, J D, Johnson, H, Quinn, M F, and Gatliff, R W. 2008. The effects of Cenozoic compression within the Faroe–Shetland Basin and adjacent areas. 121–136 in The nature and origin of compression in passive margins. Johnson, H, Doré, A G, Gatliff, R W, Holdsworth, R W, Lundin, E R, and Ritchie, J D (editors). Geological Society of London Special Publication, No. 306.

Ritchie, J D, Ziska, H, Johnson, H, and Evans, D (editors). 2011. Geology of the Faroe–Shetland Basin and adjacent areas. British Geological Survey Research Report, RR/11/01. Jard¯feingi Research Report, RR/11/01.

Roberts, D G. 1969. New Tertiary volcanic centre on the Rockall Bank, eastern North Atlantic Ocean. Nature, Vol. 223, 819–820.

Roberts, D G. 1972. Slumping on the eastern margin of the Rockall Bank, North Atlantic Ocean. Marine Geology, Vol. 13, 225–237.

Roberts, D G. 1975a. Marine Geology of the Rockall Plateau and Trough. Philosophical Transactions of the Royal Society of London, Vol. 278A, 447–509.

Roberts, D G. 1975b. Sediment distribution on the Rockall Bank, Rockall Plateau. Marine Geology, Vol. 19, 239–257.

Roberts, D G. 1989. Basin inversion in and around the British Isles. 131–150 in Inversion tectonics. Cooper, M A, and Williams, G D (editors). Geological Society of London Special Publication, No. 44.

Roberts, D G, and Backman, J. 1984. Late Paleocene–Eocene volcanic events in the Northern North Atlantic Ocean. 913–923 in Initial Reports of the Deep Sea Drilling Program, Vol. 81. Roberts, D G, and Schnitker, D et al. (editors). (Washington DC: US Government Printing Office.)

Roberts, D G, and Jones, M T. 1978. A bathymetric, magnetic and gravity survey of the Rockall Bank. H M S Hecla 1969. Admiralty Marine Science Publication, No 19.

Roberts, D G, Matthews, D H, and Eden, R A. 1972. Metamorphic rocks from the southern end of the Rockall Bank. Journal of the Geological Society of London, Vol. 128, 501–506.

Roberts, D G, Ardus, D A, and Dearnley, R. 1973. Pre-Cambrian rocks drilled from the Rockall Bank. Nature, Physical Science, Vol. 244, 21–23.

Roberts, D G, Flemming, N C, Harrison, R K, Binns, P E, and Snelling, N J. 1974. Helen’s Reef: a micro-gabbroic intrusion in the Rockall intrusive centre, Rockall Bank. Marine Geology, Vol. 16, M21–M30.

Roberts, D G, Hunter, P M, and Laughton, A S. 1979. Bathymetry of the north-east Atlantic: continental margin around the British Isles. Deep-Sea Research, Vol. 26A, 417–428.

Roberts, D G, Masson, D G, and Miles, P R. 1981. Age and structure of the southern Rockall Trough: new evidence. Earth and Planetary Science Letters, Vol. 52, 115–128.

Roberts, D G, Bott, M H P, And Ur Ski, C. 1983. Structure and origin of the Wyville Thomson Ridge. 133–158 in Structure and development of the Greenland–Scotland Ridge: new methods and concepts. Bott, M H P, Saxov, S, Talwani, M, and Thiede, J (editors). (New York: Plenum Press.)

Roberts, D G, Ginzberg , A, Nunn, K, And McQuillin, R. 1988. The structure of the Rockall Trough from seismic refraction and wide-angle reflection measurements. Nature, Vol. 332, 632–635.

Roberts, D G, Thompson, M, Mitchener, B, Hossack, J, Carmichael, S, And Bjørnseth, H-M. 1999. Palaeozoic to Tertiary rift and basin dynamics: mid Norway to the Bay of Biscay — a new context for hydrocarbon prospectivity in the deep water frontier. 7–40 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, a J, and Boldy, S A R (editors). (London: The Geological Society.)

Roberts, J M, Wheeler, A J, and Freiwald, A. 2006. Reefs of the deep: the biology and geology of cold water coral ecosystems. Science, Vol. 312, 543–547 DOI: 10.1126/science.1119861.

Ruddiman, W F, Kidd, R B, Thomas, E et al. 1987. Site 610. 351–470 in Initial reports of the Deep Sea Drilling Project, Vol. 94. Ruddiman, W F, Kid , R B et al. (editors). (Washington DC: U.S. Government Printing Office.)

Russell, M J, And Smythe, D K. 1978. Evidence for an Early Permian oceanic rift in the northern North Atlantic. 173–179 in Petrology and geochemistry of continental rifts. Neumann, E-R, And Ramber , I B. (Dordrecht: Riedel.)

Sabine, P A.1960. The geology of Rockall, North Atlantic. Bulletin of the Geological Survey, Vol. 16, 156–178.

Sandwell, D T, And Smith, W H F. 1997. Marine gravity anomaly from Geosat and ERS-1 satellite altimetry. Journal of Geophysical Research, Vol. 102, 10 039–10 054.

Saunders, A D, Fitton, J G, Kerr, A C, Norry M J, And Kent, R W. 1997. The North Atlantic Igneous Province. 45–93 in Large igneous provinces: continental, oceanic and planetary volcanism. Mahoney, J J, And Coffin, M F (editors). American Geophysical Union, Monograph, Vol. 100.

Saunders, A D, Larsen, H C, and Fitton, J G. 1998. Magmatic development of the south-east Greenland margin and evolution of the Iceland plume: geo chemical constraints from Leg 152. 479–501 in Proceedings of the Ocean Drilling Program, Scientific Results, Vol. 152.

Saunders, A D, Larsen, H C, and Wise, S W (editors). (College Station, Texas: Ocean Drilling Program.)

Scanlon, R P, Daly, J S, and Whitehouse, M J. 2003. The c.1.8 Ga Stanton Banks Terrane, offshore western Scotland, a large juvenile Palaeoproterozoic crustal block within the accretionary Lewisian Complex. Geophysical Research Abstracts, Vol 5, 13248.

Scoffin, T P. 1988. The environments of production and deposition of calcareous sediments on the shelf west of Scotland. Sedimentary Geology, Vol. 60, 107–124.

Scoffin, T P, Alexandersson, E T, Bowes, G E, Clokie, J J, Farrow, G E, And Milliman, J D. 1980. Recent temperate, sub-photic, carbonate sedimentation: Rockall bank, north-east Atlantic. Journal of Sedimentary Petrology, Vol. 50, 331–356.

Scotchman, I C. 2001. Petroleum geochemistry of the Lower and Middle Jurassic in Atlantic margin basins of Ireland and the UK. 31–60 in The petroleum exploration of Ireland’s offshore basins. Shannon, P M, Hau Ton, P D W, And Cor Oran, D V (editors). Geological Society of London Special Publication, No. 188

Scotchman, I C. and Thomas, J R W. 1995. Maturity and hydrocarbon generation in the Slyne Trough, north-west Ireland. 385–412 in The petroleum geology of Ireland’s offshore basins. Croker, P F, and Shannon, P M (editors). Geological Society of London Special Publication, No. 93.

Scotchman, I C. Griffith, C E, Holmes, A J, And Jones, D M. 1998. The Jurassic petroleum system north and west of Britain: a geochemical oil–source correlation study. Organic Geochemistry, Vol. 29, 671–700.

Scourse, J D, Haapaniemi, A I, Colmenero-Hidalgo, E, Peck, V L, Hall, I R, Austin, W E N, Knutz, P C, And Zahn, R. 2009. Growth, dynamics and deglaciation of the last British–Irish ice sheet: the deep-sea ice-rafted record. Quaternary Science Reviews, Vol. 28, 3066–3084.

Scrutton, R A. 1970. Results of a seismic refraction experiment on Rockall Bank. Nature, Vol. 227, 826–827.

Scrutton, R A. 1971. Gravity and magnetic interpretation of Rosemary Bank, north-east Atlantic. Geophysical Journal of the Royal Astronomical Society, Vol. 24, 51–58.

Scrutton, R A. 1972. The crustal structure of Rockall Plateau microcontinent. Geophysical Journal of the Royal Astronomical Society, Vol. 27, 259–275.

Scrutton, R A. 1986. The geology, crustal structure and evolution of the Rockall Trough and the Faroe–Shetland Channel. Proceedings of the Royal Society of Edinburgh. Section B, Vol. 88, 7–26.

Scrutton, R A. and Bentley, P A D. 1988. Major Cretaceous volcanic province in southern Rockall Trough. Earth and Planetary Science Letters, Vol. 91, 198–204. selby, I. 1989. The Quaternary geology of the Hebridean continental margin. Unpublished PhD thesis, University of Nottingham.

Shannon, P M, Moore, J G, Jacob, A W B, And Makris, J. 1993. Cretaceous and Tertiary basin development west of Ireland. 1057–1066 in Petroleum geology of north-west Europe. Proceedings of the 4th conference. Parker, J R (editor). (London: The Geological Society).

Shannon, P M, Jacob, A W B, Makris, J, O’Reilly, B, Hauser, F, And Vogt, U. 1994. Basin evolution in the Rockall region, North Atlantic. First Break, Vol. 12, 515–522.

Shannon, P M, Jacob, A W B, Makris, J, O’Reilly, B, Hauser, F, And Vogt, U. 1995. Basin development and petroleum prospectivity of the Rockall and Hatton region. 435–457 in The petroleum geology of Ireland’s offshore basins. Croker, P F, And Shannon, P M (editors). Geological Society of London Special Publication, No. 93.

Shannon, P M, Jacob, A W B, O’Reilly, B, Hauser, F, Reader, P W, and Makris, J. 1999. Structural setting, geological development and basin modelling in the Rockall Trough. 421–431 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Shannon, P M, Faleide, J I, Smallwood, J R, and Walker, I M. 2005. The Atlantic margin from Norway to Ireland: geological review of a frontier continental margin province. 733–737 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Sinton, C W, Hitchen, K, and Duncan, R A. 1998. 40Ar–39Ar geochronology of silicic and basic volcanic rocks on the margins of the North Atlantic. Geological Magazine, Vol. 135, 161–170.

Smallwood, J R. 2004. Tertiary inversion in the Faroe–Shetland Channel and the development of major erosional scarps. 187–198 in 3D seismic technology: application to the exploration of sedimentary basins. Davies, R J, Cartwright, J A, Stewart, S S, Lappin, M, and Underhill, J R (editors). Geological Society of London Memoir, No. 29.

Smallwood, J R, and Gill, C E. 2002. The rise and fall of the Faroe–Shetland Basin: evidence from seismic mapping of the Balder Formation. Journal of the Geological Society of London, Vol. 159, 627–630.

Smallwood, J R, and Maresh, J. 2002. The properties, morphology and distribution of igneous sills: modelling, borehole data and 3D seismic from the Faroe–Shetland area. 271–306 in The North Atlantic igneous province: stratigraphy, tectonic, volcanic and magmatic processes. Jolley, D W, and Bell, B R (editors). Geological Society of London Special Publication, No. 197.

Smallwood, J R, and White, R S. 2002. Ridge-plume interaction in the North Atlantic and its influence on continental breakup and sea-floor spreading. 15–37 in The North Atlantic igneous province: stratigraphy, tectonic, volcanic and magmatic processes. Jolley, D W, and Bell, B R (editors). Geological Society of London Special Publication, No. 197.

Smith, D B. 1976. A review of the Lower Permian in and around the British Isles. 14–22 in The continental Permian in central, west, and south Europe. Falke, H (editor). (Dordrecht: Reidel.)

Smith, K, and Ziska, H. 2011. Devonian and Carboniferous. 79–91 in Geology of the Faroe–Shetland Basin and adjacent areas. Ritchie, J D, Ziska, H, Johnson, H, and Evans, D (editors). British Geological Survey Report, RR/11/01; Jarðfeingi Research Report, RR/11/01.

Smith, K, Whitley, P W, Kimbell, G S, Kubala, M, and Johnson, H. 2009. Enhancing the prospectivity of the Wyville Thomson Ridge. 286–302 in Proceedings of the 2nd conference. Varming, T, and Ziska, H (editors). Annales Societatis Scientiarum Faeroensis, Supplementum, Vol. 50.

Smith, W H F, and Sandwell, D T. 1997. Global sea-floor topography from satellite altimetry and ship depth soundings. Science, Vol. 277, 1956–1962.

Smythe, D K. 1989. Rockall Trough — Cretaceous or Late Palaeozoic? Scottish Journal of Geology, Vol. 25, 5–43.

Smythe, D K, Dobinson, A, McQuillin, R, Brewer, J A, Matthews, D H, Blundell, D J, and Kelk, B. 1982. Deep structure of the Scottish Caledonides revealed by MOIST reflection profile. Nature, Vol. 299, 338–340.

Snyder, D, and Flack, C A. 1990. A Caledonian age for reflectors within the mantle lithosphere north and west of Scotland. Tectonics, Vol. 9, 903–922.

Snyder, D, and Hobbs, R. 1999. The BIRPS Atlas II. A second decade of deep seismic reflection profiling[CD-ROM]. (London: The Geological Society.)

Snyder, D, England, R W, and McBride, J H. 1997. Linkage between mantle and crustal structures and its bearing on inherited structures in north-western Scotland. Journal of the Geological Society of London, Vol. 154, 79–83.

Souza, A J, Simpson, J H, Harikrishnan, M, and Malarkey, J. 2001. Flow structure and seasonality in the Hebridean slope current. Oceanologica Acta, Vol. 24, S63–S76.

Spencer, A M, and MacTiernan, B. 2001. Petroleum systems offshore western Ireland in an Atlantic margin context. 9–29 in The petroleum exploration of Ireland’s offshore basins. Shannon, P M, Haughton, P D W, and Corcoran, D V (editors). Geological Society of London Special Publication, No. 188.

Spencer, A M, Birkeland, Ø, Knag, G Ø, and Fredsted, R. 1999. Petroleum systems of the Atlantic margin and north-west Europe. 231–246 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Srivastava, S P. 1978. Evolution of the Labrador Sea and its bearing on the early evolution of the North Atlantic. Geophysical Journal of the Astronomical Society, Vol. 72, 373–399.

Srivastava, S P, and Tapscott, C R. 1986. Plate kinematics of the North Atlantic. 379–404 in The Geology of North America, Vol. M. The western North Atlantic Region. Vogt, P R, and Tucholke, B E (editors). Decade of North American Geology Series. (Boulder, Colorado: Geological Society of America.)

Steel, R J, and Wilson, A C. 1975. Sedimentation and tectonism (?Permo-Triassic) on the margin of the North Minch Basin, Lewis. Journal of the Geological Society of London, Vol. 131, 183–202.

Steers, J A. 1973. The coastline of Scotland. (Cambridge: Cambridge University Press.)

Stein, A M. 1988. Basement controls upon basin development in the Caledonian foreland, north-west Scotland. Basin Research, Vol. 1, 107–119.

Stein, A M. 1992. Basin development and petroleum potential in The Minches and Sea of Hebrides Basin. 17–20 in Basins on the Atlantic seaboard: petroleum geology, sedimentology and basin evolution. Parnell, J (editor). Geological Society of London Special Publication, No. 62.

Stein, A M, and Blundell, D J. 1990. Geological inheritance and crustal dynamics of the north-west Scottish continental shelf. Tectonophysics, Vol. 173, 455–467.

Stewart, H A, Davies, J, Long, D, Stromberg, H, and Hitchen, K. 2009. JNCC Offshore Nature Survey, Anton Dohrn Seamount and East Rockall Bank areas of search, 2009/03-JNCC cruise report. British Geological Survey Commissioned Report, CR/09/113.

Stewart, M. 1933. Notes on the Geology of the Sula Sgeir and Flannan Islands. Geological Magazine, 110pp.

Stoker, M S. 1982. Old Red Sandstone sedimentation and deformation in the Great Glen fault Zone, north-west of Loch Linnhe. Scottish Journal of Geology, Vol. 18, 147–156.

Stoker, M S. 1988. Pleistocene ice-proximal glaciomarine sediments in boreholes from the Hebrides Shelf and Wyville Thomson Ridge, north-west UK Continental Shelf. Scottish Journal of Geology, Vol. 24, 249–262.

Stoker, M S. 1990. Glacially influenced sedimentation on the Hebridean slope, north-western United Kingdom continental margin. 349–362 in Glacimarine environments: process and sediments. Dowdswell, J A, Scourse, J D (editors). Geological Society of London Special Publication, No. 53.

Stoker, M S. 1995a. Shallow sampling programme 1994 — stratigraphical sites. Volume 3, Appendix 2 of Stoker, M S, and Hitchen, K. 1995. Rockall Continental Margin Project: final geological report. British Geological Survey Technical Report, WB/95/11(5 volumes).

Stoker, M S. 1995b. The influence of glacigenic sedimentation on slope-apron development on the continental margin off north-west Britain. 159–177 in The tectonics, sedimentation and paleoceanography of the North Atlantic Region. Scrutton, R A, Stoker, M S, Shimmield, G B, and Tudhope, A W (editors). Geological Society of London Special Publication, No. 90.

Stoker, M S. 1997. Mid to late Cenozoic sedimentation on the continental margin off north-west Britain. Journal of the Geological Society of London, Vol. 154, 509–515.

Stoker, M S. 1998. Sediment-drift development on the continental margin off north-west Britain. 229–254 in Geological processes on continental margins: sedimentation, mass-wasting and stability. Stoker, M S, Evans, D, and Cramp, A (editors). Geological Society of London Special Publication, No. 129.

Stoker, M S. 1999. Stratigraphical nomenclature of the UK north-west margin 3 — mid to late Cenozoic stratigraphy. (Keyworth, Nottingham: British Geological Survey).

Stoker, M S. 2002. Late Neogene development of the UK Atlantic margin. 313–329 in Exhumation of the North Atlantic margin: timing, mechanisms and implications for petroleum exploration. Doré, A G, Cartwright, J A, Stoker, M S, Turner, J P, and White, N (editors). Geological Society of London Special Publication, No. 196.

Stoker, M S, and Bradwell, T. 2005. The Minch palaeo-ice stream, north-west sector of the British–Irish Ice Sheet. Journal of the Geological Society of London, Vol. 162, 425–428.

Stoker, M S, and Holmes, R. 1991. Submarine end-moraines as indicators of Pleistocene ice-limits off north-west Britain. Journal of the Geological Society of London, Vol. 148, 431–434.

Stoker, M S, and Varming, T. 2011. Cenozoic sediments. 151–208 in The geology of the Faroe–Shetland Basin and adjacent areas. Ritchie, J D, Ziska, H, Johnson, H and Evans, D (editors). British Geological Survey Research Report, RR/11/01 and Jardfeingi Research Report, RR/11/01.

Stoker, M S, Morton, A C, Evans, D, Hughes, M J, Harland, R, and Graham, D K. 1988. Early Tertiary basalts and tuffaceous sandstones from the Hebrides Shelf and Wyville Thomson Ridge, north-east Atlantic. 271–282 in Early Tertiary volcanism and the opening of the north-east Atlantic. Morton, A C, and Parson, L M (editors). Geological Society of London Special Publication, No. 39.

Stoker, M S, Harland, R, Morton, A C, and Graham, D K. 1989. Late Quaternary stratigraphy of the northern Rockall Trough and Faroe–Shetland Channel, north-east Atlantic Ocean. Journal of Quaternary Science, Vol. 4, 211–222.

Stoker, M S, Hitchen, K, and Graham, C C. 1993. UK Offshore Regional Report: The geology of the Hebrides and West Shetland shelves and adjacent deep-water areas. (London: HMSO for the British Geological Survey.)

Stoker, M S, Leslie, A B, Scott, W D, Briden, J C, Hine, N M, Harland, R, Wilkinson, I P, Evans, D, and Ardus, D A. 1994. A record of Late Cenozoic stratigraphy, sedimentation and climate change from the Hebrides slope, north-east Atlantic Ocean. Journal of the Geological Society of London, Vol. 151, 235–249.

Stoker, M S, Akhurst, M C, Howe, J A, and Stow, D A V. 1998. Sediment drifts and contourites on the continental margin off north-west Britain. Sedimentary Geology, Vol. 115, 33–51.

Stoker, M S, van Weering, T C E, and Svaerdborg, T. 2001. A Mid to Late Cenozoic tectonostratigraphic framework for the Rockall Trough. 411–438 in The petroleum exploration of Ireland’s offshore basins. Shannon, P M, Haughton, P D W, and Corcoran, D V (editors). Geological Society of London Special Publication, No. 188.

Stoker, M S, Praeg, D, Hjelstuen, B O, Laberg, J S, Nielsen, T, and Shannon, P M. 2005a. Neogene stratigraphy and the sedimentary and oceanographic development of the north-west European Atlantic margin. Marine and Petroleum Geology, Vol. 22, 977–1005.

Stoker, M S, Praeg, D, Shannon, P M, Hjelstuen, B O, Laberg, J S, Nielsen, T, van Weering, T C E, Sejrup, H P, and Evans, D. 2005b. Neogene evolution of the Atlantic continental margin of north-west Europe (Lofoten Islands to south-west Ireland): anything but passive. 1057–1076 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Stoker, M S, Hoult, R J, Nielsen, T, Hjelstuen, B O, Laberg, J S, Shannon, P M, Praeg, D, Mathiesen, A, van Weering, T C E, and McDonnell, A. 2005c. Sedimentary and oceanographic responses to early Neogene compression on the north-west European margin. Marine and Petroleum Geology, Vol. 22, 1031–1044.

Stoker, M S, Holford, S P, Hillis, R R, Green, P F, and Duddy, I R. 2010. Cenozoic post-rift sedimentation off north-west Britain: recording the detritus ofepisodic uplift on a passive continental margin. Geology, Vol. 38, 595–598.

Stoker, M S, Balson, P S, Long, D and Tappin, D. 2011. An overview of the lithostratigraphical framework for the Quaternary deposits on the United Kingdom, continental shelf. British Geological Survey Research Report, RR/11/03.

Štolfová, K, and Shannon, P M. 2009. Permo-Triassic development from Ireland to Norway: basin architecture and regional controls. Geological Journal, Vol. 44, 652–676.

Stone, J O, and Ballantyne, C K. 2006. Dimensions and deglacial chronology of the Outer Hebrides Ice Cap, north-west Scotland: implications of cosmic ray exposure dating. Journal of Quaternary Science, Vol. 21, 75–84.

Storetvedt, K M, and Steel, R J. 1977. Palaeomagnetic evidence for the age of the Stornoway Formation. Scottish Journal of Geology, Vol. 13, 263–269.

Stow, D A V, and Mayall, M. 2000. Deep-water sedimentary systems: new models for the 21st Century. Marine and Petroleum Geology, Vol. 17, 125–135.

Stow, D A V, Armishavo, J E, and Holmes, R. 2002. Holocene contourite sand sheet on the Barra Fan slope, NW Hebridean margin. 99–119 in Deep-water contourite systems: modern drifts and ancient series, seismic and sedimentary characteristics. Stow, D A V (editor). Geological Society of London Memoir, No. 22.

Strachan, P, and Evans, D. 1991. A local deep-water failure on the north-west slope of the UK. Scottish Journal of Geology, Vol.27, 107–111.

Strachan, R A, Smith, M, Harris, A L, and Fettes, D J. 2002. The Northern Highland and Grampian terranes. 81–147 in The geology of Scotland (4th edition). Trewin, N H (editor). (London: The Geological Society.)

STRATAGEM Partners. 2002. Stoker, M S (compiler). Neogene stratigraphy of the glaciated European margin from Lofoten to Porcupine. A product of the EC-supported STRATAGEM project. (Great Yarmouth: Svitzer.)

STRATAGEM Partners. 2003. Stoker, M S (compiler). Neogene evolution of the glaciated European margin. A product of the EC-supported STRATAGEM project. (Great Yarmouth: Svitzer.)

Surlyk, F. 1990. Timing, style and sedimentary evolution of Late Palaeozoic–Mesozoic extensional basins of East Greenland. 107–125 in Tectonic events responsible for Britain’s oil and gas reserves. Hardman, R P F, and Brooks, J (editors). Geological Society of London Special Publication, No. 55.

Stuart, F M, Bluck, B J, and Pringle, M S. 2001. Detrital muscovite 40Ar/39Ar ages from Carboniferous sandstones of the British Isles: provenance and implications for the uplift of orogenic belts. Tectonics, Vol. 20, 255–267.

Sutherland, D G, Ballantyne, C K, and Walker, M J C. 1984. Late Quaternary glaciation and environmental change on St Kilda, Scotland, and their palaeoclimatic significance. Boreas, Vol. 13, 261–272.

Sutton, J, and Watson, J. 1951. The pre-Torridonian metamorphic history of the Loch Torridon and Scourie areas in the north-west Highlands and its bearing on the chronological classification of the Lewisian. Quarterly Journal of the Geological Society of London, Vol. 106, 241–307.

Talbot, L A, Paul, M A, and Stoker, M S. 1994. Geotechnical and acoustic characteristics of Plio-Pleistocene sediments from the Hebrides Slope. Geo-Marine Letters, Vol. 14, 244–251.

Talwani, M, and Eldholm, O. 1972. Continental margin off Norway; a geophysical study. Geological Society of America Bulletin, Vol. 83, 3575–3606.

Talwani, M, and Eldholm, O. 1977. Evolution of the Norwegian–Greenland Sea. Geological Society of America Bulletin, Vol. 88, 969–999.

Tappin, D R, Tulloch, G J, and Wild, J B L. 2001. DTI 2000 Cruise: sea-bed geology, geohazards and sea-bed images (including former white zone, 19th round acreage). British Geological Survey Commissioned Report, CR/01/43.

Tate, M P. 1993. Structural framework and tectono-stratigraphic evolution of the Porcupine Seabight Basin, offshore western Ireland. Marine and Petroleum Geology, Vol. 10, 95–123.

Tate, M P, and Dobson, M R. 1989. Pre-Mesozoic geology of the western and north-western Irish continental shelf. Journal of the Geological Society of London, Vol. 146, 229–240.

Tate, M P, Dodd, C D, and Grant, N T. 1999. The North-east Rockall Basin and its significance in the evolution of the Rockall–Faroes/East Greenland rift system. 391–406 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Thiede, J, and Eldholm, O. 1983. Speculations about the paleodepth of the Greenland–Scotland Ridge during late Mesozoic and Cenozoic times. 445–456 in Structure and development of the Greenland–Scotland Ridge: new methods and concepts. Bott, M H P, Saxov, S, Talwani, M, and Thiede, J (editors). (New York: Plenum Publishing Corporation.)

Thiersten, H R, Geitznauer, K R, Molfino, B, and Shackleton, N J. 1977. Global synchroneity of late Quaternary coccolith datum levels: validation by oxygen isotopes. Geology, Vol. 5, 400–404.

Thomson, K. 2005. Extrusive and intrusive magmatism in the North Rockall Trough. 1621–1630 in Petroleum geology: north-west Europe and global perspectives. Proceedings of the 6th conference. Doré, A G, and Vining, B A (editors). (London: The Geological Society.)

Thomson, K, and Hutton, D. 2004. Geometry and growth of sill complexes: insights using 3D seismic from the North Rockall Trough. Bulletin of Volcanology, Vol. 66, 364–375.

Torsvik, T H and seven others. 1996. Continental breakup and collision in the Neoproterozoic and Palaeozoic of Baltica and Laurentia. Earth Science Reviews, Vol. 40, 229–258.

Trewin, N H, and Rollin, K E. 2002. Geological history and structure of Scotland. 1–25 in The geology of Scotland (4th edition). Trewin, N H (editor). (London: The Geological Society.)

Trewin, N H, and Thirlwall, M F. 2002. Old Red Sandstone. 213–249 in The geology of Scotland (4th edition). Trewin, N H (editor). (London: The Geological Society.)

Trueblood, S, and Morton, N. 1991. Comparative sequence stratigraphy and structural styles of the Slyne Trough and Hebrides Basin. Journal of the Geological Society of London, Vol. 148, 197–201.

Tuitt A, Underhill, J R, Ritchie, J D, Johnson, H, and Hitchen, K. 2010. Timing, controls and consequences of compression in the Rockall–Faroe area of the north-east Atlantic Margin. 963–977 in Petroleum geology: from mature basins to new frontiers. Proceedings of the 7th conference. Vining, B A, and Pickering, S C (editors). (London: The Geological Society.)

Turnbull, M J M, Whitehouse, M J, and Moorbath, S. 1996. New isotope age determination for the Torridonian. Journal of the Geological Society of London, Vol. 153, 955–964.

Turner, J D, and Scrutton, R A. 1993. Subsidence patterns in western margin basins: evidence from the Faroe–Shetland Basin. 975–983 in Petroleum geology of north-west Europe. Proceedings of the 4th conference. Parker, J R (editor). (London: The Geological Society.)

UKOOA. 2000. Atlantic margin environmental surveys of the sea floor 1996 & 1998. CD-ROM.

Unnithan, V, Shannon, P M, McGrane, K, Readman, P W, Jacob, A W B, Keary, R, and Kenyon, N H. 2001. Slope instability and sediment redistribution in the Rockall Trough: constraints from GLORIA. 439–454 in The petroleum exploration of Ireland’s offshore basins. Shannon, P M, Haughton, P D W, and Corcoran, D V (editors). Geological Society of London Special Publication, No. 188.

Upton, B G J. 1988. History of Tertiary activity in the North Atlantic borderlands. 429–453 in Early Tertiary volcanism and the opening of the north-east Atlantic. Morton, A C, and Parson, L M (editors). Geological Society of London Special Publication, No. 39.

Upton, B G J, Fitton, J G, and Macintyre, R M. 1987. The Glas Eilean lavas: evidence of a Lower Permian volcano-tectonic basin between Islay and Jura, Inner Hebrides. Transactions of the Royal Society of Edinburgh, Earth Sciences, Vol. 77, 289–293.

Vanneste, K, Henriet, J-P, Posewang, J, and Theilen, F. 1995. Seismic stratigraphy of the Bill Bailey and Lousy Bank area: implications for subsidence history. 125–139 in The tectonics, sedimentation and palaeoceanography of the North Atlantic region. Scrutton, R A, Stoker, M S, Shimmield, G B, and Tudhope, A W (editors). Geological Society of London Special Publication, No. 90.

Vanney, J R, and Stanley, D J. 1983. Shelfbreak physiography: an overview. 1–24 in The shelfbreak: critical interface on continental margins. Stanley, D J, and Moore, G T (editors). Society of Economic Paleontologists and Mineralogists Special Publication, No. 33.

Verhoef, J, Roest, W R, Macnab, R, Arkani-hamed, J. et al. 1996. Magnetic anomalies of the Arctic and North Atlantic Oceans and adjacent land areas. Geological Survey of Canada, Dartmouth NS. Open File Report No. 3125a.

Viana, A R, Kowsmann, R O, and Caddah, L F G. 1994. Architecture and oceanographic controls on the sedimentation of the Campos Basin continental slope. 14th International Sedimentary Congress, Abstracts Volume.

Viana, A R, Faugeres, J-C, and Stow, D A V. 1998. Bottom current controlled sand deposits–a review from modern shallow to deep water environments. Sedimentary Geology, Vol. 115, 53–80.

Vining, B A, and Pickering, S C (editors). 2010. Petroleum geology: from mature basins to new frontiers. Proceedings of the 7th Petroleum Geology Conference, Volumes 1 and 2. (London: The Geological Society.)

Vogt, P R. 1971. Asthenosphere motion recorded by the ocean floor south of Iceland. Earth and Planetary Science Letters, Vol. 13, 153–160.

Vogt, U, Makris, J, O’Reilly, B M, Hauser, F, Readman, P W, Jacob, A W B, and Shannon, P M. 1998. The Hatton Basin and continental margin: crustal structure from wide-angle seismic and gravity data. Journal of Geophysical Research, Vol. 103,12 545–12 566.

Waagstein, R. 1988. Structure, composition and age of the Faroe basalt plateau. 225–238 in Early Tertiary volcanism and the opening of the north-east Atlantic. Morton, A C, and Parson, L M (editors). Geological Society of London Special Publication, Vol. 39.

Waagstein, R, Morton, A C, Prægel, N-O, and Taylor, P N. 1989. Highly alkaline lapilli tuffs from Rosemary Bank: evidence for continental lithosphere beneath the northern part of the Rockall Trough. Terra Abstracts, Vol. 1, 31–32.

Waddams, P, and Cordingley, T. 1999. The regional geology and exploration potential of the north-east Rockall Basin. 379–390 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Walsh, A, Knag, G, Morris, M, Quinquis, H, Tricker, P, Bird, C, and Bower, S. 1999. Petroleum geology of the Irish Rockall Trough — a frontier challenge. 433–444 in Petroleum geology of north-west Europe. Proceedings of the 5th conference. Fleet, A J, and Boldy, S A R (editors). (London: The Geological Society.)

Warner, M, Morgan, J, Barton, P, Morgan, P, Price, C, and Jones, K. 1996. Seismic reflections from the mantle represent relict subduction zone within the continental lithosphere. Geology, Vol. 24, 39–42.

Warrington, G, Audley-Charles, M G, Elliott, R E, Evans, W B, Ivimey-cook, H C, Kent, P E, Robinson, P L, Shotton, F W, and Taylor, F M. 1980. A correlation of Triassic rocks in the British Isles. Geological Society of London Special Report, No. 13.

Westbrook, G. 1974. The South Harris magnetic anomaly. Proceedings of the Geologists Association, Vol. 35, 1–12.

White, N, and Lovell, B. 1997. Measuring the pulse of a plume with the sedimentary record. Nature, Vol. 387, 888–891.

White, R S. 1988. A hot-spot model for early Tertiary volcanism in the North Atlantic. 3–13 in Early Tertiary volcanism and the opening of the north-east Atlantic. Morton, A C, and Parson, L M (editors). Geological Society of London Special Publication, No. 39.

White, R S. 1992. Crustal structure and magmatism of North Atlantic continental margins. Journal of the Geological Society of London, Vol. 149, 841–854.

White, R S. 1997. Rift–plume interaction in the North Atlantic. Philosophical Transactions of the Royal Society of London, Vol. 355, 319–339.

White, R S, and McKenzie, D. 1989. Magmatism at rift zones: the generation of volcanic continental margins and flood basalts. Journal of Geophysical Research, Vol. 94, 7685–7729.

White, R S, Westbrook, G K, Fowler, S R, Spence, G D, Barton, P J, Joppen, M, Morgan, J, Bowen, A N, Prescott, C, and Bott, M H P. 1987. Hatton Bank (north-west UK) continental margin structure. Geophysical Journal of the Royal Astronomical Society, Vol. 89, 265–272.

Whitehouse, M J. 1990. An early Proterozoic age for the Ness anorthosite, Lewis, Outer Hebrides. Scottish Journal of Geology, Vol. 26, 131–136.

Whitehouse, M J. 1993. Age of the Corodale gneisses, South Uist. Scottish Journal of Geology, Vol. 29, 1–7.

Whitehouse, M J, and Bridgwater, D. 2001. Geochronological constraints on Palaeoproterozoic crustal evolution and regional correlations of the northern Outer Hebridean Lewisian complex, Scotland. Precambrian Research, Vol. 105, 227–245.

Whitmarsh, R B, Avedik, F, and Saunders, M R. 1986. The seismic structure of thinned continental crust in the northern Bay of Biscay. Geophysical Journal International, Vol. 86, 589–602.

Whitmarsh, R B, Manatschal, G, and Minshull, T A. 2001. Evolution of magma-poor continental margins from rifting to sea-floor spreading. Nature, 413, 150–154.

Whittington, G and Hall, A M. 2002. The Tolsta Interstadial, Scotland: a correlation with D–O cycles GI–8 to GI–5? Quaternary Science Reviews, Vol. 21, 901–915.

Williams, D F, Thunell, R C, Tappa, E, Rio, D, and Raffi, I. 1988. Chronology of the Pleistocene oxygen isotope record; 0–1.88 my BP. Palaeogeography, Palaeoclimatology, Palaeoecology, Vol. 64, 221–240.

Wilson, J B. 1979a. Biogenic carbonate sedimentation on the Scottish continental shelf and on Rockall Bank. Marine Geology, Vol. 33, M85–M93.

Wilson J B. 1979b. ‘Patch’ development of the deep-water coral Lophelia pertusa (L.) on Rockall Bank. Journal of the Marine Biology Association of the UK, Vol. 59, 165–177.

Wilson, L J, Austin, W E N, and Jansen, E. 2002. The last British ice sheet: growth, maximum extent and deglaciation. Polar Research, Vol. 21, 243–250.

Winchester, J A. 1988. Later Proterozoic environments and tectonic evolution in the northern Atlantic lands. 253–270 in Later Proterozoic stratigraphy of the Northern Atlantic regions. Winchester J A (editor). (Glasgow: Blackie.)

Wold, C N. 1994. Cenozoic sediment accumulation on drifts in the northern North Atlantic. Paleoceanography, Vol. 9, 917–941.

Wold, C N. 1995. Palaeobathymetric reconstruction on a gridded database: the northern North Atlantic and southern Greenland–Iceland–Norwegian sea. 271–302 in The tectonics, sedimentation and palaeoceanography of the North Atlantic region. Scrutton, R A, Stoker, M S, Shimmield, G B, and Tudhope, A W (editors). Geological Society of London Special Publication, No. 90.

Wright, S C. 1985. The study of depositional environments and diagenesis in the Durness Group of north-west Scotland. Unpublished PhD Thesis, University of Oxford.

Zachos, J, Pagani, M, Sloan, L, Thomas, E, and Billups, K. 2001. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science, Vol. 292, 686–693.

Ziegler, P A. 1988. Evolution of the Arctic–North Atlantic and the Western Tethys. Memoir of the American Association of Petroleum Geologists, No. 43.

Ziegler, P A. 1990. Geological Atlas of Western and Central Europe (2nd edition). (Bath: Geological Society for Shell International Petroleum Maatschappij, B V.)

Ziska, H, and Varming, T. 2008. Palaeogene evolution of the Ymir and Wyville Thomson ridges, European North Atlantic margin. 153–168 in The nature and origin of compression in passive margins. Johnson, H, Dore, A G, Gatliff R W, Holdsworth R, Lundin E R, and Ritchie J D (editors). Geological Society of London Special Publication, No. 306.

Figures and tables

Figures

(Figure 1) The report area in relation to the north-east Atlantic Ocean.

(Figure 2) Location of BGS boreholes and commercial wells within, and adjacent to, the report area.

(Figure 3) Bedrock geology within, and adjacent to, the report area. In the offshore basinal area, the map represents the subcrop to the C10 unconformity (intra-Early Pliocene).

(Figure 4) Timescale and associated terminology used in this report.

(Figure 5) Gravity anomaly map (free-air anomolies offshore and Bouger anomolies onshore). The percentages on Rockall Bank are confidence limits of possible natural oil seeps (see Economic chapter). Structural features picked out by the gravity data are: AC=Ardnamurchan Volcanic Centre, ADS=Anton Dohrn Seamount, DC=Darwin Volcanic Centre, DGH=Darwin Geikie High, EGBC=East George Bligh Volcanic Centre, ERHB=East Rockall High Basin, FB=Flannan Basin, FBCK=Faroe Bank Channel Knoll Volcanic Centre, GK=Geikie Volcanic Centre, HB=Hatton Basin, HTS=Hebrides Terrace Seamount, LH=Lousy High, MB=Minch Basin, MUC=Mull Volcanic Centre, NERB=North-east Rockall Basin, NLB=North Lewis Basin, NRB=North Rockall Basin, NWRB=North-west Rockall Basin, OHH=Outer Hebrides High, RB=Rónán Basin, RBS=Rosemary Bank Seamount, RC=Rockall Volcanic Centre, RH=Rockall High, RHC=Rum Volcanic Centre, SGC=Sula Sgeir Volcanic Centre, SH=Stanton High, SKC=St Kilda Volcanic Centre, SRB=south Rockall Basin, SS=Sigmundur Volcanic Centre, SWC=Swithin Volcanic Centre, SYC=Skye Volcanic Centre, WFB=West Flannan Basin, WGBC=West George Bligh Volcanic Centre, WLB=West Lewis Basin, WLH=West Lewis High, WTR=Wyville Thomson Ridge, YR=Ymir Ridge.

(Figure 6) Magnetic (total field) anomaly map. Abbreviations as for (Figure 5).

(Figure 7) Structural elements within the Rockall report and adjacent area. Mainly based on British Geological Survey (2002; 2007), British Geological Survey/Petroleum Affairs Division (Ireland) (2002), Fyfe et al. (1993), Stoker et al. (1993), Waddams and Cordingley (1999), Tate et al. (1999), Keser Neish (2003), Kimbell et al. (2005) and Naylor et al. (1999).

(Figure 8) Generalised summary of the tectonic and plate tectonic events that affected the north-east Atlantic margin including the Rockall Basin and adjacent areas. Timescale is based on ‘A Geologic Time Scale 2004’ by Gradstein et al., 2004, with additions. Abbreviations: FSB=Faroe–Shetland Basin, RB=Rockall Basin, WTRC=Wyville Thomson Ridge Complex.

(Figure 9) Evolution of the Caledonides with regard to the North Atlantic area (modified reproduction from Coward et al. (2003) by permission of the Geological Society, London). Abbreviations: RB=Rockall Basin area.

(Figure 10) a) Palaeogeographical and palaeofacies plate reconstruction of the Rockall Basin and surrounding area for early Permian times (modified reproduction from Coward et al., 2003 by permission of The Geological Society, London). b) Global early Permian plate reconstruction (after Coward et al., 2003). Abbreviations: FSB=Faroe–Shetland Basin, MB=Møre Basin, MF=Moray Firth, NPB=Northern Permian Basin, RB=Rockall Basin, SPB=Southern Permian Basin, VB=Vøring Basin, WOB=West Orkney Basin, WSB=West Shetland Basin.

(Figure 11) Palaeogeographical and palaeofacies plate reconstruction of the Rockall Basin and surrounding area for Mid Triassic times (modified reproduction from Coward et al., 2003 by permission of The Geological Society, London). Abbreviations as in (Figure 10) except: CNS=Central North Sea, EB=Erris Basin, FB=Flannan Basin, PB=Porcupine Basin, SB=Slyne Basin, VG=Viking Graben, UB=Unst Basin, WFB=West Flannan Basin, WLB=West Lewis Basin.

(Figure 12) Palaeogeographical and palaeofacies plate reconstruction of the Rockall Basin and surrounding area for Late Jurassic times (modified reproduction from Coward et al., 2003 by permission of The Geological Society, London). Abbreviations as for (Figure 10) and (Figure 11) except: HT=Halten Terrace, SHLMB=Sea of Hebrides–Little Minch Basin.

(Figure 13) Palaeogeographical and palaeofacies plate reconstruction of the Rockall Basin and surrounding area for Early Cretaceous times (modified reproduction from Coward et al., 2003 by permission of The Geological Society, London). Abbreviations as for Figure 10) and (Figure 11) except: BR=Barra Volcanic Ridge Complex, CSB=Celtic Sea Basin, EWF=End of the World Fault, MF=Moray Firth.

(Figure 14) (opposite) Crustal structure of the Rockall Basin area interpreted from the results of wide-angle seismic reflection/refraction experiments. Sections: AMP-E from Klingelhöfer et al. (2005). (b) Line 86-002 and (c) 86-005 from Roberts et al. (1988). (d) CDP87-1/2/3 from Keser Neish (1993). (e) RAPIDS from Shannon et al. (1999). Soundings: (f) Line A of Bunch (1979). (g) PUMA from Powell and Sinha (1987).

(Figure 15) Line drawings from deep near-normal-incidence seismic reflection profiles across the eastern side of the report area. (a) GECO MP-1 after Stein and Blundell (1990). Reprinted from Tectonophysics, Vol. 173, 455–467. 'Geological inheritance and crustal dynamics of the north-west Scottish continental shelf', with permission from Elsevier. (b) DRUM after McGeary et al. (1987). (c) and (d) WINCH after Brewer et al. (1983). Reprinted with permission from MacMillan Publishers Ltd: Nature, Vol. 305, 206–210.

(Figure 16) Regional crustal structure from 3D modelling (Kimbell et al., 2004, 2005). The geometry of the model components was optimised to provide the best match between observed and calculated gravity anomalies. (a) Modelled thickness of the sedimentary layer. (b) Modelled thickness of crystalline crust. (c) Modelled depth to Moho. (d) Apparent extension factor (based on an initial crystalline crustal thickness of 30 km). Abbreviations as in (Figure 7).

(Figure 17) (a) Restoration of late Caledonian strike-slip deformations modified after Snyder et al. (1997) and Snyder and Hobbs (1999). Blue labels and contours indicate depth to mantle reflectors detected by BIRPS (Snyder and Hobbs, 1999) and LISPB (Barton, 1992) experiments. Abbreviations: GGF=Great Glen Fault, HBF=Highland Boundary Fault, FCL=Fairhead Clew Bay Line, FSB=Faroe Shetland Basin, WOB= West Orkney Basin, WTR=Wyville Thomson Ridge. (b) Locations of seismic lines on which the mantle reflectors have been detected.

(Figure 18) Generalised BGS interpretation of deep seismic profile M89-WB-02. The west-trending profile obliquely traverses the north-west-trending Anton Dohrn Lineament that is interpreted as separating the north and south Rockall basins. For location, see (Figure 7).

(Figure 19) Isostatically corrected Bouguer gravity anomaly over the report area. The thick yellow line indicates the boundary between satellite-derived gravity data to the west and marine and land data to the east. The bathymetry used in estimating the gravity corrections was based on the model of Smith and Sandwell (1997), which employs a combination of ship soundings and satellite altimetry. For abbreviations, see (Figure 7).

(Figure 20) Generalised geological cross-sections across the North-east Rockall Basin, north and south Rockall basins and adjacent areas constructed from deep seismic reflection profiles (a) DGH95-10, M89-WB-02, and (c) WESTLINE, respectively (all modified after Nadin et al., 1999). For locations, see (Figure 7).

(Figure 21) Generalised geological cross-sections across (a) the north Rockall Basin, Anton Dohrn Seamount and Outer Hebrides High, (b) the Rockall High, south Rockall Basin, Outer Hebrides High and Barra Basin and (c) the Rockall High, Rónán Basin and Ladra High (based on BGS and PAD, 2002, cross-sections 1, 2, and 3 respectively). For location, see (Figure 7).

(Figure 22) Idealised model regarding the development of the north and south Rockall basins and adjacent areas based on deep seismic profile M89-WB-02 (modified after Musgrove and Mitchener, 1996). For location, see (Figure 7).

(Figure 23) Interpreted BGS seismic profile across the Rockall High/East Rockall High Basin and the south Rockall Basin. For location, see (Figure 7).

(Figure 24) Interpreted seismic profile across the north Rockall Basin, Darwin Volcanic Centre, North-east Rockall Basin, West Lewis High and West Lewis Basin (modified after Tate et al., 1999; Archer et al., 2005 and A Tuitt, personal communication). For location, see (Figure 7). Abbreviations: NL=Ness Lineament.

(Figure 25) Generalised geological cross-section constructed from potential field interpretations across the Darwin–Geikie High, North-east Rockall Basin, West Lewis High, West Lewis Basin and Sula Sgeir High (modified from Waddams and Cordingley, 1999). For location, see (Figure 7).

(Figure 26) Interpreted seismic profile across the north Rockall Basin, Wyville Thomson Ridge, Ymir Ridge and Faroe Bank Channel Basin area (modified from Johnson et al., 2005). For location, see (Figure 7).

(Figure 27) Generalised geological cross-section, interpreted seismic profile and gravity and magnetic response across the Flannan Basin and Flannan High (modified from BGS, 1990, cross-section 3). For location, see (Figure 7).

(Figure 28) Generalised geological cross-section across the Outer Hebrides High, Sula Sgeir High and North Lewis Basin (modified from BGS 1990, cross-section 2). For location, see (Figure 7).

(Figure 29) Generalised geological cross-section and gravity and magnetic response across the North-east Rockall Basin, West Flannan Basin and Flannan High (modified from BGS, 1990, cross-section 1). For location, see (Figure 7).

(Figure 30) Conceptual geological cross-section (not to scale) across the North-east Rockall Basin, West Lewis High, West Lewis Basin and western edge of the Sula Sgeir High (modifid from Isaksen et al., 2000). For general location, see (Figure 7).

(Figure 31) Structural model for the formation of the Munkagrunnur, Wyville Thomson and Ymir ridges (modified from Tate et al., 1999).

(Figure 32) The distribution of basement terranes within and immediately adjacent to the Rockall regional report area (inset b, partly modified terrane distribution model after Kinny et al., 2005). Abbreviations as in (Figure 7) except: ASZ=Alasdair Shear Zone, CG=Corodale Gneiss, ESZ=Ensay Shear Zone, GCT=Gairloch Terrane, GHT=Grampian Highlands Terrane, GT=Gruinard Terrane, IT=Ialltaig Terrane, LFSZ=Langavat–Finsbay Shear Zone, NWHT=North-west Highlands Terrane, NST=Nis Terrane, OIF=Outer Isles Fault, RLT=Roineabhal Terrane, ROT=Rona Terrane, SSZ=Shieldaig Shear Zone, TT=Tarbet Terrane, UBB=Uist/Barra Block.

(Figure 33) Photomicrographs of metamorphic basement core in crossed polarised light from (a) a metabasic gneiss from well 164/25-2 containing plagioclase, clinopyroxene, garnet, orthopyroxene, hornblende, quartz and minor biotite, apatite, opaque minerals (P808713), (b) an amphibolite from well 154/03-1 containing amphibole, plagioclase, quartz, biotite and accessory titanite, opaque minerals, apatite (P808714) and (c) a tonalitic granulite from BGS shallow sample site 56-15/18 containing plagioclase, quartz, clino-pyroxene, orthopyroxene, K-feldspar, hornblende and minor opaque minerals, apatite, rutile, biotite, zircon (P808715). Field of view = 2 mm. For location, see (Figure 32).

(Figure 34) Photographs of metamorphic basement core from (a) an interbanded amphibolite/quartzo-felspathic gneiss from BGS shallow drill site 56-08/920 (P808716) and (b) a dioritic gneiss basement core from 57-09/537 (P808717). For location, see (Figure 32).

(Figure 35) North-east Atlantic margin mid Palaeozoic reconstruction showing terrane juxtapositions between offshore north-west UK and East Greenland. Plate reconstructions were produced using Atlas plate reconstruction software, developed by Cambridge Paleomap Services Limited (CPSL). The reconstruction is based on a best fit using the 1000 m bathymetric contour. Grey and hachured areas represent plate overlap and gap mismatches, respectively. Abbreviations as in (Figure 7) except AMB=Ammassalik Mobile Belt, CGC=Central Greenland Craton, LGC=Lewisian Gneiss Complex, NAC=North Atlantic Craton.

(Figure 36) Evolution of the Caledonides in the North Atlantic region. Modified from Coward et al., (2003).

(Figure 37) Palinspastic maps of active fault structures and sediment facies in the North Atlantic region during the (a) Devonian, (b) early Carboniferous, and (c) late Carboniferous. Modified and simplified after Coward et al. (2003). Abbreviations: CB=Clair Basin, GG=Great Glen Fault Zone, MTF=Møre–Trøndelag Fault Zone, OB=Orcadian Basin, SF=Solund Fault Zone.

(Figure 38) Distribution of Cambro-Ordovician rocks, and location of main Devono-Carboniferous basins, smaller outcrops, and other indicators referred to in the text, within and adjacent to the report area, including the West Shetland region (inset map). Structural framework is simplified from (Figure 7).

(Figure 39) Stratigraphical range chart for the Cambrian to Carboniferous, showing the relative ages of the Palaeozoic sequences preserved adjacent to the report area. Timescale after Gradstein et al. (2004).

(Figure 40) Series and global/regional stage boundaries for the Carboniferous Period. Regional stage names are those utilised in Western Europe. After Gradstein et al. (2004).

(Figure 41) Pangaean orogenic belts and Permo-Triassic basins, slightly modified after Doré et al. (1999). Note that this reconstruction depicts a rather limited distribution of Permo-Triassic basins and contrasts with those of Coward et al. (2003) and Zeigler (1990) (see (Figure 10) and (Figure 11). Abbreviations: WSB=West Shetland Basin, SH=Sea of the Hebrides Basin, EB=Erris Basin.

(Figure 42) Distribution of Permo-Triassic rocks within the Rockall report area. Abbreviations: see (Figure 7).

(Figure 43) Summary lithological and wireline log character of the (presumed) Permo-Triassic succession in well 164/25-1Z and the Permian reservoir succession in Irish well 12/2-1Z (the ‘Dooish’ condensate discovery).

(Figure 44) The distribution of Jurassic strata in, and adjacent to, the report area.

(Figure 45) Lithological and age data for Jurassic strata recovered from offshore wells and bore-holes in and around the report area.

(Figure 46) Regional palaeogeographical maps showing distribution of Early and Late Cretaceous deep-water sediments (based on Millennium atlas maps; Evans et al., 2003). AD=Anton Dohrn, BVRS=Barra Volcanic Ridge System, RB=Rosemary Bank.

(Figure 47) Wells and boreholes proving Cretaceous strata: simplified lithologies and age ranges on a summary stratigraphical chart.

(Figure 48) Distribution of Cretaceous rocks in the report area; with location of wells and boreholes proving Cretaceous. EB=Erris Basin, NERB=North East Rockall Basin, WFB=West Flannan Basin, WLB=West Lewis Basin.

(Figure 49) Geoseismic section from the Irish sector showing the structural setting of the Cretaceous in the Erris Basin in the vicinity of Irish well 12/13-1A (redrawn from Chapman et al. (1999).

(Figure 50) Geoseismic section across the eastern margin of the Rockall Basin in the UK sector, passing through well 132/15-1. The interpretation of expanding spread profile ESP-12 (Joppen and White, 1990) from the centre of the Rockall Basin in the Irish sector is shown for comparison.

(Figure 51) Reconstruction of the northern Atlantic region at 55 Ma, shortly after the onset of sea-floor spreading (modified after White, 1992). Coloured area shows known extent of lava flows and sills emplaced during continental breakup with dykes shown as thin lines. Note that the extent of dykes emplaced beneath Greenland is unknown due to ice cover. The reconstruction uses an equal-area Lambert stereographic projection centred on the core of the Iceland Plume, and encompasses an area with a radius of 1500 km.

(Figure 52) Characterisation of crust in the north-east Atlantic (modified after White, 1997). FI=Faroe Islands, FIR=Faroe–Iceland Ridge, GB=Great Britain, GIR=Greenland–Iceland Ridge, IB=Iceland Basin, ICE=Iceland, IrB=Irminger Basin, KR=Kolbeinsey Ridge, RB=Rockall Basin, RH=Rockall Plateau, RR=Reykjanes Ridge.

(Figure 53) Distribution of igneous rocks within the Rockall Basin area. The wells and boreholes shown penetrated igneous intervals. The short sea-bed cores and dredges recovered igneous material. ADS=Anton Dohrn Seamount, AC=Ardnamurchan Volcanic Centre, BBC=Blackstones Bank Volcanic Centre, DKC=Drekaeyga Volcanic Centre, DRC=Darwin Volcanic Centre, EGB=East George Bligh intrusion, GC=Geikie Volcanic Centre, HTS=Hebrides Terrace Seamount, MC=Mull Volcanic Centre, RBS=Rosemary Bank Seamount, RKC=Rockall Volcanic Centre, RC=Rum Volcanic Centre, SGC=Sigmundur Volcanic Centre, SC=Skye Volcanic Centre, SKC=St Kilda Volcanic Centre, SSC=Sula Sgeir Volcanic Centre, SWC=Swithin Volcanic Centre, WGB=West George Bligh intrusion.

(Figure 54) BGS high-resolution seismic line 03/01-1 across Rosemary Bank Seamount (located in (Figure 53)). The former volcano appears to have been eroded and tilted to the south-east. BGS borehole BH90/18 proved 23.53 m of Neogene and ?Maastrichtian sediments overlying the volcanic rocks.

(Figure 55) Photograph of lavas (part of short sea-bed short core 59-11/12) obtained from the top of Rosemary Bank Seamount. For location, see (Figure 53). Five individual lava flows were penetrated by the core (one flow top is shown, left end of lower core sample). The lavas are highly potassic and some of the vesicles are flow aligned (P808718).

(Figure 56) Photograph of a conglomerate (part of short sea-bed short core 57-12/18) obtained from the top of Anton Dohrn Seamount. For location, see (Figure 53). Angular clasts of basaltic lava, one of which has been dated at 54 Ma (Ypresian, Early Eocene), are contained in a Middle to Upper Eocene carbonate matrix (P808719).

(Figure 57) Seismic line across the Outer Hebrides High (located in (Figure 53)) illustrating large-scale lava progrades (probably formed in a submarine environment) overlain by more horizontal lavas flows (probably subaerial). The Hebridean Escarpment may mark the position of a contemporary shoreline.

(Figure 58) Logs through the igneous section penetrated by well 164/07-1 in the North-east Rockall Basin (located in (Figure 53)). The well proved c. 1287.5 m of extrusive lavas, breccias and tuffs overlying a Cretaceous interval intruded by at least 70 sills (Archer et al., 2005). MD=Measured depth, TVDSS=True vertical depth sub sea.

(Figure 59) BGS seismic line 93/02-D2 (located in (Figure 53)) from the western side of the Rockall Basin. The short, discontinuous high-amplitude reflections are caused by sills intruded into Upper Cretaceous sedimentary rocks. The forceful nature of the intrusions is indicated by disruption of the contemporary overburden. Imaging of the deeper Mesozoic section is degraded by the presence of the sills.

(Figure 60) Distribution of Cenozoic strata at or near to outcrop.

(Figure 61) Geoseismic and schematic sections across the Hebrides–Rockall region showing the general relationships of the Cenozoic megasequences. (b) is a composite across-basin profile. Sections 1, 3 and 4 are modified from BGS North and Central Rockall Basin Solid Geology map sheets; section 2 is modified from STRATAGEM Partners (2003). Discrete Palaeogene structural elements are highlighted in red text. Abbreviations: GU=Glacial Unconformity, BN=Base Neogene boundary.

(Figure 62) Cenozoic event stratigraphy for the Rockall Basin. Timescale is from Gradstein et al. (2004); the sea-level curves and indication of global ice extent is from Abreu and Anderson (1998); tectonic and other events derived from a variety of sources (see text for details).

(Figure 63) Boreholes and other features referred to in the text, located beyond the report area. Abbreviation: RSG=Irish Rockall Studies Group.

(Figure 64) (a) Palinspastic map for the Paleocene–Early Eocene interval. Abbreviations: VB=Vøring Basin, MB=Møre Basin, FSB=Faroe–Shetland Basin, NERB=North-east Rockall Basin, NRB=north Rockall Basin, SRB=south Rockall Basin, HB=Hatton Basin, NSB=North Sea Basin, PB=Porcupine Basin, RBS=Rosemary Bank Seamount, ADS=Anton Dohrn Seamount, HTS=Hebrides Terrace Seamount. (b) Global view of Paleocene tectonics and ocean circulation at about 60–65 Ma. Maps based on information derived from Saunders et al. (1998), Naylor et al. (1999), Brekke (2000), Redfern (2000), Faleide et al. (2002), Mosar et al. (2002), Coward et al. (2003) and BGS Central and North Rockall Basin Solid Geology sheets (BGS and PAD, 2002; BGS, 2007).

(Figure 65) (a) Palinspastic map for the Late Eocene–Oligocene interval. Abbreviations as in (Figure 64) except: GSR=Greenland–Scotland Ridge, JMFZ=Jan Mayen Fracture Zone, IB=Iceland Basin, IRB= Irminger Basin. (b) Global view of Oligocene tectonics and ocean circulation at about 30 Ma. Maps based on information derived from Thiede and Eldholm (1983), Jansen and Raymo (1996), Andersen et al. (2000), Redfern (2000), Faleide et al. (2002), Coward et al. (2003), Johnson et al. (2005) and Stoker et al. (2005a, b).

(Figure 66) (a) Palinspastic map for the Mid to Late Miocene interval. Abbreviations as in Figures C4 and C5 except: FSC=Faroe–Shetland Channel, FBC=Faroe Bank Channel, MR=Munkagrunnur Ridge, WTR=Wyville Thomson Ridge, JMR=Jan Mayen Ridge, GB=Greenland Basin, LB=Lofoten Basin, NB=Norwegian Basin, DS=Denmark Strait. (b) Global view of Late Miocene tectonics and ocean circulation at about 10 Ma. Maps based on information derived from Jansen and Raymo (1996), Andersen et al. (2000), Redfern (2000), Galloway (2001), Faleide et al. (2002), Huuse (2002), Coward et al. (2003), STRATAGEM Partners (2003) and Stoker et al. (2005a, b).

(Figure 67) a) Palinspastic map for the Pliocene–Pleistocene interval, after 4 Ma. Abbreviations as in (Figure 63)(Figure 64)(Figure 65) except: BIF=Bear Island Fan, MNW=Mid Norwegian Wedge, NSF=North Sea Fan, FW=Foula Wedge, RW=Rona Wedge, EFW=East Faroe Wedge, WFW=West Faroe Wedge, SSF=Sula Sgeir Fan, BF=Barra Fan, DF=Donegal Fan, ERW=East Rockall Wedge. (b) Global view of Pliocene tectonics and ocean circulation at about 4 Ma, but incorporating the maximum extent of northern hemisphere glaciation since about 2.74 Ma. Maps based on information derived from CLIMAP (1981), Eiriksson (1981), Thiede and Eldholm (1983), Jansen and Raymo (1996), Japsen and Chalmers (2000), Redfern (2000), Faleide et al. (2002), Huuse (2002), Coward et al. (2003), STRATAGEM Partners (2003) and Stoker et al. (2005a, b).

(Figure 68) Distribution of Palaeogene sedimentary and volcanic rocks.

(Figure 69) Stratigraphical range chart, thickness (metres) and generalised lithology of the Palaeogene. Data from BGS boreholes and released commercial wells. Timescale is from Gradstein et al. (2004). Sites are located in (Figure 63) and  (Figure 68).

(Figure 70) Summary of Palaeogene stratigraphical and lithological data obtained from short rock cores. All are BGS samples, except cores 44 (Jones et al., 1986) and 7710 (Ferragne et al., 1984). Cores are located in (Figure 68).

(Figure 71) (a) BGS airgun profile 92/01-18 showing Eocene sediments (RPd megasequence) exposed at sea bed in the north-west Rockall Basin. (b) BGS airgun profile 92/01-8 showing collapsed and deformed Eocene shelf-margin succession adjacent to Rockall Bank. Red text highlights Palaeogene sub-basin terminology. Sections located in (Figure 68).

(Figure 72) Simplified log of BGS borehole BH94/3 and interpreted line drawing of part of BGS airgun profile 92/01-26, which penetrated a Lower to Middle Eocene prograding wedge (East Rockall Wedge) developed on the eastern flank of the Rockall High.

(Figure 73) Core photographs showing: (a) gravelly bioclastic sandstone incorporating rounded pillow lava fragments with chilled margins (Lower Eocene, BGS borehole BH94/3, 207.6 m) (P808720), (b) bioturbation of tuffaceous sandstone and mudstone (Lower to Middle Eocene, BGS borehole BH94/3, 124.2 m) (P808721), (c) bivalves in sandstone, possibly in life position (Middle to Upper Eocene, BGS borehole BH94/2, 16.3 m) (P808722) and (d) photomicrograph of bioclastic limestone showing open framework of poorly sorted debris, including echinoderm fragments that display syntaxial calcite overgrowths on echinoderm fragments (Upper Oligocene, BGS borehole BH94/4, 47.8 m) (P808723).

(Figure 74) BGS sparker profile 84/06-17 and interpreted line drawing across the upper Hebrides Slope showing the shelf-margin succession calibrated to BGS borehole BH88/7,  BH88/7A, and core 44 (Jones et al., 1986). Section located in (Figure 68). Modified from Stoker (2002).

(Figure 75) Distribution and thickness of Miocene–Lower Pliocene sediments (megasequence RPb).

(Figure 76) Distribution and thickness of Lower Pliocene–Holocene sediments (megasequence RPa).

(Figure 77) Stratigraphical range chart, thickness (in metres) and generalised lithology of the Neogene–Quaternary succession. Data from BGS boreholes, DSDP/ODP sites and released commercial wells. Interpretation of borehole data based on information derived from Stoker et al. (2001), Stoker (2002) and STRATAGEM Partners (2002). Timescale is from Gradstein et al. (2004). Sample sites on map and in table are located in (Figure 63), (Figure 75) and (Figure 76).

(Figure 78) Summary of selected Neogene–Quaternary stratigraphical and lithological data obtained from short cores. All are BGS samples, except core MD95-2006 (Knutz et al., 2001, 2002a, b) and core MD04-2822 (Hibbert et al., 2010). Cores are located in (Figure 75) and (Figure 76).

(Figure 79) Late Neogene–Quaternary stratigraphical framework for the Hebridean margin. Abbreviation: GU=Glacial Unconformity.

(Figure 80) (a) Interpreted geoseismic section of BGS airgun profile 92/01-24 across the western flank  of the Rockall Basin. Inset shows detail of erosional sea bed, the relict Middle Miocene to Lower Pliocene (RPb megasequence) contourite sediment drift, and locally exposed Eocene (RPd megasequence) strata.(b) BGS sparker profile 92/01-35, located south-west of profile 24, showing the Early Pliocene unconformity (C10) preserved beneath a drape of Lower Pliocene–Holocene sediments, together with the location of BGS borehole BH94/1. Locations of profiles are shown on (Figure 75). Modified from Stoker (2002).

(Figure 81) a) BGS airgun profile 83/04-6 across the Hebridean margin showing the Lower Pliocene–Holocene slope apron of the prograding Sula Sgeir Fan overlying and partly interbedded with onlapping and upslope migrating contourite drift and sediment waves, which both comprise the RPa megasequence. Submarine end moraines are preserved above the Mid Pleistocene glacial unconformity. Modified after Stoker et al. (2005a). (b) BGS airgun profile 92/01-56 extending across the lower part of the Barra Fan and into the Rockall Basin, and showing the Early Pliocene unconformity (C10) truncating the underlying Middle Miocene–Lower Pliocene strata (RPb megasequence), and the lateral relationship between the Lower Pliocene–Holocene shelf-margin and basinal strata. The insets highlight the erosive nature of C10, and the interdigitating nature of the debris flow and deep-marine deposits. Modified after Stoker (2002). SBM=sea-bed multiple, BP=bubble pulse. Locations of profiles are shown in (Figure 76).

(Figure 82) Royal NIOZ sleevegun profile 97–12 across the southern Rockall Basin showing the seismic characteristics of the Cenozoic succession, including the Feni Ridge contourite drift, smaller subsidiary drift adjacent to the Porcupine High, sediment waves at sea bed, and wide erosional moats adjacent to the highs. Insets show detail of the basin-margin and moated areas, particularly the erosional nature of C10, the onlapping character of C10 and C20, and the irregular slumped topography associated with C30. Modified after Stoker et al. (2001). Location of profile is shown in (Figure 63).

(Figure 83) BGS airgun profile 85/05-4 showing the late Neogene–Quaternary sediment drift complex that characterises the RPa megasequence in the North-east Rockall Basin adjacent to the Wyville Thomson Ridge, including elongate-mounded and broad-sheeted drifts and associated sediment waves. Inset shows sediment waves that display an upward change in geometry from asymmetric to sinusoidal, which may be indicative of a general long-term decrease in bottom-current strength. The waves retain some sea-bed expression except where locally overlain by debris-flow deposits associated with the Sula Sgeir Fan. Modified after Stoker et al. (1998). SBM=sea-bed multiple. Location of profile is shown in (Figure 76).

(Figure 84) Glacial geomorphology of the Hebridean region, based on information derived from Selby (1989), Stoker and Holmes (1991), Stoker et al. (1993), Stoker and Bradwell (2005), and Bradwell et al. (2008). Chart shows the glaciation curve for the Hebridean margin based on the stratigraphical range of the sediments recovered in BGS borehole BH88/7,  BH88/7A (Stoker et al. 1994), core MD95-2006 (Kroon et al., 2000; Knutz et al., 2001), and core MD04-2822 (Hibbert et al. (2010). British chronostratigraphical stages are based on Bowen (1999); the ages of marine isotope stage boundaries (MIS) are based on Martinsen et al. (1987) and Williams et al. (1988). Note the uncertainty in the timing of ‘Wolstonian’ events recognised in borehole BH88/7, BH88/7A.

(Figure 85) Isopachs for the interval between the sea bed and the base of the hydrate stability zone.

(Figure 86) Detail of sea-bed sediments on the Hebrides Shelf and adjacent areas.

(Figure 87) Physiography of the report area, showing the major morphological features.

(Figure 88) Slope profiles across the major morphological features. AD=Anton Dohrn Seamount, FR=Flannan Ridge, GBB=George Bligh Bank, HS=Hebrides Shelf, HT=Hebrides Terrace Seamount, OHP=Outer Hebrides Platform, RB=Rockall Bank, RBa=Rosemary Bank Seamount, RT=Rockall Trough, SB=Stanton Banks.

(Figure 89) Multibeam data (courtesy of NERC and SEA7) depicting Rosemary Bank. (a) Vertical view. (b) View from the west. Depth scales in metres (projection WG84). Rosemary Bank has been the focus of three surveys between 2003 and 2006. The first survey was carried out from the RRS James Clark Ross during 2003 (data supplied by Dr R Larter, see also Howe et al. 2006) followed by two surveys funded by then Department of Trade and Industry (now the Department of Environment and Climate Change) and the Department of Environment, Food and Rural Affairs as part of their Strategic Environmental Assessment (SEA) acquiring data in 2005 and 2006.

(Figure 90) Plot showing persistent residual circulation patterns that affect sediment mobilisation and distribution. Maximum tidal currents are shown on the Hebrides Shelf.

(Figure 91) Sea-bed sediment distribution across the report area. AD=Anton Dohrn Seamount, FR=Flannan Ridge, GBB=George Bligh Bank, HS=Hebrides Shelf, HT=Hebrides Terrace Seamount, OHP=Outer Hebrides Platform, RB=Rockall Bank, RBa=Rosemary Bank Seamount, RT=Rockall Trough, SB=Stanton Banks.

(Figure 92) Percentage biogenic carbonate in sea-bed sediments. AD=Anton Dohrn Seamount, FR=Flannan Ridge, GBB=George Bligh Bank, HS=Hebrides Shelf, HT=Hebrides Terrace Seamount, OHP=Outer Hebrides Platform, RB=Rockall Bank, RBa=Rosemary Bank Seamount, RT=Rockall Trough, SB=Stanton Banks.

(Figure 93) Location of mobile bedforms and net sediment transport pathways. AD=Anton Dohrn Seamount, FR=Flannan Ridge, GBB=George Bligh Bank, HS=Hebrides Shelf, HT=Hebrides Terrace Seamount, OHP=Outer Hebrides Platform, RB=Rockall Bank, RBa=Rosemary Bank Seamount, RT=Rockall Trough, SB=Stanton Banks.

(Figure 94) Photographs (approximate location on (Figure 90)) showing different bed-form morphology. a) Sediment covered cobble beds (after Armishaw et al., 1998, fig. 10, courtesy of the Geological Society of London). b) Parallel linear ripples with gravel-rich troughs (after Stow et al., 2002, fig. 9c, courtesy of the Geological Society of London). c) Lunate ripples (after Armishaw et al., 1998, fig 11a, courtesy of the Geological Society of London.

(Figure 95) Static sea-bed bedforms. Small inactive submarine landslides, other inactive systems of sea-bed failure and changes of bedrock hardness that have generated static bedforms are not shown. Features from these include former sea-bed faulting and former small igneous intrusions. They mainly map to areas where bedrock is less than 5 m below the sea bed, as indicated above.

(Figure 96) Profile of an elongate sediment drift, moat and scarp, north-east Hebrides Slope. Location shown on (Figure 90).

(Figure 97) Location of geological features that may pose a hazard to developments or activities within and adjacent to the Rockall report area.

Tables

(Table 1) Comparison of structural and bathymetric terminology.

(Table 2) Acronyms used for deep seismic investigations.

(Table 3) Nomenclature of the major structural features referred to within the report area.

(Table 4) Inferred approximate maximum thickness of basin infill within the north and south Rockall basins as derived from the results of deep, near-normal incidence and wide-angle seismic reflection/refraction geophysical data. 1 Roberts et al. (1988); 2 Nadin et al. (1999); 3 Keser Neish (1993); 4 England and Hobbs (1997); 5 Shannon et al. (1999), 6 Mackenzie et al. (2002) and Morewood et al. (2005). For location of the geophysical profiles and geological cross-sections, see (Figure 7) and (Figure 14).

(Table 5) Summary of lithological and Sm-Nd and U-Pb isotopic data derived from commercial wells, BGS boreholes, shallow drill sites and dive/dredge sites that penetrate Archaean to Proterozoic metamorphic basement within the Rockall report area (partly compiled from Chambers et al., 2005; Morton and Taylor, 1991; Roberts et al., 1973).

(Table 6) Table of Ar-Ar ages obtained from igneous rocks in the Rockall Basin area. For location of samples, see (Figure 53).

(Table 7) Dimensions of proposed models (see text) to explain the gravity anomaly associated with the Geikie Volcanic Centre (located in (Figure 53)).

(Table 8) Dimensions of proposed models (see text) to explain the gravity anomaly associated with the Darwin Volcanic Centre (located in (Figure 53)).

(Table 9) Summary of commercial wells drilled within the report area.

(Table 10) Published total organic carbon analyses from BGS boreholes in the West Lewis Basin (WLB) and West Flannan Basin (WFB).

Tables

(Table 1) Comparison of structural and bathymetric terminology

(Table 2) Acronyms used for deep seismic investigations

(Table 3) Nomenclature of the major structural features referred to within the report area

(Table 4) Inferred approximate maximum thickness of basin infill within the north and south Rockall basins as derived from the results of deep, near-normal incidence and wide-angle seismic reflection/refraction geophysical data

(Table 5) Summary of lithological and Sm-Nd and U-Pb isotopic data derived from commercial wells, BGS boreholes, shallow drill sites and dive/dredge sites that penetrate Archaean to Proterozoic metamorphic basement within the Rockall report area.

(Table 6) Table of Ar-Ar ages obtained from igneous rocks in the Rockall Basin area

.

(Table 7) Dimensions of proposed models (see text) to explain the gravity anomaly associated with the Geikie Volcanic Centre (located in (Figure 53)).

(Table 8) Dimensions of proposed models (see text) to explain the gravity anomaly associated with the Darwin Volcanic Centre (located in (Figure 53)).

(Table 9) Summary of commercial wells drilled within the report area

(Table 10) Published total organic carbon analyses from BGS boreholes in the West Lewis Basin (WLB) and West Flannan Basin (WFB).