A Geological and Geophysical Description of the Arc
Mounds, southwest Porcupine Bank
Fiona Stapleton
A thesis submitted for the degree of Master of Science
Supervisors: Dr. John Murray (primary), Garret Duffy, Anthony Grehan, Mike Williams
Head of School: Prof. Vincent O'Flaherty
Earth and Ocean Sciences,
School of Natural Sciences,
National University of Ireland, Galway.
January 2014
ii
Abstract
Cold-water coral carbonate mounds occur worldwide and are prominently developed along
the European margin. Such ecosystems are currently the focus of considerable research
efforts - in particular, trying to understand their initiation and maintenance processes. This
project involves a description of what have been termed the ‘Arc Mounds’; a previously
undescribed region which features coral mounds situated on the south-western margin of the
Porcupine Bank. The geomorphological setting of these bioherms was characterised
geophysically using ship- and ROV-mounted multibeam and sub-bottom profiler data.
Mapping the location revealed over 40 carbonate build-ups, hundreds of metres long and
tens of metres high, aligning N-S along scarps and occurring in W-E orientated clusters in
water depths of between 630 m and 850 m. Both individual conical mounds and elongated
mound complexes were recognised.
The geophysical information was complemented with sedimentological data from two cores
recovered from the summit of a carbonate mound and from off-mound sediments. The
mounds are partly Recent in age, and appear to be actively forming today, but their history
extends back to the Pleistocene. The off-mound core showed that bottom currents have
deposited apparent contourite drifts, since at least as far back as the Pleistocene. Sub-bottom
profiler data shows that the drift deposits have onlapped basal portions of the mounds, and
that the mounds formed on top of a strong reflector. Sediment recovered from the equivalent
depth of the sub-mound reflector indicates that the mounds initially developed on a
carbonate hardground. Additionally, the provision of slope breaks by scarps appears to be an
important factor in the development of the carbonate mounds. The development of these
coral mounds appears to be the result of an interaction between contourite drift deposition
and biological growth processes, which are influenced by both the local topography and
current regime.
Keywords: cold-water coral; carbonate mound; sediment; ice rafted debris; multibeam; NE
Atlantic; Porcupine Bank
iii
iv
Dedications
This thesis is dedicated to my mother, who taught me that the best kind of knowledge
is that which is learned for its own sake.
The thesis is also dedicated to my father, who, along the same vein, told me that I
didn't have to study.
Finally, this thesis is dedicated to Iris - for not being a "granny" granny.
v
vi
Acknowledgements
I wish to acknowledge INFOMAR and a GRIFFITH Geoscience Award for providing the
main financial support for this research endeavour. Additionally a Thomas Crawford Hayes
Research Award awarded by the NUI Galway School of Natural Sciences was crucial to
travel to the IFREMER institute to collect data.
The data presented in this report was largely collected by IFREMER on the BoBEco cruise,
onboard the RV Pourqoui Pas?. I wish to acknowledge IFREMER for making this data
available. I am very grateful to Jean-François Bourillet of IFREMER for welcoming me to
IFREMER and providing considerable assistance and expert advice with the sediment core
analysis at Brest. I also warmly thank Matthieu Veslin, Emili Mon, Adriana Constantinescu,
and Gwen Craig for their help and friendship during my stay at IFREMER. Also of the
IFREMER institute I wish to thank Mickael Rovere for processing MSCL data, and the Unit
Research of Marine Geosciences, 'Laboratory of Sedimentary Environments’ for processing
the XRF data.
I wish to acknowledge and sincerely thank Jens Fohlmeister and Norbert Frank of the
Institute of Environmental Physics (University of Heidelberg) for the invaluable radiocarbon
dates and U/Th dates respectively.
I also wish to thank the following for their kind and expert help: Prof. David Harper of
Durham University for his identification of brachiopods; Prof. Paul Pearson (Cardiff
University) for thin-section identification of planktonic foraminifera; Dr. Tom Dunkley
Jones (University of Birmingham) for nannofossil identification; Louise Allcock and Patrick
Collins of the Ryan Institute in NUI Galway for identification of shell material; and Alex
Costanzo for preparation of thin-sections and access to the petrographic analysis lab. I
sincerely thank Martin White for valued and interesting discussion regarding oceanographic
aspects of the project.
I wish to warmly thank my supervisors, John Murray, Garret Duffy, Anthony Grehan and
Mike Williams for their guidance and encouragement. The quality of this thesis was
significantly improved by their suggestions. I wish to particularly thank my primary
supervisor John. John your supportive, enthusiastic and motivational discussions were
critical to me throughout the entire masters - but especially towards the end. I am indebted to
you for all the extra effort you put in to various aspects of the project. I also am grateful for
the many stories, and I am very glad I had the experience of being your student. Garret I
wish to pay tribute to your absolute expertise in advice and help regarding the geophysical
aspect of the project, as well as your constant friendly support. I will always greatly
appreciate the opportunity of ship time that was granted to me by Anthony Grehan. It was a
fantastic learning experience that was vital to my understanding of cold-water coral mounds.
I also want to acknowledge A. Grehan's key role in the provision of the funding for acquiring
the dates, and for his foresight and friendly advice. I greatly appreciated Mike William's
astute observations in the preparation for the IGRM and for the friendly help he didn't
hesitate in providing.
I thank the postgraduate students and everyone in the department that have been a friend
and/or provided advice - especially Peter Fallon for his unbelievable generosity with help,
time, and finally even his computer. I also thank Cathal Clarke for his help with investigating
the Chirp data. I wish to thank Shane Rooney who I called upon countless times with
computer troubles and I wish to warmly thank Tiernan Henry for his advice throughout the
thesis, and for provision of a laptop when everything 'went'.
Finally I wish to thank my family and friends, who never waver to support me and always
have faith in me. I am very looking forward to spending more time with you again.
vii
viii
‘…he must fall,
Who thinks all made for one, not one for all.’
An Essay on Man
Alexander Pope
ix
x
Table of Contents
Abstract ....................................................................................................................... iii
Dedications................................................................................................................... v
Acknowledgements .................................................................................................... vii
List of Figures ............................................................................................................ xv
List of Tables ......................................................................................................... xxvii
List of Appendices .................................................................................................. xxix
Declaration .............................................................................................................. xxxi
Chapter 1: Introduction ................................................................................................ 1
1.1: Overview ........................................................................................................... 1
1.2: Study area .......................................................................................................... 2
1.3: Project motivation ............................................................................................. 3
1.4: Project objectives .............................................................................................. 5
Chapter 2: Materials and Methods ............................................................................... 7
2.1: Overview ........................................................................................................... 7
2.2: Geophysical investigation ............................................................................... 10
2.2.1: Multibeam ................................................................................................. 10
2.2.2: Sidescan .................................................................................................... 37
2.2.3: Chirp ......................................................................................................... 37
2.3: Sediment core analysis .................................................................................... 38
2.3.1: Core acquisition ........................................................................................ 38
2.3.2: GEOTEK Multi-sensor Core Logger Analysis (MSCL) .......................... 40
2.3.3: Sedimentological logging ......................................................................... 43
2.3.4: AVATECH XRF ...................................................................................... 44
2.3.5: Grain size analysis .................................................................................... 47
2.3.6: Ice-rafted debris counting ......................................................................... 51
2.3.7: Core CS07 grain composition analysis ..................................................... 53
xi
2.3.8: Dating........................................................................................................ 54
2.3.9: Core CS06 mound base reflector sediment............................................... 55
2.3.10: Core dropstone analysis .......................................................................... 57
Chapter 3: Background............................................................................................... 59
3.1: Cold-water corals............................................................................................. 59
3.2: Cold-water coral mounds ................................................................................ 62
3.2.1: Definition of a cold-water coral mound .................................................... 62
3.2.2: Cold-water coral mound development ...................................................... 66
Chapter 4: Environmental Context and Previous Research ....................................... 73
4.1: Geological & geomorphological setting .......................................................... 73
4.1.1: Regional development .............................................................................. 73
4.1.2: The Porcupine Bank.................................................................................. 74
4.2: Oceanographic setting ..................................................................................... 75
4.2.1: Modern ...................................................................................................... 75
4.2.2: Palaeo-oceanographic considerations ....................................................... 78
4.3: North Atlantic Carbonate Mounds .................................................................. 78
4.3.1: General introduction ................................................................................. 78
4.3.2: Examples of NE Atlantic carbonate mounds ............................................ 79
4.3.3: Mound base of the NE Atlantic buildups.................................................. 82
4.3.4: Mound initiation and development in the NE Atlantic ............................. 83
Chapter 5: Geophysical Investigation ........................................................................ 85
5.1: Multibeam........................................................................................................ 85
5.2: Sidescan backscatter ...................................................................................... 107
5.3: Chirp .............................................................................................................. 108
5.4: Discussion and conclusions ........................................................................... 124
Chapter 6: Off-mound Core (CS06)......................................................................... 135
6.1: Core description............................................................................................. 135
xii
6.2: Integrated description of facies ..................................................................... 142
6.3: Mound base reflector sediment (Bas Ogive) ................................................. 150
6.4: Discussion and conclusions ........................................................................... 156
Chapter 7: On-mound Core (CS07) ......................................................................... 165
7.1: Core description ............................................................................................ 165
7.2: Integrated description of facies ..................................................................... 177
7.3: Discussion and conclusions ........................................................................... 191
Chapter 8: Summary ................................................................................................ 202
8.1: Discussion ..................................................................................................... 202
8.2: Conclusions ................................................................................................... 205
8.3: Recommendations for future work ................................................................ 207
References ................................................................................................................ 208
APPENDICES ......................................................................................................... 220
xiii
xiv
List of Figures
Fig. 2.1 1: Bathymetric map showing data acquisition during Leg two of the BobEco
cruise on the French Research Vessel Pourqoui Pas? .................................................. 7
Fig. 2.2. 1: A GPS tide file ......................................................................................... 12
Fig. 2.2. 2: Example GPS tide file being multiplied by -1 in the Generic Data Parser.
.................................................................................................................................... 13
Fig. 2.2. 3: The pre-corrected DEM surface (ROV_10m) shows deeper depths
(1208.6m); and the corrected DEM surface (test_tide_10m), created using remerged
lines with the GPS tide multiplied by -1, shows the true shallower depths (655.82m).
.................................................................................................................................... 14
Fig. 2.2. 4: Left: DEM of incorrect depth, 10 m resolution, showing groove with
deeper soundings. Right: Depth-corrected DEM, 10 m resolution, showing groove
greatly reduced. .......................................................................................................... 15
Fig. 2.2. 5: Left shows the axial noise pattern caused by gyro anomalies (highlighted
in yellow in the Attitude editor). Right shows the Cleaned data after the anomalies
were manually removed (with interpolation). ............................................................ 16
Fig. 2.2. 6: Example of the effect of cleaning using the swath editor (left uncleaned,
right cleaned). Removing the spikes removed the circular artefacts seen around the
mound......................................................................................................................... 17
Fig. 2.2. 7: 'Lumpy' texture (circular features) on the mound is shown as smooth
undulations in swath editor (top), as opposed to the spikes associated with circular
artefacts off the mound. This indicates that the ‘cauliflower’ texture represents real
features – interpreted as coral colonies based on video evidence from the area........ 18
Fig. 2.2. 8: A 20 m resolution DEM of the noisy 7111 100 kHz data (top) and the
associated swath profiles in swath editor in plan and rear view (bottom). ................ 19
Fig. 2.2. 9: The subset editor shows this part of the survey line has little/no
information and that the noise is shallower than the true seabed surface, which can be
observed in the adjacent survey line. Reson 7111 (100 kHz) shipborne multibeam.
Top: Subset Editor aerial view of DEM surface (25 m resolution) showing the
difference between the normal line and a line with noise. Bottom: lateral view. The
red line was rejected due to poor signal quality in the deeper regions. Data density
decreases with depth as a consequence of the beam geometry and attenuation of
frequency. ................................................................................................................... 20
xv
Fig. 2.2. 10: 7111: 25 m DEM with many gaps in data coverage. ............................. 21
Fig. 2.2. 11: Reson 7111, 50 m DEM shows less data gaps than the 25m DEM but
loss of resolution (compare with Fig. 3.1.13). Blue line is ship track which had bad
multibeam data and was ignored. Colour gradation symbolises depth with red-blue
representing shallow to deep (~500-900 m). .............................................................. 22
Fig. 2.2. 12: The Reson 7111 (100 kHz) 75m DEM produces the cleanest image
without data gaps, but resolution is poor. Blue line is ship track which had bad
multibeam data and which was ignored. Colour gradation symbolises depth with redblue representing shallow to deep (~500-900 m). ...................................................... 23
Fig. 2.2. 13: ROV 7125 (400 kHz) data: Left - Hillshaded DEM 20 cm. Right Hillshade of the resampled bathymetry data (using cubic technique); produced a
smoother image with little loss of information .......................................................... 24
Fig. 2.2. 14: Workflow to define mound footprints: A slope raster (B) was created
from the bathymetry (A) and the 7° slope contour was chosen to define the mound
bases (red line around zoomed-in mounds). C. Polygons created from the 7° slope
contour were filtered to contain only polygons with areas greater than 5,625 m2
(polygons <5,625m2 highlighted in blue). Polygons in the zoomed-in example show
that the 7° slope contour is conjoining the mounds with the scarp. D. Polygons were
manually modified to better represent the identified mounds after consulting focal
maximum statistics.
Local maxima occur where overlapping pixels of the two
surfaces are the same. This raster highlighted the mound shapes aiding mound
boundary delineation. It also showed some of the mounds to have two or more local
maxima (red dots highlight peaks in zoomed in example)......................................... 26
Fig. 2.2. 15: Mounds had to be manually modified after the automatic extraction
process in cases where the 7° slope contour (purple line in zoomed-in example)
adjoined adjacent mounds within one polygon. ......................................................... 27
Fig. 2.2. 16: Comparison of 3D bathymetry (above) to the 7° slope contour polygons
(below) showed that some smaller mounds were not enclosed by the polygons (red
dots) and noise associated with the bathymetric data created false polygons
(highlighted in blue). .................................................................................................. 28
Fig. 2.2. 17: Summary workflow of steps involved in mound extraction. ................. 29
Fig. 2.2. 18: Bathymetric map with zoomed-in area inset showing almost half of the
polygons (48%) created from the 7° slope contours (dark grey) did not correlate to
mound features in the 3D DEM or local maximum raster. Contours that define
xvi
mound footprints are shown in pink. The high level of noise is due to the quality of
the multibeam data and the nature of the area because the scarps are picked out by
the slope contour. ....................................................................................................... 31
Fig. 2.2. 19: The polygon created from the 7° slope contour (highlighted in blue)
creates an overall northwest-southeast orientation (black line) quite different from
the expected east-west orientation indicated by the mound ridge morphology (closely
spaced slope contours indicating mound ridge). This indicates limitations of using
the 7° slope contour.................................................................................................... 32
Fig. 2.2. 20: The 7° slope contours for the ROV data was very noisy (polygons >9 m2
shown). Also note how the mounds are broader than the width of the swath. .......... 33
Fig. 2.2. 21: Steps to calculate mound heights: (A) Data enclosed by the mound
footprint polygons were removed from the original DEM. (B) This DEM was then
reinterpolated to produce a hypothetical DEM without mounds. Subtraction of a
DEM of only the mounds (C) from the interpolated DEM (B) generated a raster of
mound heights (D). Zonal statistics were then performed to find the maximum height
associated with each polygon. .................................................................................... 35
Fig. 2.2. 22: Workflow to extract mound heights. ..................................................... 36
Fig. 2.3. 1: ROV photo from the summit of the carbonate mound (from which core
CS07 was acquired) showing live coral growth (white) and infilled dead coral
framework. Scale of photo is ~2m wide. The poor quality of the photograph is due to
scattering by sediment particles. ................................................................................ 38
Fig. 2.3. 2: Seafloor locations of cores CS06 and CS07. CS07 was taken from the
summit of a coral mound, and core CS06 was taken from the adjacent seafloor 966 m
to the southwest. ROV track, Reson 7125 400 kHz multibeam data (hillshaded DEM
at 1m resolution), overlain on shipborne multibeam data, Reson 7150 24 kHz
(hillshaded DEM at 50 m resolution). ........................................................................ 39
Fig. 2.3. 3: The MSCL-S. Image courtesy of www.geotek.co.uk.............................. 41
Fig. 2.3. 4: Simplified diagram showing the principle of XRF core logging (Richter
et al., 2006)................................................................................................................. 45
Fig. 2.3. 5: The AVATECH XRF scanner at the Royal Netherlands Institute for Sea
Research (NIOZ, Texel, The Netherlands). ............................................................... 46
Fig. 2.3. 6: The piston corer set-up with the location of the Peau d'Orange (catcher)
and Bas Ogive (cutter) highlighted by the red box. ................................................... 56
xvii
Fig. 3.1. 1: Distribution of cold-water and tropical (warm) coral reefs. The global
distribution of cold-water corals shows that they occur at all latitudes. Note that this
distribution is probably biased due to concentration of research in the Atlantic.
Source: UNEP/GRID-Arendal Maps and Graphics Library, February 2008,
http://maps.grida.no/go/graphic/distribution-of-coldwater-and-tropical-coral-reefs. 60
Fig. 3.1. 2: Drawings of Lophelia pertusa (left) and Madrepora oculata fragments
from photograph in Zibrowius (1980). Scale bar = 8.3 mm for L. pertusa and 4.3
mm. for M. oculata. .................................................................................................... 61
Fig. 3.2. 1: The Main biological groups responsible for reef construction on the earth
through time. Horizontal grey hatched bars represent major extinction episodes.
(Taken from Benton & Harper 2009, their figure 11.32, and based on an original
scheme by Wood, 2001). ............................................................................................ 63
Fig. 3.2. 2: Global distribution of cold-water coral mounds from Eisele (2010),
showing that they do not form at all locations where cold-water corals are known to
exist (compare with Fig. 3.1.1). The distribution may be underestimated due to
uncertainty about whether a reef constitutes as a mound (Roberts et al. 2009). ........ 65
Fig. 3.2. 3: Hypothetical development sequence for cold-water coral mounds from
Mullins et al. (1981) based on Squires (1964) and Neumann et al. (1977). .............. 69
Fig. 4.1. 1: East-west seismic section (from DCNER/PAD 2007) illustrating structure
of the Porcupine Basin and its margins (Porcupine Bank to the west). ..................... 74
Fig. 4.2. 1: Schematic diagram, following the concepts of van Rooij et al. (2007a)
and Toms (2010), showing water masses in the Rockall Trough (RT) in the vicinity
of the western Porcupine Bank (PB). Dashed lines indicate water mass boundaries
modified from Toms (2010). Carbonate mounds occur between 500 – 1200 m
(Klages et al., 2004). Enhanced hydrodynamics and thermocline (600 – 1000m) are
interpreted following White and Dorschel (2010). NAD – North Atlantic Drift,
NACW – North Atlantic Central Water, ENAW – Eastern North Atlantic Water,
MOW- Mediterranean Outflow Water, LSW- Labrador Seawater, NADW – North
Atlantic Deep water, and the SEC – Shelf Edge Current. .......................................... 77
.................................................................................................................................... 79
Fig. 4.3. 1: Carbonate mounds (red ellipses) and CWC species Lophelia pertusa
(green dots) from Foubert (2007), modified to show location of Arc Mounds (J). (A)
Belgica Mound Province at eastern margin of Porcupine Seabight, (B) Magellan and
Hovland Mound Provinces at northern margin of Porcupine Seabight, (C) Porcupine
xviii
Bank Canyon Mounds at western margin of Porcupine Bank, (D) Pelagia Mounds
along north-western flank of Porcupine Bank, (E) Logachev Mounds on southeastern slope of Rockall Bank, (F) mounds on western Rockall Bank, (G) Darwin
Mounds, (H) Galicia Bank, (I) Moroccan margin mounds ........................................ 79
Fig. 5.1. 1: Hillshaded Reson 7150 (24 kHz) 50 m DEM. Multibeam data show many
mounds to be aligned along the scarp to the east. ...................................................... 86
Fig. 5.1. 2: Hillshaded Reson 7111 (100 kHz) 75 m DEM. ....................................... 87
Fig. 5.1. 3: Hillshaded Reson 7125(400 kHz) 1 m DEM with mound example inset
and associated scarp highlighted. ............................................................................... 88
Fig. 5.1. 4: Slope of Reson 7150 (24 kHz) 50 m DEM. The lower the slope value, the
flatter the terrain; the higher the slope value, the steeper the terrain. Slope is average
slope of seabed within a square of 25 by 25 m. ......................................................... 89
Fig. 5.1. 5: Reson 7125 (400 kHz) 1 m DEM with slope (left) and hillshaded 1 m
DEM (right) of mound example inset. The lower the slope value the flatter the
terrain; the higher the slope value the steeper the terrain........................................... 90
Fig. 5.1. 6: Reson 7150 (24 kHz) 50 m DEM with polygons of the identified mounds
outlined and numbered. .............................................................................................. 91
Fig. 5.1. 7: Reson 7150 (24 kHz) 50 m DEM showing orientations of the convex hull
polygons. .................................................................................................................... 92
Fig. 5.1. 8 part 1: Bathymetric profiles of mounds 1-10, based on their orientations
parallel to the long axes and footprint edges. See Fig. 5.1.6 for mound numbering. 93
Fig 5.1.8 part 3: Bathymetric profiles of mounds 21-30, based on their orientations
parallel to the long axes and footprint edges. See Fig. 5.1.6 for mound numbering. 95
Fig. 5.1.8 part 4: Bathymetric profiles of mounds 31-42, based on their orientations
parallel to the long axes and footprint edges. See Fig. 5.1.6 for mound numbering. 96
Fig. 5.1. 9: 3D bathymetric model of the Arc Mounds study area. There is a raised
terrace bounded by two along-slope crescent-shaped scarps to the east and the west,
with mounds aligning the eastern scarp (sc2) and occurring west of the western scarp
(sc1). Note how the scarps are joined in the south but the large isolated mound with
associated moat appears to cause reoccurrence of Sc2. Depth scale indicated by
colour bar in metres. Data vertically exaggerated by 4.............................................. 97
Fig. 5.1. 10: High resolution bathymetric 3D model shows sediment draping off the
mounds over the scarp. Data vertically exaggerated by 4. ......................................... 98
xix
Fig. 5.1. 11: 3D bathymetric map showing mound morphology varies from elongated
ridge mound types to conical forms. Colour scale showing depth. Data vertically
exaggerated by 4......................................................................................................... 99
Fig. 5.1. 12: Bathymetric profile of mounds following the route of the ROV (SENW). The mounds are mostly over 50 m in elevation from the adjacent seafloor.
Note mounds appear spikey as they are highly vertically exaggerated...................... 99
Fig. 5.1. 13: Plot of mound footprint area versus exceedance probability. .............. 103
Fig. 5.1. 14: Plot of mound footprint area versus mound height. There is a positive
correlation between mound area and mound height................................................. 103
Fig. 5.1. 15: Plot of mound height versus depth shows a very slight trend for larger
mounds occurring at greater depths ......................................................................... 104
Fig. 5.1. 16: Mound footprint area versus mound footprint shape. Mound shape is
based on the Principal Axes Ratio (PAR): values approaching one indicate circular
shapes, values approaching zero indicate elongated forms. The data indicate that
there is a trend of mounds becoming more elongated as their size increases. ......... 104
Fig. 5.1. 17: Mound polygons of the top five PAR results are highlighted in blue,
indicating that smaller more isolated mounds are also more circular in form. ........ 105
Fig. 5.1. 18: Unidirectional rose diagram showing mound footprint orientation in
percent of occurrences. Mean (105°) shown by black line. Standard deviation
(±50.95°) shown by red line. There is a northwest-southeast trend and a slight
northeast-southwest trend. ........................................................................................ 106
Fig. 5.2. 1: Map of the sidescan data acquired along the ROV transect with mound
example highlighted. The mound complexes have a higher backscattering signal. 107
Fig. 5.2. 2: Map of part of the ROV sidescan data with the scarp highlighted. ....... 108
Fig. 5.3. 1: Chirp profile A-A’ (line BOBECO122A) orientated west-east with
bathymetric map and bathymetric profile (inset). The acoustically homogenous
hyperbolic features indicate the presence of mounds. The dashed red line indicates
core CS06 location; the red line indicates intersection of Chirp line with profile B-B’.
Sc- Scarp. TWT - two way travel time. SP – shotpoint. Depth at 1 sec (TWT) is 750
m. .............................................................................................................................. 110
Fig. 5.3. 2: Identifying the shotpoints of the apices of the mound features on Chirp
profile A-A’ (line BOBECO122A) with bathymetric profile (inset). The shotpoints of
the apices of the mound features were mapped to their corresponding coordinates.
The dashed red line indicates core CS06 location; the red line indicates intersection
xx
of Chirp line with profile B-B’. TWT - two way travel time. SP – shotpoint. Depth at
1 sec (TWT) is 750 m. ............................................................................................. 111
Fig. 5.3. 3: North-south Chirp profile B-B’ (line BOBECO119A) with bathymetric
map and bathymetric profile (inset). Scales: horizontal 1/25000, vertical 50 cm/sec.
TWT – Two way travel time. Depth at 0.90 sec (TWT) is 675 m. .......................... 112
Fig. 5.3. 4: Identifying the shotpoints of the apices of the mound features on Chirp
profile B-B’ (line BOBECO119A) with bathymetric profile inset. The shotpoints of
the apices of the mound features were mapped to their corresponding coordinates.
The dashed red line indicates core CS07 location; the red line indicates intersection
of Chirp line with profile A-A’. TWT - two way travel time. SP – shotpoint. Depth at
0.80 sec (TWT) is 600 m. ........................................................................................ 113
Fig. 5.3. 5: Composite plot of the coordinates of the mound-like features in the Chirp
data with the mound structures imaged by the multibeam data. The Chirp mound
coordinates are offset from the actual mounds imaged in the multibeam data due to
sideswipe. CH6 appeared to show a buried mound in the Chirp data and it does not
show any surface representation. 7150 hillshaded shipborne multibeam data (24
kHz). Purple dots represent CH1-CH8 coordinates from line BOBECO122A (A-A’;
West-East);
black
triangles
represent
CH8-CH12
coordinates
from
line
BOBECO119A (B-B’; South-North) ....................................................................... 115
Fig. 5.3. 6: Identification of seismic units. The directly detected mound appears as an
acoustically homogenous dome-shaped structure rooting on the strong reflector R1.
R1 has been interpreted as the mound base. Three seismostratigraphic units U1, U2,
and U3 were identified U1 is acoustically transparent but there is a detection of an
irregular reflector within it (r2). U1 underlies an irregular, erosional surface: the R1
reflector that underlies the mounds. U2 is acoustically transparent and conformably
underlies U3. U3 shows draping, conformable, stratified, sub-bottom reflectors
which onlap the flanks of the mounds. This unit usually directly overlies U1 on the
R1 reflector but also occurs overlying U2. .............................................................. 116
Fig. 5.3. 7: Mound CH5 with drift sediment build-up to the east of the mound and an
exposed flank on the western side............................................................................ 118
Fig. 5.3. 8: Multibeam data indicates that the steep western flank of mound CH5
(indicated by the red dot) is associated with the mound build up and is not a seafloor
characteristic such as a scarp.................................................................................... 118
xxi
Fig. 5.3. 9. Possible interpretation (scenario 1) to describe deposition chronology of
units. The mounds are established on an erosion surface (R1), followed by deposition
of unit 2 and 3. The mounded nature of unit 2 could be explained by moating
associated with the two adjacent mounds (CH5 and the now buried mound). BM –
buried mound............................................................................................................ 119
Fig. 5.3. 10: Possible interpretation (scenario 2) to describe deposition chronology of
units. U2 was eroded and there was subsequent development of the mounds on the
R1 reflector of U1. The middle section represents 2 adjacent buried mound structures
with the middle buried mound possibly related to Ch5. .......................................... 120
Fig. 5.3. 11: Possible interpretation (scenario 3) to describe deposition chronology of
units. Mounds both developed on R1 and on U2 where it remained after being mainly
eroded (following interpretations by Van Rooij et al., 2003). This would mean the
mound base developed with upslope and downslope sides initiating at different
depths. ...................................................................................................................... 121
Fig. 5.3. 12: Interpreted Chirp profile A-A’ (line BOBECO122A) with original Chirp
results and bathymetric profile inset. The hyperbolic features indicate the presence of
mounds (coloured grey). The strong reflector underlying the mound structure R1 is
interpreted as the hard mound base and has been correlated to the topmost reflector
of U1. r2 is an irregular reflector within U1. The U3 interpreted drift deposits cover
the area. U2 is interpreted to occur mid-section and to the east. The dashed red line
indicates core CS06 location; the red line indicates intersection with the north-south
Chirp line profile B-B’. Blue lines indicate possible faults (F). BM – buried mound.
Sc – Scarp. TWT - two way travel time. Depth at 1 sec (TWT) is 750 m. .............. 123
Fig. 5.4. 1: Possible water deflection of Sc1 could be influencing the E-W mound
clusters (black arrows) or the mounds could be occurring on Variscan E-W trending
ridges. ....................................................................................................................... 130
Fig. 5.4. 2: 3D representation of the mounds flanking Sc2. There appears to be
erosion occurring in between the mounds, indicating that the scarp’s development is
part-synchronous with the mounds. The prevention of erosion by the mounds could
then explain their location on the projections of the terraced seafloor. ................... 132
Fig. 6.1. 1: Summary log of core CS06. .................................................................. 137
Fig. 6.1. 2: Photograph of stippled mud in core CS06. The stipples are interpreted as
possible mm-scale burrows; they are beige and contain a coarser fill (from fine to
xxii
coarse sand) than the surrounding orange mud matrix (three examples are outlined).
The stippled deposits also display cm-scale burrows and bioturbated contacts. ..... 139
Fig. 6.1. 3: CS06 facies (B = bioturbation). (a) Beige foraminiferal muddy sand at
top of core with irregular, gradational lower contact. (b) Bioturbated greyish brown
foraminiferal muddy sands and sandy muds with irregular contacts. (c) Heavily
bioturbated yellowish brown stippled foraminiferal muds and muddy sands with
erosive contacts (two stipples outlined by black ellipses). (d) Granule-rich
structureless very pale brown to reddish pale brown foraminiferal muddy sands with
mud clast outlined. (e) Structureless reddish white foraminiferal muddy sands with
irregular, gradational upper contact with facies D. Width of core is 11 cm. ........... 141
Fig. 6.2. 1: Results of core CS06: Photograph of core alongside graphic lithology,
dates, sieve-based grain size analysis results, mean and sorting (derived from the
laser-based results), % IRD, Magnetic Susceptibility, and Ca/Fe and Ca/Sr ratios. For
key to graphic lithology see Fig. 6.1.1. .................................................................... 143
Fig. 6.2. 2: Articulated terebratulid brachiopod with intact brachidium in silty muds
belonging to Facies C at 140.4 cmdc in CS06. Scale to the right is in cm. ............. 147
Fig. 6.3. 1: Bas Ogive sediment (a) Sieved sediment in beaker (sediment >63μm); (b)
Lithics >2mm (ruler in cm), note angularity (except large pebble top left) and red and
green colouration; (c) Prepared thin-sections; (d) Top surface of pebble showing
colonised surface and bottom surface showing cement; (e) Foraminiferal wackestone
clast in thin-section (PPL). ....................................................................................... 151
Fig. 6.3. 2: Packstone clast with abundant planktonic foraminifera in Bas Ogive. The
specimen highlighted by red box is Globorotalia menardii or a closely related form.
Many thin shell fragments are also present. Thin section BG2 (PPL). .................... 152
Fig. 6.3. 3: Fossils in Bas Ogive. (a) Globigerinid in BG1 (PPL); (b) Spines showing
a pseudo-uniaxial cross and a possible crustacean shell (homogenous shell structure
showing undulose extinction) in BG1, (XPL); (c) Possible Cibicides benthic foram
(left), mollusc shell (wavy elongated S form above) and well-rounded chalk
intraclast in BG1 (PPL); (d) Possible echinoderm spine in BG2 (PPL); (e) Possible
bryozoan or calcareous algae in BG3 (XPL). .......................................................... 153
Fig. 6.3. 4: Photomicrographs of lithified matrix found between the Bas Ogive
limestone clasts. Left – PPL, right XPL. (A-D) from thin-section BG3, (E-F) from
thin-section BG2. Green colouration ascribed to glauconite. .................................. 154
Fig. 6.3. 5: Iron rich clast in thin section from Bas Ogive. ...................................... 155
xxiii
Fig. 6.4. 1: Diagenetic sequence of hardground formation as described by Noé etal.
(2006). ...................................................................................................................... 162
Fig. 6.4. 2: Well-rounded intraclasts (within a larger clast) in the BOg. Left is PPL
and right is XPL. ...................................................................................................... 162
Fig. 7.1. 1: Summary log of core CS07 ................................................................... 167
Fig. 7.1. 2: Coarse unit in Section 6 containing partially lithified, possible biogenic
rubble (middle left) and mud clasts (bright white in lower right) as well as branch
size coral (a well preserved coral branch beside a heavily dissolved branch in upper
right), a gastropod, and a bivalve shell (beneath mud clast) and pebbles. ............... 170
Fig. 7.1. 3: CS07 Section 6 mottling evident at 520 cmdc. ..................................... 171
Fig. 7.1. 4: Coral-rich unit located at 471.5-483.5 in Section 5 before (left) and after
sampling; bivalve is the longitudinal shell fragment in the resultant cavity. Scale in
cm. ............................................................................................................................ 172
Fig. 7.1. 5: Coarser unit located at 313.5-323 cmdc in Section 4, showing erosive
upper and lower boundaries; with an upper contact consisting of a 3cm thick layer of
large coral fragments (granules- branches) that pinches to the left. Each line
represents a cm in scale to right ............................................................................... 173
Fig. 7.1. 6: Coral bands in CS07 Section 3 at 216-220 cmdc. ................................. 174
Fig. 7.1. 7: Irregularly shaped non-lithified bright grey blotches; upper blotches with
clayey infill and lower with silt-fine sand. ............................................................... 175
Fig. 7.1. 8: Left: Lithified irregularly shaped concretions found in Section 1 in CS07
between 55 and 65 cmdc. Right: sieved fraction >500 µm, interpreted as calcitecemented biogenic rubble......................................................................................... 176
Fig. 7.2. 1: Results of core CS07: Photograph of core alongside graphic lithology,
dates, sieve-based grain size analysis results, mean and sorting (derived from the
laser-based results), % IRD, Magnetic Susceptibility, and Ca/Fe and Sr/Ca ratios. For
key to graphic lithology see Fig. 7.1.1. .................................................................... 181
Fig. 7.2. 2: Framework-building coral embedded in unlithified sandy carbonate
matrix in Facies A of CS07 Section 1 Unit 1 (5-8 cmdc). ....................................... 183
Fig. 7.2. 3: Detail of CS07 (Facies A) showing a pronounced Sr/Ca decrease at the
termination of the branch-sized coral and onset of the muddy sediments (Facies B) at
35 cmdc .................................................................................................................... 184
Fig. 7.2. 4: Correlation of coral bands (identified during logging) with peaks in XRF
Sr/Ca ratio element profiles...................................................................................... 188
xxiv
Fig. 7.3. 1: Scheme to describe possible fissure-fill sediment. (a) Mound growth
during transitional or warm conditions with water currents (indicated by blue
arrows). Multibeam bathymetry shows the mound to have multiple peaks and
irregular topography (see mound number 17 in Fig 5.1.8). (b) Increased bottom
current activity generates a fissure on the mound. (c) Destabilisation of the flanks
causes downslope movement of the IRD-rich layer and coral debris from above,
infilling the fissure in the process. (d) Further erosion and mass wasting causes
downslope transport of the IRD-rich sediment to the base of the mounds; but the
sediment within the fissure is preserved, including young coral. (e) Mound growth
continues for the remainder of the Holocene. Coring of the mound (cylinder)
recovers old mound deposits interrupted by the young fissure-fill. ......................... 194
Fig. 7.3. 2: Structural classification of Organic Reefs and Carbonate Mud Mounds
(from Riding 2002). Red boxes suggest appropriate classification for deposits in core
CS07......................................................................................................................... 196
xxv
xxvi
List of Tables
Table 2.1. 1: List of the methods and where they are described in this chapter. ......... 9
Table 2.2. 1: Echosounder attributes .......................................................................... 10
Table 2.2. 2: Performance attributes of the ROV multibeam .................................... 11
Table 2.3. 1: CS06 and CS07 sediment core attributes. ............................................ 40
Table 2.3. 2: Folk (1954) sediment classification, incorporating the modification of
Flemming (2000). This scheme was used for visual classification............................ 44
Table 2.3. 3: Grain size fractions based on the Udden-Wentworth grain-size
classification of terrigenous sediments (Wentworth, 1922) ....................................... 48
Wet sieving was carried out in one run and split into four size fractions: <63 µm, 63250 µm, 250-500 µm and 500-2000 µm ( .................................................................. 50
Table 2.3. 4: Wet-sieve stack configuration with Udden-Wentworth grade, μm and
Phi (ɸ) values. ............................................................................................................ 51
Table 2.3. 5: Material from CS06 and CS07 that were radiocarbon dated using an
Accelerator Mass Spectrometer (AMS). .................................................................... 54
Table 2.3. 6: Attributes of the coral clast samples submitted for U-Th dating. ......... 55
Table 3.2. 1: Proposed processes controlling mound initiation and development: The
Hovland et al. (1998) Hydraulic Theory, with initiation and growth generated by
hydrocarbon seeps; versus the Environmental Control theory, with mechanisms
governed by optimal environmental conditions. Modified from Roberts et al (2009).
.................................................................................................................................... 67
Table 5.1. 1: Summary of the morphometric parameters used to quantify mound
spatial patterns (adapted from Correa et al., 2012). ................................................. 102
Table 5.3. 1: Coordinates of mound-like features identified in the Chirp data for line
BOBECO122A. SP-shotpoint .................................................................................. 114
Table 5.3. 2: Coordinates of mound-like features identified in the Chirp data for line
BOBECO119A. SP-shotpoint .................................................................................. 114
Table 5.4. 1: Characteristics of mounds of examples in the North East Atlantic, listed
by latitude from south to north. ................................................................................ 126
Table 6.1. 1: Sedimentary facies identified in CS06. .............................................. 140
Table 6.2. 1: Sedimentary facies summary for CS06. ............................................. 142
xxvii
Table 6.3. 1: Sieving results for the 'Bas Ogive' sediment from the base of core
CS06. ........................................................................................................................ 150
Table 6.4. 1: Comparable facies to those documented herein in CS06, with those
identified by previous studies on the Porcupine Bank. ............................................ 160
Table 7.2. 1: Characteris-tics of facies of the on-mound core CS07 ....................... 179
Table 7.3. 1: Characteristics of the type reefs, within the category of matrixsupported reefs, which may describe the deposits of the on-mound core CS07 (from
Riding, 2002). ........................................................................................................... 195
xxviii
List of Appendices
APPENDIX A - Detailed sedimentological logs ..................................................... 220
APPENDIX B - MSCL ............................................................................................ 232
APPENDIX C - Grain size analysis ......................................................................... 236
APPENDIX D - IRD count ...................................................................................... 241
APPENDIX E - Sediment grain composition .......................................................... 242
APPENDIX F – XRF analysis ................................................................................. 249
APPENDIX G – Dating ........................................................................................... 250
APPENDIX H - Dropstone analysis ........................................................................ 255
xxix
xxx
Declaration
I hereby declare that this thesis is my own work, except when otherwise stated. It has
not previously been submitted to this or any other University. I agree that this thesis
may be lent in accordance with the National University of Ireland Galway
regulations.
_____________________
Fiona Stapleton
xxxi
xxxii
Chapter 1: Introduction
1.1: Overview
Cold-water corals [CWC] have been known to science since Linnaeus’ Systema
Naturae in 1758; yet, much remains to be understood about their biology, ecology,
behaviour and distribution. Within the last few decades, due to the increased use of
deep-water acoustics, the spatial extent of their distribution is being realised. It has
been discovered that corals are actually primarily cold-water inhabitants (Roberts et
al., 2009). CWC's form reefs from the polar to tropical habitats (Wood, 1999)
including: continental shelves, slopes, banks, on seamounts, in fjords and on
submarine canyon walls (van der Land, 2011).
It was only in the 1960s that deep-sea exploration lead to the discovery of moundlike structures colonised by corals (Stetson et al., 1962). CWC mounds develop
through successive build-ups of in-situ (autochthonous) coral debris and related
trapped or baffled externally supplied (allochthonous) sediment (Mazzini et al.,
2012). CWC mounds represent unique habitats in terms of biodiversity, providing
records of climate fluctuations and also act as carbon sinks (Klages et al., 2004). As
much information about the palaeoenvironmental conditions in which the CWC's
existed is needed in order to adequately predict their ability to survive in the future.
Additionally, as carbonate mounds are well known from the fossil record (see Wood,
1999 and references therein), studying contemporary examples may aid in their
interpretation.
This work integrates multibeam, sidescan sonar, sub-bottom profiling and MSCL and
XRF of piston cores; from a site of CWC mounds (the Arc Mounds) on the
southwest Porcupine Bank. No work, thus far, has studied palaeoclimates using
marine sedimentary cores from the Arc Mounds region.
1
CHAPTER 1: INTRODUCTION
1.2: Study area
The area examined during this research was located in the Arc Mounds province in
the southwest Porcupine Bank of the North Atlantic. The Porcupine Bank forms a
raised area of seabed lying between the Porcupine Seabight and Rockall Trough
water basins (Fig. 1.2.1), positioned between 51°N – 54°N and 12°W – 15°W and
covering an area of approximately 10,000 km2 (Scoffin & Bowes, 1988). The
~154 km2area of seafloor that was examined, located from 51°15’N – 51°25’N and
14°55’W – 14° 35’W, is characterised by CWC carbonate mounds and scarps in
water depths of 500 – 1000 m.
Fig. 1.2. 1: Location of study area. Bathymetry of seafloor using GEBCO (General
Bathymetric Chart of the Ocean) data with horizontal grid spacing of 30 arc seconds, WGS
1984). The location of the detailed survey map Reson 7150 50 m DEM overlain on hillshade
(on left) is indicated by the red inset box in the general location map (on right).
2
CHAPTER 1: INTRODUCTION
1.3: Project motivation
Possible significance to the rock record
It has only relatively recently come to light that scleractinian bioherms are not
uniquely shallow-water in origin, and the increasing accounts of deep-water coral
bioherms demonstrate that these mounds are not biological oddities. In fact, Newton
(1987) suggested that they may be
“considered characteristic of regions at intermediate depths (200-1200 m) where local hard
substrates are available for colonization by frame- building ahermatypes”.
Teichert (1958) warned that the fact hermatypic corals are capable of constructing
bioherms in deep water suggests that similar structures may be present in the rock
record. The distinction between deep- and shallow-water bioherms is critical for
palaeoenvironmental and stratigraphic interpretations. Therefore these (modern)
deep-water buildups must be investigated and described in order to properly assess
comparable bioherms in the rock record.
Designated Area of Conservation
This work aims to support designating the Arc Mounds region as a Special Area of
Conservation under the Habitats Directive. Threats to these habitats include:

Immediate threats: Coral has a slow habitat recovery rate (Connell, 1997).
Human activities such as trawling, oil and gas exploration, and undersea pipelaying constitute some of the greatest threats to CWC's. The documented
destruction of deep-water corals by trawling (e.g. Hall-Spencer et al., 2002)
necessitates the implementation of protective measures in order to ensure the
long-term survival of these habitats.

Longer-term threats: Carbonate chemistry determines whether it is physically
possible for corals to calcify, thus future global warming and ocean
acidification also pose a threat to these ecosystems. Orr et al. (2005)
demonstrated that the depths at which hard corals are able produce their
skeletons are likely to decrease within the timescale of the present century.
Guinotte et al. (2006) overlaid the predicted shallowing on the current
3
CHAPTER 1: INTRODUCTION
distribution of cold-water corals, and found that many of the CWC habitats
may soon be in corrosive seawater.
Reasons for conserving such habitats include:
(i)
The mound habitats are biodiversity centres
(ii)
These bioherms play an important role in the carbon cycle and
(iii)
The skeletons of the corals and the mounds they help construct are
important archives for climate and environmental change
These three reasons are elaborated upon below:
(i) Biodiversity centre
CWC reefs occurring on CWC carbonate mounds are as rich in associated fauna as
their tropical reef counterparts making them high biodiversity centres (e.g. van der
Land, 2011; Wheeler et al., 2011). Sponges, molluscs, bryozoa, gastropods,
anemones, and fishes use the coral framework as a nursery, refuge or stable substrate
to grow upon (van der Land, 2011); and CWC reefs are also characterised by a more
diverse meio-epifaunal community than that in the underlying sediments (Raes &
Vanreusel, 2005).
(ii) Role in the carbon cycle
The carbonate content of CWC mounds can be several times higher than the
carbonate content of the surrounding sea floor sediments (Dorschel et al., 2007;
Titschack et al., 2009). Lindberg and Mienert, (2005) concluded that CWC
ecosystems may contribute >1% of global carbonate production, indicating that
CWC mounds may represent an important carbon sink for consideration in global
carbon budget models.
(iii) Corals and their mounds as archives
Coral skeletons record the temperature and chemical characteristics of the ambient
seawater in which they developed (Wood, 1999). Therefore deep-water corals are
one of a limited number of archives that record deep-sea conditions. In addition, the
mounds which the corals develop also record environmental variability through
signals in their make-up. Understanding how the earth's climate has changed
naturally in the past will aid in prediction of the future but will also help in
quantifying how much man may be influencing the environment.
4
CHAPTER 1: INTRODUCTION
1.4: Project objectives
In 2005, in the Porcupine Seabight, the IODP [Integrated Ocean Drilling Program]
Expedition Leg 307 drilled for the first time through a coral mound (Challenger
Mound) and revealed the origin and depositional processes within this sedimentary
structure (IODP 307 Expedition Scientists, 2005). However, globally, the controls on
the growth and decline of corals on carbonate mounds, as well as mound growth
trigger mechanisms, have yet to be fully understood. In order to elucidate these
processes mound genesis, topography and morphology must be investigated. This
work aims to describe the Arc Mounds region in this context and to interpret to what
extent the Arc Mounds are representative of the mound model derived from drilling
Challenger Mound. This project involves the analysis of a combination of
geophysical and geological data acquired from the Arc Mound site in the Porcupine
Bank. The multi-disciplinary approach will allow a holistic description of the mound
province. The aims are to:
(i)
Create multi-scale digital terrain models of the mound structures and
conduct geospatial analysis on mound statistics using GIS: Investigating
the nature of the type and distribution of the mounds will help
interpretations on the governing controls on mound growth - i.e. the
geological setting and associated hydrographic influence.
(ii)
Conduct an analysis of the sedimentological content of the cores and
quantify Ice Rafted Debris (IRD) fluxes to the mound and off-mound
areas: Analysing the core material will allow interpretations of the
palaeoenvironmental conditions that have affected both the mounds and
the off-mound areas.
(iii)
Investigate whether AVAATECH XRF geochemical profiles and MSCL
results for the mound and off-mound areas can provide useful signals of
environmental change: Investigating these additional proxies will
corroborate the lithological logging and sediment analysis interpretations.
(iv)
Investigate the nature of the base-cutter sediment of core CS06, the depth
of which correlates to the depth of a strong Chirp reflector that may be the
mound base reflector: The objective of studying the possible mound base
substrate is to provide an indication of the timing of the onset of mound
growth and the associated environmental conditions.
5
CHAPTER 1: INTRODUCTION
6
Chapter 2: Materials and Methods
2.1: Overview
A program of multibeam sounding, sidescan sonar, sub-bottom profiling, coring, and
video dives was completed on the west Porcupine Bank During Leg two of the
BoBEco [Bay Of Biscay – ECOlogy] cruise on the French Research Vessel
Pourqoui Pas? from the 23rd September until the 10th October 2011 (Fig.2.1.1). Two
shipborne and one ROV [remotely operated vehicle] multibeam datasets were
acquired. ROV multibeam backscatter data were acquired simultaneously with the
bathymetric data.
Fig. 2.1 1: Bathymetric map showing data acquisition during Leg two of the BobEco cruise
on the French Research Vessel Pourqoui Pas?
7
CHAPTER 2: MATERIALS AND METHODS
The various methods described in this chapter are listed in Table 2.1.1. The
bathymetric data were processed at NUI Galway [NUIG] in Caris to produce Digital
Elevation Models [DEMS] and then added to the ArcMap Geographic Information
System (GIS) for map creation, geospatial manipulation and analysis. Chirp profiles,
and sidescan backscatter mosaics were also provided for interpretation.
AVAATECH X-ray Florescence [XRF] down-core elemental profiles and
multisensor core logger [MSCL] measurements were acquired by the author in
IFREMER Brest. The processed data from these analyses at IFREMER were
provided in Excel format for integration into this thesis.
Coral specimens were sampled from the split cores in IFREMER for mass
spectrometric 230Th/U dating, which was performed by Norbert Frank (Heidelberg).
Grain-size measurements of bulk sediment samples were performed in IFREMER
using a Coulter LS grain-size analyser. Sediment samples were also wet-sieved in
NUIG in order to compare with the laser grain-size analyses and to obtain a >250 µm
fraction for IRD counting, composition analyses and also to pick planktonic
foraminifera for 14C-AMS dating.
Thin-section slides of the lithified base cutter sediment from core CS06 and of two
dropstones (one from each core) were produced in NUIG and provided for
petrographic analysis.
The interpretations presented in this thesis from these various collaborations are
those of the author.
8
CHAPTER 2: MATERIALS AND METHODS
Section
Geophysical
Creation of multi-scale bathymetric terrain models, analysis
rasters and GIS procedure
2.2.1
Interpretation of sidescan data
2.2.2
Interpretation of chirp data
2.2.3
Deployment of Multi-Sensor Core Logger
2.3.2
Logging (cm-scale) of the cores
2.3.3
Sub-sampling of sediments and coral from the cores
Geological
Deployment of XRF scanner
2.3.4
Laser & sieve-based grain size analysis of the core sediment
subsamples
2.3.5
Ice-Rafted Debris counting
2.3.6
Core CS07 grain composition analysis
2.3.7
Dating
2.3.8
Thin-section petrography of base-cutter sediment of core CS06
2.3.9
Thin-section petrography of dropstones
2.3.10
Table 2.1. 1: List of the methods and where they are described in this chapter.
9
CHAPTER 2: MATERIALS AND METHODS
2.2: Geophysical investigation
2.2.1: Multibeam
Data acquisition
Three bathymetric data sets were acquired, one by remotely operated vehicle (ROV)
and two shipborne. Tide corrections were applied onboard in real time for the ship’s
multibeam, using SHOM (Service Hydrographique et Océanographique de la
Marine) predictions. The ROV was deployed on the 12th September 2011 at 1633 hrs
and reached the bottom at 1720 hrs. Photographic images were acquired at 10 m
from the bottom in order to determine coral cover. Following coral presence
confirmation, multibeam data were acquired at 50 – 70 m altitude by VICTOR 6000
using the Reson 7125/400 kHz echosounder. The ROV dive was approximately 23.5
hours in duration and distance travelled was 18 km. Bathymetric datasets from the
ship’s multibeams, Reson7150/24 kHz and Reson7111/100 kHz, were then obtained
in the same area. The echosounder attributes are listed in Table 2.2.1 (below).
Manufacturer
Model
Frequency
(kHz)
Max Depth
No.
Beams
Swath
Coverage
Reson
SeaBat
7150
24 kHz
15 km
880
150°
Reson
SeaBat
7111
100 kHz
1 km
301
150°
Reson
SeaBat
7125
(ROV)
400 kHz
0.6 km
512
128°
Table 2.2. 1: Echosounder attributes
ROV multibeam
The SeaBat 7125 multi-beam echo-sounder system has up to 128° swath coverage to
an altitude of at least 100 m. The maximum slant range is greater than 200 m. 512
equally spaced bathymetry soundings are generated per ping. The soundings may
then be corrected for refraction, mechanical offsets between the projector and
hydrophone, sensor offsets, attitude, heading, depth and tide.
attributes of the ROV multibeam are listed in Table 2.2.2 (below)
10
The performance
CHAPTER 2: MATERIALS AND METHODS
Depth
6000 m
Frequency
400 kHz
Beams
512 equidistant
Bathymetric accuracy
0.2% altitude ROV
Resolution
5% altitude ROV
Swath width
3.4 × altitude ROV (< 200 m)
Table 2.2. 2: Performance attributes of the ROV multibeam
Processing
Cleaned multibeam data form the basis for digitally modelling the seafloor
topography - Digital Elevation Model (DEM) creation. A DEM is a raster dataset
containing an elevation value for each of its cells. The program CARIS HIPS & SIPS
7.1 (Computer Aided Resource Information System Hydrographic Information
Processing System & SIPS) was used in NUI Galway to automatically and manually
clean the survey data, and to create DEMs for 3D viewing of the sea floor and
incorporate into a GIS environment. After investigating preliminary results the
multibeam data were re-imported with a navigation filter to contain only readings
within W14°35’ – W14°53’, and a depth filter in the range of 400 – 1100 m. Quality
soundings other than the “0” quality flag were also rejected. The bathymetric data
were cleaned using the subset editor (area-based cleaning), swath editor (line-based
cleaning) and motion data were cleaned using the attitude editor.
The sounding density of the shipborne Reson 7150 multibeam was sufficient to
model DEMs of between 25 m and 75 m resolution. Gaps in the DEMs were
interpolated with a moving average filter (3 x 3 cell size). DEMs with 20 cm to 1 m
grid size were created for the ROV data depending on the quality of the data and the
ROV altitude. At 70 m altitude and ~200 m corridor, DEMs were processed to a
50 cm grid size. The CARIS DEM files were exported as ESRI ASCII grid files and
opened in ArcMap.
11
CHAPTER 2: MATERIALS AND METHODS
Multibeam processing issues were dealt with as follows.
Correcting the ROV depth information
Creating a DEM surface in CARIS showed water depths to be greater than expected.
The GPS height (actual depth) was found to be correct in the attitude editor but the
DEM showed the area to be twice as deep. Depth information in the raw data (s7k
file) includes the depth of the ROV vehicle. However, the depth of the vehicle data is
also included in the delta draft telegram. During the Merge process in CARIS the
delta draft data was added to the measured depths. This resulted in the depth of the
vehicle being added twice (once during creation of the s7k file and again within
CARIS). Correction involved using the Generic Data Parser as follows.
Correcting for GPS tide using Generic Data Parser
1. In CARIS all lines were selected and the process function was used to
compute the GPS tide. In the Attitude Editor, the sensor layout was opened
and the GPS tide was brought over to ‘active’. The GPS tide was then
exported using the Export Wizard ‘HIPS tide’ and opened to view in TextPad
(Fig. 2.1.1)
Fig. 2.2. 1: A GPS tide file
12
CHAPTER 2: MATERIALS AND METHODS
2. In CARIS the Generic Data Parser (GDP) was opened in the Import functions
and the raw data (i.e. the exported GPS tide file) was opened. The File
Header, File Date and Time Stamps were defined and the GDP was then used
to multiply the GPS tide data by -1 (Fig 2.2.2). The GDP could then be run,
updating the existing survey lines. The GDP was closed and the session in
CARIS was saved.
-
Fig. 2.2. 2: Example GPS tide file being multiplied by -1 in the Generic Data Parser.
13
CHAPTER 2: MATERIALS AND METHODS
3. The CARIS session was reopened and the lines were remerged, applying the
GPS tide. A new DEM surface was created. The depths were checked and
found to be correct (Fig 2.2.3.). Correcting the depth reduced the error
threshold. The pre-corrected DEM had a pronounced ‘groove’ in the
southeast which was greatly lessened by correcting the depth (Fig. 2.2.4).
Fig. 2.2. 3: The pre-corrected DEM surface (ROV_10m) shows deeper depths (1208.6m);
and the corrected DEM surface (test_tide_10m), created using remerged lines with the GPS
tide multiplied by -1, shows the true shallower depths (655.82m).
Correcting axial noise using attitude editor
Spiral artefacts in the ROV navigation data were removed by filtering the heading
(rejecting the gyro information). (Fig. 2.2.5)
14
Fig. 2.2. 4: Left: DEM of incorrect depth, 10 m resolution, showing groove with deeper soundings. Right: Depth-corrected DEM, 10 m resolution,
showing groove greatly reduced.
CHAPTER 2: MATERIALS AND METHODS
15
Fig. 2.2. 5: Left shows the axial noise pattern caused by gyro anomalies (highlighted in yellow in the Attitude editor). Right shows the Cleaned data
after the anomalies were manually removed (with interpolation).
CHAPTER 2: MATERIALS AND METHODS
16
CHAPTER 2: MATERIALS AND METHODS
Removing spikes in the data using subset editor and swath editor
The subset and swath editors were both used to remove spikes in the data which, if
left uncleaned, produced circular artefacts in the DEMs (Fig. 2.2.6). As the subset
editor is area-based it was the preferred cleaning method because it could remove
anomalous soundings more efficiently. However, as the mounds are 3-dimensional
structures they could interfere with cleaning, because in some cases it was not
possible to select the anomalies without also selecting part of the mound (i.e. the
noise features were too close to the true sounding surface). In these instances the
spikes were cleaned line by line in swath editor. The mounds themselves also
showed circular features displaying a lumpy ‘cauliflower-like’ texture. In swath
editor these lumps do not show up as anomalous spikes but as coherent undulations
(Fig. 2.2.7), indicating that these are real topographical or geomorphological features
– in this case, coral colonies.
Fig. 2.2. 6: Example of the effect of cleaning using the swath editor (left uncleaned, right
cleaned). Removing the spikes removed the circular artefacts seen around the mound.
17
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 7: 'Lumpy' texture (circular features) on the mound is shown as smooth undulations
in swath editor (top), as opposed to the spikes associated with circular artefacts off the
mound. This indicates that the ‘cauliflower’ texture represents real features – interpreted as
coral colonies based on video evidence from the area.
Noise in the Shipborne 7111
Theoretically Reson 7111 (100 kHz) multibeam produces better resolution images in
comparison to the 7150 (24 kHz) multibeam as it is higher frequency data. However,
the Reson 7111 Multibeam data was discovered to be noisy and contained large gaps.
The DEM surface showed the noise to be mainly within the outer beams and this was
verified by observing the swaths in the swath editor (Fig. 2.2.8). Cleaning the data
required omission of the outer beams. One of the lines was also rejected due to poor
signal quality in the deeper regions due to data density decrease with depth as a
consequence of the beam geometry and attenuation of frequency (Fig. 2.2.9). In the
cruise report the poor quality is noted as ‘signal quality and breadth of coverage
degrades very rapidly to a depth not exceeding 400 m.’ Thus, even though a 25 m
DEM of the 7111 data (Fig. 2.2.10) showed the mounds in higher resolution in
comparison to the 7150 25 m DEM, it was largely unusable due to data gaps. A 50 m
18
CHAPTER 2: MATERIALS AND METHODS
DEM shows fewer data gaps than the (Fig. 2.2.11) but inherently involved loss of
resolution. A 75 m DEM of the 7111 data produced the cleanest image (Fig.2.2.12)
but resolution was poor.
Fig. 2.2. 8: A 20 m resolution DEM of the noisy 7111 100 kHz data (top) and the associated
swath profiles in swath editor in plan and rear view (bottom).
19
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 9: The subset editor shows this part of the survey line has little/no information and
that the noise is shallower than the true seabed surface, which can be observed in the
adjacent survey line. Reson 7111 (100 kHz) shipborne multibeam. Top: Subset Editor aerial
view of DEM surface (25 m resolution) showing the difference between the normal line and a
line with noise. Bottom: lateral view. The red line was rejected due to poor signal quality in
the deeper regions. Data density decreases with depth as a consequence of the beam
geometry and attenuation of frequency.
20
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 10: 7111: 25 m DEM with many gaps in data coverage.
21
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 11: Reson 7111, 50 m DEM shows less data gaps than the 25m DEM but loss of
resolution (compare with Fig. 3.1.13). Blue line is ship track which had bad multibeam data
and was ignored. Colour gradation symbolises depth with red-blue representing shallow to
deep (~500-900 m).
22
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 12: The Reson 7111 (100 kHz) 75m DEM produces the cleanest image without data
gaps, but resolution is poor. Blue line is ship track which had bad multibeam data and which
was ignored. Colour gradation symbolises depth with red-blue representing shallow to deep
(~500-900 m).
GIS
After importing the grids into the ArcMap raster grid format, a low pass 3-by-3 filter
was used over the raster to smooth the entire input raster, reduce the significance of
anomalous cells and fill in data gaps. Algorithms were run to calculate slope
derivatives and contours, and to generate shaded relief images (hillshades). The slope
is calculated from the maximum change in elevation over the distance between the
cell and its eight neighbours (Burrough et al., 1998). Creating hillshades of
23
CHAPTER 2: MATERIALS AND METHODS
resampled bathymetry (cubic technique) produced the best maps, due to their
smoother appearance, and involved little loss of information (Fig. 3.1.16).
Fig. 2.2. 13: ROV 7125 (400 kHz) data: Left - Hillshaded DEM 20 cm. Right - Hillshade of
the resampled bathymetry data (using cubic technique); produced a smoother image with
little loss of information
Investigation of the mound morphologies was also undertaken in the GIS
environment. In order to carry out quantitative morphometric analyses each mound
first needed to be defined systematically as follows.
Mound boundary definition
Correa et al. (2012) constrained the minimum mound area based on a 3 x 3 pixel
matrix, as smaller matrices do not contain sufficient pixels. For the 25 m DEM this
entails that the minimum mound area is 5,625 m2, whereas the ROV 1 m DEM could
theoretically resolve footprint areas down to 9 m2. As the whole study area was not
covered by the ROV, 5,625 m2 was chosen to represent the minimum mound area.
Mound perimeters are difficult to discern by eye because some mounds are coalesced
(mound complexes) and/or have scours, scarps and sediment onlap associated with
them. It is not acceptable to define the base of the mounds from depth contours or
cross sections because the seafloor slope confounds observations. Dorschel et al.
24
CHAPTER 2: MATERIALS AND METHODS
(2010) used the 6° slope contour to delineate the beginning of the mound compared
to the seafloor. 6° was chosen as a compromise between the greater influence of
artefacts at smaller slope angles and the under-detection of mounds at larger slope
angles (Dorschel et al., 2010). Correa et al. (2012) used the 8° slope contour based
on attempts to manually delineate the mound perimeters. In this study a number of
slope degree contours were created (3-10°) to analyse which best depicted the mound
bases for the study area. The agreement of mound shapes with the 7° slope contour
entailed that this work uses an intermediate between the thresholds used by Dorschel
et al. (2010) and Correa et al. (2012) The steps in defining the mound boundaries can
be viewed in Fig. 2.2.14. The first step was to create a slope raster from the
bathymetry and extract the 7° slope contour from this raster. Polygons created from
these slope contours were filtered to contain only polygons with an area greater than
5,625 m2. Polygons within these polygons were removed. From the resulting
polygons, mounds were identified by consulting 3D bathymetry DEMs and focal
maximum statistic rasters. Local maxima - maxima of a grid are found by comparing
a 3 × 3 pixel (75 × 75 m) focal maximum of that grid to itself i.e. by subtracting the
focal statistics maxima from the original DEM. Local maxima occur where
overlapping pixels of the two surfaces are the same. The difference raster effectively
highlighted the mound shapes as well as showing that some mounds have two or
more local maxima (Fig. 2.2.14, D). It was hoped that polygons of the 7° slope
contour polygons would approximate the mound footprint shapes and thus areas.
However in most cases the polygons had to be manually modified due to the
following: noise (Fig. 2.2.20); mounds were not separated from the scarp (Fig.
2.2.15, C); the 7° slope contour conjoined adjacent mounds within one polygon (Fig.
2.2.15); and smaller mounds were not enclosed by the polygon (Fig. 2.2.16). In
essence, in most cases manual delineation of mound boundaries was necessary, but
the selection was based on standardised information (base defined by 7° slope
contour; area >5,625 m2) in order to be as objective as possible. The defined mound
footprint polygons were processed to extract statistics of area and perimeter. Convex
hull polygons of the defined mound boundaries were created in order to obtain the
length and orientation of the geometric shapes. A convex hull is the smallest convex
polygon that encloses an input feature. The measurements of the orientation are
created by drawing an imaginary line along the longest distance between two vertices
of the convex hull polygons. The convex hull width was not acceptable because it is
25
CHAPTER 2: MATERIALS AND METHODS
the shortest distance between any two vertices of the convex hull, and therefore
minimum-area bounding rectangles were created to calculate the width. The overall
steps in mound extraction is summarised in Fig.2.2.17.
Fig. 2.2. 14: Workflow to define mound footprints: A slope raster (B) was created from the
bathymetry (A) and the 7° slope contour was chosen to define the mound bases (red line
around zoomed-in mounds). C. Polygons created from the 7° slope contour were filtered to
contain only polygons with areas greater than 5,625 m2 (polygons <5,625m2 highlighted in
blue). Polygons in the zoomed-in example show that the 7° slope contour is conjoining the
mounds with the scarp. D. Polygons were manually modified to better represent the
identified mounds after consulting focal maximum statistics. Local maxima occur where
overlapping pixels of the two surfaces are the same. This raster highlighted the mound
shapes aiding mound boundary delineation. It also showed some of the mounds to have two
or more local maxima (red dots highlight peaks in zoomed in example).
26
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 15: Mounds had to be manually modified after the automatic extraction process in
cases where the 7° slope contour (purple line in zoomed-in example) adjoined adjacent
mounds within one polygon.
27
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 16: Comparison of 3D bathymetry (above) to the 7° slope contour polygons (below)
showed that some smaller mounds were not enclosed by the polygons (red dots) and noise
associated with the bathymetric data created false polygons (highlighted in blue).
28
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 17: Summary workflow of steps involved in mound extraction.
29
CHAPTER 2: MATERIALS AND METHODS
Limitations of the mound extraction process:
The mound extraction process highlighted the high level of noise associated with the
7° slope contours because almost half of the polygons (48%) did not correlate to
mound features in the 3D DEM (e.g. Fig. 2.2.18) or the local maximum raster. This
is due to the quality of the multibeam data (gaps and spikes in the data that remained
after cleaning). In addition there were extra polygons created due to the nature of the
area because the scarps were picked out by the slope contour.
As with any automatic extraction method there are inherent errors associated with the
choice of thresholds. The 7° slope contour may not be the best suited to define the
base for all the mounds because each mound represents a natural system whose
characteristics (e.g. geomorphology) are affected by its’ local environment. It was
observed that the orientation calculated from the convex hull polygon did not always
represent the orientation that would be expected from observing the mound ridge
orientation. This was due to the 7° slope contour creating an irregular polygon with a
broad base and differing to the trend of the ridge (Fig. 2.2.19). Correcting for this
effect would involve modifying the mound polygon to represent the more ‘userintuitive’ shape of the mound. The mound base could be defined by the user as the
point where the slope contours indicate flattening of the topographic profile. This,
besides involving more time, would entail a lack of consistency and bring
subjectivity in to the results. Also, the mound size could be that which was picked by
the mound slope contour and the ridge orientation may not reflect the overall
orientation of the mound as it accretes laterally. Thus the predicted mound shapes, as
chosen by the 7° slope contours, were kept.
It must be noted that due to the 25 m resolution it is probable that there is under
detection of small mounds. If small mounds were broad with low relief they would
not be observable: some 7° slope contours that created polygons >5,625 m2 were not
included as mounds because they were not obvious in the 3D bathymetry, nor were
they highlighted in the slope or focal maximum rasters. The ROV 1 m DEM could
theoretically resolve footprint areas down to 9 m2. Thus an attempt to groundtruth the
method was made by performing the same mound extraction process on the ROV
data. The 7° slope contours created from this bathymetry were very noisy however
and the sizes of the mounds were wider than the ROV swath (Fig. 2.2.20).
Nonetheless the high resolution of the ROV data allowed comparison of the number
30
CHAPTER 2: MATERIALS AND METHODS
of mounds observed in the ROV 1 m 3D bathymetry to the number that were
resolved by the mound extraction technique using the 25 m bathymetry, and this
acted as a groundtruthing mechanism for the method. The agreement of the number
of mounds identified allowed confidence in the mound extraction method.
Fig. 2.2. 18: Bathymetric map with zoomed-in area inset showing almost half of the polygons
(48%) created from the 7° slope contours (dark grey) did not correlate to mound features in
the 3D DEM or local maximum raster. Contours that define mound footprints are shown in
pink. The high level of noise is due to the quality of the multibeam data and the nature of the
area because the scarps are picked out by the slope contour.
31
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 19: The polygon created from the 7° slope contour (highlighted in blue) creates an
overall northwest-southeast orientation (black line) quite different from the expected eastwest orientation indicated by the mound ridge morphology (closely spaced slope contours
indicating mound ridge). This indicates limitations of using the 7° slope contour.
32
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 20: The 7° slope contours for the ROV data was very noisy (polygons >9 m2
shown). Also note how the mounds are broader than the width of the swath.
Calculating mound height
Mound height calculation from the defined mound footprints followed the same
principle of Correa et al. (2012) whereby an original DEM is compared to a
hypothetical DEM without mounds. The steps to produce a raster of mound heights
are illustrated in Fig. 2.2.21 and the overall workflow to extract the height values is
shown in Fig 2.2.22. Data enclosed by the mound footprint polygons were first
removed from the original DEM. This DEM with holes where the mounds used to be
was then converted to points and then reinterpolated using the natural neighbour
method (Sibson, 1981) with an output cell size of 25 m (to match that of the original
DEM). The result was the production of a hypothetical DEM without mounds.
33
CHAPTER 2: MATERIALS AND METHODS
Subtraction of this interpolated DEM from a DEM of only the mounds generated a
raster of mound heights. Zonal statistics were used to calculate the highest elevation
within each mound footprint and produce a raster with mound polygons representing
the maximum height found in each. The mound height raster was compared to the
maximum height polygon raster to determine which cells in the mound height raster
correspond to the highest elevation found within each polygon feature. This was
done using a conditional statement in the raster calculator of ArcMap where the
output is a raster depicting only the mound height raster cells with the highest
elevation values and all other cells set to no data:
Con("mound height raster" == " maximum height polygon raster "," mound height
raster ")
The output mound height raster cells with the highest elevation values were
converted to point features and assigned XY coordinates. These maximum height
feature points were then spatially joined to the mound footprint polygons in order to
give the maximum thickness associated with each mound.
34
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 21: Steps to calculate mound heights: (A) Data enclosed by the mound footprint
polygons were removed from the original DEM. (B) This DEM was then reinterpolated to
produce a hypothetical DEM without mounds. Subtraction of a DEM of only the mounds (C)
from the interpolated DEM (B) generated a raster of mound heights (D). Zonal statistics
were then performed to find the maximum height associated with each polygon.
35
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.2. 22: Workflow to extract mound heights.
36
CHAPTER 2: MATERIALS AND METHODS
Identifying mound peaks
The depths to the mounds’ peaks were investigated to see if there was any spatial
relationship between the location of the mounds in the water column and their
(vertical) sizes, which could indicate control by water currents. The mound peaks
were identified by calculating the focal maximum statistics of the bathymetric DEM.
The highest points (peaks) were then spatially joined to their associated mound
footprint polygon.
2.2.2: Sidescan
Backscatter data was recorded within Reson *.s7k raw data files. CARIS uses the
snippet format (which incorporates the bathymetric data to remove water column
noise); therefore the S7K raw data files were unusable in the Caris Geocoder
program. The backscatter was thus processed separately by Garret Duffy (NUI
Galway) in PDS2000 and made available in a raster format for inclusion in ArcMap.
2.2.3: Chirp
A chirp is a signal in which the frequency increases or decreases with time. A
matched filter collapses the received swept frequency modulated signal into a pulse
of short duration, maximizing the signal-to-noise-ratio.
Acquisition
The Chirp sediment sounder was used simultaneously with the multibeam on the
29th September 2011, The SUBOP acquisition system was used with a circular array
of 7 Tonpilz transducers used for transmission and reception (the diameter of each
transducer was 30 cm). Vertical resolution is finer than 30 cm. The frequencies
ranged from 1.7 – 5.3 kHz. The maximum acoustic penetration is ~ 300 ms (TWT).
The chirp data was processed onboard using MATLAB and offshore in the
IFREMER institute by Mathieu Veslin. Pdf results of the data from line
BOBECO122A (7 km, from west to east) and an intersecting line BOBECO119A (2
km, from south to north), and the associated navigation data txt. files, were made
available to the author for this study.
37
CHAPTER 2: MATERIALS AND METHODS
Method
The shotpoints from the apices of the mound-like features were assigned coordinates
from their navigation data and these coordinates were plotted on the multibeam data.
This was completed in order to ensure the features correlated to the mounds seen on
the multibeam and were not artefacts of acquisition. Bathymetric maps of the chirp
transects were produced and inset in to the chirp results. The results were then
interpreted and seismic stratigraphic units were defined.
2.3: Sediment core analysis
2.3.1: Core acquisition
Two Calypso piston cores (CS06 and CS07) with core diameters of 110 mm were
acquired on the 30th September 2011. Core CS07 was acquired 584 m below sea
level [bsl] from the summit of a carbonate mound which showed live cold-water
coral cover (Fig. 2.3.1) and 625.5 cm of section was recovered.
Fig. 2.3. 1: ROV photo from the summit of the carbonate mound (from which core CS07
was acquired) showing live coral growth (white) and infilled dead coral framework. Scale of
photo is ~2m wide. The poor quality of the photograph is due to scattering by sediment
particles.
38
CHAPTER 2: MATERIALS AND METHODS
Core CS06 was acquired 655 m bsl from the adjacent seafloor sediments 966 m to
the southwest, and produced 297.5 cm of off-mound stratigraphy. Both cores were
cut into 1 m or less sections on board. The core locations are shown in Fig 2.3.2 and
the core attributes are provided in Table 2.3. 1. The cores were logged at the
IFREMER (Brest) institute in May 2012.
Fig. 2.3. 2: Seafloor locations of cores CS06 and CS07. CS07 was taken from the summit of
a coral mound, and core CS06 was taken from the adjacent seafloor 966 m to the southwest.
ROV track, Reson 7125 400 kHz multibeam data (hillshaded DEM at 1m resolution),
overlain on shipborne multibeam data, Reson 7150 24 kHz (hillshaded DEM at 50 m
resolution).
39
CHAPTER 2: MATERIALS AND METHODS
Core ID
CS07
CS06
Lat
N 51 °21.243
N 51 °21.017
Long
W 014 °43.044
W 014 °43.693
Water depth
584 m
655 m
Section
Length (interval)
Length (interval)
1
97.5 cm (0 – 97.5 cmdc)
100 cm (0 – 100 cmdc)
2
100.5 cm (97.5 - 198 cmdc)
100 cm (100 – 200 cmdc)
3
100 cm (198-298 - cmdc)
97.5 cm (200 – 297.5 cmdc)
4
100.5 cm (298 – 398.5 cmdc)
5
100.5 cm (398.5 – 499 cmdc)
N/A
6
100.5 cm (499 – 599.5 cmdc)
7
26 (599.5 – 625.5 cmdc)
Total Core
length
625.5 cm
297.5 cm
Table 2.3. 1: CS06 and CS07 sediment core attributes.
2.3.2: GEOTEK Multi-sensor Core Logger Analysis (MSCL)
A GEOTEK Multi-Sensor Core Logger Standard [MSCL-S] (Fig. 2.3.3) was deployed on the
two recovered sediment cores in order to provide continuous measurements of:

Gamma-ray attenuation,

P-wave travel time, and

Magnetic susceptibility.
40
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.3. 3: The MSCL-S. Image courtesy of www.geotek.co.uk.
The technical specifications in the following descriptions were obtained from the online
GEOTEK MSCL-S description1.
Gamma-ray attenuation: The gamma ray source (Caesium-137) and detector are both
mounted on the sensor stand and are aligned with the centre of the core. Emission of a narrow
beam of collimated parallel gamma rays from the source (with energies principally at 0.662
MeV) causes photons to pass through the core and these are detected at the other side of the
sample. The density of the material is determined by measuring the number of transmitted
gamma photons that pass through the core unattenuated. Depending upon count time the
density resolution may be better than 1%. Gamma density or GRAPE [gamma ray attenuation
porosity evaluator] data therefore provides a high-resolution record of bulk density
(indicating how tightly the sediment is packed together), which may be used as a broad
lithology indicator and highlight porosity changes. Also, depending on the water content and
the degree of compaction, the average bulk density of wet mud is commonly between 1 and 2
g cm-3, whereas the density of coral aragonite is 2.7g cm-3 (Douarin et al., 2013). Therefore,
MSCL results may also be used to identify possible accumulations of coral in sediment cores
by highlighting regions of higher density (e.g. Douarin et al., 2013).
1
www.geotek.co.uk/products/mscl-s.
41
CHAPTER 2: MATERIALS AND METHODS
P-wave velocity: This mainly varies with the lithology, porosity, and bulk density of the
sediment, but it is also controlled by a variety of other factors such as consolidation,
lithification, fracturing and the occurrence of free gas (e.g. Blum, 1997). P-wave velocity is
measured by passing an ultrasonic pulse through the sediments at 250-500 kHz using piezoelectric ceramic transducers that are spring-loaded against the sample. The p-wave velocity
is accurate to approximately 0.2%, depending on the condition of the core.
Magnetic susceptibility: This refers to the degree of magnetization of a material in response
to an applied magnetic field. The magnetic susceptibility was acquired with a Bartington
looped sensor and the results have 5% calibration accuracy. Material in the vicinity of the
sensor that have a magnetic susceptibility cause a change in the oscillator frequency and this
pulsed frequency information is converted into magnetic susceptibility values. Sediment
containing a high proportion of ferro- or paramagnetic minerals such as magnetite and iron
sulphide has high magnetic susceptibility; biogenic material, such as silica, calcite, quartz and
feldspar show low or even negative magnetic susceptibility values (Rothwell, 2006).
Thus multi-sensor core logger values and trends can be important parameters relating to
sediment character, provenance, palaeoclimate and/or diagenetic environment.
Method
The GEOTEK MSCL-S was deployed at the IFREMER institute in Brest on the 31st May
2012 to measure the three parameters (outlined above) of the unopened core sections. The
room temperature was measured and used to calibrate the MSCL. As the cores contained
water-saturated sediments the calibration section consisted of a cylindrical piece of
aluminium of varying thickness surrounded completely by water in a sealed liner. Once
calibrated, the core sections were carefully placed with the halfway horizontal line (marked
during acquisition) parallel to the rack and scanned separately at a scanning interval of 1 cm.
The pushing element, driven by a stepper motor, can position the core with an accuracy of
better than 0.5mm. All the data were automatically correlated because the computer
controlling the stepper also controls the sensors. As there were breaks due to the sections the
data is corrupted by air gaps, but these are obvious artefacts that are easily correlated to their
depth downcore. It took approximately 22 minutes to scan a single 1 m section of sediment.
42
CHAPTER 2: MATERIALS AND METHODS
The MSCL data was processed by Mickael Rovere at IFREMER and provided in Excel
spreadsheet format for further analysis. The preliminary graphs were also consulted at
IFREMER in order to investigate whether the cores required freezing prior to splitting.
The data of section 7 at the very base of core CS07 (26 cm in length - see Table 2.3.1) was
rendered unusable due to anomalous p-wave velocity readings and was disregarded in data
production. This was possibly due to bad contact of the P-wave transducers with the core
during the measurement.
2.3.3: Sedimentological logging
Both cores were split in half lengthwise and visually inspected and carefully logged at
IFREMER. Core CS07 Section 1 was frozen before splitting due to the observed variability
in the p-wave velocity MSCL readings (see Section 7.2 and Fig. 7.2.2) which indicated a
possible coral accumulation. Freezing prevented the coral branches from being dragged down
the core by the wire cutter, which would have undoubtedly destroyed sedimentary structures
and textures in the process. The rest of the core sections were split as per normal procedure
using wire cutters. Individual core sections were numbered according to their position in the
core. The nomenclature used utilised the core number first, followed by the section number,
from the top to the base. One half of each core was stored for archival purposes in the core
repository at IFREMER.
Method
The 'working' split core section was laid out on a bench and the upper surface of the sediment
carefully skimmed (cleaned) using a broad palate knife to ensure accurate representation of
sedimentological features. The working core section was then immediately photographed
alongside a scale using a high-resolution camera. Detailed (cm-scale) stratigraphic logging of
the core was then undertaken. Lithology, colour and sedimentary structures were described.
Colour was determined visually using the Munsell colour chart. Coral fragments for dating
were also extracted at this time.
As the sediments commonly included admixtures of sedimentary materials the Folk (1954)
classification, incorporating the modification of Flemming (2000) was used, which
43
CHAPTER 2: MATERIALS AND METHODS
distinguishes six sediment types on the basis of mud and sand content as shown in Table 2.3.
2:
Sand
<5% mud
Slightly muddy sand
5-25% mud
Muddy sand
25-50% mud
Sandy mud
50-75% mud
Slightly sandy mud
75-95% mud
Mud
>95% mud
Table 2.3. 2: Folk (1954) sediment classification, incorporating the modification of Flemming (2000).
This scheme was used for visual classification.
The resultant hand-drawn logs were digitised at NUIG using Adobe Illustrator. The presented
logs were subsequently modified and improved through integration with the results of
additional analyses (grain size, microscopic observations, XRF, and MSCL).
2.3.4: AVATECH XRF
During X-ray Fluorescence [XRF] analysis a prism is lowered onto a sediment surface and
elements in a given sample are exposed to a beam of photons (high intensity X-rays at a 45°
angle).
The ejection of electrons from inner atomic shells creates a vacancy that is
subsequently filled by electrons from the outer shells, and this release of energy is emitted as
a pulse of secondary X-radiation. Different elements generate characteristic fluorescent Xrays (Fig. 2.3.4). The intensity of the peak in the XRF spectrum, i.e. the total counts, is also
known as net area peak and is proportional to the concentration of the corresponding element
in the analysed sample. Thus XRF core scanning rapidly (and non-destructively) measures
down-core changes in the chemical composition (major, minor and trace elements) of split
sediment cores.
44
CHAPTER 2: MATERIALS AND METHODS
Fig. 2.3. 4: Simplified diagram showing the principle of XRF core logging (Richter et al., 2006).
The emitted X-rays contain information only for a thin superficial layer of sediment, of about
4 mm (Toms, 2010); the penetration depth depends on the wavelength of the fluorescent
radiation and the chemical composition of the matrix (Jenkins et al., 1969). Therefore,
surface roughness can affect the element signal (Croudace et al., 2006). Also, as the response
depth varies from element to element, large particles attenuate the fluorescent radiation of
elements other than those within the particle. Thus useful results can be expected from the
XRF scanning of clays and silts, but data from sandy sediments involves more careful
interpretation. Additionally, large ice-rafted granules that coincide with the XRF track can
introduce spikes in the data. Poor signal can be identified by abnormal fluctuations in the
total counts per second [CPS] (Toms, 2010).
The primary data produced from the AVAATECH XRF scanner are elemental profiles; and,
in this way, XRF results provide geochemical proxy data for measuring environmental
change. It is established that calcium carbonate records in the Atlantic Ocean can be related
to
glacial–interglacial
cycles,
with
interglacial periods having higher
carbonate
concentrations (e.g. Balsam & McCoy, 1987). This is because pelagic and neritic carbonate
accumulation generally increases during interglacials, whereas siliciclastic input is much
higher during glacial periods. Thus, XRF data allows quantitative changes in carbonate and
siliciclastic sedimentation to be determined based on changes in carbonate (Ca, Sr) and
terrestrial (Fe, Al, K) sourced elements. Normalising the signal using element ratios produces
clearer peaks and troughs to highlight variations in particular components. In the North
Atlantic, the Ca/Fe ratio offers a good proxy for discriminating glacial/interglacial cycles
(Balsam & McCoy, 1987). The Ca/Fe compares the detrital Fe signal with the mostly
biogenic Ca signal (e.g. Toms, 2010). Hodell et al. (2008) identified that Ca/Sr is a potential
proxy for detrital carbonate (dolomite) that can be used to recognise Heinrich Layers.
45
CHAPTER 2: MATERIALS AND METHODS
Additionally, as aragonite generally has high Sr and calcite low Sr, the Sr/Ca ratio can
indicate relative aragonite contribution to particular sediments (Rothwell, 2006).
Method
XRF scanning on the archival half of the split sediment cores was completed using the
AVAATECH XRF instrument (Fig. 2.3.5) at IFREMER on the 30th April 2012, covering the
atomic mass range from Al to U.
Fig. 2.3. 5: The AVATECH XRF scanner at the Royal Netherlands Institute for Sea Research (NIOZ,
Texel, The Netherlands).
XRF scans were obtained for core CS06 Sections 1 - 3. Time constraints entailed that only
Section 1 and Section 3 were scanned from core CS07. These sections were chosen in order
to investigate preliminary stratigraphic interpretations (a perceived glacial–interglacial cycle).
The core samples were prepared by scraping the split core surface to remove irregularities
from core slicing and therefore minimise scattering affects, and to remove dust particles or
contaminants that may have settled since the initial core opening. A 40 µm ultralene film was
46
CHAPTER 2: MATERIALS AND METHODS
applied to further diminish surface roughness, prevent contamination of the prism unit during
logging and to prevent the core from drying-out. The scanner was set to skip the first four cm
of both CS07 section 1 (because the first 2.5 cm was void) and CS07 section 3 (due to the
rough nature of sediment). Element intensities were analysed at 1 cm intervals, with each
measurement taken over an area of 1 cm2 and a count time of 30 seconds, with:
1. An X-Ray tube voltage of 10 kV, X-Ray current of 0.6mA and no filter for counts of:
Al, Si, S, Cl, K, Ca, Ti, Mn, Fe, and Rh; and
2. An X-Ray voltage of 30 kV, X-Ray current of 1mA and Pd filter, for counts of: Fe,
Ni, Zn, Br, Rb, Sr, Zr, Rh and Pb.
The measured XRF spectra were processed by the IFREMER institute and the processed data
were provided as Excel spreadsheets with results given in total counts per element. Results
are presented as percentages of the total element counts (% TC). It must be noted that,
although the unit concentrations (parts per million) represent down-core elemental variations,
they are semi-quantitative relative values and have not been calibrated to standards. However,
studies comparing XRF surface scanner data against bulk sample XRF analysis (e.g. Jansen
et al., 1998; Croudace et al., 2006) indicate that the data are indeed comparable. Curves were
produced for all of the elements (as percentage of total counts) and ratio profiles. Several
ratios and element intensities were compared, however, only the Ca/Fe, Sr/Ca and Ca/Sr
ratios will be presented. The Ti, K, Mn and Al element intensities all showed the same trend
as the Fe intensity, but as the Ca/Fe ratio showed the clearest signal it was selected as a
geochemical proxy for the input of terrestrial material. An ideal XRF sample is
homogeneous, dry, and has a smooth surface. Due to the inhomogeneous and rough nature of
the surface of both the split sediment cores, the semi-quantitative records of the relative
variability in elemental composition of sediments downcore must be carefully interpreted.
The elemental and ratio profiles are displayed against the core photographs and graphic logs
to help identify perturbations created by local surface variations or anomalous grain effects.
2.3.5: Grain size analysis
Grain size analysis was carried out to examine trends in processes related to the dynamic
conditions of sediment transportation and deposition. It cannot be distinguished whether an
increase in grain size is due to an increase in the mean speed or an increase in the variability
of speed; nevertheless, it is a good indicator of the energy of an environment.
47
CHAPTER 2: MATERIALS AND METHODS
The grain size fractions used (Table 2.3. 3) are based on the Udden-Wentworth grain-size
classification of terrigenous sediments (Wentworth, 1922). The sediment is divided into its
coarse (>63 µm) and fine (<63 µm) components, which separates the sand-sized sediment
from the mud (silts and clays). Five subsequent subdivisions of the coarse fraction are
differentiated:
Udden-Wentworth term:
Very coarse sand, granules,
pebbles & above
Scale:
>1000 µm
Coarse sand
500-1000 µm
Medium sand
250-500 µm
Fine sand
125-250 µm
Very fine sand
63-125 µm
Silt & mud
<63 µm
Coarse component
Fine component
Table 2.3. 3: Grain size fractions based on the Udden-Wentworth grain-size classification of
terrigenous sediments (Wentworth, 1922)
Two grain-size analytical methods were used to assess the sedimentology: Laser Diffraction
Grain Size Analysis and Sieve Based Grain Size Analysis. The relative proportions of mud
(<63 μm), sand (>63 μm & <2 mm) and gravel (>2 mm) were displayed as a stacked plot
alongside the graphic core log to portray vertical textural variations. As well as providing
separate insights into the composition of the sediments, the results of the two techniques were
also graphed for comparative purposes (Appendix C).
Laser Diffraction Grain Size Analysis (LDGSA)
Laser particle sizing was used to analyse the bulk texture of the sediments. LDGSA was
undertaken using the Granulometric Laser: Coulter LS200 version 3.39 for both cores. The
LS200 measures grain sizes within a size range from 0.375 µm - 2000 µm (i.e. it does not
measure gravel or pebbles). Laser Diffraction Particle Size Analyzers measure particle size
48
CHAPTER 2: MATERIALS AND METHODS
using the classic Mie theory of light scattering (e.g. Hahn, 2006) and the LS200 uses reverse
Fourier lens optics incorporated in a binocular lens system. The LS200, typically, has better
than 1% reproducibility2.
The grain-size data included carbonate as well as siliciclastic grains. Many studies remove
the carbonate component in order to investigate the sortable silt fraction, which is used as a
proxy for bottom current intensity on fine-grained silts and clays. However, as discussed by
Toms (2010), the Porcupine Bank deposits are composed of a large textural range and
therefore the sortable silt content was not undertaken as part of this study. In addition, the
LGSA results were also later used to compare to the SBGSA results; the SBGSA samples had
to retain the carbonate content in order to undertake Ice-Rafted Debris counting (involving
the ratio of lithic fragments to planktonic foraminifera).
Method
A blank was run through the LS200 for calibration at 0 % obscuration. The pA values of the
blank were checked to verify that 50 < 2000 pA and 62 < 4000 pA. A two mm sieve was
placed over the sample input entrance in order to retain larger particles that could obstruct the
flow circuit of the analyser. A small amount of representative sediment was skimmed using a
small spatula, dispersed in tap water and was then mixed by vibration using the Retsch
mm200 for two minutes at 19Hz. The sample was washed into the analyser with distilled
water and the obscuration of the laser beam was measured. If the obscuration lay outside 8-20
% more of the sample was added (for < 8 %) or the sample was diluted (for > 20 %). The
sieve was washed in distilled water before the introduction of each sample.
Sieve Based Grain Size Analysis (SBGSA)
SBGSA was carried out at NUIG for both cores in order to fully assess the sedimentary
characteristics of each sample (including particles >2000 µm) and to facilitate the collection
of the foraminiferal sample set.
2
www.beckmancoulter.com
49
CHAPTER 2: MATERIALS AND METHODS
Method
Sediment samples, approximately 1 cm in stratigraphic thickness, were taken at IFREMER at
5 cm intervals for CS06 and Section 1 of CS07. The remaining sections of CS07 were
sampled at 10 cm intervals. The material from the outer edges (~1 cm) was discarded in all
cases, to avoid contamination by disturbed and reworked sediment due to friction from the
metal coring cylinder during recovery. The sediment samples were removed using two semicircular knives and stored in plastic zip-lock bags. The voids in the core section were refilled
with foam blocks as sampling progressed down-core, to ensure minimal disturbance of the
adjacent sediments.
At NUIG, labelled beakers were weighed (on a scale sensitive to 0.1 g) before the samples
were loaded for drying. The samples were then oven dried at 50-60°C (but never above 70°C
as destruction of foraminiferal shells due to cracking may occur). The samples were slowly
dried until completely desiccated (i.e. there was no further loss of weight with time). The
beakers with the dried samples were then reweighed. Subtraction of the original beaker
weight from this value produced a dry weight for the sediment sub-sample.
Wet sieving was carried out in one run and split into four size fractions: <63 µm, 63-250 µm, 250500 µm and 500-2000 µm (
Table 2.3. 4). The <63 μm fraction was not collected. This was due to the practical recovery
problem of very long settling times required to recover this portion of each sample. Samples
were poured onto the sieve stack and washed gently with water. The separated fractions from
each sample were collected, redried (at 60°C) and then reweighed. The raw data were
converted into weight percentages (wt %):
Dry weight (g) retained on a sieve
Dry weight (g) of total sample.
× 100
The size of the unrecovered <63 μm fraction (silt and mud) was calculated by subtracting the
combined weight of the separated coarser fractions from the overall weight of the dry sample.
Grade
Aperture
Aperture
(µm)
(ɸ value)
50
CHAPTER 2: MATERIALS AND METHODS
Gravel
2000
-1
Coarse Sand
500
1
Medium Sand
250-500
2
Fine sand
63-250
4
Top of stack
Base of stack
Table 2.3. 4: Wet-sieve stack configuration with Udden-Wentworth grade, μm and Phi (ɸ) values.
2.3.6: Ice-rafted debris counting
‘Ice-Rafted Debris in marine sediments is a direct indicator of the presence of glacial ice extending to sea
level on adjacent landmasses, and, therefore, is an important palaeoclimatic signal from the mid- to high
latitudes.’ (St John, 1999)
Increases in the abundance of Ice Rafted Debris [IRD] indicate periods of continental
glaciation, and decreases reflect periods of climatic warming (e.g. Ruddiman, 1977; Heinrich,
1988; Krissek, 1985). It is generally assumed that icebergs transport sediment containing
particles of all grain sizes and have a grain size distribution similar to a till (Grobe, 1987).
Current erosion of the finer fraction may modify this sediment and subsequent addition of
particles may produce a compound glacial marine sediment (Grobe, 1987). IRD has been
defined by many authors as any detrital material >125 μm in a marine environment that was
transported by floating ice (e.g. Allen & Warnke, 1991; Breza, 1992). Apart from turbidity
currents transporting coarse sediment down continental slopes, there are no other
sedimentological processes capable of carrying grains of this size so far offshore and deposit
them along with hemipelagic mud (Breza, 1992). However, transport of sand by bottom
currents (e.g. contourites) at the Porcupine sample site is likely. Using scanning electron
microscopy investigations (e.g. Shi et al., 2010) have attempted to discriminate grains of
51
CHAPTER 2: MATERIALS AND METHODS
glacial origin from those which have undergone extensive transport and reworking by
seafloor currents. This problem was overcome in this particular study by only considering the
>250 µm fraction. Other studies in the Pacific (von Huene et al., 1973; Krissek, 1985), the
Sub Antarctic/South Atlantic (Allen & Warnke, 1991), and the North Atlantic (Knutz et al.,
2002) have also shown that the > 250 μm fraction is a valid indicator of IRD abundance.
The lithogenic components of deep marine sediments are usually either ice-rafted or volcanic
in origin. Volcanic particles were not classed separately herein as only grains > 250 μm were
investigated and any source for eruptive fall out was deemed far enough away as to not be
deposited in this size fraction (Dr Shane Tyrell, NUIG, pers. comm. 2013). Thus in this
present study, due to the size of the fraction and the isolation of the area, all terrigenous
grains >250 µm are assumed to have originated by ice-rafting, although it is recognised that
smaller sized sediment particles may have undergone transport by bottom currents
subsequent to deposition.
There are a different methods to determine the IRD content using >250 μm fraction. Some
studies count and compare only the clastic component (e.g. Breza 1992) and many use the
volumetric concentration of IRD within samples to determine IRD flux and bulk mass
accumulation rates (e.g. Allen and Warnke, 1991; Scourse et al., 2009). Other studies (e.g.
Heinrich, 1988; Fronval et al., 1998; Toms, 2010) have shown a clear relationship between a
high IRD content and a low foraminifera abundance and vice versa: interglacial sediments
having an increase in planktonic foraminifera and decreased IRD; glacial episodes
demonstrating a decrease in planktonic foraminifera and an increase in IRD. It was decided to
follow the method of Toms (2010) because of the recognised effect of bottom current
winnowing on the Porcupine Bank (Øvrebø, 2005; Toms, 2010). The latter author used the
work of Poore & Berggren (1975), whereby an IRD Index was calculated on the basis of the
ratio of IRD grains to planktonic foraminifera:
IRD
IRD Index (%IRD)
× 100
IRD + Total planktonic foraminifera
Due to the very different nature of the two cores the IRD analysis undertaken for core CS06
was not feasible to apply to core CS07 due to the large quantities of coral hash diluting the
samples.
52
CHAPTER 2: MATERIALS AND METHODS
CS06 IRD count
For each sediment sub-sample the >250 to 500 μm fraction was combined with the >500 μm
<2 mm fraction (both obtained from the wet sieving procedure (see Section 2.3.6). The
resultant >250 μm to 2 mm fraction was then used for the IRD characterisation. Samples
were split using a microsplitter in order to achieve a representative fraction. The sample
residues were sprinkled onto a 45 square micropalaeontological picking tray and a random
number generator picked squares to be analysed until 500 counts were made under a
stereomicroscope. Every IRD grain and planktonic foraminifera in the field of view was
counted and recorded until the required tally of +500 specimens was achieved.
CS07 IRD count
It became apparent after viewing the sediment of CS07 that the method of Toms (2010) was
not suitable for IRD assessment, because the samples were very much diluted by the high
quantity of biogenic debris. Thierens et al. (2013), who investigated the Challenger mound,
reported the sediment proxy ‘percentage planktonic carbonate’ which is based on counts of
300 >150 μm planktonic foraminifera grains and the associated lithics. For core CS07 this
proxy was adopted except a minimum of 200 planktonic foraminifera were counted instead of
300. This was because the larger size fraction used in this study (>250 μm in comparison to >
150 μm) contains much less planktonic foraminifera as well as the aforementioned issue of
dilution of the sample by the mound organisms. Therefore the same procedure was
undertaken for IRD characterisation as that for CS06 except that for CS07 every IRD grain
and planktonic foraminifera in view was counted and recorded until the required tally of +200
specimens was achieved.
2.3.7: Core CS07 grain composition analysis
The sediment grain composition of core CS07 was investigated in order to qualitatively
describe different facies and to assess coral input. The procedure of the grain composition
analysis followed that of the IRD count (described above) except that all particles under the
stereomicroscope were counted until a tally of +300 was reached. Depending on whether 200
planktonic foraminifera were included in this tally, the IRD count either continued or ceased
53
CHAPTER 2: MATERIALS AND METHODS
after completion of the grain composition analysis count. The grains were classified as
planktonic biogenic, benthic biogenic (with a number of categories e.g. bryozoan, ostracods,
serpulid etc.) or terrestrial in origin, and presented as percentage of the total number of grains
counted.
2.3.8: Dating
Radiocarbon dating
To obtain an Accelerator Mass Spectrometry [AMS] 14C age ~10 mg of foraminifera are
required (e.g. Andree, 2006). Therefore 500 shells of adult (250–500 μm) surface dwelling
planktonic foraminifera (Globigerina sp.) were picked from the sieved sediments using a fine
brush and a binocular microscope; for three intervals of core CS06 and four intervals of
CS07. Additionally, brachiopods from five intervals of core CS06 were recovered from the
>2 mm fraction. The samples (Table 2.3. 5) were submitted to the University of Heidelberg
Institute of Environmental Physics facility for radiometric dating in September 2013.
Core, section, cmdc
Material
Number of shells
CS06 S1 55
Vial of brachiopod material
1
CS06 S2 100
Vial of brachiopod material
2
CS06 S2 140
Vial of brachiopod material
8
CS06 S2 165
Vial of brachiopod material
8
CS06 S3 210
Vial of brachiopod material
4
CS06 S1 25
Slide of planktonic forams
500
CS06 S2 165
Vial of planktonic forams
500
CS06 S3 290
TOTAL CS06
Vial of planktonic forams
8
500
CS07 S2 140
Vial of planktonic forams
500
CS07 S3 250
Vial of planktonic forams
500
CS07 S3 290
Slide of planktonic forams
500
CS07 S6 595
TOTAL CS07
Vial of planktonic forams
4
500
Table 2.3. 5: Material from
CS06 and CS07 that were
radiocarbon dated using an
Accelerator Mass
Spectrometer (AMS).
The brachiopods were
leached with 4% HCl prior hydrolysis, with a sample mass loss between about 20 and 30%.
The sampled were then hydrolised with 3N HCl under a vacuum and dried with water traps.
The CO2 was then converted into graphite by reducing with ultrapure hydrogen (H2) gas in
the presence of an iron catalyst (Alfar Aesar, -325mesh) at a temperature of 575°C.
54
CHAPTER 2: MATERIALS AND METHODS
The foraminifera and brachiopod samples were measured at the Klaus-Tschira Laboratory for
Scientific Dating, Mannheim (Germany) with a MICADAS-AMS system (Synal et al., 2007;
Kromer et al., 2012). The 14C results were normalised -25permil according to their
individual d13C signature measured contemporaneously with the AMS. Values were
calibrated using the CalPal software of Weninger (2004) using the CalCurve:
CalPal_2007_HULU3.
U-Th dating
Coral clasts (> 1 cm; Table 2.3. 6) were sampled from the freshly split and opened cores at
IFREMER and sent for dating to the CNRS laboratory, Gif Sur Yvette in France. Data was
acquired using the ICP Q mass spectrometer (iCAP ThermoFisher) after mechanical and
chemical coral cleaning according to Douville et al. (2010).
Length x Width (mm)
Genera/species:
BobEco CS07 S4, 311
15 x 15
Lophelia pertusa
BobEco CS07 S4, 311
20 x 15; 25 x 25
Madrepora oculata
BobEco CS07 S5, 423.5
26 x 15
Lophelia pertusa
BobEco CS07 S5, 440
25 x 20
Lophelia pertusa
BobEco CS07 S6, 556
17 x 15
Lophelia pertusa
Sample, CMDC
Table 2.3. 6: Attributes of the coral clast samples submitted for U-Th dating.
2.3.9: Core CS06 mound base reflector sediment
Geotechnical results showed that the coring of CS06 was abruptly stopped after 1.5 seconds
due to encountering a firm and highly resistant layer. The ~400-500 cm of penetration
corresponds to the depth of the mound base reflector highlighted in the chirp data (see Fig.
5.3.1). The recovered base cutter and catcher sediment (named ‘Bas Ogive’ and 'Peau
d'Orange respectively after their locations on the piston corer, see Fig 2.3.6) contained
moderately lithified reddish to green pebbly breccio-conglomeratic sediment. Carbonate
3
www.calpal-online.de
55
CHAPTER 2: MATERIALS AND METHODS
cement was observed adhering to the pebbles and gravels that were encased in carbonate silty
mud and the presence of chalk was also noted.
It is assumed that the rock broke up due to the release of pressure during coring. The Bas
Ogive [BOg] sediment was wet-sieved in NUI Galway for grain size analysis. In order to
investigate the composition of the clasts and the nature of the carbonate cement seven 40 µm
thin sections (BG1-BG7) were prepared from the >63 µm fraction of sieved sediment, for
analysis under normal and polarised light - plane polarised (PPL) and cross polarised (XPL).
Samples of the cement were scraped off clasts of the core catcher Peau d'Orange and sent to
Dr. Tom Dunkley Jones (the School of Geography, Earth and Environmental Sciences,
University of Birmingham) in order to investigate the nannofossil content.
Fig. 2.3. 6: The piston corer set-up with the location of the Peau d'Orange (catcher) and Bas Ogive
(cutter) highlighted by the red box.
56
CHAPTER 2: MATERIALS AND METHODS
2.3.10: Core dropstone analysis
Two large granitic dropstones were recovered during logging, one from CS06 and one from
CS07. The hand specimens were described and 30 µm thick thin sections were prepared to
identify the minerals present, document their textural relationships and to assess the
petrography of the parent rocks.
57
CHAPTER 2: MATERIALS AND METHODS
58
Chapter 3: Background
3.1: Cold-water corals
‘Appearing as solitary forms in the fossil record more than 400 million years ago, corals are extremely
ancient animals’ Raaz and Kumar (2012)
The word ‘coral’ is an umbrella term which applies to seven discrete cnidarian taxa, defined
by Cairns (2007) as:
‘Animals in the cnidarian classes Anthozoa and Hydrozoa that produce either: calcium carbonate
(aragonitic or calcitic) secretions resulting in a continuous skeleton or as numerous microscopic,
individualised sclerites; or that have a black, horn-like proteinaceous axis.’ (p. 312)
Cold-water corals [CWC] are found in five of these taxa: stony corals (Scleractinia), soft
corals (Octocorallia), black corals (Antipatharia), zoanthids (Zoanthidea) and hydrocorals
(Stylasteridae) (Roberts et al., 2006). As of 2007 there were approximately 5160 species of
corals recognised, with 65% of these occurring in water deeper than 50 m (Roberts et al.,
2009). Thus, corals should be considered as primarily cold-water inhabitants. This work
utilises the term ‘cold-water corals’, as opposed to ‘deep corals’, because although CWCs
are typically found at water depths of ~500 m to 2500 m (Robinson et al., 2005), they are
also known to occur as shallow as 39 m (Freiwald et al., 2004).
CWCs can exist at greater depths and at colder temperatures (typically between 4-15°C;
Frank et al. (2011)) than their tropical counterparts; but like the tropical forms they are also
constrained by the calcite compensation depth. The CWCs differ from the majority of tropical
forms by the fact that they are asymbiotic (i.e. do not contain zooxanthellae) and therefore do
not require sunlight to live. Thus, unlike their shallow water relatives, they are not
constrained by latitude and are worldwide in their distribution (Fig. 3.1.1).
59
CHAPTER 3: BACKGROUND
Fig. 3.1. 1: Distribution of cold-water and tropical (warm) coral reefs. The global distribution of
cold-water corals shows that they occur at all latitudes. Note that this distribution is probably biased
due to concentration of research in the Atlantic. Source: UNEP/GRID-Arendal Maps and Graphics
Library, February 2008, http://maps.grida.no/go/graphic/distribution-of-coldwater-and-tropicalcoral-reefs.
CWCs are suspension feeders that use their tentacles to actively capture organic particles
from the water column (Freiwald et al., 2004). They feed on dissolved organic carbon
[DOC], phytodetritus and/or zooplankton, as suggested by stable carbon and nitrogen isotope
measurements (e.g. Duineveld et al., 2004). To facilitate discrimination between deep-water
and shallow-water corals, Wells (1933, p. 109) proposed the following bipartite division of
the group:

Hermatypic (reef-builders, possessing zooxanthellae) and

Ahermatypic (non-reef building, lacking zooxanthellae).
This simple distinction is hard to apply in practice because:
1. Shallow-water, reef-building asymbiotic (i.e. lacking zooxanthellae) tropical corals
are known, which co-exist with tropical symbiotic corals (e.g. Fabricius et al., 1995).
2. Zooxanthellate corals exist that do not contribute to reef structures and
3. Cold-water azooxanthellate framework-constructing corals exist.
The latter coral grouping are the focus of this thesis.
The presence of CWCs depends on a variety of external forcing conditions, including
temperature, salinity, dissolved oxygen concentration, aragonite saturation (and associated
carbonate compensation depth) and current speed of the ambient water mass (e.g. Freiwald et
60
CHAPTER 3: BACKGROUND
al., 2004; Roberts et al., 2009). However, the presence of a hard substrate and elevated
current velocities have been recognised as imperative for reef initiation and development
(Freiwald et al., 2004.). Flowing water provides food to the coral, but also prevents
blanketing or smothering of the corals through excessive sediment deposition (e.g. Dorschel
et al., 2005).
CWCs exist at the Porcupine Bank in well-ventilated, near-thermocline waters ranging in
temperature between 4 – 12 °C, close to, or within, the oxygen minimum zone (Frank et al.,
2009; Wheeler et al., 2007). High surface productivity has been observed over the bank and
is linked to water circulation around the bank causing organic matter accumulation (White et
al., 2005). It has been suggested that there is enhanced nutrient availability for coral growth,
due to this high surface productivity and downslope transport of the resultant organic matter
to the benthic flank regions (White et al., 2005). Intermediate nepheloid layers (Dickson &
McCave, 1986; Kenyon et al., 2003) have also been cited as a mechanism of providing a food
supply.
Lophelia pertusa is currently believed to be the dominant framework-building CWC (Pirlet,
2010) and the Arc Mound Province is characterised by their growth, along with (less
commonly) Madrepora oculata (Fig.3.1.2). The genus Lophelia is known from Late Miocene
times, and may possibly have first appeared as far back as the early Palaeocene (based on
poorly preserved fossil material; see Roberts et al., 2009). Lophelia records remain scarce
until the Plio-Pleistocene, when Lophelia pertusa becomes common in the Atlantic and the
Mediterranean (Pirlet, 2010). Growth rate estimates of Lophelia reefs have been reported as
~ 1 mm per year (Fossa, 2002).
Fig. 3.1. 2: Drawings of Lophelia
pertusa (left) and Madrepora oculata
fragments from photograph in
Zibrowius (1980). Scale bar = 8.3 mm
for L. pertusa and 4.3 mm. for M.
oculata.
61
CHAPTER 3: BACKGROUND
3.2: Cold-water coral mounds
3.2.1: Definition of a cold-water coral mound
CWC ecosystems have been known since the century before last (e.g. Duncan,
1870), but there is an unfortunate overlap in descriptive terminology regarding
growth and form of colonies. Ambiguity regarding the terms ‘reef’ and ‘mound’ has
resulted in these related, but quite different, CWC ecosystems being considered as
equivalent entities. CWCs have the potential to form a variety of colonial
aggregations, which may be termed mounds, banks, bioherms, knolls, patches,
massifs, thickets or groves. These aggregations, although different, are often
considered under the general term of ‘reefs’. This work distinguishes between CWC
reefs and CWC carbonate mounds [often referred to as coral banks (De Mol et al.,
2002; Stetson et al., 1962) and carbonate knolls (Hovland et al., 1994)].
Carbonate bioherms are documented from the fossil record since the Proterozoic over
3.4 Ga (Wood, 2001; Grotzinger et al., 2000) and both modern and ancient ‘reefs’
encompass a broad spectrum of forms. Corals, calcifying algae and microbes along
with a host of other sessile invertebrates have the biological potential to form reefs in
a range of environmental settings, and these roles have changed and evolved through
time (Fig. 3.2.1).
62
CHAPTER 3: BACKGROUND
Fig. 3.2. 1: The Main biological groups responsible for reef construction on the earth
through time. Horizontal grey hatched bars represent major extinction episodes. (Taken
from Benton & Harper 2009, their figure 11.32, and based on an original scheme by Wood,
2001).
Thus, there is no universally accepted or agreed definition of a reef, and any
definition, by nature, cannot be specific to coral. Nonetheless the following
definition is suitably encompassing and appropriate for the purposes of this work: A
63
CHAPTER 3: BACKGROUND
reef is a discrete carbonate structure formed by in-situ or bound organic
components, which develops topographic relief upon the sea floor (Wood, 1999).
Only a few species of CWC develop reef structures (Freiwald et al. 2004). CWC
reefs initially develop through colonisation by coral larvae on a hard substrate (De
Haas et al., 2009); for example rock, shell, allochthonous coral rubble, locally
lithified sediments or even man-made objects (van der Land, 2011). Reef formation
is dependent upon a combination of inorganic and organic in-situ factors (Wood,
1999), including:

Biomineralisation, producing calcareous skeletons;

Formation of sediment grains by skeletal disintegration and destruction;

Baffling, binding, or trapping of loose sediment by organisms; and

Precipitation of carbonate cement and mud (micrite).
Variety in the shape and form of reefs results from the fact that bioherm development
involves complex interaction of the above biological, physical and chemical
constructional concerns (skeletal growth, sediment build-up, early cementation) as
well as physical and bioerosional destructive processes.
Le Danois (1948) noted "massifs coralliens" occurring in deep water settings along
the European continental margins, and unequivocal mound-like structures (colonised
by corals) were discovered in the 1960s (Stetson et al., 1962). Cold-water coral
carbonate mounds are composed of loose frameworks of azooxanthellate
scleractinian corals and associated benthic organisms within enclosing sediments that
are mainly hemipelagic in origin (e.g. Titschack et al., 2009; De Mol et al., 2002;
Dorschel et al., 2005). According to Roberts et al. (Roberts et al., 2006; Roberts et
al., 2009) CWC carbonate mounds, in comparison to coral reefs, are defined as:
‘topographic seafloor structures that have accumulated through successive periods of coral
reef development, sedimentation and (bio)erosion. They may or may not support
contemporary reefs and typically contain stratified sequences of reef depositions separated
by non-reef (typical seafloor) sedimentary units and erosion surfaces’.
Thus CWC carbonate mounds are stipulated herein as: Topographic features which
form as a result of the build-up of coral, or which were directly influenced in
their development by the presence of coral.
64
CHAPTER 3: BACKGROUND
Topographic highs which have been colonised by coral reefs are not considered true
CWC carbonate mounds as they were not developed or influenced by coral growth
throughout their history (Roberts et al. 2006). Coral reefs positioned on these
topographic highs may, of course, subsequently develop into mounds. As noted
previously, carbonate mounds generally contain interstratified (often alternating)
sequences of sediments, which may be either reef or non-reef in origin; therefore, the
base of any given mound may be defined as the lowest recorded reef development
within the succession.
CWC mounds are biodiversity hotspots that provide a nursery and substrate for a
wide variety of cold-water species (Roberts et al., 2006). Although CWCs are
worldwide in their distribution (Fig. 3.1.1), they curiously only form mound
structures at specific locations (Fig. 3.2.2) including: the western European and
northwest African continental margins, the Chilean and Norwegian fjords, the Blake
Plateau, the Straits of Florida, the Little Bahama Bank, the Mississippi Delta, and a
sub-Antarctic seamount (Teichert, 1958; Moore & Bullis, 1960; Stetson et al., 1962;
Neumann et al., 1977; Cairns, 1979; Newton et al., 1987; Mullins et al., 1981;
Hovland, 1994; Popenoe, 1994; Freiwald et al., 1997; Treude et al., 2005 ).
Fig. 3.2. 2: Global distribution of cold-water coral mounds from Eisele (2010), showing that
they do not form at all locations where cold-water corals are known to exist (compare with
Fig. 3.1.1). The distribution may be underestimated due to uncertainty about whether a reef
constitutes as a mound (Roberts et al. 2009).
65
CHAPTER 3: BACKGROUND
As noted by Roberts et al. (2009), the distribution of CWC mounds is probably
underestimated due to the discussed uncertainty about whether a reef constitutes as a
mound (i.e. the successive nature of reef build-up has not yet been proven) as well as
the considerable areas of the seabed left to be documented. Notwithstanding, the
difference in geographic distribution is striking, suggesting that mound occurrence
may indeed be constrained by local conditions.
3.2.2: Cold-water coral mound development
The process whereby mounds grow and develop through ‘sediment trap and growth
by organisms was proposed by Mullins in 1981, but it was only when the IODP
Expedition 307 drilled the Challenger Mound in the Porcupine Seabight that it was
verified that CWC species existed throughout the structure. Thus mound build up
was proven to be intrinsically linked to coral growth.
210
Pb measurements also
showed higher sedimentation rates on the mounds themselves, due to the trapping of
sediment by the baffling coral framework (De Haas et al., 2009), thus explaining
their pronounced vertical accretion in comparison to the surrounding seafloor. More
than 50% of the sediment within Challenger Mound was found to consist of
terrigenous material (Pirlet et al., 2011). Thus, for the Challenger Mound at least, a
terrigenous sediment input appears to be as important for mound build-up as the
corals themselves. A hard substrate is also deemed necessary for CWC colonisation
(De Haas et al., 2009) and IODP 307 found the mound base to comprise of an
erosion boundary formed by a clay-rich firmground (Titschack et al., 2009).
Mound initiation and build-up theories
CWC mounds have shown age ranges up to 2.6 Ma (Kano et al., 2007) but their
origin has not yet been conclusively established.The main controls underlying mound
development are still under debate, but there are two main theories for mound
initiation and development: the Hydraulic Theory formulated by Hovland et al.
(1998) with initiation and growth generated by hydrocarbon seeps; versus the
Environmental
Control
Theory,
with
mechanisms
governed
by
optimal
environmental conditions and associated currents (e.g. Freiwald et al., 1997;
Freiwald et al., 2004; De Mol et al. 2002;) (Table 3.2.1).
66
Hydraulic Theory
Environmental Control Theory
Substrate
Gas seepage may lead to the development
of methane-derived authigenic carbonates
that act as a substratum for coral
settlement.
Erosion of the seabed by strong currents may
generate suitable coarse-grained substrata for
coral settlement.
Food
chain
Gas seepage at the seabed fuels a
microbial ecosystem based on the
microbial consumption and oxidation of
methane, increasing biomass along the
food chain.
Surface primary productivity underpins the food
chain. This may become concentrated at water
mass interfaces and transported to the coral
reef, possibly assisted by internal waves.
Infill
Along-slope and hemipelagic
sedimentation provides a lithic sediment
infill to the reef structure.
Along-slope and hemipelagic sedimentation
provides lithic sediment infill to the reef
structure.
Additional
CHAPTER 3: BACKGROUND
Seepage prevents the coral polyp
smothering by sediment
Benthic hydrodynamics help to enhance food
flux and prevent coral polyp smothering by
sediment
Table 3.2. 1: Proposed processes controlling mound initiation and development: The
Hovland et al. (1998) Hydraulic Theory, with initiation and growth generated by
hydrocarbon seeps; versus the Environmental Control theory, with mechanisms governed by
optimal environmental conditions. Modified from Roberts et al (2009).
Ultimately both theories (Table 3.2.1) specify along-slope and hemipelagic
sedimentation to provide lithic sediment source to form the mound structure; and
both involve coral larvae colonisation of a firm surface. Where the theories
essentially differ is at the base of the food chain –the Hydraulic Theory is not
controlled by surface productivity meaning that corals could/should grow anywhere
hydrocarbon seeps occur. The Environmental Control Theory requires surface
productivity and the means of transporting the labile carbon. Following these
theories, periods of reduced coral growth must be ascribed to either reduced seepage
activity or a change in environmental conditions.
It is probable that mound initiation may arrive by a variety of processes, but the now
widely accepted theory of mound development is that CWC carbonate mounds form
in the same way as any other CWC reef. Lines of evidence to support the
Environmental Control Theory include:
1. Mounds are not exclusively found above hydrocarbon reservoirs. Where this
appears to be the case it may be because the areas of continental margin
where mound formation is favoured by local hydrographic factors are similar
67
CHAPTER 3: BACKGROUND
in scale to the coinciding underlying hydrocarbon reservoirs (Roberts et al.,
2009).
2. The correlation between mounds and fault systems in the underlying bedrock
is weak. When the association is present it is possibly due to the faults
elevating the seabed topography, which favours reef formation (Roberts et
al., 2009).
3. The stable carbon isotopic signatures of coral skeletons and tissues suggest
that they metabolise contemporary carbon derived from surface ocean
productivity, and are not reliant upon ancient carbon from hydrocarbon seeps
(e.g. Mienis, 2008).
4. Wehrmann et al. (2008) found extremely low methane concentrations (<0.5
μM) at all depths and sites at the coral reefs along the Norwegian margin.
5. The mounds are generally found at a depth range along the continental
margin where there are strong benthic currents (e.g. Frederiksen et al., 1992;
White, 2007).
Mound developmental sequence
Mound development requires repeated occurrences of the processes previously
mentioned for reef formation. However, a description is necessary to describe the
developmental sequence. Squires (1964) described a sequence for CWC structures
whereby they develop from a colony to a bank (reef). Mullins et al. (1981) modified
the sequence to include the suggestion that CWC structure development begins with
the colonisation of a sea-floor 'perturbation' -e.g. hardground, or an allocththonous
block, to form a single ~ 1m in height colony (Fig 3.2.3). An aggregation of colonies
forms a thicket, increasing ecologic diversity and causing baffling of sediment from
the ambient water (Mullins et al., 1981). It is not established whether the final
dimension of the reef reflects the initial coverage of the closely associated coral
colonies or if the reef grows wider over the course of time (possibly both scenarios
occur). Nonetheless, the thicket develops into a coppice through a combination of
sediment trapping and in situ accumulation of skeletal debris to produce new
surfaces for coral colonisation (Mullins et al., 1981). Wilson (1979) suggests that
under uniform, relatively quiet conditions, the coppice form nearly circular structures
('Wilson rings'). However, development may alter to produce elongate structures
68
CHAPTER 3: BACKGROUND
which have been linked to the influence of currents (Lang & Neumann, 1980). The
final bank (reef) stage is essentially defined as when the structure that is capped by
living coral is 'large', where 'the total volume of dead coral in the entire structure
may far exceed that volume of living specimens' (Mullins et al., 1981).
Fig. 3.2. 3: Hypothetical development sequence for cold-water coral mounds from Mullins et
al. (1981) based on Squires (1964) and Neumann et al. (1977).
With changing external conditions, cold-water coral reef growth can decrease or
disappear. This can result in a coral graveyard/fossil reef or in the reef becoming
completely buried. Upon return of suitable conditions for coral growth, a
recolonsisation of the dormant structure by new corals restarts the process, building
up the coral mound architecture (Roberts et al., 2006, 2009). Whilst interglacial
conditions have been documented as important for the development of cold-water
coral mounds in the North Atlantic (e.g. Dorschel et al., 2005; Frank et al., 2009;
Mienis et al., 2009; Roberts et al., 2006; Rüggeberg et al., 2007), the mounds in
lower latitudes have shown the opposite pattern - with coral growth occurring during
the relatively cold periods of MIS 2, 3 and 4 (Wienberg et al., 2009).
69
CHAPTER 3: BACKGROUND
Mound build-up and water mass interfaces
As discussed above, mound build up appears to be controlled by the co-occurrence of
environmental conditions optimal for both coral growth and sediment
accumulation (Thierens et al., 2010). These parameters are mainly water
temperature, current strength, sediment and food availability (e.g. Dorschel et al.,
2007), and the presence of these conditions at coral mounds have been linked to
water-mass interfaces (Dullo et al., 2008). It is postulated that the water-mass
interface density gradient traps sinking organic material and benthic dynamics
(contour currents, internal tides and waves) transport and redistribute this material
laterally along this density-driven interface, feeding the corals in the process (Mienis
et. al., 2007). Evidence supporting this theory includes the correlation of the depth of
the permanent thermocline in the NE Atlantic (600 – 1000 m) to the mean water
depth where carbonate mounds occur (White & Dorschel, 2010), and bottom lander
observations by Mienis (2008), who concluded that the presence of living corals on
and around the summit of carbonate mounds appeared to be linked to the process of
internal waves and tidal currents.
Topography and Nepheloid layers
The water interface process discussed above does not take into account an important
related geologically driven factor – topography, which acts as an underlying control
on current flow, as well as influencing benthic hydrodynamics. Steep slopes enhance
the flux of organic material by increasing current velocities and/or breaking internal
waves (Frederiksen et al., 1992). Currents are locally enhanced in areas of critical
slope, i.e. where the slope of the ray path of tidal waves equals the slope of the
seafloor (Huthnance, 1981), due to intensification of internal tides. This is most
simply given by:
(Thorpe & White 1988), where
for the internal tide,
is the Coriolis parameter and N is buoyancy, given by
is the local bottom slope,
( )
, where
is the in situ density. The key variables in this model are the bottom slope and the
70
CHAPTER 3: BACKGROUND
potential density gradient.Thus density gradients in the water column may intersect
the slope at a point favouring internal wave amplification (McCave et al., 1995). As
the critical slope is inversely proportional to the density gradient it is more attainable
where the density gradient is largest (i.e. the permanent thermocline). Across-slope
directed tidal currents induce resuspension of slope sediments on the northwestern
Porcupine Bank to form nepheloid layers, as described by Dickson & McCave
(1986), Mienis et al. (2007), and White & Dorschel (2010). The local enhancement
of these nepheloid layers also correlates to the same depth of the permanent
thermocline and mounds (Mienis et al., 2007).Thus nepheloid layers may be the
sediment and nutrient-providing source for coral growth and subsequent mound
development on the Porcupine Bank (Kenyon et al., 2003; Mienis et al., 2007).
In conclusion, variation in CWC mound, distribution shape and form may be related
to a number of external elements with different factors influencing to varying
degrees, but current regime (with associated sedimentation/nutrients), water mass
interfaces (and associated thermocline), and underlying topography appear to have
particular significance.
71
CHAPTER 3: BACKGROUND
72
Chapter 4: Environmental Context and Previous Research
4.1: Geological & geomorphological setting
4.1.1: Regional development
The Irish seabed is a westward extension of the continental crust that underlies
Ireland, with a shelf break located at ~600 m water depth (Øvrebø et al., 2006). The
overall geomorphology of the area is structured into a series of northeast/southwest
(NE/SW) orientated banks and basins (Dorschel et al., 2010). The Caledonian
Orogenic Cycle (c.475– 400 Ma; Chew & Stillman, 2009) was principally
responsible for generating NE to SW trending structures. During the later Variscan
Orogeny (c. 300 Ma), folds and thrusts with an E-W to ENE – WSW trend were
produced (Coward, 1990; Graham, 2009). These events provided a structural
framework for future rifting phases, facilitating the development of the Porcupine
and Rockall Basins (Doré et al., 2007). Thick sedimentary successions were
deposited in these depocentres from the latest Palaeozoic to the present (Fig. 4.1.1)
and during this interval the area has undergone a complex geological history (Toms,
2010). Episodes of rifting, related to the splitting of the Pangaean supercontinent,
occurred during the Permo-Triassic and continued through to the Early Cretaceous
(Shannon, 1991).
The Permian stratigraphy of the region includes lacustrine claystones and siltstones
interbedded with aeolian sandstones (Tate & Dobson, 1989), and succeeding Triassic
deposits are composed of terrestrial sandstones overlain by mudstones (Naylor &
Shannon, 2009). A marine transgression followed an Early Jurassic hiatus,
represented by a succession of bioclastic-rich sandstone overlain by marine shales
interbedded with limestones (Sinclair et al., 1994). Following the Late Cretaceous
global transgression, the Porcupine/Rockall region then underwent three post-rifting
episodes during the Cenozoic:

Episode 1: Basinward progradation of sediment wedges, driven by
Palaeocene to early Eocene epeirogenic tilting, followed by

Episode 2: Epeirogenic sagging ~ 35–25Ma, driving the onset of deepwater
processes on the Rockall-Porcupine margin (Praeg et al., 2005).
73
CHAPTER 4: ENVIRONMENTAL CONTEXT AND PREVIOUS RESEARCH

Episode 3: Epeirogenic tilting then recommenced with basinward
progradation of shelf-slope wedges from ~4 Ma (Pliocene) to present (Praeg
et al., 2005).
The result of these phases was the development of three regional unconformitybound 2-4 km thick megasequences (Toms, 2010), made up of deep marine marls,
sandstones, mudstones and siltstones:

C30 (Palaeocene to -early Eocene megasequence)

C20 (mid-Miocene megasequence) and

C10 (early Pliocene megasequence).
The C20 unconformity is suggested to have formed through lithification and
diagenetic processes, whereas both C30 and C10 are associated with submarine
erosion (Dolan, 1986; Stoker et al., 2001).
Fig. 4.1. 1: East-west seismic section (from DCNER/PAD 2007) illustrating structure of the
Porcupine Basin and its margins (Porcupine Bank to the west).
4.1.2: The Porcupine Bank
The Porcupine Bank is a remnant of the palaeocontinent of Pangaea; that split-off as
a failed rift during the opening of the proto-North Atlantic in the Middle to Late
Jurassic (Naylor & Mounteney, 1975). The bank is still connected to the Irish
continental margin in the northeast by the underlying Slyne Ridge (Dorschel et al.,
2010).The bedrock geology of the Porcupine Bank is poorly constrained at present. It
consists of a metamorphic basement overlain by Upper Palaeozoic siltstones and
sandstones, with small remnants of Mesozoic (Lower Cretaceous) shallow-marine
deposits (Masson et al., 1989). The only cored section of the basement, recovered in
a MEBO (“Meeresboden-Bohrgerät”, German for “sea floor drill rig”) shallow
borehole in 2006, showed it to be a granitic orthogneiss. A crystallisation age of
74
CHAPTER 4: ENVIRONMENTAL CONTEXT AND PREVIOUS RESEARCH
~1.3 Ga was indicated by U-Pb Zircon geochronology, suggesting that it was
metamorphosed during the Grenvillian Orogeny (Daly et al., 2008).
The southern, western and northern margins of the Porcupine Bank are steep,
whereas the eastern margin is gentle, but the Porcupine Bank slope as a whole is
relatively stable (Øvrebø, 2005). Previous studies on the Quaternary stratigraphy of
the Western Porcupine Bank (Øvrebø, 2005; Toms, 2010), based on gravity cores,
revealed a sedimentary succession characterised by alternations between lighter and
darker sediments, corresponding to higher carbonate and siliciclastic inputs
respectively, and interpreted as representing glacial/interglacial deposits. Øvrebø
(2005) described contourite deposits related to bottom current variations during the
late Pleistocene and Holocene, and suggested the geometry to be similar to plastered
sheet drifts (low relief, near-constant thickness and occurring where there is smooth
topography). Toms (2010) described three categories of deposits from the Porcupine
Bank: pelagic, hemipelagic and contourite sediments; and suggested that a stronger
Antarctic Bottom Water [AABW] and a weaker North Atlantic Deep Water
[NADW] occurring during glacial periods, with the opposite scenario during
interglacial periods, possibly explains observed contourite alternations.
The surrounding deeper waters to the west, north and south, and the broad Irish Shelf
to the east (Fig 1.2.1) results in the Porcupine Bank being significantly removed from
the influence of terrigenous input, resulting in carbonate sediment accumulation and
concentration on the bank at present and mixing with relict quartz sands and other
glacial material (Scoffin & Bowes, 1988).
4.2: Oceanographic setting
4.2.1: Modern
Present North Atlantic oceanic circulation is made up principally of four components
(Fig. 4.2.1):
1. Elements of the North Atlantic Drift, which is a northeast continuation of the
Gulf Stream, bring warm water into the upper layers of the North East
Atlantic (Schmitz, 1996).
75
CHAPTER 4: ENVIRONMENTAL CONTEXT AND PREVIOUS RESEARCH
2. North Atlantic Central Water [NACW] underlies this and is the body of water
in the upper 500 - 1000 m of the ocean whose T/S characteristics have been
formed primarily by air-sea interaction across a wide band of latitudes (Iselin,
1939). The NACW is principally Eastern North Atlantic Water [ENAW] at
the Porcupine margin which is that part of NACW that is more saline than the
Western North Atlantic Water (Harvey, 1982; New et al., 2001). ENAW
forms during the winter months in the Bay of Biscay (Pollard et al., 1996)
and is driven northwards by the mid-slope Shelf Edge Current [SEC] (New &
Smythe-Wright, 2001). The SEC flows with average speeds of 15 – 30 cm/s-1
in depths of up to 1000 – 1200 m (New & Smythe-Wright, 2001). The mean
flow of this water mass of mixed origins is a quasi-geostrophic current,
driven by temperature and salinity differences (Huthnance, 1986).
3. Underlying the ENAW are intermediate water masses Mediterranean Outflow
Water [MOW] and Labrador Sea Water [LSW]. The MOW is sourced from
the Straits of Gibraltar. It flows northwards in an eastern boundary
undercurrent, typically at depths 1000–1200 m (Reid, 1979; New & SmytheWright, 2001) and is characterized by an increase of salinity and potential
temperature between 800 and 1000 m water depth (Dorschel et al., 2005).
LSW is sourced from the Labrador Sea and0 typically occurs between water
depths of 1600 - 1900 m and is a cold (2.8–3.0° C) dense body of water (New
& Smythe-Wright, 2001).
4. North Atlantic Deep Water [NADW] is found below and is composed of
several water masses (Frew et al., 2000).
Water depths on the Porcupine Bank range from less than 200 m to greater than
1000 m (Shannon et al., 2001). Water masses associated with the depth range of
carbonate mounds in the NE Atlantic are the Eastern North Atlantic Water [ENAW]
at 200 – 700 m depth and the underlying Mediterranean Outflow Water [MOW]
(White & Dorschel, 2010), both of which are advected northwards along the
Porcupine Bank into the southern Rockall Trough. Mixing between ENAW and
MOW occurs between 600 and 700 m (Rüggeberg et al., 2005). This upper boundary
of MOW is associated with enhanced current dynamics and the permanent
thermocline (Dullo et al., 2008). In the NE Atlantic, the permanent thermocline lies
between 600 and 1000 m and correlates to the depths of carbonate mounds (White &
76
CHAPTER 4: ENVIRONMENTAL CONTEXT AND PREVIOUS RESEARCH
Dorschel, 2010). North of 53° N the MOW re-circulates west of the Porcupine Bank
and its signal rapidly diminishes (e.g. New et al., 2001).
Fig. 4.2. 1: Schematic diagram, following the concepts of van Rooij et al. (2007a) and Toms
(2010), showing water masses in the Rockall Trough (RT) in the vicinity of the western
Porcupine Bank (PB). Dashed lines indicate water mass boundaries modified from Toms
(2010). Carbonate mounds occur between 500 – 1200 m (Klages et al., 2004). Enhanced
hydrodynamics and thermocline (600 – 1000m) are interpreted following White and
Dorschel (2010). NAD – North Atlantic Drift, NACW – North Atlantic Central Water, ENAW
– Eastern North Atlantic Water, MOW- Mediterranean Outflow Water, LSW- Labrador
Seawater, NADW – North Atlantic Deep water, and the SEC – Shelf Edge Current.
The magnitude of the slope along which the geostrophic currents flow exerts a strong
hydrodynamic influence (see Section 3.2.2). The presence of the Porcupine Bank
accelerates these contour-following currents as they flow along the margin (Wheeler
et al., 2005) and on scarps fast currents are inferred geostrophically (McCave et al.,
1995). The bottom currents resulting from contour current activity may be also be
locally superimposed by internal waves and trapped tides (e.g. Chatwin, 1996). The
Porcupine Bank is under the influence of topographically-steered south to north
bottom-current enhancement (Øvrebø, 2005), inferred from:
(i)
Bedform geometries - Kenyon et al. (2003) noted medium-sized sand
waves between mounds on the Porcupine Bank, indicating possible
current velocities of 40-50cm/s.
77
CHAPTER 4: ENVIRONMENTAL CONTEXT AND PREVIOUS RESEARCH
(ii)
Current meter data (e.g. White & Dorschel, 2010) and modelled bottom
currents (New & Smythe-Wright, 2001)
(iii)
Pattern of scour around the upper-slope carbonate mounds (Øvrebø,
2005).
4.2.2: Palaeo-oceanographic considerations
The Porcupine Basin and its surrounding margins have been below sea level since
the late Cretaceous, the exception being a period of exposure during the early Eocene
(Shannon et al. 2001). Thus, it was only in the Late Eocene that modern North
Atlantic Ocean current circulation initiated, causing major change in deep water
processes along the North East Atlantic Margin [NEAM] (Toms, 2010). Northern
Hemisphere glacial/interglacial fluctuations are believed to be intrinsically linked to
variations in NEAM water masses. During glacial periods the present ice sheets in
the Arctic Ocean and Greenland advanced much further south to cover large parts of
northern Europe and North America (e.g. Sejrup et al., 2005). Suppression of oceanic
circulation, by increased freshwater pulses into the North Atlantic, is believed to
have altered the heat distribution across the oceans and directly influenced climate
(e.g. Broecker, 1997; Ballini et al., 2006). Also, during the Last Glacial Maximum
the lower sea level reduced the Atlantic and Mediterranean exchange, resulting in the
MOW not passing further north than 40º N (Schönfeld & Zahn, 2000). The MOW
could therefore possibly be an important climatic factor influencing mound
development for the Arc Mound site, a concept first postulated by Freiwald (2002)
for mounds in the Porcupine Seabight.
4.3: North Atlantic Carbonate Mounds
"The discovery of giant deep-water carbonate mounds along the European continental margin is
among the most spectacular findings of the past decade" (Dorschel et al., 2005)
4.3.1: General introduction
Investigation of carbonate mounds progressed in earnest only recently in the last
three decades (Margreth et al., 2009) after Hovland et al. (1994) described seabed
78
CHAPTER 4: ENVIRONMENTAL CONTEXT AND PREVIOUS RESEARCH
mounds suspected to be modern bioherms in the Porcupine Seabight. Research
efforts accelerated through this period, through various European and International
projects
(ECOMOUND,
GEOMOUND,
ACES,
MOUNDFORCE,
MICROSYSTEM, HERMES, HERMIONE, etc.). CWC mounds were discovered
along the north-eastern Atlantic margin, from Norway (Freiwald et al., 2004) to
Mauritania (Colman et al., 2005) in water depths ranging from 40 m (Norway), to
more than 2000 m along the Irish margin (De Mol et al., 2002) (Fig. 4.3.1).
Fig. 4.3. 1: Carbonate mounds (red
ellipses) and CWC species Lophelia
pertusa (green dots) from Foubert (2007),
modified to show location of Arc Mounds
(J). (A) Belgica Mound Province at eastern
margin of Porcupine Seabight, (B)
Magellan and Hovland Mound Provinces
at northern margin of Porcupine Seabight,
(C) Porcupine Bank Canyon Mounds at
western margin of Porcupine Bank, (D)
Pelagia Mounds along north-western flank
of Porcupine Bank, (E) Logachev Mounds
on south-eastern slope of Rockall Bank, (F)
mounds on western Rockall Bank, (G)
Darwin Mounds, (H) Galicia Bank, (I)
Moroccan margin mounds
4.3.2: Examples of NE Atlantic carbonate mounds
On a regional scale, different mounds have different characteristics, with their size
and shape varying considerably - from simple cones hundreds of metres wide to
amalgamated ridge features, with some having a 5 km wide base and rising to 300 m
79
CHAPTER 4: ENVIRONMENTAL CONTEXT AND PREVIOUS RESEARCH
above the seafloor (e.g. De Mol et al., 2002; White & Dorschel, 2010). The
following short descriptions serve to illustrate mound distribution in the NE Atlantic
and the variation between several mound provinces.
Porcupine Bank
On the Porcupine Bank, the mound provinces have a somewhat irregular distribution
of buildups (White et al., 2007); with carbonate mounds developed between 500 –
1200 m water depths (Klages et al., 2004). The Pelagia Mounds in the north
Porcupine Bank are noted as existing in a strong current regime (potentially 40-50
cm/s) that rise 50-200 m above the seabed, with a maximum length of 1-2 km
(Kenyon et al., 2003; van Weering et al., 2003). Like the Arc Mounds (described
below), the Pelagia mounds are often found associated with scarps which also run
subparallel to the isobaths (Wheeler et al., 2007). Kenyon et al. (2003) found the
Pelagia mounds to be made up of interbedded layers of foraminiferal ooze and coral
debris, with thickets of living coral and associated benthic organisms covering the
upper parts of the mounds.
Arc Mounds
The Arc Mounds (the focus of the present study) were first discovered in association
with scarps on the southwest Porcupine Bank in 2005, during a survey undertaken by
the RV Celtic Explorer4. The EU-funded CoralFISH program conducted ROV video
surveys and long-term in-situ measurements of fish abundance and current meter
data using benthic landers in, and outside, the coral reef framework .As far as can be
ascertained, at the time of writing, the only work completed on the Arc Mounds is
1. Habitat suitability modelling (Rengstorf et al., 2012; Rengstorf et al., 2013),
which identified terrain attributes such as slope and bathymetric position
index as important predictive parameters, and
2. Hydrodynamic modelling, which showed intensified near-bottom currents in
areas where living corals were observed (Mohn et al., 2014).
Rengstorf et al. (2012) described the Arc mound province as being characterised by
isolated medium-sized mounds, approximately 50 - 100 m high with base lengths
seldom exceeding 500 m, qualifying them as ‘medium-sized mounds’.
4
HERMES Hotspot Ecosystem Research on the Margins of European Seas: Periodic Activity Report: Month 12 (2006)
80
CHAPTER 4: ENVIRONMENTAL CONTEXT AND PREVIOUS RESEARCH
Porcupine Seabight
Hovland et al. (1994) first published seismic profiles which resolved mound
structures up to 200 m high and more than 1,500 m in length from the Porcupine
Seabight. De Mol et al. (2002) provided the first description of three Porcupine
Seabight mound provinces, interpreting the mounds as being formed by a framework
of CWC and associated other macrobenthos, baffling sediments in an
oceanographically dynamic environment. Along the eastern slope of the Seabight are
the southernmost CWC mounds, making up the Belgica mound province. The
province is characterised by conical mounds which are asymmetrically buried by
contourites (Van Rooij et al., 2003) and some north-south trending elongated mound
clusters (De Mol et al., 2002). The mounds are up to 150 m high with lengths of up
to 2000 m (Wheeler et al., 2007) and occur between 500–1,000 m water depths, over
a distance of some 20 km (De Mol et al., 2002). Associated depressions, often on the
steep downslope side of the mounds, are thought to be as a result of MOW
generating local enhanced bottom currents (Van Rooij et al., 2003; Van Rooij et al.,
2010).
In the north of the Seabight is the Magellan mound province, in water depths of ~650
m (Foubert et al., 2007). This province features many medium-sized buried mounds,
averaging 72 m in height and 250 m in width, and varying geomorphologically from
single, conical mounds to (N-S directed) elongated shapes (Huvenne et al., 2003). To
the south of these are the Hovland mounds, featuring complex structures along
channels (Wheeler et al., 2007) which are up to 250 m high (De Mol et al., 2002) in
water depths of 400 to 1100 m (Foubert et al., 2007).
Rockall Bank
The carbonate mounds of the Rockall Bank in terms of elevation are the largest
known worldwide (Dorschel et al., 2010). The Logachev mound province, on the
southeastern margin of the Rockall Bank in water depths of 500 to 1200m, consists
of a belt of ‘giant’ carbonate mounds, predominantly arranged in down-slope
oriented clusters, with heights greater than 350 m and widths of 100s of meters to
several kilometres (Kenyon et al., 2003; Mienis et al., 2006; Wheeler et al., 2007).
81
CHAPTER 4: ENVIRONMENTAL CONTEXT AND PREVIOUS RESEARCH
Moroccan Margin
The Pen Duick Escarpment mounds, in water depths of 530-580 m in the Gulf of
Cadiz, are relatively small (~ 30 m high) compared to the mounds of the Porcupine
Seabight, Porcupine Bank and Rockall Bank (Van Rooij et al., 2011). There is
apparently no documented live coral growth currently on these mounds (Van Rooij et
al., 2011).
Mini-mounds in the NE Atlantic
Mini-mound provinces, consisting of hundreds of small CWC mound developments
~ 5m high, have been reported from water depths ~ 1000m in the Porcupine Seabight
(Moira Mounds) and the northeast corner of the Rockall Trough (Darwin Mounds)
(Foubert et al., 2007). The Darwin Mounds are each about 100 m in diameter,
whereas the Moira Mounds only reach between 20 to 50 m in aerial extent (Foubert
et al., 2007). Similar mini-mounds (up to 5 m high with bases 50-150 m diameter)
have also recently been discovered at three locations along the European continental
margin:

The Macnas mounds in the Porcupine Seabight (Wilson et al., 2007)

The Dangeard and Explorer canyon Mounds on the Celtic margin (Davies et
al., 2007), and

The Guilvinec Canyon mounds in the Bay of Biscay (De Mol et al., 2011),
All of these recently discovered mini-mound provinces are located at the shelf edge
in shallower water depths (<500 m) and are currently considered extinct.
4.3.3: Mound base of the NE Atlantic buildups
Seismic results showing a strong and regionally extensive reflector (Huvenne et al.,
2003; van Weering et al., 2003; Mienis et al., 2006; Van Rooij et al., 2007a) indicate
that the carbonate mounds of the Rockall Trough margin (Rockall Bank and
Porcupine Bank) and Porcupine Seabight are directly underlain by an unconformity.
Van Rooij et al. (2010) noted the onset of Belgica mound growth at 2.6 Ma and the
conspicuous feature which the Belgica Porcupine Seabight coral mounds are
established directly on has been termed the RD1 unconformity (Van Rooij et al.,
2003). According to Shannon et al. (2005) the RD1 discontinuity should be
82
CHAPTER 4: ENVIRONMENTAL CONTEXT AND PREVIOUS RESEARCH
correlated with the C10 unconformity of the Early Pliocene (4 Ma) (see Section 4.1).
However, Van Rooij et al. (2009) dated this regional structural feature to the Late
Miocene, and it is thus estimated to represent a hiatus of some 7 million years,
between Late Miocene and Early Pliocene times (Van Rooij et al., 2010), with the
onset of drift sedimentation estimated at 1.2 Ma (Van Rooij et al., 2003). The
associated phase of erosion and non-deposition is attributed to extensive
oceanographic changes at the onset of northern hemisphere glaciation, which created
a firmground (Thierens et al., 2010). Van Rooij et al. (2003) ascribe this event to
replacement of a slow water mass with the more vigorously flowing water of the
MOW.
4.3.4: Mound initiation and development in the NE Atlantic
Both the Hydraulic Theory and Environmental Control Theory (Section 3.2.2, Table
3.2.1) have been postulated to explain the origin of the mounds and their
development in the NE Atlantic. Due to lack of evidence for hydrocarbon seepage
(De Mol et al., 2002; Van Rooij et al., 2009) most studies have concluded that
mound distribution in the NE Atlantic is controlled by factors of the latter theory,
involving principally currents and food supply (e.g. Frederiksen et al., 1992;
Freiwald et al., 1997; Mortensen et al., 2001; De Mol et al., 2002; Kenyon et al.,
2003; Duineveld et al., 2004; White et al., 2005; Wheeler et al. 2005; William et al.,
2006; Wheeler et al., 2007).
De Mol et al. (2005) suggest that the development of the mounds at the Porcupine
Sea Bight margins may be linked to the aforementioned flow of MOW; with the
start-up phase following a period of erosion and non-deposition during the late
Pliocene. They suggest that the MOW may have been responsible for transporting
coral larvae, as well as a sediment load, which could then be baffled by corals and
lead to carbonate mound build-up. However, the Pen Duick Escarpment Mounds are
believed to be related to both environmental and seepage pathways, with seepagerelated carbonate crusts suggested to have offered a suitable substrate for coral
colonisation (Van Rooij et al., 2011). Additionally, Van Weering (2003) postulated
that pore-water fluid, perhaps related to cooling of volcanic sediments following
83
CHAPTER 4: ENVIRONMENTAL CONTEXT AND PREVIOUS RESEARCH
initial formation of the Rockall Trough, might explain a first phase of mound
formation in the southeast Rockall Bank.
It is not known whether the mini-mounds represent initial stages of mound growth
(i.e. they are precursors of the larger coral mounds) or if they are the product corals
surviving under stressed conditions (Foubert et al., 2011). Masson et al. (2003)
postulated that the Darwin mini-mounds formed by fluid escape transporting sand to
the seabed, arguing that this positive relief was enough for corals to colonise.
However Foubert et al. (2007) ascribed the formation of the Moira mini-mounds to
an interaction between CWC growth and sediment baffling.
Thus, there is yet to be a general consensus for the initiation and development of
CWC mounds in the NE Atlantic and it may, and probably will, be brought to light
that the precise development of each of the different provinces is associated with
different and varying factors.
84
Chapter 5: Geophysical Investigation
5.1: Multibeam
A range of maps were produced for the Arc Mound study area and these are
presented in Figs. 5.1.1-5.1.5. As discussed earlier in Section 2.2.1, the poor quality
of the Reson 7111 shipborne data (Fig. 5.1.2) entailed that the Reson 7150 shipborne
results (Fig. 5.1.1) were used for investigating the overall survey area, with detailed
results for a smaller area obtained from ROV Reson 7125 data (Fig. 5.1.3). The
scarps were found to be best highlighted by the slope raster (Fig. 5.1.4). The slope
was also used in conjunction with the multibeam hillshaded bathymetry in order to
discern geomorphology of the mound features (Fig. 5.1.5). Using the extraction
approach described in the methods (Chapter 2: Section 2.2.1) 42 mounds were
identified and their footprints defined (Fig. 5.1.6). The orientations measured from
the convex hull polygons of the mound footprints are shown in Fig. 5.1.7. Individual
mound bathymetric profiles along the orientation direction, starting and finishing at
the edge of the footprint polygons, were also created (Figs. 5.1.8 part 1-4). The
profile graphs show the same depth and horizontal scales (100 m and 1500 m
respectively) in order to compare their widths and heights.
85
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.1. 1: Hillshaded Reson 7150 (24 kHz) 50 m DEM. Multibeam data show many mounds
to be aligned along the scarp to the east.
86
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.1. 2: Hillshaded Reson 7111 (100 kHz) 75 m DEM.
87
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.1. 3: Hillshaded Reson 7125(400 kHz) 1 m DEM with mound example inset and
associated scarp highlighted.
88
89
steeper the terrain. Slope is average slope of seabed within a square of 25 by 25 m.
Fig. 5.1. 4: Slope of Reson 7150 (24 kHz) 50 m DEM. The lower the slope value, the flatter the terrain; the higher the slope value, the
CHAPTER 5: GEOPHYSICAL INVESTIGATION
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.1. 5: Reson 7125 (400 kHz) 1 m DEM with slope (left) and hillshaded 1 m DEM
(right) of mound example inset. The lower the slope value the flatter the terrain; the higher
the slope value the steeper the terrain.
90
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.1. 6: Reson 7150 (24 kHz) 50 m DEM with polygons of the identified mounds outlined
and numbered.
91
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.1. 7: Reson 7150 (24 kHz) 50 m DEM showing orientations of the convex hull
polygons.
92
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.1. 8 part 1: Bathymetric profiles of mounds 1-10, based on their orientations parallel
to the long axes and footprint edges. See Fig. 5.1.6 for mound numbering.
93
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig 5.1.8 part 2: Bathymetric profiles of mounds 11-20, based on their orientations parallel
to the long axes and footprint edges. See Fig. 5.1.6 for mound numbering.
94
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig 5.1.8 part 3: Bathymetric profiles of mounds 21-30, based on their orientations parallel
to the long axes and footprint edges. See Fig. 5.1.6 for mound numbering.
95
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.1.8 part 4: Bathymetric profiles of mounds 31-42, based on their orientations parallel
to the long axes and footprint edges. See Fig. 5.1.6 for mound numbering.
96
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Geomorphology of the study area
The water depths in the study area range from ~500 m in the east to ~1000 m in the
west. The geomorphology of this gently sloping area is characterised by a raised
terrace bounded by two along-slope scarps, and carbonate mounds. The western
scarp is punctuated by concave indentations while the eastern is punctuated by
convex projections, with mounds located at the apices of the projections (Fig. 5.1.9).
Fig. 5.1. 9: 3D bathymetric model of the Arc Mounds study area. There is a raised terrace
bounded by two along-slope crescent-shaped scarps to the east and the west, with mounds
aligning the eastern scarp (sc2) and occurring west of the western scarp (sc1). Note how the
scarps are joined in the south but the large isolated mound with associated moat appears to
cause reoccurrence of Sc2. Depth scale indicated by colour bar in metres. Data vertically
exaggerated by 4.
The western scarp (Sc1) is segmented into curvilinear segments up to 30 m high with
slopes up to 16° and extends from south to north for 12 km, in between 14° 46’W 14° 44’W at ~700 m water depth. The upslope scarp (Sc2) is ~9 km long located
between 14° 43’W- 14° 42’ W in water depths of 650 to 700 m, and is also up to 30
m high with slopes up to 16°. The scarps are connected by a 1.2 km long converging
branch at 51° 17’ 19”N, 14° 43’ 54”W. Approximately 2 km to the southeast there is
an isolated mound ridge with an associated moat that has a similar expression to Sc2.
The scarps clearly occur in association with the mounds. Mound complexes flank
scarp Sc2, which delimits their occurrence to the east. The multibeam maps show
97
CHAPTER 5: GEOPHYSICAL INVESTIGATION
sediment from the mound draping off the scarp (Fig. 5.1.10). Three mounds are
located on Sc1 whereas the rest of the mounds are concentrated in three downslope
E-W orientated clusters, west of scarp Sc1. These may be depth delimited or
delimited by distance from the scarp, as the most western mound in each cluster
occurs at a depth of ~820 m and ~ 5 km from the scarp. Aside from the scarps the
slope gradients across the intermound region range between 0.5 - 3°.
Fig. 5.1. 10: High resolution bathymetric 3D model shows sediment draping off the mounds
over the scarp. Data vertically exaggerated by 4.
Due to their spatial distribution, and the seabed dipping gently towards the west, the
mound summits are located at depths ranging from 590 to 780 m and bases at depths
from 630-850 m. Bathymetric profiles (Fig 5.1.8) and 3D bathymetric models of the
mounds (Fig. 5.1.11) show mound morphology varying from elongated ridges with
several peaks (e.g. Fig.5.1.8, mound 1) to conical forms (e.g. Fig.5.1.8, mound 6). A
topographic cross-section that follows the ROV route from NE to SW demonstrates
that the mounds are generally over 50 m in elevation, in comparison to the adjacent
seafloor, but elevations for the survey area range from 10 to 80 m (Fig 5.1.12). The
mounds have steepest slopes on the upper flanks becoming smoother towards the
base. The bases of the mounds are broader than the rest of their structure.
98
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.1. 1: 3D bathymetric map showing mound morphology varies from elongated ridge
mound types to conical forms. Colour scale showing depth. Data vertically exaggerated by
4.
Bathymetric profile
-600
-620
Elevation (m)
-640
-660
-680
-700
-720
-740
-760
-780
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
Distance (m)
Fig. 5.1. 2: Bathymetric profile of mounds following the route of the ROV (SE-NW). The
mounds are mostly over 50 m in elevation from the adjacent seafloor. Note mounds appear
spikey as they are highly vertically exaggerated.
99
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Morphometric analysis of mapped mounds
Tropical reef landscapes have been reported to adhere to scaling laws - whereby the
statistical pattern of size and shape is predictable by virtue of the existence of fractal
scaling (Purkis et al., 2007), that is, there is a degree of consistency across scale.
Remote sensing is an efficient way to quantify the spatial complexity of depositional
systems. The average mound density was found to be 0.35 km-2 for the surveyed area
of ~ 154 km2. The mounds were quantitatively analysed in terms of their height, sizefrequency distribution, footprint area, shape and orientation (Table 5.1.1).
Mound size and frequency distribution
Footprints of individual mounds varies between 22,700 – 630,000 m2 (with an
average of ~106, 300 m2) and mound heights range from 6 to 84 m (with an average
of 33 m). Lengths range from 220 m to 1,400 m (with an average of ~400 m) and
widths range from 166 m to 626 m (with an average of ~300 m).
There are no small mounds (<10, 000 m2), a large number of medium-sized mounds
(76% are less than 150,000 m2) and a small number of larger structures
(>150,000 m2). The mound footprint threshold was 5,625 m2 but the smallest mound
defined was ~17,700 m2. As discussed in the methods (Chapter 2, Section 2.2.1) it is
probable that there is under detection of small mounds. The results are subject to
noise and user specifications and interpretation. The results were groundtruthed by
comparison of the number of mounds observed in the ROV 1 m 3D bathymetry with
the number that were resolved by the mound extraction technique using the 25 m
bathymetry. The number of mounds counted in the ROV data matched the number
counted for the area covered by the ROV track in the 25 m data mound extraction
technique. This indicates that the overall survey results are credible.
The mound size frequency distribution was determined by plotting the exceedance
probability versus area (Table 5.1.1). This parameter represents the probability that a
given mound will be equal to or greater than a given area. For example, all mounds
were greater than 17,700 m2 in area and these mounds therefore have a probability of
1.0; whereas all mounds greater than 600,000 m2 have a probability of 0 because the
largest mound was 588,000 m2. The frequency of the mounds is found to decrease
with increasing mound area (Fig. 5.1.13) and there is a positive correlation between
mound height and mound area with the highest mound having the largest footprint
100
CHAPTER 5: GEOPHYSICAL INVESTIGATION
(Fig. 5.1.14). A very slight trend for larger mounds to occur in deeper water was
found (Fig. 5.1.15).
Mound shape and orientation
By quantifying the shape of mound footprint we can determine whether trends in
geometric form are present. The descriptor used to characterize the shape of the
mound footprint polygons was the principle axes ratio (PAR) (e.g. Correa et al.,
2012; Purkis et al., 2007) between the width of the minimum bounding rectangles
and the length of the convex hull polygons, both created from the mound footprints,
where dmin is the diameter of the minimum bounding rectangle and dmax is the
length of the convex hull polygon. The PAR metric scales between 0 and 1. High
values indicate circular shapes, elongated ellipsoidal mounds are represented by a
ratio approaching 0.5 (Correa et al., 2012) and low values indicate elongate forms.
The Principle Axes Ratio ranges from 0.4 – 0.9 indicating that the mounds range
from elongated ellipsoidal forms to circular forms. For the smaller bioherms there is
near-equal probability for it to be circular or elongated ellipsoidal, but there is a
slight trend for mounds to become more elongated as their size increases, with the
largest mound having the most elongate form (Fig. 5.1.16). It was observed that the
mounds with the top five PAR ratios (i.e. more circular in form) were generally small
and isolated mounds (Fig. 5.1.17).
Mound orientation varies from 0 - 174°. Plotting a rose diagram of the percent
occurrences of the orientations shows a northwest-southeast trend (Fig. 5.1.18) and a
slight northeast-southwest trend, with a large standard deviation (±50.95°) around the
mean (105°). The orientations of the mounds in the east reflect the shape of the scarp
whereas mainly forms approaching 180° occur away from the scarp in the west (Fig.
5.1.7).
101
102
orientation
Degree (0 - 180°)
Dimensionless
m
m
m
Dimensionless
m2
Unit
Table 5.1. 1: Summary of the morphometric parameters used to quantify mound spatial patterns (adapted from Correa et al., 2012).
hull polygon
Azimuth of the longest axis of the convex
northwest-southeast trend
Axis along convex hull length
to become more elongated
Mound footprint
to length
Slight trend for mounds
220 – 1,400
as their size increases
principle axis ratio (PAR)
(circular versus
Ratio of the convex hull polygon width
vertices of the convex hull
The longest distance between any two
vertices of the convex hull
166 - 626
6 – 84
mounds
Mainly medium-sized
22,700 – 630,000
Results
elongated)
Convex hull
Convex hull length
Mound length
Mound footprint shape
Convex hull width
Mound width
The shortest distance between any two
maximum Z value within mound footprint
Height
Probability that a given mound will be of area ≥ of
Mound height
Exceedance probability
Size-frequency
off
Area of mound perimeter based on 6 degree cut-
a given area (Purkis et al. 2007)
Area
Mound footprint size
Definition
distribution
Parameter
Objective
CHAPTER 5: GEOPHYSICAL INVESTIGATION
CHAPTER 5: GEOPHYSICAL INVESTIGATION
1.0
Exceedance Probability
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0
100000
200000
300000
Area
400000
500000
600000
(m2)
Height (m)
Fig. 5.1. 13: Plot of mound footprint area versus exceedance probability.
90
80
70
60
50
40
30
20
10
0
0
100000
200000
300000
Area
400000
500000
600000
(m2)
Fig. 5.1. 14: Plot of mound footprint area versus mound height. There is a positive
correlation between mound area and mound height.
103
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Height (m)
80
60
40
20
0
-800
-750
-700
-650
-600
-550
Depth to peak (m)
PAR (-)
Fig. 5.1. 15: Plot of mound height versus depth shows a very slight trend for larger mounds
occurring at greater depths
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
10000
110000
210000
310000
Area
410000
510000
(m2)
Fig. 5.1. 16: Mound footprint area versus mound footprint shape. Mound shape is based on
the Principal Axes Ratio (PAR): values approaching one indicate circular shapes, values
approaching zero indicate elongated forms. The data indicate that there is a trend of mounds
becoming more elongated as their size increases.
104
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.1. 17: Mound polygons of the top five PAR results are highlighted in blue, indicating
that smaller more isolated mounds are also more circular in form.
105
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.1. 18: Unidirectional rose diagram showing mound footprint orientation in percent of
occurrences. Mean (105°) shown by black line. Standard deviation (±50.95°) shown by red
line. There is a northwest-southeast trend and a slight northeast-southwest trend.
106
CHAPTER 5: GEOPHYSICAL INVESTIGATION
5.2: Sidescan backscatter
The multibeam backscatter data augmented the multibeam data by showing that the
mound complexes and the scarp are characterised by higher backscattering
amplitudes (Fig 5.2.1) and (Fig. 5.2.2) than the surrounding seafloor. The higher
backscattering (Fig 5.2.1) of the mounds is associated with a combination of specular
returns from steep slopes and a rough texture. The higher backscattering linear
feature of the scarp indicates that it is composed of a harder substrate than the
surrounding seafloor, suggesting slight lithification. The lower backscatter that
characterises the seafloor is attributed to a homogenous soft silty seafloor.
Fig. 5.2. 1: Map of the sidescan data acquired along the ROV transect with mound example
highlighted. The mound complexes have a higher backscattering signal.
107
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.2. 2: Map of part of the ROV sidescan data with the scarp highlighted.
5.3: Chirp
The deepest penetration of the Chirp signal was seen on the W-E Chirp line (Fig.
5.3.1) and is ~280 ms or ~21 m into the subsurface. The single-channel sub-bottom
profiles are unmigrated; therefore mounds appear as hyperbolic reflectors with the
mound peaks corresponding to the apices of the hyperbolae. The west to east Chirp
data from line BOBECO122A (Figs. 5.3.1-2), and the intersecting south to north line
BOBECO119A (Figs. 5.3.3-4), show eight and five mound-like features respectively,
with one mound directly detected in both lines. Therefore the mounds are named CH
108
CHAPTER 5: GEOPHYSICAL INVESTIGATION
1 – CH 12 (CH to indicate detection by Chirp) and include a buried mound-like
feature (CH6) in line BOBECO122A.
Since the acoustic energy travels from the transducer in a broad cone, with a
footprint in the order of 50 to 100 m, steep topographic features that are not directly
below the device can be detected by sideswipe. This occurs in both transects and is
clearly observed in Fig. 5.3.3 of the S-N Chirp line where the bathymetric data shows
the line to be west of the mounds. The S-N line passes 90 m west of core CS07, i.e.
still within the Chirp acoustic footprint, and the W-E line passes ~18 m north of the
core CS06.
In addition to the seismic data, bathymetry information was used to aid interpretation
of the sub-surface. Plotting the coordinates of the apices of the mound-like features
in the Chirp data (Tables 5.3.1 and 5.3.2) showed that they correlated to the mounds
resolved by the multibeam data (Fig. 5.3.5). Thus the bathymetric data showed that
the acoustically homogenous hyperbolic features, illustrated in Figs. 5.3.1-4, indicate
the presence of mounds.
Depositional processes determine sediment facies which exhibit different acoustic
responses arising from their contrasting physical properties, allowing identification
of seismic units. Correlation and interpretation of seismic units was made difficult by
the small amount of data, the occurrence of sideswipe, and the non-distinct nature of
some of the reflections. The resulting interpretation is necessarily tentative, but three
seismostratigraphic units (U1, U2, and U3) were identified as well erosion surfaces
(R1 and r2) and the mounds (Fig. 5.3.6).
109
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.3. 1: Chirp profile A-A’ (line BOBECO122A) orientated west-east with bathymetric map and bathymetric profile (inset). The acoustically homogenous
hyperbolic features indicate the presence of mounds. The dashed red line indicates core CS06 location; the red line indicates intersection of Chirp line with
profile B-B’. Sc- Scarp. TWT - two way travel time. SP – shotpoint. Depth at 1 sec (TWT) is 750 m.
110
Fig. 5.3. 2: Identifying the shotpoints of the apices of the mound features on Chirp profile A-A’ (line BOBECO122A) with bathymetric profile (inset). The
shotpoints of the apices of the mound features were mapped to their corresponding coordinates. The dashed red line indicates core CS06 location; the red line
indicates intersection of Chirp line with profile B-B’. TWT - two way travel time. SP – shotpoint. Depth at 1 sec (TWT) is 750 m.
CHAPTER 5: GEOPHYSICAL INVESTIGATION
111
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.3. 3: North-south Chirp profile B-B’ (line BOBECO119A) with bathymetric map and bathymetric profile (inset). Scales: horizontal 1/25000, vertical 50
cm/sec. TWT – Two way travel time. Depth at 0.90 sec (TWT) is 675 m.
112
Fig. 5.3. 4: Identifying the shotpoints of the apices of the mound features on Chirp profile B-B’ (line BOBECO119A) with bathymetric profile inset. The
shotpoints of the apices of the mound features were mapped to their corresponding coordinates. The dashed red line indicates core CS07 location; the red line
indicates intersection of Chirp line with profile A-A’. TWT - two way travel time. SP – shotpoint. Depth at 0.80 sec (TWT) is 600 m.
CHAPTER 5: GEOPHYSICAL INVESTIGATION
113
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Mound-like feature
Chirp Line
SP
LAT
LONG
CH1
BOBEC0122A
550
51.34843890
-14.81562780
CH2
BOBEC0122A
625
51.34843890
-14.81038890
CH3
BOBEC0122A
820
51.34852220
-14.79832220
CH4
BOBEC0122A
1050
51.34903610
-14.78473330
CH5
BOBEC0122A
1310
51.34985000
-14.77026940
CH6
BOBEC0122A
1710
51.34986390
-14.74983610
CH7
BOBEC0122A
2275
51.35039720
-14.72048330
CH8
BOBEC0122A
2330
51.35036940
-14.71754170
Table 5.3. 1: Coordinates of mound-like features identified in the Chirp data for line
BOBECO122A. SP-shotpoint
Mound-like feature
Chirp Line
SP
LAT
LONG
CH8
BOBEC0119A
240
51.35116940
-14.71754440
CH9
BOBEC0119A
620
51.36309720
-14.72141390
CH10
BOBEC0119A
715
51.36611390
-14.72249720
CH11
BOBEC0119A
820
51.36938330
-14.72385000
CH12
BOBEC0119A
880
51.37111110
-14.72466940
Table 5.3. 2: Coordinates of mound-like features identified in the Chirp data for line
BOBECO119A. SP-shotpoint
114
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.3. 5: Composite plot of the coordinates of the mound-like features in the Chirp data with the mound structures imaged by the multibeam data.
The Chirp mound coordinates are offset from the actual mounds imaged in the multibeam data due to sideswipe. CH6 appeared to show a buried
mound in the Chirp data and it does not show any surface representation. 7150 hillshaded shipborne multibeam data (24 kHz). Purple dots
represent CH1-CH8 coordinates from line BOBECO122A (A-A’; West-East); black triangles represent CH8-CH12 coordinates from line
BOBECO119A (B-B’; South-North)
115
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.3. 6: Identification of seismic units. The directly detected mound appears as an
acoustically homogenous dome-shaped structure rooting on the strong reflector R1. R1 has
been interpreted as the mound base. Three seismostratigraphic units U1, U2, and U3 were
identified U1 is acoustically transparent but there is a detection of an irregular reflector
within it (r2). U1 underlies an irregular, erosional surface: the R1 reflector that underlies
the mounds. U2 is acoustically transparent and conformably underlies U3. U3 shows
draping, conformable, stratified, sub-bottom reflectors which onlap the flanks of the mounds.
This unit usually directly overlies U1 on the R1 reflector but also occurs overlying U2.
Mounds and the R1 reflector
The CHIRP data show the mounds to have a different geophysical structure to the
surrounding sediment of unit 3 [U3], indicating a different composition to the offmound areas (Fig. 5.3.6). The directly detected mound appears as an acoustically
homogenous dome-shaped structure overlying a strong reflector (R1). The lack of
internal reflectors indicates a uniform geophysical facies with little or no internal
acoustic impedance differences. The detection of the very strong sub-bottom
reflector, R1, observed beneath the directly detected coral mound demonstrates that
not all the seismic energy is absorbed or dispersed inside the mound structure. R1 has
been interpreted as the mound base and can be seen in both Chirp lines (Fig. 5.3.1-4).
116
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Unit 1
The nature of unit 1 [U1] cannot be distinguished due to the limit of penetration of
the Chirp signal. Its acoustic transparency may possibly be due to signal attenuation
at R1 however, there is a faint detection of a reflector (r2) within this unit (Fig.
5.3.6). The upper boundary of U1 is an erosional surface that is interpreted as
correlating to the R1 mound base reflector.
Unit 2
Unit 2 [U2] ranges from less than a metre in thickness to 16 m thick. It is
acoustically transparent and conformably underlies U3 (Fig. 5.3.6). It is assumed to
be a massive facies of pure muds or sands due to its acoustic transparent nature.
There is widespread removal of this unit and a higher energy environment is assumed
to have caused major erosion of this facies so removing it in elevated areas. The
relationship of the mounds with U2 is not clear as it has been removed in the area
where the mound base is clearly identified (Fig. 5.3.6). It remains to be identified
whether mound development occurred pre-deposition or post-erosion of U2, because
the mound bases and most of U3 develop on the hard reflector R1of U1. Mound CH5
and its adjacent sediments were investigated in an attempt to describe the chronology
of deposition. Mound CH5 shows sediment build up to the east of the mound with
the western side characterized by an exposed flank (Fig. 5.3.7). This is corroborated
by multibeam data (Fig. 5.3.8) indicating that the steep western flank is associated
with the mound build up and is not a seafloor characteristic (i.e. a scarp).
117
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.3. 7: Mound CH5 with drift sediment build-up to the east of the mound and an
exposed flank on the western side.
Fig. 5.3. 8: Multibeam data indicates that the steep western flank of mound CH5 (indicated
by the red dot) is associated with the mound build up and is not a seafloor characteristic
such as a scarp.
The sediment (U3) ponding to the east of mound CH5 overlies two acoustically
transparent mound-like features (Fig. 5.3.7). The mound-like feature to the east has a
well delineated topography and has been interpreted as a buried mound, but its base
is not detected. It is not clear whether the buried feature to the west is mounded U2
sediment formed from moats, or another buried mound. There are three possible
scenarios:
1. Scenario 1 is illustrated in Fig. 5.3.9. The mounds were established on an
erosion surface (R1) and deposition of U2 and U3 followed. The mounded
nature of U2 could be explained by moating associated with the two adjacent
118
CHAPTER 5: GEOPHYSICAL INVESTIGATION
mounds (CH5 and the now buried mound). If this were to be the case then the
mound CH5 would appear to be onlapped by U2, indicating it was established
pre- or during deposition of that unit
Fig. 5.3. 9. Possible interpretation (scenario 1) to describe deposition chronology of units.
The mounds are established on an erosion surface (R1), followed by deposition of unit 2 and
3. The mounded nature of unit 2 could be explained by moating associated with the two
adjacent mounds (CH5 and the now buried mound). BM – buried mound.
2. Scenario 2 is illustrated in Fig. 5.3.10. It is suggested that U2 was eroded and
there was subsequent development of the mounds on the R1 reflector of U1.
The middle section represents 2 adjacent buried mound structures with the
more western buried mound possibly related to CH5.
119
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.3. 10: Possible interpretation (scenario 2) to describe deposition chronology of units.
U2 was eroded and there was subsequent development of the mounds on the R1 reflector of
U1. The middle section represents 2 adjacent buried mound structures with the middle
buried mound possibly related to Ch5.
3. Scenario 3 is illustrated in Fig. 5.3.11. It is suggested that mounds both
developed on R1 and on U2 where it remained after being mainly eroded
(following interpretations by Van Rooij et al., 2003). This would mean the
mound base developed with upslope and downslope sides initiating at
different depths.
120
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.3. 11: Possible interpretation (scenario 3) to describe deposition chronology of units.
Mounds both developed on R1 and on U2 where it remained after being mainly eroded
(following interpretations by Van Rooij et al., 2003). This would mean the mound base
developed with upslope and downslope sides initiating at different depths.
It doesn’t seem likely that the feature represents a buried mound (Scenario 2) simply
due to the indistinct nature of the topmost reflector in comparison to the adjacent
interpreted buried mound. However the lack of appreciable slope may have resulted
in softer reflectors. The fact that the reflector associated with the top of U2 is weak
also indicates that scenario 3 is unlikely because it doesn’t seem to suggest a suitable
substrate for coral colonisation. Thus, this work interprets the mounded transparent
unit as representing mounding of U2 sediments (Scenario 1). Scenario 1 indicates
that the mounds were present for an extended period of time where the U2 that
developed around the mound was subsequently mostly eroded, and deposition of U3
around the mounds then followed. Van Rooij et al. (2003) found mounds both
developed on the RD1 unconformity and on a unit P2, implying that the P2
sediments were already deposited before mound growth, but these results suggest
that mounds formed before or during deposition of U2. This indicates that the
environmental conditions controlling mound initiation and U3 deposition may not be
that related. The above discussion presents theories for consideration, but in essence
121
CHAPTER 5: GEOPHYSICAL INVESTIGATION
this work finds that a clear seismic stratigraphic interpretation was precluded by the
fact that there were only profiles.
Unit 3
The most recent unit, U3, ranges from less than a metre in thickness to 21 m thick.
U3 covers the survey area and represents the off-mound sediments that were sampled
by CS06 (Chapter 6). U3 shows draping, conformable, stratified, sub-bottom
reflectors which onlap the flanks of the mounds. This unit usually directly overlies
U1on the R1 reflector but also occur overlying U2 (Fig. 5.3.6). From the Chirp data
and observations of the core material these draping deposits are interpreted as
contourites, with low-amplitude reflectors interpreted as representative of finegrained sediment and high-amplitude reflectors interpreted as being coarser in nature.
Contourite drifts are manifested by the action of bottom currents in deep water and
are especially common along continental margins (Stow et. al., 2002). They are
characterised internally by sub-parallel moderate to low amplitude reflectors which
change gradationally between seismic facies (Koenitz et. al., 2008).
The retention of U3’s thickness at Sc2 (Fig. 5.3.1) indicates that scarp formation
predates the deposition of U3’s sediments, because the deposits don’t appear
deformed by faulting or erosion. The definite onlapping nature of lowermost
reflectors of U3 on the directly detected mound (Fig. 5.3.6) yields information about
the start-up of mound growth as by inference the mound was already present at the
moment of deposition and therefore must be older than the sediment of U3. The
reflectors pinch out as they approach the mound, indicating higher energy conditions
occurred around the structure and/or the slopes of the mound prevented deposition.
It is recognised that the small amount of data (two lines) means that any
interpretation has a lot of associated uncertainty, but the overall interpretation and
correlation of seismic units is shown in Fig. 5.3.12.
122
Fig. 5.3. 12: Interpreted Chirp profile A-A’ (line BOBECO122A) with original Chirp results and bathymetric profile inset. The hyperbolic features indicate
the presence of mounds (coloured grey). The strong reflector underlying the mound structure R1 is interpreted as the hard mound base and has been
correlated to the topmost reflector of U1. r2 is an irregular reflector within U1. The U3 interpreted drift deposits cover the area. U2 is interpreted to occur
mid-section and to the east. The dashed red line indicates core CS06 location; the red line indicates intersection with the north-south Chirp line profile B-B’.
Blue lines indicate possible faults (F). BM – buried mound. Sc – Scarp. TWT - two way travel time. Depth at 1 sec (TWT) is 750 m.
CHAPTER 5: GEOPHYSICAL INVESTIGATION
123
CHAPTER 5: GEOPHYSICAL INVESTIGATION
5.4: Discussion and conclusions
Origin of the scarps
Scarps may occur on fractures of mid-ocean ridges; on margins undergoing tectonic
faulting; through erosion on carbonate-depositional continental margins; or by
gravitational slump and slide mass-wasting processes (Harris & Baker, 2011).
Øvrebø et al. (2005) suggest that the occurrences of scarps in slope and basin-floor
settings indicate that slope failures can be largely unrelated to slope stability; instead,
bottom currents may have an important role in controlling the location of slope
failures. Thus, the general gentle slope in the area indicates that the scarps could be
either the surface expression of vertical displacement as a result of faulting or the
production of erosive-based scours. De Haas et al. (2002) suggest that the sinuous
ridges of the north -eastern Porcupine Bank represent the edges of erosional scours
where shallow bedding planes are exposed. This is supported by Klages et al. (2004)
who suggests that the scarps may have formed by preferential seabed erosion due to
a lack of evidence for neo-tectonic activity in the area. Van Rooij et al. (2003) have a
similar argument because the scarps in the Porcupine Seabight are contour-parallel
and depth -restricted it means that they are likely current generated. However,
commercial seismic data suggests a correlation between Rockall Trough boundary
faults and scarps on the Porcupine Bank (Croker et al., 2002). Indeed, the subparallel topographic ridges on which the north -eastern Porcupine Bank mounds
(52°N - 54°N) predominantly occur are described by Wheeler et al. (2005) as fault
scarps that overlie Rockall Trough boundary faults. In addition, Mazzini et al. (2012)
suggest that the crescent-shaped geometry of the scarps at the Porcupine Bank
Canyon coral mounds indicates that they are the seabed expression of a regional
listric fault system. The interpreted seismic units and reflectors of the Chirp data
indicate vertical displacement occurred at the scarps but the raised morphology
suggests a horst and graben explanation for the scarps related to the surrounding
extensional Rockall Trough and Porcupine basins (see Chapter 4, Section 4.1). The
Caledonian Orogeny, which resulted in SW-NE trending structures, may explain the
Arc Mounds scarps’ orientation.
124
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Characteristics of the Arc Mound Province
The Arc Mounds are 6-84 m in height with footprints (areal extent) between 29,000 –
630,000 m2 (Table 5.1.1). The mounds vary from conical to elongate in morphology
and are onlapped by apparent drift deposition. The site is characterised by N-S
elongate mounds flanking a scarp and E-W trending mound clusters. The lack of
small mounds (Table 5.1.1), and the elongate form of many of the structures,
indicates that the site represents a mature stage of coral carbonate mound
development. The water depths of mound occurrence shows that the Arc Mounds are
occurring in a depth range close to the permanent thermocline (White & Dorschel,
2010) and are similar to the depth range of other carbonate mounds in the NE
Atlantic (Table 5.4.1). It is difficult to compare the mound provinces as their
characteristics are often reported in a broad, general sense. In most cases the base
diameter is reported but the mounds will often be described as elongate (e.g. Mazzini
et al. 2012; Foubert et al., 2007). Nonetheless, comparing the base diameters
(bearing in mind that they may represent width or length) and descriptions of the
other mounds and mound provinces indicates that the Arc Mounds are medium-sized
in both their elevation and areal extent (Table 5.4.1).
125
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Pen Duick Escarpment
Mounds/Province
Gulf of Cadiz
Location
15-20
Height (m)
diameter
given
350-450
Length (m)
50-150 [d]
200-300
Width/diameter [d] (m)
<500
530-580
Base depth
(m)
Characteristics
Conical to elongated features
Source
Van Rooij et al.,
2011
De Mol et al., 2011
<5
Mini-mounds
Bay of Biscay
Van Rooij et al.,
2003;
Wheeler et al., 2007
Guilvinec Canyon
Conical mounds asymmetrically buried by contourites;
some NNE–SSW trending elongated mound clusters
Foubert et al., 2011
500-1000
Mini-mounds
Wilson et al., 2007
300-1800
900-1080
Mini-mounds
up to 2000
20 to 50 [d]
300-500
Conical individual mounds and elongate ridges;
associated with elongated depressions or blind channels
De Mol et al., 2002;
Foubert et al. 2007
Foubert et al., 2007;
Huvenne et al., 2003
100 -150
50-150 [d]
400 - 1100
Mostly buried mounds; N-S elongated and conical mounds
This work.
Eastern Porcupine
Seabight
up to 1500 [d]
500-700
Conical and elongated mounds onlapped by contourites,
N-S elongate mounds flanking scarp
and E-W trending mound clusters
Mazzini et al., 2012
Belgica
300-800
630-850
Most are elongated in shape but conical mounds also
identified; along canyon edge or escarpments. Some E-Welongated ridge-like features.
Kenyon et al., 2003;
van Weering et al.,
2003
~5
up to 1450
166-626
600–1,150
Often found on top of scarps; distinctly linear and oriented
SSW/NNE; Variety of sizes
Davies et al., 2007
Porcupine Seabight
(average) 72
220-1400
300-2000 [d]
650-900
Mini-mounds
Kenyon et al., 2003;
Mienis et al., 2006;
Wheeler et al., 2007
Moira
6-84
diameter
given
up to 700
<500
Giant mounds in coalescing mound clusters aligned
north-south (downslope) and east-west (along-slope)
Masson et al., 2003
<5
Southwest Porcupine
Bank
50-300
up to 2000
50-150 [d]
500-1200
Predominantly quartz sand, indicating that coral debris is
not a major
contributor to mounds
Porcupine Seabight
Arc Mounds
West margin of
Porcupine Bank
50 - 200
diameter
given
100s-km [d]
900-1060
Macnas
Porcupine Bank
Canyon
Northwest Porcupine
Bank
<5
diameter
given
(average) 75 [d]
Magellan
diameter
given
diameter
given
diameter
given
Pelagia
Celtic Margin
>350
diameter
given
150-250
Dangeard & Explorer
Canyon
Southern Rockall Bank
~5
Hovland
Logachev
Northern Rockall Trough
Northern Porcupine
Seabight
Northern Porcupine
Seabight
Darwin
Table 5.4. 1: Characteristics of mounds of examples in the North East Atlantic, listed by latitude from south to north.
126
CHAPTER 5: GEOPHYSICAL INVESTIGATION
The mound density at the Arc Mounds of 0.35 km-2 is lower than that found for the
Great Bahama Bank (14 km-2; Correa et al., 2012) and the Magellan mounds (1 km-2;
Huvenne et al., 2003) but similar to estimates for the Florida-Hatteras slope (0.3 km2
; Paull et al., 2000). The high densities found by Correa et al. (2012) and Huvenne
et al. (2003) are probably due to the increased resolution of those datasets that
allowed mounds as small as 81 m2 and 600 m2, respectively, to be detected.
Mounds and drift accumulation
The acoustic characteristics and thickness of U3 is similar to sediment infill between
mounds in the north west Porcupine Bank (van Weering et al., 2003; Kenyon et al.,
2003). Although it is clear that the mounds pre-existed before the interpreted
contourite drift deposition (U3), it is not apparent at what stage of mound
development the mound was at during deposition of U3. Without dates it cannot be
clear whether,
(a) the mound pre-existed and later influenced the deposition of U3.
(b) the change in palaeotopography (mound development) changed the bottom
current regime and induced drift deposits
(c) the same conditions ideal for mound growth are associated with current flow
conditions, the resulting balance between sediment supply and bottom
current strength drove competitive growth between the mound and the drift
deposition around it (Van Rooij et al., 2010; Huvenne et al., 2009)
Following the interpretation regarding U2 (Fig. 5.3.9) it appears that the mound preexisted for a substantial period of time and later influenced the deposition of U3.
Drift sedimentation around mounds in the Porcupine Seabight is estimated to have
started only from the middle Early Pleistocene (Van Rooij et al., 2009). It is possible
however that all the above factors came into effect. In the later stages of mound
development, as the mounds grew to greater dimensions, the change in
palaeotopography could have locally changed the bottom current regime and induced
drift deposits. Also, conditions favouring mound growth are probably associated with
the current flow conditions and drive competitive growth between the mound and the
drift deposition. The buried mound could be an example of a lost mound-drift
competition as described at the Enya mounds in the Porcupine Seabight by Van
Rooij et al. (2009).
127
CHAPTER 5: GEOPHYSICAL INVESTIGATION
The mound base reflector
R1 is similar to the unconformity reported underneath mounds in the north west
Porcupine Bank by van Weering et al. (2003) and Kenyon et al. (2003) and it may
correlate to the RD1 unconformity (Van Rooij et al., 2003) discussed in Section
4.3.3. The presence of the reflector surface (R1) at the base of the bioherms,
especially obvious in the N-S line (Fig. 5.3.3), suggests it may have played a
prominent role in mound initiation. It indicates that mound growth started in a single
‘event’ confined in time. Core CS06 is believed to have penetrated R1 and its
firmground composition is discussed in Chapter 6 (Section 6.3). If the reflector
correlates to this feature then the fact that coral colonization is documented at the
Belgica Mounds in the Porcupine Seabight by 2.6 Ma (Van Rooij et al., 2010) could
suggest a similar period for mound initiation at the Arc Mounds.
Controls on mound morphology and distribution
A steeper bed slope supports a locally faster current (e.g. McCave & Robinson,
1995), which probably explains the occurrence of the mounds on the scarp S2, due to
increased nutrient supply and sweeping of the coral from sedimentation. The
Stjersund–Sill (Freiwald et al., 1997) and Florida–Hatteras Slope (Paull et al., 2000)
cold-water coral communities also develop along a break in slope associated with
deep-water terraces. Guinan et al. (2011, Chapter 47) report scarps on the flank of
the Rockall Plateau at water depths from 200 m to 3000 m, colonised by Lophelia
pertusa and De Mol et al. (2011, Chapter 46) describe a coral carbonate mound
occurring at the summit of the Pen Duick Scarp in the Gulf of Cadiz. The Pelagia and
the Belgica Mounds in the Porcupine Seabight, Porcupine Bank Canyon coral
mounds, and West Florida Slope mounds also occur in association with scarps
(Newton et al., 1987; Van Rooij et al., 2003; Wheeler et al., 2007).
Sedimentary deposits originating from a single source and advected by
hydrodynamic flow result in deltaic sedimentary body shapes that expand distally
(Huber et al., 2006). Reef growth is largely a sedimentary process but the spatial
distribution of carbonate mounds is governed by interplay between both the sediment
deposition and biotic action. Also, over geologic timescales, chemical effects play a
major role in modifying carbonate deposition architecture (Purkis et al., 2007). Thus
carbonate mounds are moulded by multiscale and temporal processes, and lack a
single sedimentary source –explaining their differing shapes, with the PAR results
128
CHAPTER 5: GEOPHYSICAL INVESTIGATION
showing elongated and circular forms. However, there is a trend for larger mounds to
become more elongated (Fig. 5.1.16). When accommodation space of a facies is less
constrained it allows the deposition of more isotropic patches (Purkis et al., 2007),
i.e. circular mounds, and the top six PAR ratios (more circular forms) were
associated with the more isolated mounds (Fig. 5.1.17). This appears to indicate that
elongate mounds are the conjoined effect of the clustering of previously single
isolated mounds. However, Correa et al. (2012) suggest that mounds become
elongate as they increase in area in order to become more streamlined and Mienis
(2008) suggests the shapes of the mounds reflect the influence of biotic interaction
with strong near-bed currents - with elongation in the direction of the strongest
current (growth by accretion into the current). Paull et al. (2000) also found
individual lithoherms trending parallel with regional current flow on the FloridaHatteras Slope. This suggests that the elongate mounds aligning the scarp may reflect
the predominant north-south topographically-controlled current regime. In addition,
elongation may also be explained by bioherm units adopting the form of the shelf on
which it sits, as the size of the bioherm unit approaches the dimension of the host
shelf (Purkis et al., 2007). As the scarp and prevailing current have similar
orientations it is difficult to establish a causal link for elongation but the fact that
many of the mounds aligning the scarp reflect the shape of the scarp could indicate
that bedrock control is the dominant factor in the elongation direction for most of the
eastern mounds. There is less compelling evidence for the differing dominant
orientation of the elongate non-scarp mounds (more east-west than north-south; Fig.
5.1.7). They may be influenced by water deflection off Sc1 (Fig. 5.4.1) or it is
speculated that they may be occurring on E-W trending ridges which developed
during the Variscan Orogeny (Section 4.1.1).
129
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Fig. 5.4. 1: Possible water deflection of Sc1 could be influencing the E-W mound clusters
(black arrows) or the mounds could be occurring on Variscan E-W trending ridges.
In essence, the elongation of the Arc Mounds may be attributed to:

clustering of single mounds (conjoining effect)

hydrodynamic control
o through streamlining, or
o biotic growth by accretion into the current

bedrock control (adopting the form of the underlying topography)
It is veritable that all of the above factors are involved to different degrees at the Arc
Mound site, but the existence of off-scarp mounds not having a N-S orientation
(which is the prevalent hydrodynamic regime) seems to indicate that bedrock is the
dominant control; with individual mounds benefiting from increased current activity
by their location on the scarp and probably clustering to form elongate mounds.
The above discussion suggests that there is an interaction of topography with local
current dynamics (with associated biotic interactions) and that this may be
influencing the coral mound distribution and plan morphology. Wheeler et al. (2007)
distinguished cold-water coral carbonate mounds as those exhibiting:
130
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Inherited morphology
Developed morphology
Where
carbonate
mound
morphology
reflects
the
morphology of the (antecedent) seafloor topography.
Where carbonate mound assume their own gross
morphology mainly reflecting hydrodynamic controls.
The Arc Mounds appear to exhibit both inherited and developed morphological
characteristics. The alignment parallel to the scarps indicates a morphology
influenced by the antecedent seafloor topography. However, the reason the mounds
are occurring on the scarp is deemed related to biotic interaction with the prevailing
N-S current regime; therefore it could be argued that this same alignment is therefore
also a developed morphology. The Arc mounds do not apparently conform to either
morphological division, but rather are incorporated under both categorisations and
must be considered as such.
Influence of mounds on scarp morphology
Finally, following the above discussion, it could be argued that the mounds’
locations on the apices of the convex projections of Sc2 (Fig 5.1.9) are because these
areas are the most optimal for current delivery i.e. the scarp’s antecedent topography
controlled the mound locations. However, this could also appear to show the
influence of the mounds on the scarps’ development i.e. the presence of the mounds
prevents erosion, shaping the scarp due to topographic steering of the current.
Topographic steering of the mound(s) whereby the mounds deflect bottom currents
to force local accelerations and scouring may also have induced deeper scouring of
the seafloor. This, and the similar expression of the southernmost mounds’ moat to
Sc2 (Fig. 5.1.9), could be used as evidence that erosion is occurring in association
with the scarp and that scarp development and mound development could be partsynchronous. Although there is a lack of bedforms in the erosive areas that would
indicate sediment transport (attributed to data resolution) there are apparent erosion
features in the multibeam data (Fig. 5.4.2). Erosion associated with mounds is also
suggested to be occurring on the western margin of the Rockall trough (Øvrebø et
al., 2005) with deep fringing mound moats ascribed to current scour operating
parallel to the slope. Therefore, it is suggested that the scarp to the east (Sc2) may
131
CHAPTER 5: GEOPHYSICAL INVESTIGATION
have been modified by the presence of the mounds through topographically induced
erosion.
Fig. 5.4. 2: 3D representation of the mounds flanking Sc2. There appears to be erosion
occurring in between the mounds, indicating that the scarp’s development is partsynchronous with the mounds. The prevention of erosion by the mounds could then explain
their location on the projections of the terraced seafloor.
132
CHAPTER 5: GEOPHYSICAL INVESTIGATION
Conclusions
Spatial variability in mound characteristics was investigated. Mound shape
differences probably reflect accretion architecture governed by spatially varying
environmental conditions during development (differing local topography,
hydrodynamic regime and sedimentation) but appear to be overall strongly
dominated by bedrock control.
The scarps appear to be fault-based and it is believed that the presence of the scarp in
the east provided an elevated surface, with associated currents, for initial coral
colonisation. There also appears to be erosion occurring in association with the scarp
to the east.
Three scenarios were provided to explain the seismic profiles but without more
results the relative chronology of the mounds and unit 2 cannot be distinguished. The
mounds were established prior to the most recent drift deposition and influenced
their sedimentation, but whether they are in competition or induced their deposition
is unclear.
The Chirp results showed that the mounds root on a reflector (that correlates to a
firmground penetrated by core CS06) which may be related to Neogene (Late
Miocene-Pliocene) erosion attributed to oceanographic changes at the onset of
Northern Hemisphere glaciation.
133
CHAPTER 5: GEOPHYSICAL INVESTIGATION
134
Chapter 6: Off-mound Core (CS06)
The core description (derived from the detailed logs in Appendix A) which allowed
the identification of the sedimentary facies is given in Section 6.1. A summary log of
core CS06 (with the facies divisions) is first presented in Fig 6.1.1 in order to give an
initial overview of the core. An integrated description of the facies using the results
from the various sedimentological investigations is presented in Section 6.2,
followed by a description of the mound base reflector sediment in Section 6.3. A
discussion of the results is in Section 6.4. The individual results are presented in
Appendices A-H.
6.1: Core description
135
CHAPTER 6: OFF-MOUND CORE (CS06)
136
CHAPTER 6: OFF-MOUND CORE (CS06)
The logging results (Appendix A and Fig. 6.1.1) showed the bottom of CS06 (at
297.5 cmdc) to be made up of bright grey, slightly iron-stained, structureless, poorlysorted, muddy fine sands. These sediments rapidly pass upwards into reddish light
grey, structureless, very poorly sorted, muddy fine to medium sands which contain
mud lenses, dispersed black granules and a large dropstone (Fig. 6.1.1). The basal
contact (at 260 cmdc) of the latter unit is irregular and erosional. An abrupt change in
lithology then follows, at 212.5 cmdc, to heavily bioturbated, very poorly-sorted, red
silty mud, which alternates with very poorly-sorted, bioturbated beige muddy sands.
The muds contain brachiopods and often display green staining and most notably
they are characterised by beige stipples which are mm in scale and contain a coarser
fill (from fine to coarse sand) than the surrounding red mud matrix (Fig. 6.1.2).
Fig. 6.1. 2: Photograph of stippled mud in core CS06. The stipples are interpreted as
possible mm-scale burrows; they are beige and contain a coarser fill (from fine to coarse
sand) than the surrounding orange mud matrix (three examples are outlined). The stippled
deposits also display cm-scale burrows and bioturbated contacts.
These stippled muds with muddy sands continue from 212.5 cmdc in Section 3 to
186 cmdc in Section 2. At this point, a sharp colour change occurs, at the very
bioturbated, irregular, upper contact, to an 11 cm thick unit of chaotic, structureless,
very pale brown, muddy medium sand. The latter unit contains mud lenses and
grades upwards to become a light grey silty fine sand (Fig. 6.1.1). This pale brown to
light grey unit is comprised mainly of planktonic foraminifera, with a high lithic
component. This unit has a sharp colour change at the upper boundary (at 153 cmdc)
139
CHAPTER 6: OFF-MOUND CORE (CS06)
to pale brown silty fine sands that normally grade to red silt with a second horizon of
brachiopods (Fig. 6.1.1). There then follows grading to red silty clays, which are
again characterised by beige stipples. Larger biogenic structures are very common,
with vertical burrows (of presumed crustacean origin) particularly conspicuous. The
top of this unit is obscured by the downcore movement of the clast logged at 65
cmdc (6.1.1). However, there appears to be an abrupt lithological change at the
irregular, bioturbated boundary, at 55 cmdc, from the stippled muds to brown muddy
sands. These sands alternate with brown sandy muds in a similar manner to the
deposits beneath and the sediments also show vertical burrows, but there are no
stipples. The sediments reverse grade to an irregular upper contact at 4 cmdc, with a
colour change from light brownish grey muddy fine sand to beige muddy medium
sand.
Ice rafted debris (IRD) was observed throughout CS06 and was also noted at 0 cmdc.
Microscope observations during the IRD count showed that the carbonate sand-size
material is predominantly composed of planktonic and benthic foraminifera with
varying amounts of shells of molluscs, echinoderms, crustacean, brachiopods,
gastropods, etc.
Facies
Five sedimentary facies are evident from the logged succession in CS06 (Fig. 6.1.3
and Table 6.1.1).
Facies A
Beige foraminiferal muddy sand
Facies B
Greyish brown foraminiferal muddy sands and sandy muds
Facies C
Yellowish brown stippled foraminiferal muds and muddy sands
Facies D
Very pale brown to reddish pale brown foraminiferal muddy sands
Facies E
Reddish white foraminiferal muddy sands
Table 6.1. 1: Sedimentary facies identified in CS06.
140
CHAPTER 6: OFF-MOUND CORE (CS06)
Fig. 6.1. 3: CS06 facies (B = bioturbation). (a) Beige foraminiferal muddy sand at top of
core with irregular, gradational lower contact. (b) Bioturbated greyish brown foraminiferal
muddy sands and sandy muds with irregular contacts. (c) Heavily bioturbated yellowish
brown stippled foraminiferal muds and muddy sands with erosive contacts (two stipples
outlined by black ellipses). (d) Granule-rich structureless very pale brown to reddish pale
brown foraminiferal muddy sands with mud clast outlined. (e) Structureless reddish white
foraminiferal muddy sands with irregular, gradational upper contact with facies D. Width of
core is 11 cm.
141
CHAPTER 6: OFF-MOUND CORE (CS06)
6.2: Integrated description of facies
Fig. 6.2.1 shows the facies alongside the results of dating, grain-size analysis, IRD
content, magnetic susceptibility and X-ray fluorescence. The date of the sample at
165 cmdc is highlighted with an asterisk (*) as it is near the background limit. In the
XRF data the first 15 cm of CS06 showed lows total counts per second (CPS),
indicating a low signal. The core was unevenly split at this particular interval and
these results were discounted during graph production.
These additional results allowed an integrated description of each of the sedimentary
facies. The characteristics of each facies are described below and summarised in
Table 6.2.1 below.
Characteristics
Facies A
Facies B
Facies C
Munsell Colour
2.5Y 7/2
2.5Y 6/2 -2.5Y
5/2 Light
10YR 5/6 - 2.5Y
6/2 Yellowish
Light Grey
Brownish GreyGreyish Brown
Brown - Light
Brownish Grey
Lithology
Muddy
medium sand
Sandy mud &
muddy sand
Sedimentary
structures
Reverse
grading
Contacts
Facies D
Facies E
10YR 8/2
2.5Y 8/1
Very pale brown
White
Sandy mud &
muddy sand
Muddy sand
Muddy sand
Some normal &
reverse
grading;
bioturbated
Some normal &
reverse
grading;
bioturbated
Structureless/
slight normal
grading
None
Gradational
Sharp,
bioturbated
Sharp to
gradational;
bioturbated
Gradational
Irregular,
gradational
Ca/Fe
N/A
2-12
0-20
4-28
20-62
% IRD
Mag. Susc.
(M.S.I)
12
8-22
6-47
8-13
3-9
23-34
23-60
18-103
13-130
5-13
Ice-rafting +
hemipelagic
settling;
possibly
remobilised.
Hemipelagic
settling:
possibly
bottom current
reworked.
Depositional
mechanism
Ice rafting +
hemipelagic
settling; strong
bottom
currents.
Ice-rafting & hemipelagic settling;
very weak bottom currents
alternating with strong bottom
currents.
Table 6.2. 1: Sedimentary facies summary for CS06.
142
CHAPTER 6: OFF-MOUND CORE (CS06)
Facies A: Beige foraminiferal muddy sand
Occurrence: 0-4 cmdc.
This facies is composed of beige, poorly-sorted, muddy medium grade sand with an
irregular, gradational transition to Facies B beneath (Fig. 6.1.1). It is composed of
80% sand, 19% mud and 1% gravel; and has average IRD content (12%) and
magnetic susceptibility values (23-34 M.S.I) (Fig. 6.2.1).
Context: The presence of IRD at the very top of CS06 indicates high current flow
and very little sedimentation at this location, resulting in glacial sediments occurring
in the near surface. This would suggest the Holocene portion of the sequence may be
highly reduced and possibly winnowed.
Facies B: Greyish brown foraminiferal muddy sands and sandy muds
Occurrence: 4-55 cmdc
The bioturbated poorly sorted brown muddy sands and sandy muds of the Facies B
show normal and reverse grading with bioturbated contacts (Fig. 6.1.1). The muddy
sands show 24-40% mud, 60-76% sand and 1-8% gravel and the sandy muds show
42-65% mud 35-40% sand and 0-2% gravel; and the sand horizons show better
sorting in comparison to the muds (Fig. 6.2.1). Sediments from Facies B have low
Ca/Fe values (2-12) and high magnetic susceptibility (23-60 M.S.I) and IRD content
(8-22%). The Ca/Sr ratio increases with the muddy units, whereas the Ca/Fe ratio
decreases.
The planktonic foraminifera C14 date of ~28 ka at 25 cmdc indicates that the
succession, at this point in CS06, belongs to the top of MIS 3 (27 ka – 60 ka). It is
possible, therefore, that the continuation of Facies B above this level (up to 4 cmdc)
represent MIS 2. The 'out of range' date produced from a sample taken at the
transition between Facies B and C. suggests an age greater than 45 ka.
Context: Mud is a significant constituent of Facies B and the brown colour may
suggest a high input of terrigenous material. The latter contention is supported by the
presence of dropstones, high counts of both IRD and magnetic susceptibility and also
low Ca/Fe values. These factors suggest Facies B reflects glaciomarine
sedimentation.
145
CHAPTER 6: OFF-MOUND CORE (CS06)
Facies C: Stippled yellowish brown foraminiferal muds and muddy sands
Occurrence: 55-153 cmdc; 182.5-220 cmdc
Facies C occurs twice in CS06 and is comprised of heavily bioturbated, poorlysorted, stippled silty muds, alternating with poorly-sorted muddy sands; with
erosional and bioturbated contacts (Fig. 6.1.1). The foraminiferal sand units are made
up of 70-80 % sand 15-20% mud and 5-7% gravel and the muddy units are
composed of 63-83% mud, 18-37 % sand and less than 5% gravel; and the sands
show better sorting in comparison to the muds (Fig. 6.2.1). The IRD content and
magnetic susceptibility values are high (6-47 % IRD; and 18-103 M.S.I). The highest
peaks of IRD occur in Facies C, at 200 cmdc (47% IRD) and 105 cmdc (42%). The
lower IRD counts of Facies C occur at the transition zones between Facies C and D
at 140 cmdc and 210 cmdc. The magnetic susceptibility peaks at 74 cmdc with the
coarser interval of granule material in the foraminiferal sands logged at that depth.
The facies shows the lowest Ca/Fe values, with a minimum of 0.4 at 199 cmdc. The
Ca/Fe ratio fluctuates with the foraminiferal sands (positively; between 5-11) and the
stippled muds (negatively; between 1-6). The Ca/Sr ratio shows maximum values in
this facies (~3.5) in correlation with the muddy units.
Both Facies B and C are composed of a mixture of sand and silt; however, they are
sufficiently distinct to merit separation – they have different colours and internal
structures (the stipples are particularly characteristic of Facies C).
Brachiopod Shell Accumulations
Facies C included several unusual and highly conspicuous terebratulid brachiopod
horizons. Three of these brachiopod accumulations were identified in CS06, all
within the stippled muds (at 105 cmdc, 130-147.5 cmdc and 210 cmdc Fig. 6.1.1).
Samples of shell material were kindly examined by Professor Dave Harper (Durham
University) and determined as Macandrevia cranium (Müller, 1776). This particular
terebratulide is characterised by a smooth, oval, ventribiconvex shell with a longlooped brachidium which extends forward, anteriorly, to occupy much of the length
of the dorsal valve and about half the width of the shell. The shells of these
146
CHAPTER 6: OFF-MOUND CORE (CS06)
pedunculate brachiopods are quite thin and delicate and many are still intact and fully
articulated in CS06 (see Fig. 6.2.2).
Fig. 6.2. 2: Articulated terebratulid brachiopod with intact brachidium in silty muds
belonging to Facies C at 140.4 cmdc in CS06. Scale to the right is in cm.
In life, the pedicle of M. cranium terminates in a series of branching root-like
processes (i.e. they are rhizopedunculate) which helps attachment, allowing it to root
into soft substrates or attach to dead shells. The brachiopods were sampled at 55, 100
and 140 cmdc for dating, but all were found to be out of range of the radiocarbon
dating method (Appendix G) indicating that they are older than 45 ka.
Today M. cranium has been reported from around the Hebrides, Orkneys and
Shetlands, off northwest and southwest Ireland and in the Western Approaches
(Brunton & Curry, 1979). Generally the species is reported from water depths of
between 10 and 2,900 metres in the modern Atlantic Ocean but there are few data on
the mode of occurrence of these brachiopods in the Holocene, or indeed today (pers.
comm. Dave Harper, 2014).
Context: The red colouration of the sediment of Facies C indicates well-oxygenated
seafloor conditions and a moderate to high iron-rich mineral content. This is
interpreted as a reflection of iron-bearing terrigenous (ice-rafted) material to the core
site, corroborated by the high IRD content. This suggests that Facies C may represent
147
CHAPTER 6: OFF-MOUND CORE (CS06)
glaciomarine sedimentation. The Ca/Fe ratio indicates that the lighter colour of the
sands reflects higher biogenic carbonate content, which is attributable largely to
foraminiferal content. The Ca/Sr ratio indicates increased detrital carbonate input for
the muddy deposits. The intense bioturbation may mark zones where there was a
slow down or reduction in sedimentation.
The mode of preservation of the terebratulid brachiopods, with thin shells fully
articulated and delicate loops intact, coupled with the very fine-grained composition
of the enclosing sediments (Fig. 6.2.2), unequivocally indicates that they are
autochthonous. The three levels at which they were observed in CS06 may represent
colonisation surfaces. Most brachiopod species avoid locations with very strong
currents or waves, perhaps explaining their occurrence in the muds of Facies C. The
shell accumulations of M. cranium suggest a once thriving population at depth, free
from predation, and attached to debris and other shells. The in situ assemblages were
subsequently blanketed, possibly by muddy contourites periodically invading the
placid deeper-water environment.
Facies D: Very pale brown to reddish pale brown foraminiferal muddy sands
Occurrence: 153-182.5 cmdc; 220-260 cmdc
Facies D comprises very poorly-sorted, structureless, pale brown to reddish brown
muddy sands with ‘blotches’ and lenses of mud. Upper boundaries are gradational
and irregular, whereas lower contacts are sharp to gradational and bioturbated (Fig.
6.1.1). Upon visual inspection, the sediments contain conspicuous IRD granules.
Microscopic observations showed a high foraminiferal content, which included polar
forms (Neogloboquadrina pachyderma [sinistral]); however, there is a dominance of
warmer forms (e.g. Globigerina bulloides). The sediment grain composition ranges
from 56-83% sand, 18-54% mud and 1-17% gravel (Fig. 6.2.1). The Ca/Fe ratio
shows low to intermediate values from 4-28. The IRD content is intermediate at 813% and the magnetic susceptibility ranges from intermediate to very high (13-130
M.S.I). A magnetic susceptibility spike of 53 M.S.I at 180 cmdc correlates to a
concentration of granule material noted in the log. The very large magnetic
susceptibility spike (130 M.S.I) is due to the leucocratic granite dropstone (Appendix
H) at 235 cmdc. The lower values of the magnetic susceptibility are found with the
148
CHAPTER 6: OFF-MOUND CORE (CS06)
gradual transition to Facies E beneath at 260 cmdc. The planktonic foraminifera C14
date of 44.736 ± 1.560 kyr at 165 cmdc (Appendix G) indicates the unit in the upper
occurrence of Facies C represents Marine Isotope Stage 3 [MIS 3].
Context: Facies D has an intermediate composition, reflecting a mixture of both
carbonate and siliciclastic components. The intermediate IRD component in
association with abundant warm planktonic foraminifera types suggests that the
deposits are interstadial in nature (i.e. glacial conditions, but the fauna imply the
conditions were not polar) and this is also suggested by the MIS 3 date of the upper
occurrence of the facies (153-182.5 cmdc). The poorly-sorted and disorganised
nature of this facies may reflect some degree of localised sediment remobilisation; or
the 'blotches of clay' could be rip-up mud clasts, whereby contourite sands are
contaminated by re-suspended mud during slope erosion; this may suggest strong
currents.
Facies E: Reddish white foraminiferal muddy sands
Occurrence: 260-297.5 cmdc
Facies E consists of bright white, slightly iron-stained, structureless, muddy fine to
medium sands that are conspicuously cleaner than Facies A-D. The facies has an
irregular, gradational contact with Facies D above (Fig. 6.1.1). Some dispersed black
granules were noted during logging and the IRD content is low (3-9%) (Fig. 6.2.1).
The sediments are composed of 77-81% sand, 19-23% mud and <4% gravel. This
facies shows a step-like increase in the Ca/Fe ratio in comparison to Facies A-D,
with values from 20-62 and a high average of 42. There is also a marked decrease in
the magnetic susceptibility which shows low values ranging from only 5-13 M.S.I.
Microscopic observations showed abundant large globigerinids in the >250 µm
fraction.
Context: The cleaner nature of the sands suggests some degree of sorting by bottom
current activity. The similarities in lithology of Facies E with Facies A (the
uppermost sediments, interpreted as condensed Holocene), as well as the low
magnetic susceptibilities and IRD, and high Ca/Fe values, suggest the
commencement of an interglacial period, or the final stages of a warm period before
149
CHAPTER 6: OFF-MOUND CORE (CS06)
the interpreted colder period in Facies D above. This indicates that the sediments are
possibly representative of MIS 5.
6.3: Mound base reflector sediment (Bas Ogive)
The firmground sediment (Bas Ogive [BOg]) which marked the termination of CS06
in the core catcher sediment (Peau d’Orange) was correlated to the mound base
reflector (R1) seen in the chirp data (Section 5.3).
Composition of the Bas Ogive sediment
Wet-sieving (Table 6.3.1) showed the BOg sediments to be composed mainly of
gravel-pebble clasts >2 mm (69 wt%). Of the remaining sediment 12 wt% of the
grains are greater than or equal to coarse sand (>500 µm); 10 wt% makes up the
fine/medium sand and there is an 8 wt% mud fraction. The gravel and pebbles are
angular, with red and green staining (Fig. 6.3.1). Thin-sections showed these clasts to
be composed of bioclastic (predominantly foraminiferal) wackestone and packstone
limestones (Figs. 6.3.1 E, 6.3.2, and 6.3.3), as well as chalks, and some intraclastic
mudstones.
CS06
(wt)% MUD
(wt)% >63
µm
(wt)% 250
µm
(wt)% 500
µm
(wt)% >
2mm
Bas Ogive
8
6
4
12
69
Table 6.3. 1: Sieving results for the 'Bas Ogive' sediment from the base of core CS06.
The Globorotalia menardii, or a closely related form, in Fig. 6.3.2 suggests that the
sample cannot be older than middle Miocene (pers. comm. Prof Paul Pearson, 2013).
Two of the recovered larger iron-stained clasts from the BOg have a surface covered
in small circular impressions (e.g. Fig. 6.3.1 D). These are interpreted as the remains
of surface colonisation by benthic invertebrates. One possible candidate is a
cnidarian hydroid, such as Hydractinia echinata, with individual hydranths joined to
the colony by stems producing the circular pattern.
150
CHAPTER 6: OFF-MOUND CORE (CS06)
Fig. 6.3. 1: Bas Ogive sediment (a) Sieved sediment in beaker (sediment >63μm); (b) Lithics
>2mm (ruler in cm), note angularity (except large pebble top left) and red and green
colouration; (c) Prepared thin-sections; (d) Top surface of pebble showing colonised surface
and bottom surface showing cement; (e) Foraminiferal wackestone clast in thin-section
(PPL).
151
CHAPTER 6: OFF-MOUND CORE (CS06)
The carbonate grains are made up of planktonic foraminifera (globigerinids are
particularly abundant, Fig. 6.3.2), benthic foraminifera and also benthic invertebrate
skeletons including bryozoans, brachiopods, bivalves, and coral (Fig. 6.3.3).
Fig. 6.3. 2: Packstone clast with abundant planktonic foraminifera in Bas Ogive. The
specimen highlighted by red box is Globorotalia menardii or a closely related form. Many
thin shell fragments are also present. Thin section BG2 (PPL).
152
CHAPTER 6: OFF-MOUND CORE (CS06)
Fig. 6.3. 3: Fossils in Bas Ogive. (a) Globigerinid in BG1 (PPL); (b) Spines showing a
pseudo-uniaxial cross and a possible crustacean shell (homogenous shell structure showing
undulose extinction) in BG1, (XPL); (c) Possible Cibicides benthic foram (left), mollusc shell
(wavy elongated S form above) and well-rounded chalk intraclast in BG1 (PPL); (d)
Possible echinoderm spine in BG2 (PPL); (e) Possible bryozoan or calcareous algae in BG3
(XPL).
153
CHAPTER 6: OFF-MOUND CORE (CS06)
During the wet-sieving process the presence of a lithified carbonate matrix adhering
to the pebbles and surrounding (loose) carbonate mud became apparent and indicated
at least two phases of carbonate production. The thin-sections allowed the nature of
this feature to be investigated further. Some of the clasts showed an outer coating of
bioclastic micrite. This carbonate mud may also form matrix between clasts (Fig.
6.3.4). There were also instances of a thick red outer coating (Fig. 6.3.5) interpreted
as an iron crust.
Fig. 6.3. 4: Photomicrographs of lithified matrix found between the Bas Ogive limestone
clasts. Left – PPL, right XPL. (A-D) from thin-section BG3, (E-F) from thin-section BG2.
Green colouration ascribed to glauconite.
154
CHAPTER 6: OFF-MOUND CORE (CS06)
Fig. 6.3. 5: Iron rich clast in thin section from Bas Ogive.
Core catcher Peau d'Orange nannofossil content
Nannofossils from the matrix scrapings of a sample from the core catcher Peau
d'Orange included:

small reticulofenestrids (<4µm R. minuta)

Coccolithus pelagicus

Calcidiscus leptoporus

Syracosphaera pulchra

Pontosphaera discopora

Helicosphaera carteri, and

H. sellii.
The presence of S. pulchra suggests a Pliocene (rather than Miocene) age, but it is
difficult to prove due to the lack of diagnostic nannolith species (i.e. discoasters,
sphenoliths) (pers. comm. Dr Tom Dunkley Jones, 2013).
155
CHAPTER 6: OFF-MOUND CORE (CS06)
Context: The BOg analysis showed the rock to be composed of clearly reworked
limestone clasts. These displayed wackestone and packstone textures and they
contained planktonic foraminifera and benthic bioclasts. These lithified carbonate
clasts were redeposited with a carbonate mud, which subsequently lithified forming a
breccio-conglomerate. The green colouration of the pebbles and small green clasts
seen within the thin sections (e.g. Fig. 6.3.4) may possibly be ascribed to the
formation of glauconite. Detrital glauconite grains are a typical feature of
hardgrounds (Carson & Crowley, 1993) and their presence suggests that the area
contained iron-rich muds where sedimentation rates were relatively low. The red
colouration of some clasts is ascribed to iron-crust formation (Fig. 6.3.5).
Manganese- and ferromanganese-rich horizons are often found in association with
contourites (e.g. González et al., 2012) with the currents preventing the accumulation
of other sediments. The circular ichnofossils on the clasts also has important
implications because it means the pebbles were exposed at the surface for a sufficient
period of time for colonisation to occur (corroborated by the iron-staining and
glauconite, which both require time to develop). The nannofossil results indicate that
the rock is Pliocene, but it is difficult to prove due to the lack of diagnostic nannolith
species.
6.4: Discussion and conclusions
Discussion of sedimentary facies
The off-mound sediment of the Arc Mound site in the Porcupine Bank appears to be
highly condensed. The C14 planktonic foraminiferal dates show the sediments to be ~
28 k at 25 cmdc and 43-45 ka at 165 cmdc (as this second date is near the
background limit it has a high error margin). The radiocarbon date of the brachiopod
at 55 cmdc is deemed spurious as that interval is the transition between Facies B and
C and appears heavily bioturbated. However, the apparently in situ brachiopods at
140 cmdc also yielded a result > 45 ka. By way of comparison, a 450 cm off-mound
core from the Hovland Mound province in the Porcupine Seabight had a basal age of
only 27 kyr BP, slightly further back than the Last Glacial Maximum (Dorschel et
al., 2005).
156
CHAPTER 6: OFF-MOUND CORE (CS06)
CS06 shows alteration between five distinct sedimentary facies, but all are
essentially composed of foraminiferal muddy sands and sandy muds (Table 6.2.1).
Syn-sedimentary structures include normal and reverse graded bedding, and erosive
contacts and reactivation surfaces are also common. The different facies indicate
different processes and/or conditions acting at the time of deposition.
Deposits comprised of foraminiferal sand are common across the Porcupine Bank
(Toms, 2010; Stow et al., 2002; Øvrebø, 2004). Deep-sea currents are rarely able to
move quartz, but can move foraminifera to form dunes (McCave & Robinson, 1995).
North-moving flows on the Porcupine Bank have been reported with velocities up to
37 cm/s (~500m; Dickson & McCave, 1986). The foraminiferal sands with erosive
bases are interpreted as a result of frequent movement of the sand by currents and
resuspension of the fine fraction, which Stow and Holbrook (1984) classify as
contourite lag deposits. In addition, these lags are associated with low accumulation
rates and lack of stratigraphic resolution (McCave & Hall, 2006). Sandy contourites
have been recognised for this part of the eastern Rockall Trough margin (Sacchetti et
al., 2012; Øvrebø, 2004; Toms, 2010). Thus periods of higher current velocities are
inferred for the muddy sandy deposits of Facies A, B, C and E - or 'sandy contourite'
deposits.
The sandy units of Facies A, B and C appear similar to the description, textural and
compositional character of facies C-1a of Øvrebø (2004) and facies R1 of Toms
(2010) (Table 6.4.1) which were recorded in cores from the upper slopes of the
Porcupine Bank. The uppermost facies of CS06 (A) may also correlate to facies CM
of Toms (2010) which was found at the top of most of the cores. Likewise Øvrebø
(2004) and Toms (2010) attributed the deposits to bottom current reworking and
deposition. The muddy units in Facies B and C of CS06 are attributed to weaker
bottom currents or 'muddy contourites'. The brown sandy muds of Facies B may
correlate to the brown sandy mud facies S2-a of Toms (2010); albeit with a lower
IRD content of 8-22% in comparison to his 25-42%. These deposits are also similar
to the pale yellowish brown sandy mud facies C1-b of Øvrebø (2004), who reported
the IRD content as rare to common. The red poorly-sorted bioturbated sandy muds of
Facies C appear analogous to the 'red muddy sands' facies S2-b of Toms (2010) and
may also correlate to the IRD-rich yellowish brown sandy mud facies IH-1 of
Øvrebø (2004).
157
CHAPTER 6: OFF-MOUND CORE (CS06)
Thus the alternation between muds and muddy sands is interpreted as representing
fluctuating periods of enhanced and reduced bottom current activity. This agrees
with Howe’s (1995) contention that sandy contourites are always found together with
muddy contourites and can appear as thin intervals (a few centimetres thick).
Furthermore, sandy contourites usually overlie muddy contourites, with gradational
or sharp, erosive contacts and they often show grading, indicating waning of the
current (e.g. Faugéres & Stow, 1993; Howe, 1995). These features are all observed in
both Facies B and C. As facies A, B and C appear to broadly represent glaciomarine
deposits these mid-latitude contourites may be referred to as 'glacigenic contourites'
(Rebesco et al., 2008).
The interpreted alternation between muddy and sandy contourites in CS06 may
possibly indicate the influence of associated climatic variation. Palaeoceanographic
studies describe reduced current activities for intermediate water masses in the North
Atlantic during glacial conditions (Manighetti & McCave, 1995). In addition, within
the last glacial (MIS 2-4) there were also variations in the position and strength of
NADW - it was less vigorous and occurred at a lower latitude during stadials than
during interstadials (Øvrebø et al., 2005). Thus the muddier deposits may possibly
reflect colder conditions with more sluggish oceanic circulation. This hypothesis is
supported by higher IRD counts in the muddy units.
Øvrebø et al. (2005) report several intervals of sandy contourites deposited during
past interglacials and interstadials on the Porcupine Bank, with prominent erosion
surfaces recorded within MIS 3; and muddy contourites, ice-rafting and/or mass-flow
deposits during colder periods. In addition, the reconstruction by Sacchetti et al.
(2012) of depositional process active along the Porcupine Bank (north of the survey
area) suggested ice rafting and downslope remobilisation occurring during glacials,
and bottom current action predominating during interstadials - as well as ice rafting.
The brown sands and muds of Facies B and C may represent associated advances and
retreats of the British Irish Ice Sheet. [BIIS] A major expansion of the BIIS occurred
~ 29 ka (Kroon et al., 2000), which may potentially correlate to the muddy unit of
Facies B (at 21-30 cmdc), as the middle of the unit was dated at ~28 ka (25 cmdc).
The peak in IRD of 22% (15 cmdc) could possibly represent Heinrich 2 - when
maximum extent of the BIIS occurred around 24 ka (Greenwood & Clark 2009). The
reverse grading from this level in CS06 may represent a gradual increase in bottom
158
CHAPTER 6: OFF-MOUND CORE (CS06)
current strength during MIS 2 as it approached deglaciation. MIS 3 includes several
abrupt climatic warming phases known as Dansgaard-Oeschger [DO] events. These
abrupt transitions from cold stadial to warmer interstadial conditions are followed by
a return to cold stadial temperatures (Dansgaard et al., 1993). From 25 cmdc in CS06
the alternation of foraminiferal sands and the muds of Facies B and C may possibly
reflect these events.
Facies D may be correlated to the structureless muddy foraminiferal sand of Toms
(2010; C3-b muddy foraminiferal sand), which also contained mud patches
interpreted as rip-up clasts, and was inferred as representing hemipelagic deposition
and winnowing (contourite). In fact, the upper occurrence of Facies D in CS06
(between 153-182.5 cmdc), which indicated an age equivalent to MIS 3, may
correlate to an interstadial winnowing event during MIS 3 observed by Toms (2010),
which was linked to the Aghnadarragh interstadial onshore. However, Facies D is
also similar to the debrite facies of Øvrebø (2004; D-1) found in intermound areas,
which showed lenses of mud and patches of coarse grains. Øvrebø et al. (2005)
suggested that the debris flows are a reflection of instability during glacial times.
Thus Facies D may either be the result of strong bottom currents ripping up mud
clasts or it could represent debris flow activity. On balance, it seems the latter
interpretation is more likely due to very similar characteristics to those described by
Øvrebø et al. (2005). The scarps of the survey site (Chapter 1) are also an indication
that the area was involved in downslope movement. As episodes of downslope
movement on the Porcupine Bank are believed to be associated with glacial times
(Sacchetti et al., 2012; Øvrebø et al., 2005) and Facies D appears to represent
slumping during glacial conditions, it is postulated that the surficial scarps in the area
are of glacial origin.
Facies E is very similar to two facies described by Øvrebø (2004):
1. The pale orange sand intermound facies C-3 ascribed to bottom current deposition
and,
2. The very pale orange sand smooth slope facies P-1 interpreted as a biogenic
contourite of interglacial origin.
159
CHAPTER 6: OFF-MOUND CORE (CS06)
The composition (abundant large globigerinids, low IRD content) of Facies E
appears to be representative of warmer conditions and it is proposed that they may
belong to the Eemian interglacial (MIS 5).
Sedimentary facies subdivisions
B
CS06
(this work)
A
C
muddy
units
sandy
units
muddy
units
sandy
units
D
E
Toms
(2010)
CM or
R1
S2-a
R1
S2-b
R1
C3-b
C3-a
Øvrebø
(2004)
C-1a
C-1b
C-1a
IH-1
C-1a
D-1
C-3 or P1
Table 6.4. 1: Comparable facies to those documented herein in CS06, with those identified
by previous studies on the Porcupine Bank.
The core sediments for CS06 (from 0 until 220 cmdc) may correspond to package II
of Øvrebø (2005), which spans MIS 2-4 with sand-mud couplets falling within MIS
3. The final sands of Facies E possibly mark the onset of package III which shows
interglacial periods characterised by increased bottom-current activity and deposition
of biogenic contourites.
Mound base reflector
Hardgrounds develop when carbonate production and sedimentation rates are low, a
condition that would be promoted during glacial periods (Dorschel et al., 2005;
Smeulders, 2011; van Weering et al., 2003). Mazzini et al. (2012) identified calcite,
dolomite and phosphatic hardgrounds in samples from a scarp framing the eastern
part of the Porcupine Bank Canyon region. They suggested that water column
stratification, enhanced bottom currents, and upwelling facilitated the deposition of
organic matter and that subsequent phosphatisation lead to the formation of
phosphate-glauconite deposits. Many occurrences of hardgrounds have been
documented on erosive mound flanks (van Weering et al., 2003) and Noé et al.
160
CHAPTER 6: OFF-MOUND CORE (CS06)
(2006) concluded that hardgrounds are an essential element of carbonate mound
growth as they provide a firm substrate for mound-building invertebrates to colonise
and stabilise the inclined mound flanks.
The mechanism for hardground formation in these circumstances is still under
investigation, but the two main hypotheses, as summarised by Noé et al. (2006), are:
1. Fine-grained pelagic sediments become trapped (baffled) within the coral
framework on the mound flanks and top and become lithified by earlydiagenetic physicochemical processes.
2. Bacteria contribute to a syndepositional mound stabilisation by metabolic
products which induce carbonate precipitation at the base and within the
mounds.
The hardgrounds associated with carbonate mounds in the Porcupine Seabight and
the Rockall Bank found by Noé et al. (2006)5 were composed of mid-Pleistocene
lithified carbonates, ranging from chalks to micritic limestones, which were exhumed
during the Holocene. Current-induced sedimentary structures, a non-luminescing
matrix, indicating oxic pore fluids, and a marine isotopic signature lacking a depleted
carbon regime indicated to these workers that these hardgrounds clearly differed
from bacterially induced authigenic carbonate crusts, typical of hydrocarbon seep
settings. Carbonate lithification was ascribed to carbonate ion diffusion from
supersaturated seawater into pore fluids, with vigorous bottom currents the ultimate
control. It is not known, at present, whether hydrocarbon seepage and associated
microbial activity were involved in the lithification of the BOg limestone breccioconglomerate, or if it occurred abiogenically through oxidization by currents and a
low sediment input; however, as the BOg calcirudites are similar to the MidPleistocene carbonates described by Noé et al. (2006), their diagenetic sequence of
hardground formation may also be broadly applicable (Fig. 6.4.1).
5
This work was completed during the RV Meteor cruises M161-1 and M161-3 in 2004.
161
CHAPTER 6: OFF-MOUND CORE (CS06)
Diagenetic
sequence:
Seafloor
stage
1
(i)
2
Subsurface
stage
Event:
Sedimentation
(i)
Physical compaction
(ii)
Rim and spar cementation
Age:
Possibly
Pliocene

(iii)
Neomorphism
3
(ii)
Exhumation & subsequent
fragmentation of limestones into
angular clasts
4
(iii)
Hardground formation: cementation
between limestone clasts
Holocene
to Recent
Fig. 6.4. 1: Diagenetic sequence of hardground formation as described by Noé et al. (2006).
A limestone clast from the mound base reflector sediment contained well-rounded
chalk intraclasts (Fig. 6.4.2) implying a prior phase of exhumation and reworking to
Step 1 in Fig. 6.4.1. However, as chalk is a relatively soft lithology, the rounding of
the clasts is not taken to necessarily imply particularly protracted transport distances.
Fig. 6.4. 2: Well-rounded intraclasts (within a larger clast) in the BOg. Left is PPL and right
is XPL.
The proposed diagenetic sequence in Fig. 6.4.1 could suggest the redeposited BOg
relates to Pleistocene scarp features presently found at the seafloor surface. The scarp
162
CHAPTER 6: OFF-MOUND CORE (CS06)
investigated by Klages et al. (2004) at the Porcupine Bank6 was composed of a
10 cm capping-layer of limestone hardground coated with a thin layer of manganese
oxide, which was protecting softer carbonate-rich rocks beneath and was littered with
IRD on its top surface. The IRD implies that the scarp predates the last glaciation and
the formation of the metal precipitate indicates that the rock was exposed at the
seabed for a considerable period of time. The non-lithified nature of the underlying
sediment suggested to Klages et al. (2004) that it was Pleistocene (‘Quaternary’) in
age. However, the nannofossil content suggests the BOg may be Pliocene in age.
This would correlate to the RD1 reflector (Van Rooij et al., 2003) discussed in
section 4.3.3 which was dated to the late Miocene-Pliocene, and to the Challenger
Mound basal firmground which separates underlying Miocene glauconitic siltstones
from the (Plio-) Pleistocene mound sequence above (Thierens et al., 2010). This RD1
reflector was separately postulated to correlate to the reflector in the Chirp data
which coincides with the mound base sediment depth of recovery (see Section 5.3).
Conclusions
The sedimentological features of CS06 were not suitable for the examination of
high-frequency climatic fluctuations due to its condensed nature, but they were
valuable in delineating major, broader long-term changes. CS06 is believed to
represent (Weichsel) glacial deposits from MIS 2-5. Therefore perhaps as much as
100,000 years is represented in only 3m of core. The high concentrations of IRD
corroborate the enhanced supply of terrigenous material diluting biogenic CaCO3,
indicated by the Ca/Fe ratio.
The five facies identified in CS06 appear to represent along-slope current reworking
of sediment during glacial-interglacial and interstadial-stadial cycles. The sandy
muds of Facies B and C are interpreted as muddy contourites deposited by weak
currents; the alternating sandier units of these facies, and also of Facies A and E, are
interpreted as the result of stronger currents winnowing and removing the fines to
produce lags, now preserved as sandy contourite sequences. Facies D may represent
localised slumping, possibly associated with downslope movement during glacial
6
, This work was part of Expedition ARKTIS XIX/3 and was conducted on the research vessel POLARSTERN in 2003.
163
CHAPTER 6: OFF-MOUND CORE (CS06)
conditions. The basal sediments of Facies E appear to represent the onset of warm
conditions in MIS 5.
The mound base reflector (Bas Ogive) made up of reworked fragments of (largely
pelagic) micritic limestones, some having iron-manganese crusts, creating a breccioconglomeratic carbonate hardground/firmground. To form the mound base sediment,
pelagic oozes with admixed benthic skeletal grains were lithified, producing
wackestones and packstones, with some containing previously reworked intraclastic
mud pebbles. The glauconite and ferromanganese crusts of this hardground are
interpreted as having formed at the exposed seabed surface in a long-term nondepositional palaeoenvironment with strong currents. The timing of hardground
development may correlate with increases in global oceanic current activity
accompanying late Pliocene ice age transitions. The correlation of the BOg
penetration depth to the (geophysically visualised) mound base reflector indicates
that the lithified carbonates provided a colonisation surface for the coral and
stabilised the mound flanks as it developed.
164
Chapter 7: On-mound Core (CS07)
A summary log of core CS07 (with the identified lithofacies) is shown in Fig 7.1.1 in order to
provide an initial overview of the succession evident. A lithological description of the core
(derived from the detailed logs in Appendix A) follows in Sections 7.1. An integrated
description of the facies using the results from the various sedimentological investigations is
presented in Section 7.2, followed by a discussion of the results in Section 7.3. The individual
results are presented in Appendices A to H. Note: The structureless nature of the core material
suggests the sediments are intensively bioturbated, but as it cannot be ascertained whether
coarser pockets are representative of burrows or associated with coral breakdown they are
more often indicated by ‘pockets of rubble’ and not by a burrow symbol.
7.1: Core description
165
CHAPTER 7: ON-MOUND CORE (CS07)
166
CHAPTER 7: ON-MOUND CORE (CS07)
The logging results (Appendix A and Fig. 7.1.1) show the bottom of the core (625.5
cmdc in Section 7) to comprise relatively clean, poorly sorted, white sands with a
small amount of dispersed coral material, which grades (normal) to a brown muddy
silt at 612.5 cmdc, the contact being distinct and irregular. This lens of silt may have
formed through pressure during coring, as it grades immediately into light olive grey
very poorly-sorted silty fine to medium sand with coral granule randomly dispersed.
The hyperbolic upper contact also appears to be an artefact of coring, but there is a
sharp transition to light brownish grey very poorly-sorted silty mud with coral
fragments. This mud unit grades (590 cmdc in Section 6) to light grey, muddy, very
poorly-sorted, very fine sand. The latter unit contains coral fragments and horizontal
burrows which are infilled with coarser sediment.
A sharp irregular erosional contact follows, which is capped by a yellowish brown
very poorly-sorted coarse horizon (Fig 7.1.2), which contains:

Pebbles,

Concretions,

Large coral branches,

Gastropod and barnacle shells
169
CHAPTER 7: ON-MOUND CORE (CS07)
Fig. 7.1. 2: Coarse unit in Section 6 containing partially lithified, possible biogenic rubble
(middle left) and mud clasts (bright white in lower right) as well as branch size coral (a well
preserved coral branch beside a heavily dissolved branch in upper right), a gastropod, and a
bivalve shell (beneath mud clast) and pebbles.
This coarse horizon most likely reflects a phase of high-energy flow. It is not clear if
this increase in energy was due to higher flow rates, or gravitational collapse. The
170
CHAPTER 7: ON-MOUND CORE (CS07)
shell material from this higher-energy unit was examined by Patrick Collins (Ryan
Institute, NUIG):

The gastropod was identified as a Troschelia ?berniciensis. This predatory
and scavenging omnivore has a wide range throughout the north-eastern
Atlantic from 56°01’N, 32°42’W to north-west Norway and southwards to
25°N at depths of 90-2700 m (Bouchet & Warén, 1985). The creamy pink
shell is 10×16 mm and comprises of swollen whorls with a moderately high
spire and blunt tip. There are growth lines and spiral ridges. The aperture of
the specimen is partially fragmented.

The barnacle shell was identified as a plate of Scalpellum ?stroemi. This
suspension feeder belongs to a group known as acorn barnacles, which lack a
peduncle and have the capitulum cemented directly to the substratum.
Despite its chaotic nature, this unit does appear to broadly normally grade with the
top 3.5 cm showing a distinctly more muddy texture. Following a sharp, erosional
contact at 555 cmdc there are greyish brown very poorly-sorted silty muds with
bluish grey mottling. During sampling it was noted that this mottling continues
through the core’s interior (Fig 7.1.3).
Fig. 7.1. 3: CS07 Section 6 mottling evident at 520 cmdc.
171
CHAPTER 7: ON-MOUND CORE (CS07)
These mottled muds show better preservation of coral, with discrete dissolved
fragments dispersed throughout and also concentrated in thin horizons which are
inclined (sloping). These mottled deposits continue until 450 cmdc in Section 5, with
a slightly coarser coral-rich unit occurring at 471.5-483.5 cmdc and associated with
this unit is an accumulation of large, heavily ribbed bivalve shell fragments (Fig.
7.1.4).
Fig. 7.1. 4: Coral-rich unit located at 471.5-483.5 in Section 5 before (left) and after
sampling; bivalve is the longitudinal shell fragment in the resultant cavity. Scale in cm.
At 450 cmdc an irregular contact occurs at a 5 mm thick coral seam and a transition
to light pinkish-brown, very poorly-sorted muds is observed. Coral fragments again
occur in inclined bands in this unit (Fig. 7.1.1). These muds darken to brownish grey,
with a gradational contact to grey very poorly-sorted muds. Coral fragments become
more common in the latter unit and a well preserved coral branch was found at 425
cmdc (Fig. 7.1.1). These muds reverse grade and lighten to yellowish very poorlysorted grey silt, and then to light grey very poorly sorted very fine sand by the top of
Section 5 which also have inclined horizons of coral. These light grey fine sands
grade to grey very poorly-sorted silty mud at 390 cmdc and from this level coral
fragments are noticeably scarcer, with many having been heavily dissolved.
172
CHAPTER 7: ON-MOUND CORE (CS07)
A transition to bright grey very poorly-sorted silts follows (the contact is reverse
graded), with pockets of rubble, which continue to reverse grade to bright grey
muddy fine sands at 343 cmdc. At 330 cmdc the sediment becomes browner and at
an irregular contact (at 323 cmdc) there is a gradual transition to a 9.5 cm thick
conspicuously coarser and very poorly-sorted unit (Fig. 7.1.5). The sediment is an
admixture of medium brown silty fine sand with coarse shell hash of and coral debris
1-4 mm granules. The upper contact (which is also erosional and irregular) consists
of a 3 cm thick layer of large coral fragments (granule scale branches).
Fig. 7.1. 5: Coarser unit located at 313.5-323 cmdc in Section 4, showing erosive upper and
lower boundaries; with an upper contact consisting of a 3cm thick layer of large coral
fragments (granules- branches) that pinches to the left. Each line represents a cm in scale to
right
This coarser unit (Fig. 7.1.4) is overlain by bright grey very poorly-sorted muddy
fine sands. The contact is sharp and this overlying unit contains dispersed, dissolved
granule-grade coral along with pockets of rubble at the top of Section 4 (298 cmdc).
173
CHAPTER 7: ON-MOUND CORE (CS07)
These bright grey fine muddy sands continue in Section 3, with reverse grading to a
partially lithified horizon occurring at 285-282 cmdc (Fig. 7.1.1). This is followed by
a sharp colour transition at 278 cmdc to brownish grey very poorly-sorted medium
sands and there is a notable increase in coral debris. Pockets of rubble persist in this
unit. From 270 cmdc these deposits grade to very poorly-sorted fine sand with coral
in discontinuous bands, along with discrete pockets of rubble. There is a slight colour
transition from brownish grey to light olive grey at 250 cmdc. This is followed by a
transition (contact is graded) to very poorly sorted silts and silty muds at 235 cmdc
with a distinct colour transition back to brownish grey. These muds show abundant
well-preserved coral fragments occurring in wide horizontal bands (Fig. 7.1. 6) and
they gradually become light olive grey at 209 cmdc. Pockets of rubble occur also at
this level.
Fig. 7.1. 6: Coral bands in CS07 Section 3 at 216-220 cmdc.
The fossiliferous light olive grey silty muds continue in Section 2 and they reverse
grade to light grey very poorly sorted muddy fine sands at 180 cmdc. The sediment
shows black staining from 175-165 cmdc and becomes more consolidated at 170
cmdc. After this stained interval there is also a notable decrease in the abundance of
coral fragments and their preservation is poorer from 160 cmdc onwards. Light grey
structureless, very poorly-sorted muddy fine sands, with intermittent pockets of
rubble and low coral content, continue with reverse grading to muddy medium sands,
174
CHAPTER 7: ON-MOUND CORE (CS07)
until a gradational contact with a 1 cm thick silt layer at 142 cmdc. This layer has a
sharp upper contact to light grey very poorly-sorted fine muddy sand, which again
shows scarce coral fragments. There is reverse grading to muddy medium sand at
138 cmdc and many pockets of rubble occur. The muddy medium sands grade to fine
muddy sands, and then to light olive grey mud at 102 cmdc. A sharp change back to
structureless light grey muddy sands occurs with the core section break and the
boundary could not be constrained. There then follows a gradual colour transition to
brown very poorly-sorted muddy sands over a colour change interval of 10 cm (8575 cmdc). There are grey blotches in this 10 cm interval, made up of soft (i.e. nonlithified) clay and silt-fine sand compositions (Fig. 7.1. 7).
Fig. 7.1. 7: Irregularly shaped non-lithified bright grey blotches; upper blotches with clayey
infill and lower with silt-fine sand.
A very poorly-sorted brown muddy fine sand unit follows, which contains a large
granodiorite gneiss dropstone (Appendix H) and large bright grey irregularly shaped
concretions, showing moderate lithification of the silty clay (Fig. 7.1. 8). Sampling of
these structures at 55 cmdc and subsequent sieving showed them to be composed of
bright grey sub-rounded granule and pebble grade calcite cemented biogenic rubble
(Fig. 7.1. 8) with a white streak. Additionally, these >2 mm grains were observed to
175
CHAPTER 7: ON-MOUND CORE (CS07)
be colonised by coral, indicating that they act as suitable colonisation sites for
benthic organisms. These may be lithified representatives of the previously
documented grey blotches.
Fig. 7.1. 8: Left: Lithified irregularly shaped concretions found in Section 1 in CS07
between 55 and 65 cmdc. Right: sieved fraction >500 µm, interpreted as calcite-cemented
biogenic rubble.
There is then reverse grading over a 3 cm interval to a coarser unit at 55.5-47.5
cmdc, which shows a high concentration of black granules. These deposits normally
grade to brown very poorly sorted sandy silt at 42.5 cmdc which also shows
dispersed black granules. There then follows an irregular, indistinct boundary at 35
cmdc with light brownish grey very poorly-sorted muddy fine to medium sands
which contain gravel-branch sized coral.
176
CHAPTER 7: ON-MOUND CORE (CS07)
Facies
There is yet to be consensus on facies description of coral mounds. For example,
Douarin et al. (2013) distinguished three reef facies:

Coral-Rich,

Sediment-Rich and

Shelly-Coral Hash.
However, Foubert (2007) utilised the Dunham (1962) modified by Embry and
Klovan (1971) limestone textural classification for reefs for the Challenger Mound,
and found it to be dominated by unlithified coral-bearing (Lophelia pertusa)
floatstone and rudstone. This textural classification scheme is also applied to the Arc
Mound facies described herein. The detailed core description in Section 7.1
facilitated the identification of four sedimentary facies in CS07:
Facies A
Muddy sands displaying floatstone textures.
Facies B
Brown muddy sands and sandy muds.
Facies C
Cw: Carbonate –white.
Carbonate rich sediment, displaying
wackestone, packstone and floatstone
Cg: Carbonate-grey.
textures.
Three sub-facies are apparent:
Cgm: Carbonate grey- mottled.
Facies D
Fissure filling sediments.
7.2: Integrated description of facies
Fig. 7.2.1 shows the four facies identified in CS07, alongside the results of the
dating, grain-size analysis, IRD content, magnetic susceptibility and X-ray
fluorescence. Dates highlighted with an asterisk (*) are flagged due to the following:

The U-isotopic composition of the coral sample from 18.5 cmdc was too high
for a modern sample and
177
CHAPTER 7: ON-MOUND CORE (CS07)

The U-isotopic composition of the coral samples from 311 cmdc was
significantly low (Appendix G).
Therefore, although all the dates are included, some are not regarded as reliable.
These additional results (as well as grain composition analysis results; Appendix E)
allowed a more integrated description of each of the lithofacies. The characteristics
of each facies are described below and are also summarised in Table 7.2.1 below.
178
Muddy sand
None
Lithology
Sedimentary structures
Sandy mud
and muddy sand
Some normal
and reverse
grading
Sandy mud
and muddy
sand
Normal
and reverse
grading
179
Ice-rafting +
hemipelagic
settling; very
weak bottom
currents
8-13
<2
13-25
Hemipelagic
settling +
baffling by
coral:
strong
bottom
currents
Ca/Fe
% IRD
Mag. Susc. (M.S.I)
Depositional mechanism
* 12 at possible hiatus
3-9
23-54
24-64
Gradational
Gradational
5Y 7/1 - 5Y8/1
Light Grey-White
2.5Y 4/2 Dark
Greyish
Hemipelagic
settling + baffling
by coral;
intermediate to
strong bottom
currents
23-67
<1
1-4
Sharp
to gradational
Facies Cw
Facies B
Contacts
Munsell
Facies A
2.5Y 6/2
Light
Brownish
Grey
Characteristics
Hemipelagic
settling +
baffling by
coral;
intermediate to
strong bottom
currents
12-22
generally <3*
3-10
Sharp
to gradational
Some normal
and reverse
grading
Facies Cg
5y 7/1 - 2.5Y
5/2
Light GreyGreyish Brown
Sandy mud
and muddy
sand
Ice rafting +
hemipelagic
settling + baffling
by coral; weak
bottom currents
N/A
3-14
5-11
Gradational
Mottled
Silty mud
Facies Cgm
10YR 6/1 - 2.5Y
5/1
Brownish GreyYellowish Grey
Mass-wasting;
very strong
bottom
currents
N/A
36-62
16-73
Sharp erosive
base,
sharp top.
Normal
grading
Coarse sand
5Y 8/4
Greyish Yellow
Facies D
Table 7.2.
1:
Characteris
-tics of
facies of the
on-mound
core CS07
CHAPTER 7: ON-MOUND CORE (CS07)
CHAPTER 7: ON-MOUND CORE (CS07)
180
CHAPTER 7: ON-MOUND CORE (CS07)
Facies A: Muddy sand (displaying floatstone texture)
Occurrence: 0-25 cmdc
Facies A is characterised by loose (i.e. no longer connected to framework) branchsized coral enclosed in an unlithified muddy sand matrix (Fig. 7.2.2), and is
classified as an unlithified floatstone, in which coral fragments are the dominant
bioclasts.
Fig. 7.2. 2: Framework-building coral embedded in unlithified sandy carbonate matrix in
Facies A of CS07 Section 1 Unit 1 (5-8 cmdc).
Microscopic observations showed other biologic components (sponge spicules,
ostracods, echinoid spines, serpulid worms etc.) are also present. IRD content is low
(<2%), but the brown colouration, intermediate magnetic susceptibility (13-25 M.S.I)
and moderate Ca/Fe ratios (8-13) indicate probable terrigenous input in the mud
fraction.
Dating of the cold-water corals yielded 230Th/U ages ranging from 8.5 – 11 ka for the
upper 25 cm (Fig. 7.2.1), indicating Facies A is Holocene. Peaks in the Sr/Ca ratio in
the upper 25 cm (0.65-1.1) correlate to the branch-sized coral, and at the precise
cessation of the branch-sized coral 35 cmdc there is a pronounced Sr/Ca decrease
(from 0.6 to values 0.36-0.32) (Fig. 7.2.3).
183
CHAPTER 7: ON-MOUND CORE (CS07)
Fig. 7.2. 3: Detail of CS07 (Facies A) showing a pronounced Sr/Ca decrease at the
termination of the branch-sized coral and onset of the muddy sediments (Facies B) at 35
cmdc
Context: The data presented above suggests that Facies A is Holocene and thus
represents sediments deposited during the current interglacial. The sands suggest a
strong current regime (at least intermittently), with baffling by the corals to retain
muddy sand deposits. If Holocene palaeoenvironmental conditions were similar to
previous interglacials, it does call into question why Facies A does not reoccur
further downcore. It is possible that the sediment of Facies C represents a
diagenetically altered form of Facies A.
184
CHAPTER 7: ON-MOUND CORE (CS07)
Facies B: Brown muddy sands and sandy muds
Occurrence: 25-75 cmdc; 430-450 cmdc
This facies is characterised by brown muds and muddy sands with high IRD counts
(23-54%), low Ca/Fe ratios (3-9) and a high magnetic susceptibility values (24-64
M.S.I) (Fig 7.2.1). These sediments have a high calcareous content; however, there is
a low contribution from foraminifera, with the bioclasts comprising mainly of shell
fragments. Facies B is observed to occur at two discrete levels within the core:
1. Taking in to consideration the 11 ka date at 25 cmdc (Fig. 7.2.1), the upper
occurrence of Facies B directly beneath (from 25 down to 75 cmdc) may
potentially represent MIS 2,
2. The second (lower) occurrence can only be postulated as >45 ka based on the
results of foraminifera dating further up the core (Fig. 7.2.1). The coral dates
from these sediments (at 423.5 and 440 cmdc) that are out of range indicate
the coral in these deposits are older than 500 ka (Fig. 7.2.1).
The concretions in the upper occurrence of Facies B (at 57, 63 and 69 cmdc) are
possibly associated with an increase in the Sr/Ca ratio (~0.7-0.9) (Fig 7.2.1),
indicating that these may possibly be composed of precipitated aragonite as opposed
to calcite. Also in Facies B, two pronounced Sr/Ca decreases (dropping to 0.3 and
0.25) at 48 and 74 cmdc (Fig 7.2.1) suggest increases of detrital carbonate input.
Both occurrences of the Facies B are associated with peaks in the magnetic
susceptibility values:

The 25-75 cmdc band has an average M.S.I of 14.5 (Fig 7.2.1). This broad
peak correlates to concentrations of IRD logged for that section. Additional
magnetic susceptibility peaks are also evident due to a large black pebble at
34 cmdc, concentration of granules between 42-45 cmdc (64 M.S.I), and the
granitic dropstone at 56-72 cmdc (Fig 7.2.1).

Between 425-450 cmdc there is an average value of 28 M.S.I, with a peak of
31 M.S.I at 433 cmdc and high IRD counts (43-54 %) (Fig 7.2.1).
Context: The high IRD, and magnetic susceptibility and low Ca/Fe values indicate
that Facies B probably represents glacial conditions.
185
CHAPTER 7: ON-MOUND CORE (CS07)
Facies C: Carbonate mudstones, wackestones, and packstones
Occurrence: 75-425 cmdc; 450-555 cmdc; 580-625.5 cmdc.
Facies C accounts for most of the sedimentological succession evident in CS07 and
is characterised by alternating lighter and darker grey deposits. As Facies C is
predominantly composed of white and grey ungraded, structureless sands and muds,
it was divided into three subfacies based on their different appearances: the white
deposits of Cw, the darker grey-brownish grey deposits of Cg and the mottled
appearance of Cgm. These subfacies all show very low magnetic susceptibility
readings (1-11 M.S.I with an average of 3 M.S.I), low to intermediate IRD content
(0-14%), intermediate to high Ca/Fe values (12-67) and varying concentrations of
coral fragments, with better preservation of the coral material in the darker interbeds.
The predominantly wackestone and floatstone textures developed are made up
mainly of planktonic foraminifera and coral, with reduction of coral content in the
whiter sediments. More consolidated horizons within Facies C show aggregates of
planktonic foraminifera and carbonate concretions which are also associated with
lower coral counts (Appendix E). The sedimentological features of Facies C are
ascribed to hemipelagic settling.

Subfacies Cw: Carbonate-white foraminiferal wackestones and packstones
Occurrence: 280 – 390 cmdc; 618-625.5 cmdc
Subfacies Cw features homogeneous, pale-grey to white sandy muds and
muddy sands and recurs three times downcore. These sediments are
characterised by the highest Ca/Fe values (23-67), lowest IRD content (<1%),
lowest magnetic susceptibility (1-4 M.S.I) (Fig 7.2.1), and poorer coral
preservation (Fig 7.1.1). The step-like increase in the Ca/Fe ratio (32)
correlating to the onset of the bright white sediments at 280 cmdc (with a
peak of 68 at 296 cmdc) (highlighted by grey box in Fig. 7.2.1) indicates that
the alternation of the lighter (Cw) and darker (Cg) deposits is probably due to
the lighter coloured sediments having higher carbonate contents and lower
siliciclastic input. This subfacies also shows a decrease in the magnetic
susceptibility and the minimum magnetic susceptibility readings in the core
with a value of 1 M.S.I occurring at 312.5 cmdc. It is interrupted once by a
chaotic rudstone made up of shell hash and coral debris at 313.5-323 cmdc
186
CHAPTER 7: ON-MOUND CORE (CS07)
(Fig. 7.1.1), which may represent a reefal collapse due to gravitational
instability. If the coral dates for the 311 cmdc level were reliable, they could
indicate that the sediment may possibly belong to MIS 7 or 9.
Context: Based on the results (low IRD and magnetic susceptibility, high
Ca/Fe) the CW sediments appear to be definitely interglacial. As they are
dissimilar to the interglacial sediments of Facies A it is possible that they
represent altered sediments with dissolution of coral. Foubert et al. (2007)
identified a facies that appeared homogenous but upon inspection of x-ray
images showed the ‘ghost’ skeleton of the coral. This facies was interpreted
as an altered coral-rich facies where the corals were completely dissolved,
leaving carbonate-rich sediments behind. The XRF results and magnetic
susceptibility results for subfacies Cw support this possible explanation. This
may be another indication of warmer conditions because related increased
current activity would enhance dissolution.

Subfacies Cg: Carbonate-grey (light grey to greyish brown) wackestones and
packstones
Occurrence: 75 -278 cmdc; 390 – 425 cmdc; 580- 618 cmdc
Subfacies Cg features homogenous light grey to greyish brown sandy muds
and muddy sands and it recurs three times down through the core. At the
transition from Facies B to subfacies Cg (at 75 cmdc) there is a step-like
increase in the Ca/Fe values (28) and decrease in the % IRD (0.1 %) (Fig.
7.2.1). This indicates a higher amount of diamagnetic carbonate minerals and
lower terrigenous input than the deposits of Facies B. In comparison to
subfacies Cw, the sediments are characterised by their darker colour, better
coral preservation, lower Ca/Fe ratios (12-22) and slightly higher IRD content
(<1-22%) and magnetic susceptibility (3-10 M.S.I) (Fig. 7.2.1). Two slight
colour transitions from light grey to browner grey (Fig. 7.2.1: 233 cmdc and
270-280 cmdc) correlate to slight increases in magnetic susceptibility and
reduced Ca/Fe values (~11), confirming that the brown colour reflects
increased iron content. The higher IRD content at 170 cmdc (12 %) (Fig.
7.2.1) was made up of many black grains and the black staining of the
187
CHAPTER 7: ON-MOUND CORE (CS07)
deposits at this depth is attributed to their presence. This thin, more
consolidated, interval may represent a hiatus with winnowing and
concentration of IRD.
Sr/Ca ratios fluctuate between 0.5-0.9 in subfacies Cg of Section 3, and,
although resolution and averaging of the signal must be considered, it was
noted that these fluctuations directly correlate to coral-rich pockets and bands
(Fig. 7.2.4).
Fig. 7.2. 4: Correlation of coral bands (identified during logging) with peaks in XRF Sr/Ca
ratio element profiles.
Context: The low IRD count and magnetic susceptibility values indicate that
subfacies Cg are interglacial; however, their darker colour, better coral
preservation, lower Ca/Fe ratios and slightly higher IRD content and
magnetic susceptibility indicate that the deposits may represent ‘intermediate’
conditions in comparison to subfacies Cw.
188
CHAPTER 7: ON-MOUND CORE (CS07)

Subfacies Cgm: Carbonate grey mottled mudstones and wackestones.
Occurrence: 450-555 cmdc
Subfacies Cgm is characterised by finer grain sizes (silty muds), light grey
mottling, a distinct brownish grey-yellowish grey colour, and overall higher
IRD content (3-14 %) and magnetic susceptibility values (5-11 M.S.I) than
both subfacies Cw and Cg (Fig. 7.2.1). The peak in % IRD (14 %) for these
deposits at 490 cmdc is intermediate of that from the interpreted interglacial
deposits of subfacies Cw (<1%) and the glacial sediments of Facies B
(>22%). Subfacies Cgm also shows a coral-rich unit with a bivalve
accumulation (Fig 7.1.3). The fact that the coral dates at 423.5 and 440 cmdc
were out of range indicate that the coral could be older than 500 ka (pers.
comm. N. Frank, 2013). A final observation is that ostracods are a
conspicuous bioclastic component of Cgm with the number of ostracods
sharply increasing and decreasing with the mottled mud deposits (Appendix
E).
Context: Based on the higher IRD content and magnetic susceptibility values,
this subfacies is interpreted as potentially representing colder climatic
conditions than those of Cg and Cw. It is not thought to be fully glacial, due
to the clear difference with the interpreted glacial deposits of Facies B. Thus
the facies is interpreted as representing colder-intermediate deposits. It is not
clear what the cause of the mottling is, it could possibly represent
bioturbation or coral dissolution. The bivalve accumulation may be an
indicator of higher energy and shallower water depths (lower sea level) at this
interval, or downslope mobilisation.
Facies D: Fissure fill
Occurrence: 550-580 cmdc
This greyish yellow facies is characterised by poorly sorted-very poorly sorted very
coarse (allochthonous) material which occurs in a localised interval, with sharp
erosive boundaries (Fig. 7.1.2). This facies is noticeably coarse for its location on the
Porcupine Bank (i.e. 300 km offshore at 584 m water depth). Mud content is low,
189
CHAPTER 7: ON-MOUND CORE (CS07)
with the grain-size results showing it to be comprised of only 7.3% mud at 575 cmdc
(Appendix C). Magnetic susceptibility values are high, ranging from 16-73 M.S.I
with an average value of 44.5 M.S.I, and a peak of 73 M.S.I at 573.5 cmdc; and the
IRD component varies from 36-62%, with a peak of 62 at 575 cmdc (Fig. 7.2.1).
The chaotic and coarse nature of Facies D initially suggested it may have been the
result of gravitational collapse and fissure fill. However, it was also plausible that it
may have resulted from a sudden higher energy contour current winnowing away
mud and producing the sharp erosive basal contact. Under this scenario the apparent
normal grading (top 3.5 cm is muddier) could be ascribed to slight waning of the
current. The concretions may be interpreted as floating mud clasts undergoing
localised cementation, or the remobilisation of local carbonate clasts. Finally the
pebbles could be interpreted as IRD.
Facies D is, however, also associated with the anomalous coral date of 8.2 ± 1.8 ka
for the specimen recovered from 556 cmdc (Fig. 7.2.1). There are two possible
scenarios to explain this date:
1. The entire core is Holocene in age.
2. Younger material has been reworked from the top of the core, towards the
base. This could be either due to bioturbation, gravitational collapse or fissure
filling.
The foraminiferal and coral dates and the variability in IRD content downcore both
strongly suggest that CS07 cannot be made up entirely of Holocene sediments.
Burrowing of an organism from the side of the mound into this depth interval is
perhaps plausible, but the entire unit appears to be related to an extremely erosive
sedimentological event, which would argue against reworking through bioturbation.
Thus, if the coral date is reliable, then it appears that the coarse nature of this unit
could be interpreted as a cavity infill caused by slumping during very erosive bottom
current activity, with wasting of upper younger sediments (see discussion Section
7.3). This does call in to question why there isn’t lithification associated with the
upper boundary of the fissure. It must be assumed that the wasting deposits infilled
far enough in from the edge of the fissure to be deposited in previously undisturbed
deposits.
190
CHAPTER 7: ON-MOUND CORE (CS07)
Context: The anomalous Holocene coral date, as well as the chaotic sedimentological
structure, suggests that Facies D represents wasting of the upper mound sediments to
fill a fissure or cavity lower down in the mound. However, it is acknowledged further
dates must be obtained in order to adequately explain this interval.
7.3: Discussion and conclusions
Discussion of sedimentary facies
The ARC Mound sediments revealed by CS07 are essentially composed of
autochthonous biogenic and allochthonous siliciclastic components. During logging
coral debris was observed throughout the entire 625.5 cm of the core (Appendix A).
The corals, mainly Lophelia pertusa and less frequently Madrepora oculata, occur as
granule-sized fragments or branches in loose uncemented frameworks, and their
concentration varies downcore. Maximum coral concentration is observed at the very
top of the core and this is coincident with maximum Sr/Ca values (Fig. 7.2.1). The
biogenic components are dominated by these CWC fragments, whereas the enclosing
matrix is composed of foraminifera (planktonic and benthic), echinoids, ostracods,
bivalves, brachiopods, gastropods and sponge spicules, in variable amounts
(Appendix E). These elements are similar to bioclastic components reported from
other CWC mounds (e.g. Mienis et al., 2009; Douarin et al., 2013).
The coral dates have to be treated with caution as there is high potential for
reworking and transport of older material from an adjacent colony/graveyard colony.
However, if the dates are reflective of the surrounding sediment, and the top
sediments are modern, it would mean that Facies A apparently shows 11 ka of
deposition in a 25 cm interval (Fig. 7.2.1). This would indicate an overall
sedimentation rate of 2.3 cm/ka for the Holocene. However the 8.5 ka date at 6.5
cmdc would indicate a reduced sedimentation rate of ~0.8cm/kyr for the top 6.5 cm
and a very high sedimentation rate of ~ 7.4 cm/kyr for the lower 18.5 cm interval,
which represents just 2.5 kyrs (11ka - 8.5 ka).
The data indicates that cessation of Holocene Facies A, with the abrupt termination
of the buried coral branches and sharply decreasing Sr/Ca-ratios, co-occurs with
enhanced terrigenous input at 35 cmdc (Fig 7.2.1) and the beginning of the Facies B
muddy sediments. Based on its apparent glacial signature and presence directly
191
CHAPTER 7: ON-MOUND CORE (CS07)
beneath the 11 ka Holocene level, Facies B is interpreted as representing MIS 2.
During Pleistocene stadials North Atlantic Deep Water [NADW] formation was less
vigorous and this could explain the mud deposits between 35-45 cmdc (Fig.7.1.1).
The possible Heinrich events (indicated by significant decreases in the Sr/Ca ratio in
Fig. 7.2.1) may be H0-H2 (12-24 ka). Additionally, the concretions in these deposits
may be calcite-cemented biogenic 'rubble' due to dissolution of coral debris and
recycling of the dissolved carbon with precipitation as interstitial micrite (e.g.
Mazzini et al., 2012) as indicated by the possible associated increases in the Sr/Ca
ratio (Fig. 7.2.1).
The majority of the core comprises Facies C and the growth and development of the
coral mound is reflected in a sequence of darker and lighter sediment layers within
this facies, suggesting possible cycling in response to palaeoenvironmental
fluctuations. This is supported by the lighter horizons displaying a Ca/Fe ratio
increase, commensurate with decreases in magnetic susceptibility and IRD content,
indicating warmer conditions. Furthermore, the inverse relationship of the Fe and Ca
values over both the investigated sections of CS07 (See Appendix F), suggests that
this site is predominantly a two-component sedimentary environment, wherein the
carbonate concentration is mainly controlled by terrigenous input (Foubert, 2007).
The overall predominantly grey and white coloration of Facies C indicated a
carbonate-rich depositional environment and the XRF, MSCL and grain composition
results also support this interpretation. This facies displays wackestone, packstone
and floatstone textures, and is primarily poorly-sorted, foraminiferal muddy sands
and muds, with laterally discontinuous horizons and bands of bio-eroded coral. A
key process in forming coral debris is bioerosion which, according to Roberts et al.
(2009):
“may then provide a substratum for renewed coral settlement, be incorporated into the reef
sediment to form wacke/packstones or be further bioeroded to produce micritic components
of reef sediment.”
This could explain the wackestone and packstones textures throughout CS07 as well
as the reduced coral in the whiter sediments (Appendix E) which may be explained
by an increase of the bioerosion of coral producing coral micrite.
192
CHAPTER 7: ON-MOUND CORE (CS07)
Considerable alternation between the sandy and muddy interbeds of Facies C is
apparent, which may be explained by varying current activity or by the changing
coral reef causing reduced or increased coral baffling and entrainment of very finegrained material. The thick mud deposits between 410-555 cmdc (Fig. 7.1.1),
however, may indicate a prolonged period of weaker current supply in the vicinity of
the mound.
The sediments of Facies C are interrupted by:

The muddy IRD-rich deposits of Facies B which are interpreted to be of
glacial origin (i.e. the previously mentioned occurrence 25-75 cmdc, and the
second occurrence at 430-450 cmdc), and,

Two conspicuously coarser units (the rudstone at 313.5-323 cmdc and Facies
D at 550-580 cmdc) with may represent reefal collapses (slumping) and/or
current winnowing. Slumping may have occurred through dissolution of the
(stabilising) coral framework initiating mobilising of sediment packages, or
by undercutting of the mound flanks by currents (de Haas et al., 2009). The
first explanation seems more appropriate for the coarse rudstone, whereas
Facies D (with its cleaner sediment and anomalous coral date) appears to be
best explained by a process involving current winnowing (Fig 7.3.1).
193
CHAPTER 7: ON-MOUND CORE (CS07)
Fig. 7.3. 1: Scheme to describe possible fissure-fill sediment. (a) Mound growth during
transitional or warm conditions with water currents (indicated by blue arrows). Multibeam
bathymetry shows the mound to have multiple peaks and irregular topography (see mound
number 17 in Fig 5.1.8). (b) Increased bottom current activity generates a fissure on the
mound. (c) Destabilisation of the flanks causes downslope movement of the IRD-rich layer
and coral debris from above, infilling the fissure in the process. (d) Further erosion and
mass wasting causes downslope transport of the IRD-rich sediment to the base of the
mounds; but the sediment within the fissure is preserved, including young coral. (e) Mound
growth continues for the remainder of the Holocene. Coring of the mound (cylinder)
recovers old mound deposits interrupted by the young fissure-fill.
194
CHAPTER 7: ON-MOUND CORE (CS07)
Riding (2002) defined three main categories of organic reefs:
(1) Matrix-supported reefs
(2) Skeleton-supported reefs and
(3) Cement-supported reefs (Fig. 7.3.2).
The Arc Mounds show a predominance of matrix in comparison to either the (coral)
skeletal or cement components, and thus should be considered as matrix-supported
reefs. Within this broad grouping, subcategories termed 'cluster reefs', 'segment reefs'
and 'high relief mud mounds' could further apply to the deposits, and their
characteristics are described in Table 7.3.1:
Classification
Characteristics
Cluster reefs
Skeletal reefs in which essentially in place skeletons are
adjacent, but not in contact, resulting in matrix-support; they
are characterized by relatively high matrix/skeleton ratios and
low volumes of extra-skeletal early cement; sediment trapping
is an important corollary of skeletal growth and cluster reef
organisms are tolerant of loose sediment; absence of
framework limits the topographic relief that cluster reefs can
attain relative to spatial extent, and may permit bedding to
develop within the reef.
Segment Reefs
Matrix-supported reefs in which skeletons are adjacent, and
may be in contact, but are mostly disarticulated and mainly
parautochthonous. Matrix abundance is high, and early
cement relatively low. Moderate relief can develop in response
to intense on-reef sediment production.
Carbonate mud-dominated deposits with topographic relief
High Relief Carbonate and few (or no) .in place skeletons; thick, and internal bedding,
slumping, stromatactis cavity systems, and steep marginal
Mud Mounds
slopes may be common.
Table 7.3. 1: Characteristics of the type reefs, within the category of matrix-supported reefs,
which may describe the deposits of the on-mound core CS07 (from Riding, 2002).
The importance of sediment entrapment in the creation of cluster reefs suggests that
most of the deposits evident in CS07 appear to fall into this category. The multibeam
results (mound number 17, Fig. 5.1.6 and 5.1.8) shows that the mound has a
height/width ratio of 40 m/400 m (0.1), which demonstrates both appreciable relief
and the potential for lateral expansion. Also, as it is an elongate structure it has a
maximum lateral extent of 700 m giving it an area of some 280000 m2. Additionally,
195
CHAPTER 7: ON-MOUND CORE (CS07)
some of the deposits appear to be made up of 'mostly disarticulated and mainly
parautochthonous' components, arguing for them to be classified as segment reef
deposits. Finally, there are parts of the core with 'few or no ...in place skeletons' and
slumping features also appear to be part of the mound build-up. Thus it is concluded
that there appears to be an alternation of cluster reef sediments with segment reef and
high relief carbonate mud mound deposits.
Fig. 7.3. 2: Structural classification of Organic Reefs and Carbonate Mud Mounds (from
Riding 2002). Red boxes suggest appropriate classification for deposits in core CS07.
Although dating of the corals yielded
230
Th/U ages ranging from 8.5 to 387 ka the
older dates are not reliable, and as previously mentioned, the coral clasts may have
been reworked. However, the ‘out of range’ foraminiferal carbon dates indicate the
mound sediments in the middle and base of the core are > 45 ka. Thus the dates
suggest that Holocene reefal deposits overlie a largely Pleistocene mound succession,
which is typical for mounds of the NE Atlantic (e.g. Kano et al., 2007; Foubert,
2007; Van der Land et al., 2010; Eisele, 2010).
Previous studies have demonstrated that mound build-up is not a continuous process
(Kenyon et al., 2003; Dorschel et al., 2005; Mienis et al., 2006; Eisele et al., 2008;
196
CHAPTER 7: ON-MOUND CORE (CS07)
Huvenne et al., 2009; Thierens et al., 2010). For example, in CS07 an age of 11 ka
occurs at 25 cmdc, whereas this is reported occurring at 160 cmdc for a Rockall
Bank mound (core M2001-28; Van der Land et al., 2010); whilst another core from
the Rockall Bank mounds shows a date ~300 ka at 317 cmdc (core M2003-23),
which is actually similar to the (albeit ‘unreliable’) date found for the Arc Mound at
the same depth (317 ka at 311 cmdc). On-mound cores from the Propeller mound in
the Porcupine Seabight were all greater than 200 kyr (Dorschel et al., 2005), with
instances of 178 kyr occurring before 1 m depth (core GeoB 6728-1). The sediment
of Mound Thérèse in the Porcupine Seabight indicates a purely interglacial growth of
deep-water corals over several climate cycles (Douville et al., 2010), but Dorschel et
al. (2005) conclude that the Propeller mound sediments (also of the Porcupine
Seabight) were deposited under intermediate climate conditions (i.e. a pure
interglacial signal was not found on the mound). Thus mounds show varying
sedimentation rates and preservation of different time intervals.
The relatively low IRD counts for CS07 (Fig. 7.2.1) suggests that the Arc mounds
have formed mainly during interglacial periods, and most of the sediments appear to
represent 'intermediate' conditions (subfacies Cg and Cgm). Van der Land (2011)
postulated that hiatuses spanning up to 200 kyrs are linked to the effect climate
change has on ocean circulation patterns. These would influence the local
hydrodynamic regime and therefore food supply and sedimentation patterns; all of
which would act to reduce coral growth and alter mound development. Dorschel et
al. (2005) suggested that hiatuses spanning almost all fully glacial and fully
interglacial sediments are caused by strong bottom currents and probably mass
wasting, condensing the sediment sequences on the mound. Facies D appears to be
evidence for such mass wasting.
Although it is recognised that the term coral carbonate mound has become widely
accepted in the scientific literature, it cannot (based on the results) be definitively
stated that the Arc Mounds are dominated by carbonate sediments. Infact, some of
the deposits appear to be the result of appreciable siliciclastic input (Facies B and
subfacies Cgm). However, it has been shown that coral carbonate is definitely a
diagnostic and a compositionally important component. Therefore the mounds may
confidently be categorised as CWC mounds. Mullins et al. (1981) described eight
197
CHAPTER 7: ON-MOUND CORE (CS07)
criteria (that can be used in combination) for recognition of deep-water coral
bioherms from shallow (warm) -water coral structures in the rock record:
(1) Presence or absence of algae
(2) Diversity of corals [warm-water coral structures show up to several hundred
different coral species (Milliman, 1973) in comparison to the relatively few
species that make up deep-water build-ups (Mullins et al., 1981)]
(3) Abundance of planktonic/pelagic components
(4) Coral morphology
(5) Microborings
(6) Surrounding facies
(7) Stable isotopes
(8) Trace element geochemistry.
From the results presented here, it appears that it would be possible to identify the
Arc Mounds as deep-water coral bioherms based on four of these criteria: the
absence of algae; low diversity of corals; abundance of planktonic/pelagic
components; and surrounding facies (from Chapter 6: the mounds grade laterally and
immediately into current winnowed sands and hemipelagic oozes, and vertically onto
a hardground).
Diagenetic consideration
Downcore variability in coral content of CS07 could indicate changing
palaeoenvironmental conditions (leading to increased or reduced coral growth) or, as
discussed, varying degrees of bioerosion. In addition, dissolution of coral fragments
could negatively affect the quantity of coral preserved and introduce a taphonomic
artefact into the record (Foubert, 2007). It has been suggested that most CWC
mounds are affected by diagenetic processes (e.g. Pirlet et al., 2010; Pirlet et al.,
2011; Van der Land et al., 2010) with coupling between microbial mediated organic
matter degradation and carbonate mineral diagenesis (e.g. Ferdelman et al., 2006). It
appears that post-depositional processes evident in the sediments of CS07 have
affected certain sediment intervals in a different way. The coral are better preserved
in the darker sediments and more heavily dissolved or reduced in content in the
lighter layers. Iron reduction, linked to microbial sulfate reduction, may enhance
198
CHAPTER 7: ON-MOUND CORE (CS07)
coral preservation in the darker sediments with higher IRD content (Ferdelman et al.,
2006). A possible process affecting the geophysical and geochemical parameters is
dissolution of coral fragments leading to precipitation of CaCO3-rich background
sediments. During burial, undersaturation with respect to aragonite may lead to
partial dissolution of aragonitic components (Van der Land et al., 2010). This could
be caused by the production of CO2 during organic matter degradation which
promotes carbonic acid generation and aids driving a system into a carbonate
undersaturated state (Golubic & Schneider, 1979), or by reduced sedimentation or
erosion shallowing redox fronts (Pirlet et al., 2010). Additionally, the precipitation of
cements within sediment may begin almost immediately following deposition. The
dissolution of aragonite may lower the Mg/Ca ratio of pore waters and induce
precipitation of low-Mg calcite cement (Reuning et al., 2006). This could possibly
explain the semi-lithified interval (at 282 cmdc in subfacies Cw) and also sections of
subfacies Cw and Cg with reduced coral content showing pockets of biogenic rubble
made up of calcium carbonate concretions and aggregates of planktonic foraminifera
(Appendix E).
There is also an increase in Ca intensity at the lithified interval which was also
observed in similar lithified intervals by Foubert (2007), and indeed alternating
unlithified coral-dominated intervals with lithified intervals are also reported for
cold-water coral mounds in the SW Rockall Trough margin (Mienis et. al, 2009; Van
der Land et al., 2010). Rapid growth and breakdown of cold-water corals living in a
very favourable environment may also explain the large amount of (presumed
carbonate) mud (Kenyon et al., 2003). However, it is considered that these sections
may equally represent surface sediments that underwent more extensive bioerosion
and current activity, to produce biogenic rubble which was incorporated into the
matrix of the coral mound prior to burial - and these explanations cannot be ruled
out.
Conclusions
The grain composition analysis of CS07 showed that the coral framework is infilled
by sediment derived from a pelagic source and from fauna living on the mounds.
Coral fragments are ubiquitous throughout CS07, ranging from microscopic debris to
199
CHAPTER 7: ON-MOUND CORE (CS07)
more complete branches. The concentration and preservation of the coral material,
however, varies throughout the mound succession examined; there are alternating
sections of better preserved corals with strongly corroded corals. These results
appear to suggest that dissolution of coral occurs during the warmest interglacial
conditions.
The mound sediments themselves are unlithified, but some horizons are more
consolidated with biogenic rubble and one semi-lithified 3 cm thick horizon at 282
cmdc. The increase in degree of lithification may be due to early diagenetic
alterations within the mound (possibly related to coral dissolution) or these sections
may represent deposits that underwent more extensive bioerosion and current activity
prior to burial.
Facies identification was based initially on sedimentological differences, but these
lithofacies were corroborated by geochemical and geophysical analyses. The
succession is dominated by structureless coral wackestones and packstones, with
variable amounts of corals in a matrix of varying composition (from slightly more
siliciclastic to more carbonate) and grain size (sands and muds) with gradational to
sharp contacts. The majority of lithological contacts within CS07 (Section 7.1) are
gradational, suggesting that palaeoenvironmental transitions were not sudden.
Following the scheme of Riding (2002) the Arc Mound may be classified as
composed of alternating cluster reef, segment reef and high relief carbonate mud
mound deposits. The occurrence of coral frameworks with irregular distribution
(patches) on coral mounds is believed to cause high lateral facies heterogeneity
(Heindel et al., 2010). This, along with climatic fluctuations and diagenetic factors, is
probably the reason for the facies variation throughout CS07.
The results show that Holocene sediments (approx. 0-25 cmdc) are overlaying
apparently glacial sediment, which at 75 cmdc is underlain by apparent interglacial
deposits that are > 45 ka. The mound appears to be mainly made up of Pleistocene
sediments, with CWC growing throughout glacial and interglacial times, but
sediments primarily composed of an 'intermediate' interglacial signal.
The mound can be categorised as a CWC mound. More ancient analogues of the Arc
Mounds would probably be recognisable as a deep-water bioherms.
200
CHAPTER 7: ON-MOUND CORE (CS07)
201
Chapter 8: Summary
8.1: Discussion
ROV video data showed mound structures at the study location in the Porcupine
Bank, with dense coral cover and soft sediments trapped by the coral frameworks.
The mound structures investigated are interpreted as CWC mounds, based on their
biological, sedimentological and surficial seismic characteristics, and they appear to
be non-cemented based on the top 6 m of the on-mound core recovered.
Multibeam data (Chapter 5) has shown the 110 km2 study area to be characterised by
over 42 conical and elongated mounded complexes that are on average ~33 m high,
400 m long, 300 m wide and 10,600 m2 in areal extent. They presently reside in
water depths of between 630–850 m. The mounds thus occur in a depth range close
to the permanent thermocline, where enhanced seabed dynamics and associated food
supply are likely to occur (White and Dorschel, 2010; Thierens et al., 2010). CWC’s
need strong bottom currents for nutrient supply and also to lessen the threat of
becoming buried by accumulating sediment, thus, as pointed out by Van Rooij et. al.
(2003), their occurrence alone may be an indicator of bottom currents.
The Porcupine Bank is currently exposed to a northward - directed current system,
and bottom currents are believed to have been active throughout its depositional
history (Øvrebø, 2005). Investigation of the mound morphologies and spatial
distribution invokes a strong interaction of topography with local current dynamics.
Many previous studies have shown that the combination of elevated topographic
features with a dynamic hydrography (and associated nutrient supply/nepheloid
layers) provides a suitable habitat for CWC’s (De Mol et al., 2011; Dorschel et al.,
2009; Huvenne et al., 2009; Mienis et al., 2007; Roberts et al., 2006; White, 2007)
and this is also believed to be occurring at the Arc Mounds. Coral mound
development seems to be related to local topographic features, with mounds located
in the eastern part of the study area elongated in a north-south direction, parallel to
the seafloor scarp (Fig. 5.1.9). However, the N-S water current direction suggests
that coral mound morphology is influenced by the local current regime as well as the
surrounding seabed morphology. The relationship for the E-W trending mound
202
CHAPTER 8: SUMMARY
clusters is more difficult to explain and requires further investigation of the local
hydrodynamic variation and the antecedent topography.
The CHIRP data (Chapter 5) showed the Arc Mounds to have a different geophysical
signature to surrounding drift sediment, indicating a different composition. This was
verified by the two sediment cores (described in detail in Chapter 6 and 7), showing
the depositional regime to differ greatly between two areas that are located ~966 m
apart.
The lack of internal reflectors in the CHIRP data indicated a uniform facies for (onmound) core CS07, without any large acoustic impedance differences. Logging and
geochemical testing of CS07 showed it to be predominantly composed of varying
quantities of coral fragments enclosed in an unlithified, grey matrix of varying
composition (from slightly more siliciclastic to more carbonate). This correlates very
well to the buried facies predicted by Scoffin & Bowes (1988) (See Section 4.1.2):
‘‘On burial this facies would eventually consolidate to form bioturbated, poorly sorted,
foraminiferal sandstones and greensands, with thin, laterally discontinuous horizons of
bioeroded coral branches’’.
It appears that sediments of an intermediate nature are mainly preserved in the
mound (i.e. they do not represent fully glacial or fully interglacial conditions);
however, they are quite distinct from their off mound equivalents (as evidenced in
core CS06). The reason for this distinction is probably a combination of various
mound-specific factors (Dorschel et al., 2005) such as the nature of the coral colonies
and mound geometry (which focuses bottom currents and causes slumping due to the
creation of seafloor topography) and the related differing local environment
producing differing chemical alteration effects.
The off-mound core showed a drift-like signal on the CHIRP data and its
composition reflects the hemipelagic background sedimentation that is also apparent
in the on-mound core. These deposits comprised of foraminifera and varying
amounts contributed by siliciclastics; as well as small amounts of invertebrate
skeletons (ostracods, echinoids, bivalves etc.). Despite its relatively short length
(c.3m), the off-mound revealed a complex sedimentological history which, due to
low sedimentation rates, included a record of the last glacial and interglacial stage
(MIS 5). The deposits of the off-mound core are largely considered sandy and muddy
203
CHAPTER 8: SUMMARY
contourites with winnowing by bottom currents. Foraminiferal lags interlayered
within glacial mud deposits in CS06 suggest short-term periods of increased bottom
current speeds possibly related to interstadial events during glacial periods. Indeed
the most significant event of the foraminiferal sands (containing rip-up mud clasts)
was constrained by an AMS 14C date to MIS 3. These Late Pleistocene glaciomarine
drift sequences show variable conditions have acted in the Arc Mound area, which
appear to have been strongly influenced by glacial events. Variability in chemical
and
magnetic
properties,
IRD
content,
and
grain-size
probably
reflect
(palaeo)oceanographic conditions fluctuating in response to the northern hemisphere
glaciations.
The Chirp data showed the Pleistocene to Recent mounds and drifts formed on top of
a strong basal reflector. The moderately lithified calcareous conglomeratic hardground recovered from the off-mound core (CS06) was correlated to this reflector,
indicating the mounds developed above an erosional surface. It is believed that this
provided an initial colonisation surface for the coral and stabilised the mound as it
developed. Hydrocarbon seeps may possibly have played a role in the genesis of this
firmground if the base cutter sediment ultimately proves to contain authigenic
methanic cement, and the relationship between the mounds and the horst and graben
structures has to be investigated. CWC’s have been documented thriving under
scarps (De Mol et al., 2011) where they are protected from sand-scour. It is
postulated that initiation of the mound growth in the east occurred by colonisation of
the scarps on this erosion surface, and that this was followed by accumulation of
coral debris and sediment to allow vertical development. The erosion surface showed
evidence that it may be Pliocene in age (rather than Miocene) possibly correlating to
the RD1 regional erosional surface attributed to the major oceanographic changes at
the onset of the northern hemisphere glaciations (Thierens et al., 2010).
The CHIRP data showed the (off mound) drift deposits to onlap the mounds
themselves. This is to be expected, given the mounds build up a topography from the
seafloor, which laterally equivalent (off mound) sediments may eventually begin to
progressively blanket and cover if the bioherm fails to continue to develop and grow
laterally. The basal age of the Arc Mounds remain unknown – failed radioarbon
dates from c.6 m depth in CS07 suggest an age in excess of 45 thousand years. If
mound formation initiated soon after development of the hardground reflector (BOg),
204
CHAPTER 8: SUMMARY
it could suggest a timing of Pliocene to Pleistocene – which would be around 2.6 Ma.
This would be similar to the mounds from the Porcupine Seabight (Van Rooij et al.,
2010).
8.2: Conclusions
This work is the first qualitative and quantitative geological and geophysical
description of the Arc mounds. The multidisciplinary approach adopted in this study
combined aspects of marine sedimentology, utilisation of geochemical and
geophysical proxies, geophysical remote sensing, and GIS for a holistic study of the
Arc Mound CWC environment and the geological processes governing their
development. The advantage of using the multi-method approach was that an
observation otherwise deemed as a postulation - using one set of results, was
corroborated by other separate results, strengthening the original hypothesis. For
example, many workers mention 'darker' sediments and/or those with higher XRF Fe
results (e.g. Øvrebø, 2005; Toms, 2010; Foubert, 2007), as representative of colder
conditions, and this work found that these two criteria contain higher quantities of
IRD. In addition, too often are geological and geophysical interpretations carried out
in relative isolation from one another. The interpretation of the Chirp data was
greatly facilitated by the results of the coring. Likewise, overall interpretations of the
governing controls on mound growth were derived from a combination of the remote
sensing results -showing their distribution, with the analysis the core material.
The main conclusions and findings of this work are as follows:
(i)
A cluster of cold-water carbonate mounds lying along the south western
margin of the Porcupine Bank offshore west of Ireland has been identified in
multibeam data. Up to 42 conical and elongated mound complexes - up to
300 m wide, 84 m high and 10,600 m2 in areal extent- have been described, in
water depths of between 630–850 m.
(ii) The presence of coral throughout the length of the on-mound core (CS07)
supports the hypothesis that corals are intrinsic to the generation of these
deep-water bioherms in the NE Atlantic.
205
CHAPTER 8: SUMMARY
(iii) The geophysical results allowed the spatial variability in mound
characteristics to be investigated, allowing an assessment of the
geomorphological setting as a contributing factor to the distribution of the
mounds. Similar to Mazzini et al. (2012) it is suggested that the mounds
initiated as individual build-ups, which developed along a topographic scarp
and merged to form elongated complexes, and growth was facilitated by
current flow along this geomorphological feature. In addition, although
ancient carbonate mud mounds in the geological record differ in composition,
they have a size and shape similar to the modern mounds (William et al.,
2006). The abundance of modern mounds in deep water suggests that
formation in deep water may be more common than previously thought and
requires re-investigation of previously interpreted ancient mud mounds.
(iv) Conclusions on the palaeoenvironmental history of the study site were
derived from combining lithological logging observations with IRD counts
and XRF and MSCL results. The IRD counts, magnetic susceptibility and
XRF results helped clarify the relative importance of calcareous and siliceous
sediment contributions in the different facies identified, providing useful
signals of palaeoenvironmental change.
(v) The cores have clearly indicated that the mounds have been growing in a
highly dynamic environment. The on- and off-mound sediments at the Arc
Mound site represent quite different depositional environments; however
bottom currents, ice rafting, and slope instability appear to be universally
influential. The work has demonstrated that since at least as far back as the
Pleistocene, bottom currents have deposited possible contourite drifts on this
part of the Porcupine Bank. The mounds are in part Recent in age, and appear
to be still actively forming today, but their history extends back to the
Pleistocene. Pleistocene to recent sediments have onlapped (and may have
already buried) older Pleistocene mound strata.
(vi) The mounds and drift sediments appear to rest on a sharp erosional surface
made up of lithified carbonates. This Neogene (possibly Pliocene) erosive
surface could correlate to the Challenger mound base deposits and the
unconformity noted by Rooij et al. (2010) that represents a hiatus of ~ 7 Ma.
206
CHAPTER 8: SUMMARY
The lithified carbonates are believed to have provided an initial colonisation
surface for the coral and stabilised the mound as it developed.
8.3: Recommendations for future work

It remains to be proven whether the base of the mound unequivocally
correlates to the acoustic reflector that corresponds to the hardground
retrieved from core CS06; as the on-mound core (CS07) only penetrates 6.26
m. Ideally the entire mound sequence would be cored.

Acquisition of more sub-bottom profiling results would allow a clear
stratigraphic interpretation to be made and show the lateral variability of the
suggested units.

Acquisition of more coral ages would allow estimation of Vertical Mound
Growth Rates (VMGRs).

Cathodoluminescence and fluorescence measurements could be undertaken in
order to investigate if there was microbial involvement in some of the early
cements formed on the mounds and if this process was in any way influenced
by the presence of hydrocarbon seeps.

An isotopic (Sr Nd) study of Ice Rafted Debris by ITRAX™ XRF
geochemical analysis would help trace the provenance of the grains and thus
assess the relative IRD contributions from different ice-sheets.

The workflow for extracting the mound footprints could be more efficiently
incorporated into Model Builder in ArcMap.

Water current meter measurements at the study site would determine current
strengths, directions and periodicities at the seabed. This would aid in
qualifying the currents’ influence on the resulting sedimentary deposits, and
in relation to the benthic habitat.

The nepheloid layers in the region could be investigated in order to identify
them as the possible food source for these deep-water suspension feeders.
207
References
ALLEN, C. P., & WARNKE, D. A. 1991. HISTORY OF ICE RAFTING AT LEG 114
SITES, SUBANTARCTIC/SOUTH ATLANTIC1 in Ciesielski, P. F., Kristoflfersen,
Y., et al. (1991) Proceedings of the Ocean Drilling Program, Scientific Results, Vol.
114
ANDREE, M. 2006. The impact of bioturbation on a AMS 14C dates on handpicked
foraminifera; a statistical model. Radiocarbon, 29(2), 169-175.
BALLINI, M., KISSEL, C., COLIN, C., & RICHTER, T. 2006. Deep‐water mass source and
dynamic associated with rapid climatic variations during the last glacial stage in the
North Atlantic: A multiproxy investigation of the detrital fraction of deep‐sea
sediments. Geochemistry, Geophysics, Geosystems, 7(2).
BALSAM, W. L. & MCCOY, F. W. 1987. Atlantic sediments: Glacial/interglacial
comparisons. Paleoceanography, 2, 531-542.
BENTON, M. J., & HARPER, D. A. 2009. Introduction to paleobiology and the fossil
record. Wiley, 592pp.
BLUM, P. 1997. Physical properties handbook: a guide to the shipboard measurement of
physical properties of deep-sea cores. ODP Tech. Note, 26.
BOUCHET, P. & WARÉN, A. 1985.Revision of the northeast Atlantic bathyal and abyssal
Neogastropoda excluding Turridae (Mollusca, Gastropoda).Boll MalacolSuppl 1,
296.
BURROUGH, P. A., MCDONNELL, R., BURROUGH, P. A. & MCDONNELL, R. 1998.
Principles of geographical information systems, Oxford university press Oxford.
BREZA, J. R. 1992. High-resolution study of Neogene ice-rafted debris, Site 751, Southern
Kerguelen Plateau.In Proc. Ocean Drill.Program Sci. Results. Vol. 120, No. 1, pp.
207-221.
BROECKER, W. S. 1997. Thermohaline circulation, the Achilles heel of our climate system:
Will man-made CO2 upset the current balance?. Science, 278 (5343), 1582-1588.
BRUNTON, C.H.C. & CURRY, G.B. 1979. British Brachiopods. Synopsis of the British
Fauna (New Series) 17, 64 pp. Linnean Society of London, London.
CAIRNS, S. D. 2007. Deep-water corals: an overview with special reference to diversity and
distribution of deep-water scleractinian corals. Bulletin of Marine Science, 81, 311322.
CAIRNS S.D. 1979. The deep-water Scleractinia of the Caribbean Sea and adjacent waters.
Studies on the Fauna of Curaçao and other Carribean Islands 67: 1- 341
CARSON, G. & CROWLEY, S. 1993. The glauconite-phosphate association in
hardgrounds: examples from the Cenomanian of Devon, southwest England.
Cretaceous research, 14, 69-89.
CHATWIN, P. G. 1996. Nearbed flows and sediment movement on the continental slope
(Doctoral dissertation, University of Plymouth).
CHEW, D.M. AND STILLMAN, C.J. 2009. Late Caledonian Orogeny and magmatism. In:
The Geology of Ireland, 2nd edition (Eds. Holland, C.H. and Sanders, I.S.) Dunedin
Academic Press, Edinburgh, 143-173.
COLMAN, J. G., GORDON, D. M., LANE, A. P., FORDE, M. J., & FITZPATRICK, J. J.
2005. Carbonate mounds off Mauritania, Northwest Africa: status of deep-water
corals and implications for management of fishing and oil exploration activities.
In Cold-water corals and ecosystems (pp. 417-441).Springer Berlin Heidelberg.
CONNELL, J. 1997. Disturbance and recovery of coral assemblages. Coral reefs, 16, 101113.
208
REFERENCES
CORREA, T., GRASMUECK, M., EBERLI, G. P., REED, J. K., VERWER, K. & PURKIS,
S. 2012. Variability of cold‐water coral mounds in a high sediment input and tidal
current regime, Straits of Florida. Sedimentology, 59, 1278-1304.
COWARD, M.P. 1990. The Precambrian, Caledonian and Variscan framework to NW
Europe. In:Hardman, R. F. P. and Brooks, J. (Eds.), Tectonic Events Responsible for
Britain's Oil and Gas Reserves. Geological Society Special Publication, 55, 1-34
CROKER, P. F., & SHANNON, P. M. 1987.The evolution and hydrocarbon prospectivity of
the Porcupine Basin, offshore Ireland. Petroleum Geology of North West Europe.
Graham and Trotman, London, 633-642.
CROUDACE, I. W., RINDBY, A. & ROTHWELL, R. G. 2006. ITRAX: description and
evaluation of a new multi-function X-ray core scanner. SPECIAL PUBLICATIONGEOLOGICAL SOCIETY OF LONDON, 267, 51.
DALY, J. S., TYRRELL, S., BADENSZKI, E., HAUGHTON, P. D. W., SHANNON, P. M.
& WHITEHOUSE, M. 2008. Mesoproterozoic orthogneiss from the north Porcupine
High, offshore western Ireland. Abstract. 51st Irish Geological Research Meeting.
University College Dublin.
DANSGAARD, W., JOHNSEN, S. J., CLAUSEN, H. B., DAHL-JENSEN, D.,
GUNDESTRUP, N. S., HAMMER, C. U., ... & BOND, G. 1993. Evidence for
general
instability
of
past
climate
from
a
250-kyr
ice-core
record. Nature, 364(6434), 218-220.
DAVIES, A. J., WISSHAK, M., ORR, J. C. & MURRAY ROBERTS, J. 2008. Predicting
suitable habitat for the cold-water coral< i> Lophelia pertusa</i>(Scleractinia). Deep
Sea Research Part I: Oceanographic Research Papers, 55, 1048-1062.
DE HAAS, H., MIENIS, F., FRANK, N., RICHTER, T. O., STEINACHER, R., DE
STIGTER, H., VAN DER LAND, C. & VAN WEERING, T. C. E. 2009.
Morphology and sedimentology of (clustered) cold-water coral mounds at the south
Rockall Trough margins, NE Atlantic Ocean. Facies, 55, 1-26.
DE HAAS, H., HUVENNE, V., WHEELER, A. & UNNITHAN, V. 2002. M2002 Cruise
Report (RV Pelagia Cruise 64PE197): A TOBI Side Scan Sonar Survey of Coldwater coral Carbonate Mounds in the Rockall Trough and Porcupine Sea Bight. 21
June–14 July 2002. Texel-Southampton-Galway. NIOZ, Texel.
DE MOL, B., VAN RENSBERGEN, P., PILLEN, S., VAN HERREWEGHE, K., VAN
ROOIJ, D., MCDONNELL, A., HUVENNE, V., IVANOV, M., SWENNEN, R. &
HENRIET, J. 2002. Large deep-water coral banks in the Porcupine Basin, southwest
of Ireland. Marine Geology, 188, 193-231.
DE MOL, L., VAN ROOIJ, D., PIRLET, H., GREINERT, J., FRANK, N., QUEMMERAIS,
F. & HENRIET, J.-P. 2011. Cold-water coral habitats in the Penmarc'h and
Guilvinec Canyons (Bay of Biscay): Deep-water versus shallow-water settings.
Marine Geology, 282, 40-52.
DE MOL, B., HENRIET, J. P., & CANALS, M. 2005. Development of coral banks in
Porcupine Seabight: do they have Mediterranean ancestors?. In Cold-Water Corals
and Ecosystems (pp. 515-533). Springer Berlin Heidelberg.
DICKSON, R. R. & MCCAVE, I. N. 1986. Nepheloid layers on the continental slope west
of Porcupine Bank. Deep Sea Research Part A. Oceanographic Research Papers,
33, 791-818.
DOLAN, J.F. 1986. The relationship between the 'R2' seismic reflector and a zone of
abundant detrital and authigenicsmectite, Deep Sea Drilling Project Hole 610,
Rockall Plateau region, North Atlantic. In: Ruddiman, W.F., Kidd, R.B., Thomas,
E. et al. (eds), Initial Reports of the Deep Sea Drilling Project, 94. Washington (U.S.
Government Printing Office), 1109-1115.
DORÉ, A. G., LUNDIN, E. R., FICHLER, C., &OLESEN, O. 1997.Patterns of basement
structure and reactivation along the NE Atlantic margin. Journal of the Geological
Society, 154(1), 85-92.
DORSCHEL, B., HEBBELN, D., FOUBERT, A., WHITE, M. & WHEELER, A. J. 2007.
Hydrodynamics and cold-water coral facies distribution related to recent
209
REFERENCES
sedimentary processes at Galway Mound west of Ireland. Marine Geology, 244,
184-195.
DORSCHEL, B., WHEELER, A. J., MONTEYS, X. & VERBRUGGEN, K. 2010. Atlas of
the deep-water seabed: Ireland, Springer.
DORSCHEL, B., HEBBELN, D., RÜGGEBERG, A., DULLO, W. C. & FREIWALD, A.
2005. Growth and erosion of a cold-water coral covered carbonate mound in the
Northeast Atlantic during the Late Pleistocene and Holocene. Earth and Planetary
Science Letters, 233, 33-44.
DOUARIN, M., SINCLAIR, D. J., ELLIOT, M., HENRY, L. A., LONG, D., MITCHISON,
F., & ROBERTS, J. M. (2013). Changes in fossil assemblage in sediment cores from
Mingulay Reef Complex (NE Atlantic): Implications for coral reef build-up. Deep
Sea Research Part II: Topical Studies in Oceanography.
DOUVILLE, E., SALLÉ, E., FRANK, N., EISELE, M., PONS-BRANCHU, E., &
AYRAULT, S. 2010. Rapid and accurate U–Th dating of ancient carbonates using
inductively coupled plasma-quadrupole mass spectrometry. Chemical Geology,
272(1), 1-11.
DULLO, W.-C., FLÖGEL, S. & RÜGGEBERG, A. 2008.Cold-water coral growth in
relation to the hydrography of the Celtic and Nordic European continental margin.
Marine Ecology Progress Series, 371, 165-176.
DUNCAN, P. M. 1870. On the Madreporaria dredged up in the expedition of H.M.S.
"Porcupine". Nature, 1, 612-614.
DUNHAM, R. J. 1962. Classification of carbonate rocks according to depositional texture.
WE Ham, 108-121.
DUINEVELD, G. C. A., LAVALEYE, M. S. S., & BERGHUIS, E. M. (2004). Particle flux
and food supply to a seamount cold-water coral community (Galicia Bank, NW
Spain). Marine Ecology Progress Series, 277, 13-23.
EISELE, M. H. 2010. The long-term development of cold-water coral mounds in the NEAtlantic (Doctoral dissertation, Bremen, Univ., Diss., 2010).
EMBRY A.F., KLOVAN J.E. 1971.A late Devonian reef tract on northeastern Banks Island.
N.W.T. Bull. Canadian Petrol Geol 19: 730-781
EISELE, M., HEBBELN, D. & WIENBERG, C. 2008. Growth history of a cold-water coral
covered carbonate mound — Galway Mound, Porcupine Seabight, NE-Atlantic.
Marine Geology, 253, 160-169.
FABRICIUS, K. E., BENAYAHU, Y. & GENIN, A. 1995. Herbivory in asymbiotic soft
corals. SCIENCE-NEW YORK THEN WASHINGTON-, 90-90.
FAUGÈRES, J. C., & STOW, D. A. 1993. Bottom-current-controlled sedimentation: a
synthesis of the contourite problem. Sedimentary Geology, 82(1), 287-297.
FERDELMAN, T., KANO, A., WILLIAMS, T. & HENRIET, J. 2006. The IODP
Expedition 307 Scientists. IODP expedition, 307, 11-16.
FLEMMING, B. 2000. A revised textural classification of gravel-free muddy sediments on
the basis of ternary diagrams. Continental Shelf Research, 20, 1125-1137.
FREW, R.D., DENNIS, P.F., HEYWOOD, K.J., MEREDITH, M.P. AND BOSWELL, S.M.
2000. The oxygen isotope composition of water masses in the northern North
Atlantic. Deep-Sea Research I, 47, 2265-2286
FOLK, R. L. 1954. The distinction between grain size and mineral composition in
sedimentary-rock nomenclature. The Journal of Geology, 344-359.
FOSSÅ, J. H., MORTENSEN, P. B., &FUREVIK, D. M. 2002. The deep-water coral
Lophelia pertusa in Norwegian waters: distribution and fishery impacts.
Hydrobiologia, 471(1-3), 1-12.
FRANK, N., FREIWALD, A., CORREA, M. L., WIENBERG, C., EISELE, M., HEBBELN,
D., VAN ROOIJ, D., HENRIET, J. P., COLIN, C. & VAN WEERING, T. 2011.
Northeastern Atlantic cold-water coral reefs and climate. Geology, 39, 743-746.
FRANK, N., RICARD, E., LUTRINGER-PAQUET, A., VAN DER LAND, C., COLIN, C.,
BLAMART, D., FOUBERT, A., VAN ROOIJ, D., HENRIET, J. P. & DE HAAS,
210
REFERENCES
H. 2009. The Holocene occurrence of cold-water corals in the NE Atlantic:
Implications for coral carbonate mound evolution. Marine Geology, 266, 129-142.
FREDERIKSEN, R., JENSEN, A. & WESTERBERG, H. 1992. The distribution of the
scleractinian coral Lophelia pertusa around the Faroe islands and the relation to
internal tidal mixing. Sarsia, 77, 157-171.
FREIWALD, A., FOSSÅ, J. H., GREHAN, A., KOSLOW, T. & ROBERTS, J. M. 2004.
Cold-water coral reefs. UNEP-WCMC, Cambridge, UK, 84.
FREIWALD, A., HENRICH, R. & PÄTZOLD, J. 1997. Anatomy of a deep-water coral reef
mound from Stjernsund, West Finnmark, northern Norway. SPECIAL
PUBLICATION-SEPM, 56, 141-162.
FREIWALD, A. 2002. Reef-forming cold-water corals, in: G. Wefer, D.S.M. Billett, D.
Hebbeln, B.B. Jbrgensen, T.C.E. van Weering (Eds.), Ocean Margin Systems, Hanse
Conference Report, Springer, Berlin, pp. 365– 385.
FOUBERT, A., ROOIJ, D., BLAMART, D. & HENRIET, J. P. 2007. X-ray imagery and
physical core logging as a proxy of the content of sediment cores in cold-water coral
mound provinces: a case study from Porcupine Seabight, SW of Ireland.
International Journal of Earth Sciences, 96, 141-158.
FOUBERT, A. 2007. Nature and significance of the carbonate mound record: The Mound
Challenger Code. PhD. University of Ghent.
FRONVAL, T., JANSEN, E., HAFLIDASON, H., & SEJRUP, H. P. 1998. Variability in
surface and deep water conditions in the Nordic seas during the last interglacial
period. Quaternary Science Reviews, 17(9), 963-985.
GOLUBIĆ, S., & SCHNEIDER, J. 1979 .4 Carbonate Dissolution. Studies in Environmental
Science, 3, 107-129.
GONZÁLEZ, F. J., SOMOZA, L., LEÓN, R., MEDIALDEA, T., DE TORRES, T., ORTIZ,
J. E., ... & MERINERO, R. 2012. Ferromanganese nodules and micro-hardgrounds
associated with the Cadiz Contourite Channel (NE Atlantic): Palaeoenvironmental
records of fluid venting and bottom currents. Chemical Geology, 310, 56-78.
GRAHAM, J.R. 2009. Variscan deformation and metamorphism in Holland, C.H. &
Sanders, I.S. (eds) The Geology of Ireland, 2nd edn. 295-310.
GROBE, H. 1987. A simple method for the determination of ice-rafted debris in sediment
cores. Polarforschung, 57, 123-126.
GROTZINGER, J. P., & JAMES, N. P. 2000. Precambrian carbonates: evolution of
understanding. SPECIAL PUBLICATION-SEPM, 67, 3-22.
GUINAN, J. , LEAHY, Y., VERBRUGGEN, K., FUREY, T. 2011. Habitats at the Rockall
Bank slope failure features, Northeast Atlantic Ocean, in: P.T. Harris, E.K. Baker
(Eds.), Seafloor Geomorphology as Benthic Habitat: GeoHab Atlas of Seafloor
Geomorphic Features and Benthic Habitats, Elsevier, Amsterdam, The Netherlands,
2011 (Chapter 47).
GUINOTTE, J. M., ORR, J., CAIRNS, S., FREIWALD, A., MORGAN, L. & GEORGE, R.
2006. Will human-induced changes in seawater chemistry alter the distribution of
deep-sea scleractinian corals? Frontiers in Ecology and the Environment, 4, 141146.
GREENWOOD, S.L., CLARK, C.D. 2009. Reconstructing the last Irish Ice Sheet 1:
changing flow geometries and ice flow dynamics deciphered from the glacial
landform record. Quaternary Science Reviews 28, 3085–3100.
HAHN, D. W. 2006. Light scattering theory. Department of Mechanical and Aerospace
Engineering, University of Florida.
HALL–SPENCER, J., ALLAIN, V. & FOSSÅ, J. H. 2002. Trawling damage to Northeast
Atlantic ancient coral reefs. Proceedings of the Royal Society of London. Series B:
Biological Sciences, 269, 507-511.
HARRIS, P. T., & BAKER, E. K. (Eds.). 2011. Seafloor Geomorphology as Benthic
Habitat: GeoHab Atlas of seafloor geomorphic features and benthic habitats.
Access Online via Elsevier.
211
REFERENCES
HARVEY, J. 1982. θ-S relationships and water masses in the eastern North Atlantic. Deep
Sea Research Part A. Oceanographic Research Papers, 29(8), 1021-1033.
HEINDEL, K., TITSCHACK, J., DORSCHEL, B., HUVENNE, V. A., & FREIWALD, A.
2010. The sediment composition and predictive mapping of facies on the Propeller
Mound—A cold-water coral mound (Porcupine Seabight, NE Atlantic).Continental
Shelf Research, 30(17), 1814-1829.
HEINRICH, H. 1988. Origin and consequences of cyclic ice rafting in the northeast Atlantic
Ocean during the past 130,000 years. Quaternary research, 29(2), 142-152.
HODELL, D. A., CHANNELL, J. E., CURTIS, J. H., ROMERO, O. E. & RÖHL, U. 2008.
Onset of “Hudson Strait” Heinrich events in the eastern North Atlantic at the end of
the middle Pleistocene transition (∼ 640 ka)? Paleoceanography, 23.
HOVLAND, M., CROKER, P. F. & MARTIN, M. 1994. FAULT-ASSOCIATED SEABED
MOUNDS (CARBONATE KNOLLS?) OFF WESTERN IRELAND AND NORTH-WEST
AUSTRALIA. MARINE AND PETROLEUM GEOLOGY, 11, 232-246.
HOVLAND, M., MORTENSEN, P. B., BRATTEGARD, T., STRASS, P. & ROKENGEN,
K. 1998. Ahermatypic coral banks off mid-Norway; evidence for a link with seepage
of light hydrocarbons. Palaios, 13, 189-200.
HOWE, J. A. 1995. Sedimentary processes and variations in slope-current activity during the
last Glacial-Interglacial episode on the Hebrides Slope, northern Rockall Trough,
North Atlantic Ocean. Sedimentary Geology, 96(3), 201-230.
HUTHNANCE, J. 1981. Waves and currents near the continental shelf edge. Progress in
Oceanography, 10, 193-226.
HUTHNANCE, J. M. 1986. The Rockall slope current and shelf-edge processes.
Proceedings of the Royal Society of Edinburgh B, 88, 83-101.
HUBER, C., LEUENBERGER, M., SPAHNI, R., FLÜCKIGER, J., SCHWANDER, J.,
STOCKER, T. F., JOHNSEN, S., LANDAIS, A. & JOUZEL, J. 2006. Isotope
calibrated Greenland temperature record over Marine Isotope Stage 3 and its relation
to CH4. Earth and Planetary Science Letters, 243, 504-519.
HUVENNE, V. A. I., VAN ROOIJ, D., DE MOL, B., THIERENS, M., O’DONNELL, R. &
FOUBERT, A. 2009. Sediment dynamics and palaeo-environmental context at key
stages in the Challenger cold-water coral mound formation: clues from sediment
deposits at the mound base. Deep Sea Research Part I: Oceanographic Research
Papers, 56, 2263-2280.
HUVENNE, V., DE MOL, B. & HENRIET, J.-P. 2003. A 3D seismic study of the
morphology and spatial distribution of buried coral banks in the Porcupine Basin,
SW of Ireland. Marine Geology, 198, 5-25.
ISELIN, C. D. 1939. The influence of vertical and lateral turbulence on the characteristics of
the waters at mid-depths. Transactions, American Geophysical Union, 20, 414-417.
IODP 307 EXPEDITION SCIENTISTS. 2005. Modern carbonate mounds: Porcupine
drilling. IODP Preliminary Report, doi: 10.2204/iodp.pr.307.2005
JANSEN, J. H. F., VAN DER GAAST, S. J., KOSTER, B., & VAARS, A. J. 1998.
CORTEX, a shipboard XRF-scanner for element analyses in split sediment
cores. Marine Geology, 151(1), 143-153.
JENKINS, R., DE VRIES, J. L. & LIEBHAFSKY, H. 1969. Practical X-ray spectrometry.
Physics Today, 22, 111.
KANO, A., FERDELMAN, T. G., WILLIAMS, T., HENRIET, J. P., ISHIKAWA, T.,
KAWAGOE, N., ... & LI, X. 2007. Age constraints on the origin and growth history
of a deep-water coral mound in the northeast Atlantic drilled during Integrated
Ocean Drilling Program Expedition 307. Geology, 35(11), 1051-1054.
KENYON, N. H., AKHMETZHANOV, A. M., WHEELER, A. J., VAN WEERING, T. C.
E., DE HAAS, H. & IVANOV, M. K. 2003. Giant carbonate mud mounds in the
southern Rockall Trough. Marine Geology, 195, 5-30.
KOENITZ, D., WHITE, N., MCCAVE, I. N., & HOBBS, R. 2008. Internal structure of a
contourite drift generated by the Antarctic Circumpolar Current. Geochemistry,
Geophysics, Geosystems 9, no. 6.
212
REFERENCES
KLAGES, M., THIEDE, J. & FOUCHER, J. 2004. Reports on Polar and Marine Research.
The Expedition ARKTS XIX/3 of the Research Vessel Polarstern in 2003; Reports of
Legs 3a, 3b and 3c. Germany: Alfred Wegener Institute for Polar and Marine
Research.
KNUTZ, P. C., JONES, E. J. W., AUSTIN, W. E. N., & VAN WEERING, T. C. E. 2002.
Glacimarine slope sedimentation, contourite drifts and bottom current pathways on
the Barra Fan, UK North Atlantic margin. Marine Geology, 188(1), 129-146.
KROMER, B., LINDAUER, S., SYNAL, H. A., &WACKER, L. 2013. MAMS–A new
AMS facility at the Curt-Engelhorn-Centre for Achaeometry, Mannheim, Germany.
Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions
with Materials and Atoms, 294, 11-13.
KRISSEK, L. A. 1985. 28. THE OCCURRENCE, ABUNDANCE, AND COMPOSITION
OF ICE-RAFTED DEBRIS IN SEDIMENTS FROM DEEP SEA DRILLING
PROJECT SITES 579 AND 580, NORTHWEST PACIFIC1.
KROON, D., SHIMMIELD, G., AUSTIN, W., DERRICK, S., KNUTZ, P., SHIMMIELD,
T., 2000. Century to millennial-scale sedimentological–geochemical records of
glacial–Holocene sediment variations from the Barra Fan (NE Atlantic). Journal of
the Geological Society of London 157, 643–653.
LANG, J., C., & NEUMANN, A., C. 1980. Lithoherm faunal zonation and mound growth:
Geol. Soc. In America Abstracts with Programs. Vol. 12, No. 7, p. 468.
LE DANOIS E. 1948. Les profondeurs de la mer. Payot, Paris
LINDBERG, B., & MIENERT, J. 2005. Sedimentological and geochemical environment of
the Fugløy Reef off northern Norway. In Cold-Water Corals and Ecosystems (pp.
633-650). Springer Berlin Heidelberg.
LINNAEUS, C., 1758: Systema Naturae per regna tria naturae, secundum classes, ordines,
genera, species, cum chararteribus, differentiis, synonymis, locis (Holmiae: Laurentii
Salvii). Tomus 1: Regnum animale. 10th Ed. Stockholm, 824 pp.
MAZZINI, A., AKHMETZHANOV, A., MONTEYS, X. & IVANOV, M. 2012. The
Porcupine Bank Canyon coral mounds: oceanographic and topographic steering of
deep-water carbonate mound development and associated phosphatic deposition.
Geo-Marine Letters, 1-21.
MCCAVE, I., MANIGHETTI, B. & ROBINSON, S. 1995. Sortable silt and fine sediment
size/composition
slicing:
parameters
for
palaeocurrent
speed
and
palaeoceanography. Paleoceanography, 10, 593-610.
MANIGHETTI, B., & MCCAVE, I. N. 1995 "Depositional fluxes, palaeoproductivity, and
ice rafting in the NE Atlantic over the past 30 ka." Paleoceanography 10.3. 579-592.
MARGRETH, S., RÜGGEBERG, A. & SPEZZAFERRI, S. 2009. Benthic foraminifera as
bioindicator for cold-water coral reef ecosystems along the Irish margin. Deep Sea
Research Part I: Oceanographic Research Papers, 56, 2216-2234.
MASSON, D., DOBSON, M., AUZENDE, J. M., COUSIN, M., COUTELLE, A., ROLET,
J. & VAILLANT, P. 1989. Geology of Porcupine Bank and Goban Spur,
Northeastern Atlantic—Preliminary results of the Cyaporc submersible cruise.
Marine Geology, 87, 105-119.
MASSON, D. G., BETT, B. J., BILLETT, D. S. M., JACOBS, C. L., WHEELER, A. J., &
WYNN, R. B. 2003. The origin of deep-water, coral-topped mounds in the northern
Rockall Trough, Northeast Atlantic. Marine Geology, 194(3), 159-180.
MAZZINI, A., AKHMETZHANOV, A., MONTEYS, X. & IVANOV, M. 2012. The
Porcupine Bank Canyon coral mounds: oceanographic and topographic steering of
deep-water carbonate mound development and associated phosphatic deposition.
Geo-Marine Letters, 1-21.
MCCAVE, I. N., & I. R. HALL. 2006. "Size sorting in marine muds: Processes, pitfalls, and
prospects
for
paleoflow‐speed
proxies." Geochemistry,
Geophysics,
Geosystems 7.10.
MIENIS, F. 2008. Environmental constraints on cold-water coral growth and carbonate
mound formation. PhD, University of Amsterdam.
213
REFERENCES
MIENIS, F., VAN DER LAND, C., DE STIGTER, H. C., VAN DE VORSTENBOSCH, M.,
DE HAAS, H., RICHTER, T. & VAN WEERING, T. C. E. 2009. Sediment
accumulation on a cold-water carbonate mound at the Southwest Rockall Trough
margin. Marine Geology, 265, 40-50.
MIENIS, F., DE STIGTER, H. C., WHITE, M., DUINEVELD, G., DE HAAS, H. & VAN
WEERING, T. C. E. 2007. Hydrodynamic controls on cold-water coral growth and
carbonate-mound development at the SW and SE Rockall Trough Margin, NE
Atlantic Ocean. Deep Sea Research Part I: Oceanographic Research Papers, 54,
1655-1674.
MIENIS, F., VAN WEERING, T., DE HAAS, H., DE STIGTER, H., HUVENNE, V. &
WHEELER, A. 2006. Carbonate mound development at the SW Rockall Trough
margin based on high resolution TOBI and seismic recording. Marine Geology, 233,
1-19.
MILLIMAN, JOHN D. 1973. "Caribbean coral reefs." Biology and geology of coral reefs 1,
1-50.
MOHN, C., RENGSTORF, A., WHITE, M., DUINEVELD, G., MIENIS, F., SOETAERT,
K., & GREHAN, A. 2014. Linking benthic hydrodynamics and cold-water coral
occurrences: A high-resolution model study at three cold-water coral provinces in
the NE Atlantic. Progress in Oceanography.
MOORE, D. R., BULLIS, J., & HARVEY, R. 1960. A deep-water coral reef in the Gulf of
Mexico. Bulletin of Marine Science, 10(1), 125-128.
MULLINS, H. T., NEWTON, C. R., HEATH, K. & VAN BUREN, H. M. 1981. Modern
deep-water coral mounds north of Little Bahama Bank; criteria for recognition of
deep-water coral bioherms in the rock record. Journal of Sedimentary Research, 51,
999-1013.
NAYLOR, D., & SHANNON, P. 2009.Geology of offshore Ireland. The geology of Ireland,
405-460.
NAYLOR, D. & MOUNTENEY, S.N. 1975. Geology of the north-west European
continental shelf, Vol. 1. 162 pp. G.T. Dudley Publishers Ltd. London.
NEUMANN A.C., KOFOED J.W., KELLER G.H. 1977. Lithoherms in Straits of Florida.
Geology 5: 4-10
NEW, A. L., BARNARD, S., HERRMANN, P., &MOLINES, J. M. 2001. On the origin and
pathway of the saline inflow to the Nordic Seas: insights from models. Progress in
Oceanography, 48(2), 255-287.
NEW, A. & SMYTHE-WRIGHT, D. 2001. Aspects of the circulation in the Rockall Trough.
Continental Shelf Research, 21, 777-810.
NEWTON, C. R., MULLINS, H. T., GARDULSKI, A. F., HINE, A. C. & DIX, G. R. 1987.
Coral mounds on the west Florida slope: unanswered questions regarding the
development of deep-water banks. Palaios, 359-367.
NOÉ, S., TITSCHACK, J., FREIWALD, A. & DULLO, W.-C. 2006. From sediment to
rock: diagenetic processes of hardground formation in deep-water carbonate mounds
of the NE Atlantic. Facies, 52, 183-208.
ØVREBØ, L. K. 2005. Spatial and temporal variations in sedimentary slope processes
along the margins of the Rockall Trough, offshore West Ireland. PhD, University
College Dublin.
ØVREBØ, L. K., HAUGHTON, P. D. W. & SHANNON, P. M. 2006. A record of
fluctuating bottom currents on the slopes west of the Porcupine Bank, offshore
Ireland—implications for Late Quaternary climate forcing. Marine Geology, 225,
279-309.
ORR, J. C., FABRY, V. J., AUMONT, O., BOPP, L., DONEY, S. C., FEELY, R. A.,
GNANADESIKAN, A., GRUBER, N., ISHIDA, A. & JOOS, F. 2005.
Anthropogenic ocean acidification over the twenty-first century and its impact on
calcifying organisms. Nature, 437, 681-686.
PAULL, C., NEUMANN, A., AM ENDE, B., USSLER III, W. & RODRIGUEZ, N.
2000.Lithoherms on the Florida–Hatteras slope. Marine Geology, 166, 83-101.
214
REFERENCES
PIRLET, H. 2010. The matrix of cold-water coral mounds : origin and early-diagenetic
interactions. Ghent University. Faculty of Sciences, Ghent, Belgium.
PIRLET, H., COLIN, C., THIERENS, M., LATRUWE, K., VAN ROOIJ, D., FOUBERT,
A., FRANK, N., BLAMART, D., HUVENNE, V. A. I., SWENNEN, R.,
VANHAECKE, F. & HENRIET, J. P. 2011. The importance of the terrigenous
fraction within a cold-water coral mound: A case study. Marine Geology, 282, 1325.
POLLARD, R. T., GRIFFTTHS, M. J., CUNNINGHAM, S. A., READ, J. F., PÉREZ, F. F.
& RÍOS, A. F. 1996.Vivaldi 1991 - A study of the formation, circulation and
ventilation of Eastern North Atlantic Central Water. Progress in Oceanography, 37,
167-192.
POORE, R. Z., & BERGGREN, W. A.. 1975. Late Cenozoic planktonic foraminiferal
biostratigraphy and paleoclimatology of Hatton-Rockall Basin; DSDP Site 116. The
Journal of Foraminiferal Research, 5(4), 270-293.
POPENOE, P. 1994. USGS Open File Rept. 93- 724.
PURKIS, S. J., KOHLER, K. E., RIEGL, B. M. & ROHMANN, S. O. 2007.The statistics of
natural shapes in modern coral reef landscapes. The Journal of Geology, 115, 493508.
PRAEG, D., STOKER, M. S., SHANNON, P. M., CERAMICOLA, S., HJELSTUEN, B.,
LABERG, J. S., & MATHIESEN, A. 2005. Episodic Cenozoic tectonism and the
development of the NW European ‘passive’ continental margin. Marine and
Petroleum Geology, 22(9), 1007-1030.
RAAZ MAHESHWARI, B. R., & KUMAR, M. 2012. Some Corals Reefs Could Survive
Global Warming. Bull. Env. Pharmacol. Life Scien. Volume, 1, 21-25.
RAES, M., & VANREUSEL, A. (2005). The metazoan meiofauna associated with a coldwater coral degradation zone in the Porcupine Seabight (NE Atlantic). Cold-water
corals and ecosystems. Springer Berlin Heidelberg. 821-847.
REBESCO, M., CAMERLENGHI, A. & VAN LOON, A.J. 2008. Contourite Reasearch: A
field in Full Development. In: Rebesco, M. and Camerlenghi, A. (Eds.)
Countourites, Developments In Sedimentology 60, Elsevier, Amsterdam. 1-10
REID, J. L. 1979. On the contribution of the Mediterranean Sea outflow to the NorwegianGreenland Sea. Deep Sea Research Part A. Oceanographic Research
Papers, 26(11), 1199-1223.
RENGSTORF, A. M., GREHAN, A., YESSON, C., & BROWN, C. 2012. Towards HighResolution Habitat Suitability Modeling of Vulnerable Marine Ecosystems in the
Deep-Sea: Resolving Terrain Attribute Dependencies. Marine Geodesy, 35(4), 343361.
RENGSTORF, A. M., YESSON, C., BROWN, C., & GREHAN, A. J. 2013. High‐resolution
habitat suitability modelling can improve conservation of vulnerable marine
ecosystems in the deep sea. Journal of Biogeography.
REUNING, L., REIJMER, J.J.G., & MATTIOLI, E. 2006. Aragonite cycles: Diagenesis
caught in the act. Sedimentology, 53, 849-866.
RIDING, R. 2002. Structure and composition of organic reefs and carbonate mud mounds:
concepts and categories. Earth-Science Reviews, 58(1), 163-231.
ROBERTS, J. M., WHEELER, A., FREIWALD, A., CAIRNS, S. & PRESS, C. U. 2009.
Cold-water corals: the biology and geology of deep-sea coral habitats, Cambridge
University Press Cambridge.
ROBERTS, J. M., WHEELER, A. J. & FREIWALD, A. 2006. Reefs of the deep: the biology
and geology of cold-water coral ecosystems. Science, 312, 543.
ROBINSON, L. F., ADKINS, J. F., KEIGWIN, L. D., SOUTHON, J., FERNANDEZ, D. P.,
WANG, S. & SCHEIRER, D. S. 2005. Radiocarbon variability in the western North
Atlantic during the last deglaciation. Science, 310, 1469-1473.
ROTHWELL, R. G. 2006. New techniques in sediment core analysis, Geological Society
Publishing House.
215
REFERENCES
RUDDIMAN, W. F., & MCINTYRE, A. 1981. The mode and mechanism of the last
deglaciation: oceanic evidence. Quaternary Research, 16(2), 125-134.
RÜGGEBERG, A., DULLO, C., DORSCHEL, B., & HEBBELN, D. 2007. Environmental
changes and growth history of a cold-water carbonate mound (Propeller Mound,
Porcupine Seabight). International Journal of Earth Sciences, 96(1), 57-72.
RÜGGEBERG, A., DORSCHEL, B., DULLO, W. C., & HEBBELN, D. 2005. Sedimentary
patterns in the vicinity of a carbonate mound in the Hovland Mound Province,
northern Porcupine Seabight. In Cold-Water Corals and Ecosystems (pp. 87-112).
Springer Berlin Heidelberg.
SACCHETTI, F., BENETTI, S., GEORGIOPOULOU, A., SHANNON, P., O'REILLY, B.,
DUNLOP, P., QUINN, R. & Ó COFAIGH, C. 2012.Deep-water geomorphology of
the glaciated Irish margin from high-resolution marine geophysical data. Marine
Geology, 291, 113-131.
SCHMITZ, W. J. 1996. On the world ocean circulation. Volume II, the Pacific and Indian
Oceans/a global update. Woods Hole Oceanographic Institution.
SCHÖNFELD, J., & ZAHN, R. 2000.Late Glacial to Holocene history of the Mediterranean
Outflow. Evidence from benthic foraminiferal assemblages and stable isotopes at the
Portuguese margin. Palaeogeography, Palaeoclimatology, Palaeoecology, 159(1),
85-111.
SCOFFIN, T. P. & BOWES, G. E. 1988. The facies distribution of carbonate sediments on
Porcupine bank, northeast Atlantic. Sedimentary Geology, 60, 125-134.
SCOURSE, J. D., HAAPANIEMI, A. I., COLMENERO-HIDALGO, E., PECK, V. L.,
HALL, I. R., AUSTIN, W. E., ... & ZAHN, R. 2009. Growth, dynamics and
deglaciation of the last British–Irish ice sheet: the deep-sea ice-rafted detritus record.
Quaternary Science Reviews, 28(27), 3066-3084.
SEJRUP, H. P., HJELSTUEN, B. O., TORBJØRN DAHLGREN, K. I., HAFLIDASON, H.,
KUIJPERS, A., NYGÅRD, A., ... &VORREN, T. O. 2005. Pleistocene glacial
history of the NW European continental margin. Marine and Petroleum
Geology, 22(9), 1111-1129.
SINCLAIR, I. K., SHANNON, P. M., WILLIAMS, B. P. J., HARKER, S. D., &MOOREN,
J. G. 1994. Tectonic control on sedimentary evolution of three North Atlantic
borderland Mesozoic basins1. Basin Research, 6(4), 193-217.
SHANNON, P. M., CORCORAN, D. & HAUGHTON, P. 2001. The petroleum exploration
of Ireland's offshore basins: introduction. Geological Society, London, Special
Publications, 188, 1-8.
SHANNON P.M., STOKER M.S., PRAEG D., VAN WEERING T.C.E., DE HAAS H.,
NIELSEN T., DAHLGREN K.I.T., HJELSTUEN B.O. 2005. Sequence stratigraphic
analysis in deepwater, underfilled NW European passive margin basins. Mar Petrol
Geol 22:1185-1200.
SHI, L., ZHANG, Y., CHEN, Y., FU, H. & LIU, G. 2010.Quartz grain SEM microtextures
analyses of sub-glacial deposits at hailuogou glacier. Beijing Daxue Xuebao (Ziran
Kexue Ban)/Acta Scientiarum Naturalium Universitatis Pekinensis, 46, 96-102.
SMEULDERS, G. 2011. Cold-water coral habitats of Rockall and Porcupine Bank, NE
Atlantic Ocean: sedimentary facies and benthic foraminiferal assemblages.
ST JOHN, K. E. 1999. Data report: Site 918 IRD mass accumulation rate record, late
miocene-pleistocene. In Proceedings of the Ocean Drilling Program. Scientific
results (Vol. 163, pp. 163-166). Ocean Drilling Program.
STETSON, T. R., SQUIRES, D. F. & PRATT, R. M. 1962.Coral banks occurring in deep
water on the Blake Plateau. American Museum novitates; no. 2114.
STOKER, M.S., VAN WEERING, T.C.E. & SVAERDBORG, T. 2001.A Mid-Late
Cenozoic tectonostratigraphic framework for the Rockall Trough. In: Shannon,
P.M., Haughton, P.D.W. & Corcoran, D. (eds) The Petroleum Exploration of
Ireland's Offshore Basins. Geological Society, London, Special Publication, 188,
411-438.
216
REFERENCES
D.A.V. Ed. 2002 Deep-water Contourite Systems: Modern Drifts and Ancient
Series, Seismic and Sedimentation Characteristics. No. 22. Geological Society
STOW, D. A. V., & HOLBROOK, J. A. 1984. "North Atlantic contourites: an overview."
Geological Society, London, Special Publications 15.1: 245-256.
SQUIRES, D.F. 1964. Fossil coral thickets in Wairarapa, New Zealand. J Paleontol 38:904915
SYNAL, H. A., STOCKER, M., & SUTER, M. 2007. MICADAS: a new compact
radiocarbon AMS system. Nuclear Instruments and Methods in Physics Research
Section B: Beam Interactions with Materials and Atoms, 259(1), 7-13.
TATE, M. & DOBSON, M. 1989.Late Permian to early Mesozoic rifting and sedimentation
offshore NW Ireland. Marine and Petroleum Geology, 6, 49-59.
TEICHERT, C. 1958. Cold-and deep-water coral banks. AAPG Bulletin, 42, 1064-1082.
TITSCHACK, J., THIERENS, M., DORSCHEL, B., SCHULBERT, C., FREIWALD, A.,
KANO, A., TAKASHIMA, C., KAWAGOE, N. & LI, X. 2009. Carbonate budget of
a cold-water coral mound (Challenger Mound, IODP Exp. 307). Marine Geology,
259, 36-46.
THIERENS, M., BROWNING, E., PIRLET, H., LOUTRE, M.-F., DORSCHEL, B.,
HUVENNE, V., TITSCHACK, J., COLIN, C., FOUBERT, A. & WHEELER, A.
2013. Cold-water coral carbonate mounds as unique palaeo-archives: the PlioPleistocene Challenger Mound record (NE Atlantic). Quaternary Science Reviews,
73, 14-30.
THIERENS, M., TITSCHACK, J., DORSCHEL, B., HUVENNE, V. A. I., WHEELER, A.
J., STUUT, J. B. & O'DONNELL, R. 2010. The 2.6 Ma depositional sequence from
the Challenger cold-water coral carbonate mound (IODP Exp. 307): Sediment
contributors and hydrodynamic palaeo-environments. Marine Geology, 271, 260277.
THORPE, S. A., & WHITE, M. 1988. A deep intermediate nepheloid layer. Deep Sea
Research Part A. Oceanographic Research Papers, 35(9), 1665-1671.
TREUDE T., NIGGEMANN J., KALLMEYER J., WINTERSTELLER P., SCHUBERT
C.J., BOETIUS A., JORGENSEN B.B. 2005. Anaerobic oxidation of methane and
sulfate reduction along the Chilean continental margin. Geochim Cosmochim Acta
69:2767–2779.
TOMS, L. T. 2010. Late Quaternary Stratigraphy and Sediment Distribution on Currentswept Slopes, West Porcupine Bank, Offshore Ireland, University College Dublin.
School of Geological Sciences, University College Dublin.
VAN DER LAND, C. 2011. Impact of diagenesis on carbonate mound formation. Royal
Netherlands Institute for Sea Research (NIOZ).
VAN DER LAND, CEES, ET AL. 2010. "Diagenetic processes in carbonate mound
sediments at the south‐west Rockall Trough margin." Sedimentology 57.3, 912-931.
VAN ROOIJ, D., HUVENNE, V., DE MOL, B., LIETAERT, B., VAN EETVELT, B.,
BLAMART, D., CACHO, I., MCCAVE, N. & HENRIET, J. 2010. The Porcupine
Contourite Depositional System: 2.6 Ma of competition between cold-water coral
mounds and sediment drifts. Deep-water circulation: processes and products,
Sociedad Geológica de España (SGE), 177-178.
VAN ROOIJ, D., DE MOL, B., HUVENNE, V., IVANOV, M. & HENRIET, J. P. 2003.
Seismic evidence of current-controlled sedimentation in the Belgica mound
province, upper Porcupine slope, southwest of Ireland. Marine Geology, 195, 31-53.
VAN ROOIJ, D., BLAMART, D., RICHTER, T., WHEELER, A., KOZACHENKO, M. &
HENRIET, J. P. 2007a. Quaternary sediment dynamics in the Belgica mound
province, Porcupine Seabight: ice-rafting events and contour current processes.
International Journal of Earth Sciences, 96, 121-140.
VAN ROOIJ, D., BLAMART, D., KOZACHENKO, M. & HENRIET, J. P. 2007b. Small
mounded contourite drifts associated with deep-water coral banks, Porcupine
Seabight, NE Atlantic Ocean. Geological Society Special Publication, 276.
STOW,
217
REFERENCES
VAN ROOIJ, D., BLAMART, D., DE MOL, L., MIENIS, F., PIRLET, H., WEHRMANN,
L. M., ... & HENRIET, J. P. 2011. Cold-water coral mounds on the Pen Duick
Escarpment, Gulf of Cadiz: The MiCROSYSTEMS project approach. Marine
Geology, 282(1), 102-117.
VAN ROOIJ, D., HUVENNE, V., BLAMART, D., HENRIET, J.-P., WHEELER, A. & DE
HAAS, H. 2009. The Enya mounds: a lost mound-drift competition. International
Journal of Earth Sciences, 98, 849-863.
VAN WEERING, T. C. E., DE HAAS, H., DE STIGTER, H., LYKKE-ANDERSEN, H. &
KOUVAEV, I. 2003. Structure and development of giant carbonate mounds at the
SW and SE Rockall Trough margins, NE Atlantic Ocean. Marine Geology, 198, 6781.
VON HUENE, R., LARSON, E., & CROUCH, J. 1973. Preliminary study of ice-rafted
erratics as indicators of glacial advances in the Gulf of Alaska. Initial Rep. Deep Sea
Drill. Proj, 18, 835-842.WENTWORTH, C. K. 1922. A scale of grade and class
terms for clastic sediments. The Journal of Geology, 377-392.
WELLS, J. W. 1933. Corals of the Cretaceous of the Atlantic and Gulf coastal plains and
western interior of the United States (Vol. 18, No. 67). Harris co.
WENINGER, B., JOERIS, O. & DANZEGLOCKE, U. 2004. Calpal: Cologne Radiocarbon
Calibration and Paleoclimate Research Package, Cologne.
WEHRMANN, L., KNAB, N., PIRLET, H., WILD, C. & FERDELMAN, T. 2008.
Biogeochemistry of Norwegian cold-water coral reef sediments. Geophysical
Research Abstracts.
WHEELER, A., KOZACHENKO, M., HENRY, L. A., FOUBERT, A., DE HAAS, H.,
HUVENNE, V., MASSON, D. & OLU, K. 2011. The Moira Mounds, small coldwater coral banks in the Porcupine Seabight, NE Atlantic: Part A—an early stage
growth phase for future coral carbonate mounds? Marine Geology, 282, 53-64.
WHEELER, A. J., BEYER, A., FREIWALD, A., DE HAAS, H., HUVENNE, V. A. I.,
KOZACHENKO, M., OLU-LE ROY, K. & OPDERBECKE, J. 2007. Morphology
and environment of cold-water coral carbonate mounds on the NW European
margin. International Journal of Earth Sciences, 96, 37-56.
WHEELER, A., BECK, T., THIEDE, J., KLAGES, M., GREHAN, A. & MONTEYS, F.
2005. Deep-water coral mounds on the Porcupine Bank, Irish Margin: preliminary
results from the Polarstern ARK-XIX/3a ROV cruise. Cold-water Corals and
Ecosystems, 393-402.
WHITE, M., MOHN, C., STIGTER, H. & MOTTRAM, G. 2005. Deep-water coral
development as a function of hydrodynamics and surface productivity around the
submarine banks of the Rockall Trough, NE Atlantic Cold-Water Corals and
Ecosystems. In: FREIWALD, A. & ROBERTS, J. M. (eds.). Springer Berlin
Heidelberg.
WHITE, M., ROBERTS, J. M., & VAN WEERING, T. 2007. Do bottom-intensified diurnal
tidal currents shape the alignment of carbonate mounds in the NE Atlantic? GeoMarine Letters, 27(6), 391-397.
WHITE, M. & DORSCHEL, B. 2010. The importance of the permanent thermocline to the
cold-water coral carbonate mound distribution in the NE Atlantic. Earth and
Planetary Science Letters, 296, 395-402.
WIENBERG, C., HEBBELN, D., FINK, H.G., MIENIS, F., DORSCHEL, B., VERTINO,
A., CORREA, M.L., FREIWALD, A. 2009. Scleractinian cold-water corals in the
Gulf of Cadiz—first clues about their spatial and temporal distribution. Deep-Sea
Research Part I-Oceanographic Research Papers 56 (10), 1873–1893.
WILLIAM, T., KANO, A., FERDELMAN, T., HENRIET, J. P., ABE, K., ANDRES, M. S.,
... & TITSCHAK, J. 2006. Cold-water coral mounds revealed. Eos, Transactions
American Geophysical Union, 87(47), 525.
WILSON, J.B. 1979. "Patch" development of the deepwater coral Lophelia pertusa (L.) on
Rockall Bank. J Mar Biol Ass UK 59: 165-177.
218
REFERENCES
WILSON, M. F., O’CONNELL, B., BROWN, C., GUINAN, J. C. & GREHAN, A. J. 2007.
Multiscale terrain analysis of multibeam bathymetry data for habitat mapping on the
continental slope. Marine Geodesy, 30, 3-35.
WOOD, R. 1999. Reef evolution, Oxford University Press, USA.
WOOD, R. 2001. Are reefs and mud mounds really so different?. Sedimentary Geology,
145(3), 161ZIBROWIUS, H. 1980. Les Scléractiniaires de la Méditerranée et de l'Atlantique nordoriental. Mémoires de l'Institut océanographique, Monaco, 11.
219
APPENDICES
APPENDIX A - DETAILED SEDIMENTOLOGICAL LOGS
Digitised logs of CS06 Section 1-3. Note: BF-black flecked, SM- stippled mud, LHS
and RHS – Left Hand Side and Right Hand Side, v. – very, conc. - concentration,
DC-down core.
220
APPENDICES
221
APPENDICES
222
APPENDICES
223
APPENDICES
Digitised logs of CS07 Section 1-7. Note: LP- Lophelia pertusa, LHS and RHS –
Left Hand Side and Right Hand Side, v. – very, conc. - concentration, DC-down
core.
224
APPENDICES
225
APPENDICES
226
APPENDICES
227
APPENDICES
228
APPENDICES
229
APPENDICES
230
APPENDICES
231
APPENDICES
APPENDIX B - MSCL
MSCL results, photographs and graphic log for core CS06. For key to graphic log see Fig. 6.1.1.
232
APPENDICES
MSCL results for core CS07 Section 3 and 4, with relevant core photographs and graphic log. Intervals of interest are shaded in grey. For key to
graphic log see APPENDIX A.
233
MSCL results for core CS07 Section 3 and 4, with relevant core photographs and graphic log.Intervals of interest are shaded in grey. For key to
graphic log see APPENDIX A.
APPENDICES
234
APPENDICES
MSCL results for CS07 Section 5-7, with relevant core photographs and graphic log. Intervals of interest are shaded in grey. For key to graphic log
see APPENDIX A.
235
APPENDICES
APPENDIX C - GRAIN SIZE ANALYSIS
CS06
Grain size analysis results for core CS06: (from left) photographs and graphic lithology,
logged mean grain size, sieve-based grain size analysis results, laser-based grain size
analysis results, mean and sorting (derived from the laser-based results).
236
APPENDICES
The ternary sand and mud diagram produced from the laser results show the
sediment to be mainly made up of varying quantities of mud and fine sand,
producing deposits of muddy sand and sandy mud.
At 60 and 65 cmdc the laser results measured ‘muddy sand’. This was probably due
to large burrow structures present in the very fine mud. The sieving results, by
contrast, found these two sections to consist of ‘sandy mud’.
Sediment 60-65 cmdc in Core
CS06 with burrows of lighter and
coarser infill than the
surrounding (enclosing) mud.
237
APPENDICES
The wt% of material > 250 μm < 2 mm (CS%) ranges from 3.48 to 33.64% with an
average of 17.26%, and there is no long-term trend of increase or decrease downcore.
The CS% was found not to correlate with the IRD content. The coarse-sand fraction
at site CS06 contains a significant component of non-ice-rafted grains. Thus, the
conclusion of von Huene et al. (1973) that the wt% of the 250 μm-2 mm fraction
serves as a reliable indicator of ice-rafting importance does not appear to be valid for
this site. The influence of these other components within the coarse-sand fraction on
the weight percent values must be removed in order to obtain an accurate measure of
ice-rafting influence.
CS06 CS%
wt%
0.00
5.00
10.00
15.00
20.00
0
50
CMDC
100
150
200
250
300
238
25.00
30.00
35.00
APPENDICES
CS07
Grain size analysis results for core CS07: (from left) the logged mean grain size, the sievebased grain size analysis results, laser-based shows grain size analysis results, mean and
sorting (derived from the laser-based results).
239
APPENDICES
CS07 laser diffraction grain size analysis results. Ternary gravel sand mud diagram
showing the majority of the sediments fall within the sandy mud and muddy sand categories.
Only one sample (575 cmdc) is sand. Note: the laser analysis does not measure gravel.
The results of the two different grain-size analysis methods employed (laser-based
and sieve-based), as well as the logged mean grain size, all appear to correlate very
well. The ternary diagram produced from the LGSA results shows that there is
generally a high component of mud, producing deposits of gravelly sandy muds and
gravelly muddy sands, with only one sample classified as ‘sand’ (575 cmdc), and
none classified as pure mud.
240
APPENDICES
APPENDIX D - IRD COUNT
50
45
% IRD for core CS06.
Markers represent
samples analysed
40
35
%IRD
30
25
20
15
10
5
0
0
50
100
150
200
250
Depth (cm)
% IRD for core CS07. Markers represent
samples analysed.
241
APPENDICES
APPENDIX E - SEDIMENT GRAIN COMPOSITION

In core CS06 the carbonate sand-size material is predominantly composed of
planktonic and benthic foraminifera with some invertebrate skeletons
(bryozoans, ostracods, spicules, etc.).

In core CS07 the carbonate sand-size material is predominantly made up of
coral fragments, shells, calcite cement rubble and foraminifera.
The pelagic foraminifera are predominantly globigerinids with some samples
containing many Orbulina universa and the deeper-water species Globorotalia
truncatulinoides is also common. Preliminary investigation shows the benthic
compositions are dominated by variations in assemblages of Discanomalina
coronata,
Cibicides
sp.,
Pyrgo
sp.,
Uvigerina
sp.,
Textularia
sp,
and
Quinqueloculina sp.
Other (macro and micro) fossils present in both cores include brachiopod shells,
gastropods, echinoid spines, ostracods, sponge spicules and serpulids. The noncarbonate component of the sand in the cores is dominantly quartz with lithic
fragments and glauconite. Glauconite present is presumed to be authigenically
formed as casts within the chambers of foraminifera tests. The lithics comprise of a
mixture of:

Sedimentary (red sandstone and dark grey limestone)

Metamorphic (schist) and

Igneous (granite and diorite) rocks.
The lithics in the cores are interpreted to have been derived from west Ireland by icerafting during the Pleistocene.
CS07
The nature of the material in CS07 posed problems when analysing the grain
composition of the sediment. Many samples contained irregular, but roughly
spherical, white grains that appeared to be composed of carbonate clay. Many of
these 'spheres' contained skeletal grains such as bryozoan fragments and shell debris
as well as planktonic foraminifera, and they were classed as ‘calcium carbonate
concretions’. Usually they appeared alongside the other grains but in some of the
samples all the grains were obscured by the covering of the bright white cement. In
242
APPENDICES
addition, aggregations of, what appeared to be, planktonic foraminifera occurred,
loosely cemented together. These were classed as ‘planktonic foraminifera
aggregates’. The clumping of planktonic foraminiferans means that calculating in
situ versus transported would not be representative for these samples because while
the grains themselves are locally produced, their composition is made up of many
essentially allochthonous planktonic foraminiferans.
It was hoped that quantifying the different categories of sedimentary components mainly skeletal debris – would allow the contribution of the various faunal groups to
mound build-up to be assessed. However, most samples contained a large component
of biogenic hash and during the counting procedure it became apparent that many of
the fragments were unidentifiable due to their broken and abraded nature. For
example, it was often not possible to definitively identify coral fragments because
they were most often present as highly fragmented coral ‘hash’. The same problem
arose with broken bivalve and brachiopod shell fragments. These fragments were
classed as UID [unidentifiable biogenic elements]. The obvious uncertainty with
assigning elements into categories produced large numbers of UID. Particles classed
into categories for samples where they are better preserved entails a loss of the signal
of their contribution to the samples containing more broken up admixes, introducing
false results into the mound contribution elements.
For the reasons described above the grain composition analysis results cannot be
reliably used to decipher trends. Nonetheless some of the results are displayed as
percentage of fraction in order to compare against each other and describe units.
243
APPENDICES
Grain composition analysis results for
core CS07, showing the material >
250 µm < 2mm to comprise of coral
fragments and associated biogenic
debris from fauna living on the
mound, as well as derived pelagic
material of planktonic foraminifera,
fish vertebrae and IRD. PF –
planktonic foraminifera, BF-benthic
foraminifera, CC –carbonate
concretions, Agg – Aggregates of PF,
MB-mudball, Bry –bryozoan,
Ostra(cod), Serp – serpulid worm,
Espine – echinoid spine, Gastro(pod),
UID- unidentified biogenic debris
including otoliths, bivalves,
brachiopods, pteropods, crustacean
claws.
244
Percentage grain composition analysis
results for core CS07, of the material
> 250 µm < 2mm. PF – planktonic
foraminifera, BF-benthic foraminifera,
CCC –carbonate concretions, Agg –
Aggregates of PF, Bry –bryozoan,
Ostra(cod), Serp – serpulid worm,
Espine – echinoid spine, Gastro(pod),
UID- unidentified biogenic debris
including otoliths, bivalves,
brachiopods, pteropods, crustacean
claws.
APPENDICES
245
APPENDICES
Apart from the sample from the high-energy coarse event at 575 cmdc, the grain
composition analysis showed IRD and planktonic foraminifera content to anticorrelate. This provides confidence in the IRD counting method used. It also
highlights the unit of 575 cmdc as rather anomalous.
The % grains of planktonic foraminifera plotted against the IRD of the CS07 grain
composition analysis results showed them to anti-correlate, with an exception at the highenergy coarse event at 575 cmdc.
From 105 cmdc the sediment was noted as ‘more lithified’ and from 110 cmdc
aggregates of foraminifera [Agg] and calcium carbonate concretions (CCC) become
conspicuous. For example, at 110 cmdc 22% of the grains were found to be
aggregates of planktonic foraminifera cemented together. The AGG and CCC are
probably related to similar processes of digenesis and the difference in their nature
appears to be related to the depositional environment. The aggregates occur in sandy
deposits (110-190 cmdc) whereas the CCs occur in clay-rich deposits (350-390 and
530 cmdc). It was also noted that the samples with many CCs and/or Aggs occurred
246
APPENDICES
where there were also less coral fragments. It is possible that the coral has dissolved
to form a micritic cement in these sediments.
Samples with many of the grains covered in white cement (Carbonate Concretions -CCC) or
made up of aggregates of planktonic foraminifera (Agg) showed an anti-correlation to coral
content in core CS07.
247
APPENDICES
The % ostracods of the grain composition analysis results of CS07 showed them to peak with
the mottled mud facies 450-555 cmdc.
248
APPENDICES
APPENDIX F – XRF ANALYSIS
CS07
Results of the two XRF- scanned cores (Section 1 and Section 3) are presented in %
total count and are shown immediately beneath one another, in order to compare
their signals.
XRF results for core CS07 Section 1 and Section 3
249
APPENDICES
APPENDIX G – DATING
CS06
Core, Section Core Depth
CS06, S1
25
CS06, S1
55
CS06, S2
100
CS06, S2
140
CS06, S2
165
CS06, S2
165
CS06, S3
210
CS06, S3
290
Type
14C age Error
Calendric Age calBC
(yr)
(yr)
Foram
25404
76
Brachiopod >45000
Brachiopod >45000
Brachiopod >45000
Foram
43042 380
Brachiopod >45000
Brachiopod >45000
Foram
>45000
Radiocarbon dates for CS06.
250
28275 ± 272
44736 ± 1560
APPENDICES
Example of the pedicle valve of the sampled brachiopods sent for dating.
251
APPENDICES
Example of the brachial valve of the sampled brachiopods sent for dating.
252
APPENDICES
CS07
Core
Core Depth
(cm)
Age
(kyr)
Comments
CS07 S2
140
>45000
CS07 S3
250
na
Failed during sample preparation
CS07 S3
290
>45000
CS07 S6
595
>45000
14C age determinations from foraminifera for core CS07
Core
Core
Depth
(cm)
CS07 Section 1
6.5
CS07 Section 1
CS07 Section 1
CS07 Section 4
18.5
25
311
Age
(uncorr.
ka)
Age corr.
(ka)
8.6
8.5
10.4
11.0
±0.2
±0.2
±0.3
Out of range
Comments
±0.2
* U-isotopic
composition too
high
for a modern sample
8.5 ±1.0
11.0 ±0.3
317
* U-isotopic
composition too low
compared to Theory
* U-isotopic
composition too low
compared to Theory
±40
CS07 Section 4
311
Out of range 387 ±89
CS07 Section 5
423.5
Out of range
CS07 Section 5
440
Out of range
CS07 Section 6
556
11.6 ±0.3 8.2 ±1.8
U/Th age determinations from coral for core CS07.
The U-isotopic composition being too high/low compared to Theory is an indication
of some alteration process leading to the gain/loss of
234
shows how the data should evolve through time with
230
U. The first plot (below)
Th/238U being 0 once the
corals growth and isotopic ratios evolve towards unity (i.e. as corals become old).
The black lines correspond to the Uranium isotopic composition of modern seawater
that evolves through time (dashed lines are ±0‰ change in seawater U-isotopic
composition). Perfect data of various ages would always fall in between the two
dashed lines (which is not the case for the data).
253
APPENDICES
200
234U (‰)
150
100
50
0
0
0.5
1
1.5
(230Th/238U)
The anomalous coral date from 556 cmdc is not perceived to be due to an error
during sampling or labelling because the photographed coral fragment (prior to
shipping for dating) can be identifed in the core picture for that interval.
Coral fragment highlighted by red box.
254
APPENDICES
APPENDIX H - DROPSTONE ANALYSIS
CS06- Leucocratic granite
Dropstone
recovered
from core
CS06 Section
3, 235 239 cmdc, as
pictured insitu in the
core. Scale of
1cm interval
to the right
In hand sample, the dropstone was subrounded and measured 6 cm by 4 cm. It was
medium grained with a mottled yellowish pink colouration. The mottles are due to
amalgamations of black crystals in the yellow-pink matrix. The crystals have a fabric
of crystal alignment.
Thin-section analysis revealed over 75% felsic minerals (quartz, alkali feldspar,
plagioclase and also an appreciable amount of microcline) along with small amount
of muscovite, biotite and some oxides. There is high quartz content (~30%) with
stringers of quartz appearing as streams of anhedral grains and some strained quartz
showing undulose extinction. Twinning was noted in plagioclase feldspars but it was
not common.
The dropstone shows heavy alteration. The dark phenocrysts producing the
porphyritic texture are amalgamations of highly altered green pleochroic biotites
(some altered to chlorite) with epidote and sericite. Sericite is a common alteration
mineral in areas that have been subjected to hydrothermal alteration.
255
APPENDICES
General thin section (XPL) view of CS06 dropstone. Microcline with cross-hatching evident,
along with anhedral interstitial quartz
Smaller anhedral quartz alongside larger quartz grains.
256
APPENDICES
Plagioclase feldspar with polysynthetic twinning.
257
APPENDICES
Dark phenocrysts producing the porphyritic texture are amalgamations of highly altered
green pleochroic biotites with epidote and sericite.
In summary, the dropstone is a granite that appears to have undergone hydrothermal
alteration. It is felsic rock as it has >75% felsic minerals and leucocratic as <30%
minerals are mafic. Therefore its preliminary classification is that of leucocratic
granite of indeterminate provenance.
CS07 - Granodiorite Gneiss
In hand specimen it is a mottled dark grey phaneritic igneous rock which measures
6.5cm by 6cm, with sub-rounded edges. One face has a very smooth surface due to
the cutting of this section by the saw and this shows a gneissic fabric. Mediumgrained interlocking crystals are evident (white and beige-yellow feldspars,
clear/vitreous dull grey quartz) along with abundant black biotite and hornblende
amphiboles to give a dark appearance.
258
APPENDICES
Clockwise from left: Granodiorite gneiss as seen in-situ 66-72 cmdc in section 1 of core
CS07; side profile of dropstone; aerial view of smooth face showing the gneissic fabric;
smooth side facing down showing the sub-rounded nature of the rock.
In thin section feldspars (plagioclase and orthoclase) and quartz minerals make up
more than 60% of the rock. Over 65% of the feldspar is plagioclase and
approximately 40% is polysnthetically twinned. There is greater than 20 % quartz.
Approximately 35% of the rock is made up of amphiboles. Other minerals include an
appreciable amount of biotite and a very small amount of muscovite. A
hypidiomorphic-granular texture is evident.
259
APPENDICES
Hypidiomorphic-granular texture in XPL. Euhedral and subhedral crystals of green and
yellow amphiboles. Subhedral feldspar crystals with plagioclase displaying albite twinning.
Clear and grey anhedral interstitial quartz. Smaller euhedral biotite crystals. Note the high
percentage of plagioclase.
Quartz with diagnostic undulose extinction forms irregular shapes between the
feldspars and provides evidence the rock has been strained. Biotite forms smaller
euhedral crystals and is commonly found as composite crystals with the euhedralsubhedral hornblende. The amphiboles are bluish green under PPL, but crosspolarization shows the two cuts have different XPL colours. The euhedral-subhedral
amphibole crystals cut along the c-axis, show blueish-green to yellow-green
pleochroism and are greenish-grey in XPL.
The basal cut amphiboles show blueish- green to straw yellow pleochroism and a
strong yellow interference colour under XPL. There is also a small amount of
muscovite with bright interference colours.
260
APPENDICES
Left: XPL; Right PPL. Anhedral quartz with undulose extinction. Euhedral & subhedral
amphiboles (yellow and dark green in XPL; pale green and dark green in PPL). The two
cleavage directions intersecting at 120 ° are clearly visible far right. Euhedral to subhedral
biotite (yellowish brown – rusty red in XPL; brown in PPL). Large grey area in XPL is void
due to thin-sectioning.
Complex plagioclase twinning. XPL view.
Biotite (brown-light
brown) is common in
the sample; usually
in association with
hornblende (green)
forming composite
crystals. PPL view.
261
APPENDICES
Muscovite in between amphiboles; with bright interference colours in XPL (left) and
colourless in PPL (right). Amphibole enclosed by interstitial quartz in top left.
In summary, the dropstone is mainly made up of feldspar, quartz and amphiboles
with appreciable amount of biotite. The dominance of the plagioclase feldspar
indicates a granodiorite and the fabric indicates a granodiorite gneiss, with the fabric
(banding) a result of deformation. Possible provenances for this lithology include
Connemara and Mayo (Belmullet) gneisses.
262