MARINE REMOTE SENSING AND SEABED CHARACTERISATION TECHNIQUES
FOR INVESTIGATING SUBMERGED LANDSCAPES OFF THE NORTHWEST
COAST OF QATAR
by
LUCIE DINGWALL
A thesis submitted to the University of Birmingham for the degree of MASTER OF
PHILOSOPHY
Department of Classics, Ancient History and Archaeology
School of History and Cultures
College of Arts and Law
University of Birmingham
June 2014
University of Birmingham Research Archive
e-theses repository
This unpublished thesis/dissertation is copyright of the author and/or third
parties. The intellectual property rights of the author or third parties in respect
of this work are as defined by The Copyright Designs and Patents Act 1988 or
as modified by any successor legislation.
Any use made of information contained in this thesis/dissertation must be in
accordance with that legislation and must be properly acknowledged. Further
distribution or reproduction in any format is prohibited without the permission
of the copyright holder.
ABSTRACT
The Arabian Gulf is a relatively recent sea that formed as a result of post-glacial sea
level rise in the Late Pleistocene and Early Holocene. Prior to this, the former Gulf
basin was an open landscape, with a substantial river flowing through it. There is
considerable potential within this former landscape for the preservation of drowned
archaeological sites and palaeoenvironmental remains from the Early Holocene, and
for the preservation of remains of shipwrecks from the Mid-Holocene onwards.
Despite the potential, there has been very little research into this submerged
landscape, largely due to the difficulty and expense involved. In order to begin to
address the gap in knowledge, this research developed and tested new
methodologies for the exploration of the submerged landscape within a defined Study
Area off the northwest coast of Qatar. The methodology utilised marine remote
sensing data, including sidescan sonar and LiDAR bathymetry, and drew on
techniques used in terrestrial historic landscape characterisation and acoustic
seabed classification, in order to zone the seabed in the Study Area and identify
broad zones of archaeological and palaeoenvironmental potential and survival. The
seabed characterisation has provided a framework within which to begin more
detailed investigations of the submerged landscape, by defining areas of potential to
target, and specifying appropriate techniques to use within those areas, in order to
maximise the chance of successful exploration of the submerged landscape.
Keywords
Seabed Characterisation; Marine Geophysics; Acoustic Classification; Qatar;
Submerged Landscapes; Arabian Gulf
ACKNOWLEDGEMENTS
I would like to thank my supervisors, Professor Vince Gaffney, and especially Dr
Richard Cuttler, who provided unstinting support in so many ways, and without whom
this research would not have been possible. I would also like to thank H.E Shaykha
al-Mayassa bint Hamad bin Khalifa Āl-Thāni (Chairperson of the Board of Trustees,
Qatar Museums Authority), H.E. Shaykh Hasan bin Mohammed Āl-Thāni (ViceChairman
of the Board of Trustees, Qatar Museums Authority) and Faisal Al-Naimi
(Director of Antiquities, Qatar Museums Authority), for their support for this research.
Many other people provided support and assistance in many different ways, and I
would like to extend grateful thanks to the following people:
The University of Birmingham/QNHER marine survey team, and in particular Kate
Bain and Robyn Pelling, for undertaking the ground-truthing programme (grab
sampling and diving) and high-resolution survey, and generally providing vast
amounts of technical, logistical and other support; Ismail Mahmoud Al-Shaikh, Abdul
Rahman Al-Obaidly and Abdel Rahman Sorour from the Environmental Studies
Centre at Qatar University for kindly carrying out the granulometry analysis on the
grab samples and providing a fantastic report; Eoghan Kieran (Geomara) for advice
and assistance with the grab sampling and granulometry programme, and for
invaluable advice on seabed classification software; Tony Tipple and Chris Elliott
from Quester Tangent Software Ltd for providing free access to their software for a
trial period, and for providing expert advice on the use of their software; Dr Richard
Bates (University of St Andrews) and Brandon Mason (Hampshire and Wight Trust
for Maritime Archaeology) for providing advice and assistance with initial processing
of the sidescan sonar data; Dr Emma Tetlow (University of Birmingham/QNHER
Project) for providing expert palaeoenvironmental advice and moral support; Tobias
Tonner (Coaptics Ltd) and Liam Delaney (University of Birmingham/QNHER Project)
for assistance with coordinate system conversions; Dr Simon Fitch (University of
Birmingham) for initial assistance with the sidescan sonar data processing; Access to
the sidescan sonar data and geotechnical reports was kindly provided by the Qatar
— Bahrain Causeway Management and access to the Bathymetric LiDAR data for
research purposes was kindly provided by the Hydrographic Office. I would also like
to thank Chris Dodds and Isaac Dingwall, who have supported me unwaveringly
throughout my research.
Contents
CHAPTER 1: CONTEXT......................................................................................................................... 1
1.1 Introduction.................................................................................................................................... 1
1.2 Overview of The Study Area ......................................................................................................... 6
1.2.1 Research Themes .................................................................................................................. 6
1.2.2 Geology and Geomorphology .............................................................................................. 10
1.2.3 Sedimentation and Marine Taphonomic Processes............................................................. 14
1.2.4 Climate ................................................................................................................................. 18
1.2.5 Tides and Salinity ................................................................................................................. 18
1.2.6 Sea Levels............................................................................................................................ 20
1.2.7 Vegetation History and Palaeo-climate ................................................................................ 24
CHAPTER 2: REVIEW OF PREVIOUS RESEARCH AND ARCHAEOLOGICAL POTENTIAL ........ 27
2.1 An overview of Archaeology in the Arabian Peninsula ............................................................... 27
2.1.1 The Palaeolithic.................................................................................................................... 27
2.1.2 The Neolithic......................................................................................................................... 30
2.1.3 Maritime History.................................................................................................................... 35
2.2 Submerged Landscapes Research............................................................................................. 40
2.2.1 Research Outside the Arabian Gulf ..................................................................................... 40
2.2.2 Research Within the Arabian Gulf........................................................................................ 43
CHAPTER 3: OVERVIEW OF METHODOLOGY................................................................................. 46
3.1 Processes of Seabed Characterisation....................................................................................... 46
3.2 Primary Characterisation............................................................................................................. 49
3.2.1 Sediment Texture Classification........................................................................................... 49
3.2.2 Topographic Mapping........................................................................................................... 50
3.2.3 Ground-Truthing ................................................................................................................... 52
3.3. Secondary Characterisation....................................................................................................... 53
3.3.1 Identification of Geophysical Anomalies .............................................................................. 53
3.3.2 Clarification of Geophysical Signatures ............................................................................... 55
3.4 Defining Character Areas and Assigning Potential ..................................................................... 55
CHAPTER 4: PRIMARY CHARACTERISATION................................................................................. 57
4.1 Sediment Texture Classification.................................................................................................. 57
4.1.1 Methodology......................................................................................................................... 57
4.1.1.1 Overview of Sidescan Sonar .......................................................................................... 57
4.1.1.2 Acoustic Techniques for Sediment Texture Classification ............................................. 60
4.1.1.3 Trialling the Classification Methodology......................................................................... 62
4.1.1.4 Classification using Swathview ...................................................................................... 65
4.1.2 Results.................................................................................................................................. 71
4.1.2.1 Area 4 ............................................................................................................................. 74
4.1.2.2 Area 2/3 .......................................................................................................................... 78
4.1.2.3 Area 1 ............................................................................................................................. 82
4.1.2.4 Generating Initial Landscape Units ................................................................................ 87
4.1.3 Discussion ............................................................................................................................ 90
4.2 Topographic Mapping.................................................................................................................. 97
4.2.1 Methodology......................................................................................................................... 97
4.2.1.1 Overview of LiDAR Bathymetry...................................................................................... 97
4.2.1.2 Generating the Surface Model ....................................................................................... 99
4.2.1.3 Mapping Features......................................................................................................... 103
4.2.2 Results................................................................................................................................ 104
4.2.2.1 Topography Within the Study Area .............................................................................. 105
4.2.2.2 Topography Outside the Study Area............................................................................ 120
4.2.2.3 Topography and Sediment Classification..................................................................... 123
4.2.3 Discussion .......................................................................................................................... 127
4.3 Ground Truthing ........................................................................................................................ 134
4.3.1 Methodology....................................................................................................................... 134
4.3.1.1 Direct Sediment Sampling............................................................................................ 134
4.3.1.2 Video and Photography................................................................................................ 140
4.3.2 Results................................................................................................................................ 141
4.3.2.1 Direct Sediment Sampling............................................................................................ 141
4.3.2.2 Video Transects............................................................................................................ 148
4.3.3 Discussion .......................................................................................................................... 151
CHAPTER 5: SECONDARY CHARACTERISATION ........................................................................ 156
5.1. Identification of Geophysical Anomalies .................................................................................. 156
5.1.1 Methodology....................................................................................................................... 156
5.1.2 Results................................................................................................................................ 160
5.1.2.1 Geophysical Anomalies................................................................................................ 160
5.1.2.2 Geophysical Anomalies and Topography .................................................................... 175
5.1.3 Discussion .......................................................................................................................... 180
5.2 Clarification of Geophysical Signatures .................................................................................... 184
5.2.1 Methodology....................................................................................................................... 184
5.2.1.1 Diver Inspections.......................................................................................................... 184
5.2.1.2 High-Resolution Geophysical Survey........................................................................... 185
5.2.2 Results................................................................................................................................ 185
5.2.2.1 Diver Inspections.......................................................................................................... 185
5.2.2.2 High-Resolution Geophysics ........................................................................................ 189
5.2.3 Discussion .......................................................................................................................... 191
CHAPTER 6: DEFINING CHARACTER AREAS AND ASSIGNING POTENTIAL ........................... 193
6.1 Refining the Initial Landscape Units into Character Areas........................................................ 193
6.2 Assigning Potential.................................................................................................................... 202
CHAPTER 7: DISCUSSION ............................................................................................................... 210
7.1 An Evaluation of the effectiveness of the methodologies, and suggestions for further work.... 210
7.2 An evaluation of the archaeological and palaeoenvironmental potential of the submerged
landscape within the Study Area..................................................................................................... 225
CHAPTER 8: CONCLUSION: IMPLICATIONS FOR RESEARCH AND MANAGEMENT OF THE
MARINE HERITAGE RESOURCE ..................................................................................................... 239
APPENDIX 1: PROCESS AND PARAMETERS USED FOR THE ACOUSTIC CLASSIFICATION OF
SIDESCAN SONAR DATA................................................................................................................. 243
APPENDIX 2: PROCESS AND PARAMETERS USED FOR PROCESSING THE LIDAR
BATHYMETRY DATA ........................................................................................................................ 247
APPENDIX 3: GRAB SAMPLE GRAIN SIZE ANALYSIS REPORT ................................................. 249
APPENDIX 4: PARAMETERS USED FOR PROCESSING AND MOSAICING SIDESCAN SONAR
DATA................................................................................................................................................... 309
APPENDIX 5: LIST OF ALL ANOMALIES LOGGED FROM THE SIDESCAN SONAR DATA....... 310
APPENDIX 6: DERIVED DATA INCLUDED ON ACCOMPANYING CD .......................................... 352
Figures
Figure 1 Location of the Study Area (illustration by Nigel Dodds). ......................................................... 4
Figure 2 Bathymetry (height data) between Qatar and Bahrain (from Al-Naimi et al., 2012). .............. 13
Figure 3 Location of the Bahrain Anticline (from Cuttler, 2014). ........................................................... 14
Figure 4 Shoreline Reconstructions at 14,000 BP after Lambeck (1996) (from Cuttler et al., 2011b). 22
Figure 5 Shoreline reconstructions at 8,200 BP after Lambeck (1996) (from Cuttler and Al-Naimi,
2013). .................................................................................................................................................... 23
Figure 6 Distribution of ‘Ubaid and ‘Ubaid -Related Sites around the Arabian Gulf (from Cuttler, 2013).
............................................................................................................................................................... 33
Figure 7 Sidescan sonar operation (from Al-Naimi et al., 2012). .......................................................... 58
Figure 8 Coverage of sidescan sonar survey lines in the Study Area. ................................................. 59
Figure 11 Sub-Images (rectangles) generated by Swathview for the purpose of generating image
statistics................................................................................................................................................. 66
Figure 13 Results of the Swathview preliminary seabed classification using catalogues based on the
entire dataset, and using catalogues based on trial areas.................................................................... 68
Figure 14 Track plot from Swathview showing classified data points (left) and interpolation of data
points using Clams (right)...................................................................................................................... 70
Figure 15 Classified data (left) and mosaic of sidescan sonar data (right) in Area 4. .......................... 74
Figure 16 Edge of brighter reflectivity demarcating a change in texture in Area 4. .............................. 75
Figure 17 Trawler scarring in Area 4..................................................................................................... 76
Figure 18 Pattern of mixed classes south of the reef in Area 4 corresponding with visual patterns in the
mosaic data. .......................................................................................................................................... 77
Figure 19 Sand ridges visible in the mosaic data in Area 4 and differentiated as Class 3 in the
classified data........................................................................................................................................ 77
Figure 20 Classified data and mosaic of sidescan sonar data in Area 2/3. .......................................... 78
Figure 21 Mosaic of sidescan sonar data around the reef in Area 3. ................................................... 79
Figure 22 Sand ridges visible in the mosaic data in Area 2 and differentiated as Class 3 in the
classified data........................................................................................................................................ 80
Figure 23 Sand ridges visible in the mosaic data in Area 3 and differentiated as Class 6 in the
classified data........................................................................................................................................ 81
Figure 24 Classified data and mosaic of sidescan sonar data in Area 1. ............................................. 82
Figure 25 Brighter reflective area immediately north of the reef in Area1............................................. 83
Figure 26 Large band of lower reflectivity in Area 1, differentiated in the classified data mainly as
Classes 9 and 4..................................................................................................................................... 84
Figure 27 Sand ripples in the north of Area 1, visible in the mosaic data. ............................................ 85
Figure 28 Sand banks in the southeast of Area 1, visible in the mosaic data and differentiated as
Classes 3 (red) and 8 (lime) in the classified data. ............................................................................... 86
Figure 29 Initial landscape units based on classification of acoustic backscatter. ............................... 89
Figure 30 Extent of available LiDAR bathymetry. ................................................................................. 98
Figure 31 3D visualisation of the seabed in the Study Area (created in Erdas Imagine, based on the
surface model generated from the bathymetry). ................................................................................. 102
Figure 32 Surface model of the Study Area created from LiDAR bathymetry. .................................. 106
Figure 33 Features mapped from the surface model.......................................................................... 107
Figure 34 Putative former shorelines in the Study Area. .................................................................... 109
Figure 35 Hillshaded surface model with putative former shorelines.................................................. 109
Figure 36 Hillshaded surface model with putative east-west former shoreline................................... 110
Figure 37 Putative east-west former shoreline with sidescan sonar data overlain. ............................ 110
Figure 38 Putative former shorelines with possible associated mapped features. ............................. 111
Figure 39 Palaeochannel possibly relating to the former course of Wadi Debayan. .......................... 112
Figure 40 Cluster of palaeochannels. ................................................................................................. 113
Figure 41 Fishtraps visible in the surface model (top), mapped during the QNHER cultural mapping
project (middle) and an example photographed by the QNHER project (bottom). ............................. 114
Figure 42 Channels and depressions in the north of the Study Area. ................................................ 115
Figure 43 Channels and depressions in the south of the Study Area................................................. 115
Figure 44 Megaripples in the north of Area 1...................................................................................... 116
Figure 45 Possible sand banks between the reef and the bay of Al-Zubārah. ................................... 117
Figure 46 Possible solution hollow visible in the surface model but not in the sidescan sonar data.. 118
Figure 47 Undulations in the seabed to the north of the putative east-west shoreline in Area 1........ 119
Figure 48 The Reef in the west of the Study Area. ............................................................................. 120
Figure 49 Potential former shorelines to the south of the Study Area. ............................................... 121
Figure 50 Seabed surface in the Gulf of Salwa................................................................................... 122
Figure 51 Surface model with the classified data in Area 4. ............................................................... 124
Figure 52 Surface model with the classified data in Area 2, West of the Ras Ushayriq Peninsula. ... 125
Figure 53 Surface model with the classified data in Area 3, north of the reef. ................................... 126
Figure 54 Surface model with the classified data in the deep channel in Area 1. .............................. 127
Figure 55 Postulated extent of dry land at 7,000 BP (based on Stanford et al,. 2011)....................... 129
Figure 56 Postulated extent of dry land at 8,000 BP (based on Jameson and Strohmenger, 2012). 130
Figure 57 Postulated extent of dry land at 7,000 BP (based on Jameson and Strohmenger, 2012). 131
Figure 58 Grab samples overlain on Initial Landscape Units.............................................................. 135
Figure 59 Exploratory boreholes drilled in 2008, overlain on Initial Landscape Units. ....................... 137
Figure 60 Vibrocores taken in north of the Study Area in 2009, overlain on Initial Landscape Units. 138
Figure 61 Vibrocores taken in south of the Study Area in 2009, overlain on Initial Landscape Units. 139
Figure 62 Boreholes drilled over the reef in 2009, overlain on Initial Landscape Units. ..................... 139
Figure 63 Grab sample locations overlain on the surface model........................................................ 141
Figure 64 Grab sample locations overlain on the classified data........................................................ 142
Figure 65 Vibrocores and boreholes overlain on the surface model (grab samples are visible in the
background)......................................................................................................................................... 144
Figure 66 Vibrocores and boreholes overlain on the classified data (grab samples are visible in the
background)......................................................................................................................................... 145
Figure 67 Video transects overlain on surface model......................................................................... 148
Figure 68 Video transects overlain on the classified data................................................................... 148
Figure 69 Graph displaying sediment types for each class as derived from ground truthing data
(expressed as a percentage of the total number of samples/observations recorded for each class). 153
Figure 70 Distribution of all recorded anomalies within the Study Area.............................................. 161
Figure 71 Anomalies interpreted as potential debris, and proven to be debris from ground-truthing. 162
Figure 72 Examples of high-confidence anomalies in Area 1: Modern debris relating to artificial reefs
(cars, tyres etc).Clockwise starting from the top left corner: QBC_Q10126, QBC_Q10270,
QBC_Q10334, QBC_Q10902. (Each image depicts an area measuring 100m x 100m.).................. 163
Figure 73 The same anomaly (QBC_Q10815) visible in different survey lines: Modern debris relating
to artificial reefs (each image depicts an area measuring 100m x 100m)........................................... 164
Figure 745 Distribution of anomalies that are possibly depressions................................................... 166
Figure 75 A selection of anomalies that were logged as possible depressions. From left to right:
QBC_Q20071, QBC_Q40064, QBC_Q40119 (each image depicts an area measuring 100m x 100m).
............................................................................................................................................................. 166
Figure 76 Anomalies logged as possible debris. From left to right: QBC_Q40089, QBC_Q10001 (each
image depicts an area measuring 100m x 100m)............................................................................... 171
Figure 77 Anomalies logged as areas of possible sediment accumulation. From left to right:
QBC_Q40055, QBC_Q20060 (each image depicts an area measuring 100m x 100m). ................... 171
Figure 78 Anomalies logged as natural features of potential topographic significance. From left to right:
QBC_Q40026, QBC_Q20046, QBC_Q10195 (each image depicts an area measuring 100m x 100m).
............................................................................................................................................................. 172
Figure 79 Anomalies logged as potentially partially-buried objects or features. From left to right:
QBC_Q10126, QBC_Q10202 (each image depicts an area measuring 100m x 100m). ................... 172
Figure 80 A potential seabed crater: QBC_Q20094 (Image depicts an area measuring 100m x 100m).
............................................................................................................................................................. 173
Figure 81 Long linear anomaly: QBC_Q10729/Bham0028 (Image depicts an area measuring 100m x
100m). ................................................................................................................................................. 173
Figure 82 Possible debris cluster in Area 1 (the green grid lines represent 500m x 500m squares). 174
Figure 83 Long linear anomaly (QBC_Q10729/BHAM0028) visible in the sidescan sonar data, and the
surface model of the seabed from the same location (Both images are at same scale and orientation).
............................................................................................................................................................. 176
Figure 84 Bright reflective anomalies in Area 2, shown in the surface model to be natural outcrops. 177
Figure 85 Seabed crater (QBC_Q20094) visible in the sidescan sonar data (image depicts an area
measuring 100m x 100m), and the surface model of the seabed from the same location, showing
anomalies QBC_Q20094 and QBC_Q20096...................................................................................... 177
Figure 86 Bright reflective anomalies in Area 3, shown in the surface model to be natural coral
outcrops............................................................................................................................................... 178
Figure 87 Bright reflective anomalies in the southwest of Area 4, shown in the surface model to be
natural outcrops................................................................................................................................... 179
Figure 88 Anomaly QBC_Q10051 (modern debris): sidescan sonar, LiDAR points and surface model.
............................................................................................................................................................. 181
Figure 89 Anomaly QBC_Q10128 (car reef) visible in the sidescan sonar but not visible in the surface
model................................................................................................................................................... 182
Figure 90 Anomaly QBC_Q10376 (topographic location of potential interest for human settlement) in
the sidescan sonar data (image depicts an area measuring 100m x 100m), and in its landscape
context in the surface model. .............................................................................................................. 183
Figure 91 Anomaly QBC_Q10634 in Area 1: Sediment Basin (Photo by QNHER marine team, 2013).
............................................................................................................................................................. 187
Figure 92 Anomaly QBC_Q40118 in Area 4: Linear Ridges (Photo by QNHER marine team, 2013).187
Figure 93 Anomaly G3/07.05.13/L8/TG1:Car reef in Area 1 (Photo by QNHER marine team, 2013).188
Figure 94 Long linear anomaly QBC_Q10729/BHAM0028 (Photo by Hampshire and Wight Trust for
Maritime Archaeology, 2011). ............................................................................................................. 189
Figure 95 Long linear anomaly (QBC_Q10729/BHAM0028) surveyed in high resolution (each image
depicts an area measuring 100m x 100m).......................................................................................... 190
Figure 96 Anomaly QBC_Q10815 (car reef) surveyed in low resolution (left) and high resolution (right)
(each image depicts an area measuring 100m x 100m)..................................................................... 191
Figure 97 Cross-correlation of different input datasets with the Initial Landscape Units. ................... 194
Figure 98 Final Character Areas (labelled with Character Area numbers). ........................................ 195
Figure 99 Character Areas (labelled with Character Area numbers) showing overall potential. ........ 206
Figure 100 Distribution map of major ʿUbaid-related sites in Qatar.................................................... 227
Figure 101 Misfer Cave, Qatar............................................................................................................ 231
Figure 102 Evidence for archaeological deposits preserved beneath marine deposits at Wadi Debayan
(Image by Richard Cuttler/Emma Tetlow). .......................................................................................... 236
TABLES
Table 1: List of acoustic classes generated by Swathview, and their preliminary interpretation 72
Table 2: Initial landscape units - polygon statistics 90
Table 3: Seabed descriptions summarised from the video survey information 149
Table 4: List of terms used for initial categorisation of anomalies 159
Table 5: List of anomalies selected for more detailed examination 168
Table 6: Summary of results of diver inspections on selected anomalies 186
Table 7: Summary of results of diver inspections on anomalies near QBC_Q10729 188
Table 8: Summary description of Final Character Areas 197
1
CHAPTER 1: CONTEXT
1.1 Introduction
The Arabian Gulf1
and Peninsula, bordered by Africa, the Levant and the Iranian
plateau, are located in a highly significant location in archaeological terms, straddling
the African and Eurasian landmasses, and strategically located on an important
ancient trade route between the civilisations of Mesopotamia and the Indus Valley
(Parker and Goudie, 2008). The sea that currently occupies the Arabian Gulf Basin is
relatively recent, having formed as a result of post-glacial sea level rise during the
Late Pleistocene and Early Holocene, reaching its highest level just before 6,000 BP
(Lambeck, 1996; Bird et al., 2010). Prior to this, the former Gulf basin would have
been an open landscape, with a substantial river flowing through it, formed by the
confluence of the Tigris and Euphrates Rivers to the north, and containing large,
possibly freshwater, lakes (Lambeck, 1996). This resource-rich basin would have
been a very attractive habitat for human communities in the region before sea-levels
rose, possibly more attractive than the landscape that is now the Arabian Peninsula.
The coastal region of the Arabian Peninsula is known to have been a focus for newlyestablished
human settlements in the 7th Millennium BP, just as sea levels were
reaching their peak (Parker and Goudie, 2008). Evidence also exists for a very early
maritime trade network emerging at that time in the Arabian Gulf (Carter and
Crawford, 2010). Research into the archaeology of the submerged continental shelf
elsewhere in the world (Faught and Donoghue, 1997; Fitch et al., 2005; Bailey et al.,
1
The Arabian Gulf is also commonly referred to as the Persian Gulf, but as this research focuses on
the Arabian side of the Gulf, it is referred to as the Arabian Gulf throughout the thesis.
2
2007; Gaffney et al., 2007; Westley et al., 2011a) suggests that the former landscape
of the Arabian Gulf basin has considerable potential for the survival of drowned
archaeological sites and palaeoenvironmental remains from periods pre-dating the
marine transgression, and for the presence of shipwrecks dating from the 7th
Millennium BP onwards.
Despite this potential, there has been very little research into the submerged
landscape of the former Gulf basin. This is due in large part to the difficulty and
expense of such exploration and research, and as a result it has often been excluded
from archaeological analyses and ignored in the literature. However, the full picture
of Prehistoric settlement in the region cannot be adequately understood whilst this
highly significant expanse of former landscape remains unexplored. Random survey
of such a large area of submerged landscape is unlikely to be either cost-effective, or
yield useful results. The need for a research framework within which to begin to
explore the marine environment in a meaningful and effective way is therefore
required. The continuing development of techniques in marine geophysics, allowing
cheaper and more rapid examination of large areas of the seabed than would be
possible using diver survey alone, has opened up new possibilities for large-scale
exploration of the submerged landscape. Also, the availability of large, remotelysensed
datasets that have been gathered by commercial companies in advance of oil
exploration and offshore infrastructure projects makes the costs of undertaking
archaeological research of the seabed more feasible.
3
This research aims to provide a starting point for effective study of the submerged
landscape in the Arabian Gulf. It focuses on developing and testing classification and
characterisation methodologies for identifying zones of high potential for the survival
of archaeological and palaeoenvironmental remains that can subsequently be
targeted for more detailed investigations. The approach developed, seabed
characterisation, utilises commercially-captured marine geophysics data and LiDAR
bathymetry, and draws on techniques used in seabed habitat classification and
terrestrial historic landscape characterisation to try and maximise the potential for
successful exploration of the submerged landscape.
A large area of the seabed off the northwest coast of Qatar was surveyed by a
commercial company using sidescan sonar (a geophysical technique that measures
the acoustic reflectivity of the seabed) in 2008, as part of assessments for the
proposed Qatar-Bahrain causeway (GEMS, 2008). The part of the survey dataset
that lies within Qatari national waters was made available to the Qatar National
Historic Environment Record (QNHER) project (Al-Naimi et al., 2012) for the
purposes of archaeological assessment. The area surveyed covers 365 square
kilometres of the seabed, and now forms the Study Area for this research (Figure 1).
Previous research into sea level change indicates that this part of the Gulf between
Qatar and Bahrain remained free from marine influence until after 8,000 BP
(Lambeck, 1996; Al-Naimi et al., 2012).
4
Figure 1 Location of the Study Area (illustration by Nigel Dodds).
The aims and objectives of the research were:
 To develop and test techniques for classifying the submerged landscape in the
Study Area into characterisation areas that could be used to target further,
more intensive study
 To create the characterisation areas and use them to generate zones of
archaeological and palaeo-environmental potential
 To evaluate the effectiveness of the techniques used
 To evaluate the archaeological and palaeoenvironmental potential of the
submerged landscape within the Study Area
5
 To assess the implications for research and management of the marine
heritage resource
This was to be achieved by:
 Undertaking acoustic sediment texture classification on the complete sidescan
sonar dataset and using the results to classify the seabed in the Study Area
into initial landscape units
 Analysing available bathymetry and Holocene sea-level data and integrating
the results into the classified data in order to refine the initial landscape units
into character zones
 Using ground-truthing data obtained by direct sediment sampling and video
footage to validate the character zones
 Analysing the complete sidescan sonar dataset in order to identify and classify
geophysical anomalies, set them in a landscape context, and use the results
to further inform the character areas
 Targeted diver survey and high-resolution geophysical survey of selected
anomalies in order to clarify signature types
 Using the final character areas to generate zones of archaeological and
palaeoenvironmental potential
6
1.2 Overview of The Study Area
The Arabian Gulf is a shallow enclosed sea bounded at the south and west by the
Arabian Peninsula, and to the north and east by the Zagros Mountains (Lambeck
1996), with the confluence of the Tigris, Euphrates and Karun rivers forming the
Shatt al-Arab and flowing into the northern end of the Gulf. The Study Area itself, an
area measuring 365 square kilometres, lies off the northwest coast of Qatar, between
the Ras ‘Ushayriq peninsula area, and extends northwest towards the northeast
coast of Bahrain. The substantial Qit’at ash Shajarah reef, bordering Bahrain national
waters, forms the western edge of the Study Area. The water depths within the Study
Area are very shallow, averaging between 1 and 7 m, with the reef lying very close to
the surface.
1.2.1 Research Themes
Despite the technical and interpretational difficulties of studying the pre-transgression
landscape of the Arabian Gulf, and its consequent marginalisation within research
and published literature, it is clear that there are highly significant regional research
themes that could be informed by studying it.
There is increasing interest in the dispersal of early humans out of Africa, and the
geographical position of the Arabian Peninsula, lying between Africa and Asia, is
highly pertinent to this. Recent research into genetics and population movements
(Petraglia and Rose, 2009) has suggested that a postulated ‘southern dispersal
7
route’ would take populations around the southern coast of the Arabian Peninsula
and into the Arabian Gulf.
Another major research theme that is relevant to the Study Area is the 'Ubaid
phenomenon', that is, the relatively sudden appearance of settlement sites yielding
Mesopotamian ‘Ubaid pottery around the coastlines of the Arabian Gulf during the
second half of the eighth millennium BP. The question as to where these coastal
settlers had migrated from is of great significance to the study of Neolithic Arabia.
The dating and nature of these sites suggests that they may represent human
populations that were displaced from the former Gulf basin by the relatively rapid
marine transgression between c.8,000 BP and 7,500 BP (Cuttler, 2013; Rose, 2010).
The submerged landscape of the Study Area lies in close proximity to several of
these ‘Ubaid-related sites, which are considered to be the most significant preIslamic
sites in Qatar. Linked to the theme of human settlement patterns in the Early
and Mid-Holocene is the likelihood that the archaeologically-significant Karst
landscape of Northern Qatar extends into the Study Area as a drowned landscape,
and also the fact that submarine freshwater springs are known to have existed off
Bahrain in recent historical times (Cheesman, 1923).
The nature of potential submerged Prehistoric archaeological remains within the
Study Area, and the conditions and locations in which such remains could reasonably
be expected to survive marine inundation and sea-level fluctuation, are key issues for
this research. In order for settlement remains such as stone walls or cairns to survive
marine incursion and remain recognisable, they would have to have been buried
8
quickly due to the rapid rate of carbonate accretion in the Gulf. The most likely
locations for quick burial conditions to occur would be within tranquil areas protected
from the full force of wave action whilst sea-levels were rising. Potential marine
accretionary locations in the Study Area include sheltered environments on former
coastlines such as bays, spits, promontories and drainage channels.
The geographical location of the Arabian Gulf made it a nexus for trade from very
early on in maritime history (Al-Naimi et al., 2012, p.255). The Study Area itself lies in
close proximity to Bahrain, which is well-documented as a very important maritime
trade centre in the Prehistoric period (Killick and Moon, 2005). More recently,
principal trade routes to Bahrain and Southern Iraq lie to the north of Qatar, relatively
close to the Study Area. There is likely, therefore, to have been a high level of
shipping passing near, or through the Study Area. These factors, taken together with
the presence of the reef and the shallow water in the Study Area suggest that there is
a relatively high probability that shipwrecks could have occurred in the Study Area
from the Prehistoric period onwards.
Another important theme is the potential for preservation of palaeoenvironmental
remains in the Study Area. The arid climate and wind-deflated landscape of Qatar is
not conducive to the preservation of organic remains, and therefore very little
palaeoenvironmental evidence survives from terrestrial sites in Qatar. Even sites
dating from 2000 years ago are unlikely to contain useful quantities of organic
remains (Cuttler, 2014). However, the nature of the Holocene marine transgression
may have enabled the preservation of palaeoenvironmental caches dating from the
9
Late Upper Palaeolithic to Neolithic periods within sediment traps in the submerged
landscape (Cuttler, 2010, p.152). This therefore represents our best chance to find
palaeoenvironmental remains to inform research into these periods in Qatar , so it is
very important that we establish effective methodologies for locating such sediment
traps. The potential locations of such sediment traps may well coincide with the
potential locations of archaeological remains.
Although the Study Area forms only a small part of the extensive sub-aerial
landscape of the former Gulf Basin, it provides a convenient starting point from which
to begin to develop methodologies for effective exploration, which may help to
provide valuable information on a number of important regional research themes.
10
1.2.2 Geology and Geomorphology
The Arabian Gulf is a long, shallow, epicontinental sea, covering an area of 210,000
square kilometres, and is connected to the Indian Ocean by the Straits of Hormuz.
The Arabian Plate is slowly subducting below the Eurasian Plate, and the Arabian
Gulf basin is a direct result of this subduction. The subduction has resulted in a
gentle southwest to northeast tilt in the plate, and much of the northern Gulf between
Kuwait and Iran is topographically lower than the southern and western Gulf. (Cuttler
and Al-Naimi, 2013). Fringing the southern side of the Gulf are coastal islands, coral
reefs, tidal channels, tidal flats, raised beaches, and coastal salt flats known as
sabkhas (Teller et al., 2000). The Gulf is almost completely surrounded by arid land,
and as a result virtually no runoff is received from the Arabian Peninsula, where
precipitation is very low (100 mm/yr). The only runoff into the Gulf comes from rivers
draining the Zagros Mountains to the northeast and from the confluence of the Tigris,
Euphrates and Karun rivers forming the Shatt-al-Arab at the northern end of the Gulf
(Teller et al., 2000).
The Qatar peninsula projects from the relatively linear Arabian coastline on the
western side of the Gulf (KSEPL, 1973). It consists of an anticlinal dome, with a
north-south main axis (Cavalier, 1970), and it has a low-relief landscape with a
maximum elevation of about 110m above sea level (Sadiq and Nasir, 2002). The
exposed geological sequence on the Qatar peninsula consists of Tertiary limestones
and dolomites with interbedded clays, shale, gypsum and marls. This sequence is
covered in places by a series of Quaternary and recent deposits. (Leblanc, 2008).
11
About 70% of the total land surface of the Qatar peninsula is Eocene Dammam
formation limestone (Al Saad, 2005). This in turn overlies the Rus formation and the
Umm er Radhuma formation (Macumber, 2011). The anticlinal structure of Qatar (the
Qatar Arch) controls the outcrop pattern of the Dammam Formation, as do smaller
longitudinal folds such as the Dukhan anticline and the Simsima Dome (Al-Saad,
2005). In the south of Qatar, Miocene marls and limestone outcrop in mesas and
hills, and significant areas of sand dunes occur in the southeast (Johnstone and
Wilkinson, 1960, p.445)
The Fuwayrit Formation is found overlying Eocene sediments at certain locations in
Qatar. This formation consists of shallow marine and marine-derived aeolian
carbonates, preserved in coastal localities (Williams and Walkden, 2002). A study in
1999 on the Quaternary carbonates along the coastline of Qatar established that the
Quaternary Carbonate sequence, which consists almost entirely of detrital
limestones, is comprised of three lithostratigraphic units which vary in character along
the coastline, and the sequence is considerably more developed along the eastern
coast (Abu-Zeid et al., 2001, p.26). The study also established that the great variation
in the nature and intensity of freshwater cementation displayed by the Quaternary
carbonate sequence was a reflection of very considerable fluctuations in groundwater
levels (Abu-Zeid et al., 2001, p.35). The stratigraphy of offshore deposits has not
been properly formalised, but studies of sea-floor cores from six localities in the
southern Arabian Gulf indicates that the uppermost Pleistocene sequence of marine
carbonates and sabkha-derived evaporites represents the offshore equivalent of the
Fuwayrit Formation (Williams and Walkden, 2002).
12
Karst features are a widespread and striking aspect of the landscape of the Qatar
peninsula. These consist of more than 9,700 large and small depressions, and
several exposed sinkholes and caves , caused by the dissolution of the surface and
sub-surface rock. These depressions, which range from 50m to 3km in diameter,
overall covering an area of c.33,500 ha, can reach depths of up to 25m (Sadiq and
Nasir, 2002, p.132; Macumber, 2011, p.1). They form catchments for surface water
runoff, and contain better soils (known as rawdha) than are present in the rest of the
peninsula, and are therefore important for agriculture (Macumber, 2011). It is
reasonable to assume that these Karst features are also present, albeit in a modified
form, in the area that is now submerged around Qatar.
The seabed within the Study Area is relatively flat, with water depths averaging
between 1 and 7m between Qatar and Bahrain, dropping to 12m at the northern
extent of the Study Area. There are two large drying reefs in the area, the Fasht al
Azm reef and the Qit’at ash Shajarah reef (Marin Mätteknik AB, 2002). The former
reef extends on a southeast-northwest alignment between the Qatar border and
Bahrain, whilst the latter is located along the western edge of the Study Area (Figure
2).
13
Figure 2 Bathymetry (height data) between Qatar and Bahrain (from Al-Naimi et al., 2012).
The geology and geomorphology of the Study Area itself is influenced by the Bahrain
anticline. This consists of an arch or ridge which runs northwest-southeast from
Bahrain to Qatar, through the Study Area (Figure 3). This anticline is a significant
hazard for shipping and as a result no deep water ports were established along the
southwestern coastline during the Late Islamic period (Cuttler, 2014).
14
Figure 3 Location of the Bahrain Anticline (from Cuttler, 2014).
Sandbanks that have been tentatively interpreted as drowned sand dunes have been
mapped from LANDSAT images in the Gulf of Salwa between Qatar and Bahrain (AlHinai
et al., 1987, p.254).
1.2.3 Sedimentation and Marine Taphonomic Processes
The evidence from sediments in the Gulf, which date from 18,000 B.P. to the present
day (Kassler, 1973, cited in Al Ghadban et al., 1998, p.24), indicates that the present
sedimentary regime was established soon after 9,000 years B.P. when sea levels
15
were still considerably lower than present day levels, and a more humid climate
developed (Uchupi et al., 1996). On the Arabian Peninsula side of the Gulf,
carbonate deposition dominated throughout the Holocene (Uchupi et al., 1999).
Impure carbonate sediments tend to occur near the centre of the Gulf, changing to
high energy sands and local muddy embayments on the coast (KSEPL, 1973). The
present-day seafloor sediments in the Arabian side of the basin are largely windborne,
as virtually no water-borne sediment reaches the Gulf from the arid Arabian
Peninsula (Marin Mätteknik AB, 2002). This lack of water-borne sediment is probably
a major factor in the predominance of almost pure carbonate sediments on the
Arabian side (Purser, 1973).
There is still insufficient evidence for a detailed picture of sedimentation rates in
specific areas of the Gulf. According to Uchupi et al. (1996, p.267), overall sediment
patterns in the Gulf are influenced by bottom water leaving the Gulf through the Strait
of Hormuz, accretion/deposition and chemical precipitation along the Arabian coast,
and also by widespread gas seeps. Although sedimentation rates are generally high
throughout the Gulf (Al Ghadban, 1998, p.24), it is clear that there are considerable
regional and local differences that are influenced by seabed topography, sediment
types, wave energy, the orientation of the coast in relation to the prevailing wind, and
the presence or absence of offshore barriers (Alsharhan and Kendall, 2003, p.193).
However, in the shallow, Arabian parts of the Gulf, waves and surface currents are
almost certainly the strongest influence on sediment transport (KSEPL, 1973).
Furthermore, in a small sea like the Gulf, waves are largely created by the wind, so
16
the prevailing northwest ‘Shamal’ wind is crucial in determining sedimentation
patterns (Herman, 1957).
The effects of aeolian deposition by the Shamal are strongest in the northern parts of
the Gulf, and along the leeward (east) coast of the Qatar Peninsula (KSEPL, 1973).
Extrapolation from profiles of Holocene sediment accumulation indicates that finegrained
sediments can be up to 10m thick in the low-energy environments of
sheltered parts of shallow western coastal areas of the Gulf, such as in the area of
Khor and Dakhira to the east of Qatar (Kassler, 1973, cited in Al Ghadban et al.
1998, p.25; KSEPL, 1973). Thick, muddy sediment occurs in sheltered coastal
lagoons and depressions, and geophysical survey in the Gulf has indicated that
topographic depressions and submerged wadis act as sediment traps for large
volumes of mud sediment, which has smoothed out the relief of the seabed.
However, this sediment accumulation has not yet completely masked the underlying
topography. (Kassler, cited in Purser, 1973, p.31).
Conversely, in the high-energy windward areas along the southwestern coastline of
the Gulf, where coarse-grained sediments are laid down, lower sedimentation rates
result in sediment thicknesses of less than 2m (Al Ghadban et al., p.25). There are
areas off the Saudi Arabian Coast, especially at relatively shallow depths, where
marine geophysics data indicates that the sediments are so thin that they barely
cover the underlying rock, probably due to the complex topography in that area
(Evans, 1966, p.295). In these areas, a thin covering (5 - 20 cm) of sand overlies the
17
rock, and the hard rock surface tends to be exposed between large mega-ripples
(Purser, 1973, p.5).
Ooid build-ups mainly occur in tidally-influenced areas on windward coasts (Williams
and Walkden, 2002, p.382). Geological studies carried out around Qatar in the 1960s
proved that submarine lithification is in progress in some areas, mainly in areas
where there is slower sediment accumulation. (Shinn, 1969, p.139).
The Study Area lies off the northwest coast of Qatar, which directly faces the full
force of the northwesterly 'Shamal' wind, a coastal orientation not normally
associated with rapid sediment accumulation. However, the picture is more complex,
as this area is not fully exposed to strong marine currents due to the protecting
barrier of Bahrain and the Saudi Arabian coast. (KSEPL, 1973). An algal coral reef
has developed 1-4 km offshore from the northwest tip of the peninsula (Taylor and
Illing, 1969, p.72).
The seabed surface, summarised by GEMS (2008), suggests that the northern extent
of the Study Area comprises mainly hard coral and rocky substrates interspersed by
sands and gravels. The central part appears to consist predominantly of relatively
featureless medium to coarse sands, with areas of broad sand ribbons stretching for
kilometres in a general north to south direction. East of the reef, the seabed is hard
and rocky with numerous coral heads and very complex localised relief. The southern
extent of the survey area appears to be characterized by medium to coarse sands
and gravels, with evidence of extensive trawler-scars in the west.
18
1.2.4 Climate
Most of the southern Arabian Gulf currently falls within the hyper-arid zone in the
UNESCO (1979) classification scheme. It falls within the subtropical trade wind belt,
and is at present dominated by the Shamal wind (Williams and Walkden, 2002).
Qatar itself has an arid to hyper-arid climate, with annual rainfall from winter
westerlies averaging 80 mm in the north and less in the south. This small amount of
rainfall recharges into the limestone aquifer system, and discharges at the coast, and
therefore there is no permanent fresh surface water on the Qatar peninsula
(Macumber, 2011). Several substantial wadis, charged seasonally, run from the
anticline towards the coast (Cuttler and Al-Naimi, 2013). Although rainfall is low, the
country is named ‘Qatar’, the Arabic word for ‘Land of Dews’, due to the frequent
occurrence of heavy night dews (Taylor and Illing, 1969, p.71).
1.2.5 Tides and Salinity
At present, the average tidal range in the Arabian Gulf is small, around 1-2 m
(Lambeck, 1996), and in Qatar itself, the normal tidal range changes from one part of
the coast to another, for example it is about 1.4m at Doha and only 0.5m in western
lagoons. However, in these areas, the movement of water due to the wind is often far
more significant than tidal movement (Taylor and Illing, 1969, p.71). A slow
circulatory current flows within the Gulf, moving from the Straits of Hormuz
anticlockwise along the Persian Coast. This affects the distribution of salinity and
temperature in the Arabian Gulf waters as it brings in new oceanic water (KSEPL,
19
1973). A small amount of freshwater flows into the Gulf, mainly from the Shatt alArab,
and also from intermittent streams on the Persian Coast (Teller et al., 2000).
However, this small amount of freshwater inflow is exceeded by the high rate of
evaporation that occurs in this restricted basin with such high temperatures. As a
result, the average salinity of the Gulf is relatively high. Water temperature and
salinity are closely linked, and both are highest in shallow embayments and lagoons
(KSEPL, 1973). The water temperature, although varying throughout the different
parts of the Gulf, is generally high, recorded in 1966 (Evans, 1966) to be up to 320
C,
and higher in the shallow waters along the coasts, although it should be noted that
the temperatures may be quite different nearly 50 years later. This same study
recorded salinities ranging from 37 to 38 %0 in the Straits of Hormuz to 38 to 41 %0
at the northern end of the Gulf, and reaching higher levels in the coastal areas along
the southwestern shore. According to Taylor and Illing (1969, p.71), salinity values of
1.5-1.8 times standard seawater have been logged in some of the more secluded
lagoons opening to the west coast of the Gulf, although, as with temperatures,
salinity may have changed in the intervening decades. A more recent study was
carried out on water temperature and salinity around the coasts of Qatar and Bahrain
in 1992 (Abdel-Monim Mubarak and Kubryakov, 2000). This study identified a thin
surface layer of water 2m in thickness with a temperature around 26 to 270
C, and
relatively low salinity for the Persian Gulf (around 38.1%0). Below this layer,
temperature decreased and salinity increased, down to 20.40
C and 39.6%0 at 24m
depth. Seasonal variations were observed, due to solar heating and wind circulation.
20
1.2.6 Sea Levels
The Arabian Gulf Basin has been occupied by numerous marine transgressions, of
which the Holocene transgression is the most recent. The retreat of glaciers in the
northern hemisphere from around 18,000 BP caused a rise in global sea levels from
more than 100m below their present levels (Cuttler and Al-Naimi, 2013). The lack of
reliable regional sea-level curves, and the considerable debate surrounding the effect
of tectonic movement and isostatic loading movements (Lambeck, 1996; Uchupi et
al., 1999), means that it is not possible to be exact about the rate and timing of
marine transgression within the Arabian Gulf. Lambeck (1996) plotted the marine
transgression across the Gulf using seabed bathymetry to create a `glacio-hydroisostatic
model’, basing his model on the assumption that the sea-level rise was
continuous without still stands or regressions, whilst acknowledging that the rise
would not have been spatially uniform (Teller et al., 2000). Sarnthein (1972)
estimated that the rising sea levels covered the land laterally at an average rate of
100-120 m per year, with periods of still stands at about 11,300 and 10,500 BP.
There must also have been periods of faster transgression, at times greater than 1km
per year, during times of rapid eustatic rises, such as are known to have occurred
around 12-11.5 BP and 9.5-8.5 BP (Teller et al., 2000).
Models of marine transgression based on bathymetry do not take into account
tectonic movement, hydrostatic loading or marine deposition. The evidence suggests
that there has not been significant tectonic movement around Qatar during the
Holocene, and that deposition levels should not affect results at a landscape scale,
21
whereas hydrostatic adjustment is likely to have significantly affected models for the
Arabian Gulf. The effect of this is that later dates occur than would be expected for
the equivalent depth in other parts of the world (Cuttler, 2014, p13-14). This effect is
likely to be more pronounced further back in time. Despite these limitations, a good
overview of the evidence for the progress of marine transgression in the Arabian Gulf
is provided by Cuttler and Al-Naimi (2013), and this overview is summarised here.
It is generally agreed that at the time of the last glacial maximum at about 18,000 BP,
the Gulf was entirely free of marine influence. At this time, instead of flowing into the
Gulf, it is reasonable to assume that the confluence of the Tigris-Euphrates Rivers
(known as the Ur Shatt River) would have flowed through the Gulf basin, discharging
into the Gulf of Oman through the Strait of Hormuz (Cuttler and Al-Naimi, 2013). The
topography of the former Ur Shatt River valley was relatively flat, and Cuttler (2013)
has suggested that the Ur-Shatt would have been a low-energy channel which was
more likely to be a corridor of marshland and lakes rather than a high-energy river.
Although there are studies that have identified morphological or sedimentological
traces of this former watercourse, few details are available (Teller et al. 2000).
Lambeck’s model (Lambeck, 1996) indicates that during the last glacial maximum,
the Ur-Shatt fed three large freshwater lakes on the floor of the basin, along the
central part and along the southern side (Figure 4), although the issue of whether
they would have been freshwater or saline has not yet been completely resolved.
The lake hypothesis is now supported by ETOPO2 data (a combination of satellite
altimetry observations, shipboard echo-sounding and data from a Digital Bathymetric
Data Base, NOAA), which confirms the presence of a western basin and a central
22
basin in the Gulf (Cuttler and Al-Naimi, 2013). It is also supported by the analysis of
3D seismic data as part of the QNHER project, which has provided evidence of
sediments relating to the eastern lake (Cuttler, 2013). However, these models do not
take into account any subsequent tectonic plate movements or extensive marine
deposition (Cuttler and Al-Naimi, 2013).
Figure 4 Shoreline Reconstructions at 14,000 BP after Lambeck (1996) (from Cuttler et al., 2011b).
The evidence suggests that the Strait of Hormuz became a narrow waterway at the
mouth of the Ur Shatt River by 14,000 BP. (Lambeck, 1996). Around 12,500 years
ago, the central basin would have been subject to marine incursion, probably causing
a change from fresh water to saline lakes (Cuttler and Al-Naimi, 2013). The Western
Basin flooded about 1,000 years later (Lambeck, 1996). By 10,000 BP, the marine
transgression would have resulted in a long narrow sea, and the Iranian coastline
23
would already resemble the present-day coastline. However, large areas of the
southern and western basin around the Emirates, Qatar and Bahrain would have
remained as open landscape at this time (Lambeck, 1996).
After 10,000 BP periods of still-stands were interspersed with periods of more rapid
marine transgression. As late as 8,200 BP extensive areas between Qatar, Bahrain
and the Emirates remained free from marine influence (Figure 5). The Bahrain
anticline would have delayed marine transgression into the Gulf of Salwa until the
final stages of sea level rise in the 8th Millennium BP (Cuttler, 2014). The presentday
shoreline was reached shortly before 6,000 BP, after which sea levels continued
to rise for a further 1-2 m above present-day levels (Lambeck, 1996). During this
period, Qatar was almost an island, connected to the remainder of the Arabian
Peninsula by a thin strip of land (Cuttler and Al-Naimi, 2013).
Figure 5 Shoreline reconstructions at 8,200 BP after Lambeck (1996) (from Cuttler and Al-Naimi,
2013).
24
Evidence for the Mid-Holocene high stand is widespread in the southern Gulf
(Williams and Walkden, 2002), and is demonstrated by dated terraces and marine
sediments along the southern shore of the Gulf (Uchupi et al., 1999; Lambeck, 1996;
Teller at al., 2000). In Qatar, cemented beach sands from the northwestern coast, at
locations between 1.5 and 2.5 m above present sea level, were identified and dated
by Taylor and Illing (1969) to between 3930 ± 130 and 4340 ± 180 BP. Similarly, VitaFinzi
identified beach rock deposits up to 2m above present sea level from several
coastal areas in Qatar, and dated them from between 4690 ± 80 and 5830 ± 70 BP.
(Vita-Finzi, 1978, cited in Williams and Walkden, 2002). The evidence indicates that
this high stand was due to eustatic change rather than tectonic activity. Both
Lambeck (1996) and Williams and Walkden (2002) agree that although the Gulf lies
close to an active plate margin, there is no evidence for significant tectonic
movement for much of the sea floor or coastline in the southern Gulf during the
Holocene, although there are exceptions, such as on the northern side of the Strait of
Hormuz (Lambeck, 1996). Recent work at Wadi Debayan, close to the west coast of
Qatar, has dated the earliest relic beach terraces there to around 6,000 BP,
indicating a sea level high from around 6,000 BP onwards, supporting the sea-level
models produced by Lambeck (1996) and Heyvaert and Baeteman (2007).
1.2.7 Vegetation History and Palaeo-climate
The impact of the rise and fall of sea levels on human populations in the Arabian Gulf
is also intimately connected with past climate change, which itself had a profound
effect on human occupation and exploitation of the area. There is evidence from
25
sediments in the region that indicate pronounced climate changes, closely linked to
global glaciation cycles and prevailing wind systems. Such changes caused phases
of increased aridity and humidity relative to the present day climate. (Petraglia,
2005). Most palaeoenvironmental experts recognise widespread evidence for a
wetter climate during the Early Holocene period in the region, between around 9,000
and 6,000 BP (Teller et al., 2000). This evidence includes radiocarbon-dated
lacustrine sediments that developed during that period in areas that are now in the
Arabian Desert regions, including the Rub' al-Khali and Nafud regions (McClure,
1976, p.755; Parker et al., 2006). The wetter phase is linked to the northward
migration of the Indian Ocean monsoon into the Gulf region between 8,500–6,000 BP
(Parker and Goudie, 2008), bringing higher rainfall, although it is not known exactly
how far north this influence reached. Fleitmann et al. (2004, p.20) speculate that the
monsoon rainfall activity did not affect areas further north than around 23-24°
latitude, whereas most of the Qatar peninsula lies at between 25-26 ° latitude.
However, further research is needed to support this hypothesis. After around 6,000
BP, the monsoon migrated southwards, and from 4,500 BP the climate became more
arid, with an intense arid period occurring at 4,100 BP. Since then, arid conditions
similar to those found in the region today have predominated (Parker and Goudie,
2008).
Faure et al. (2002, p.53-54) proposed a ‘Coastal Oasis’ model, whereby the falling
sea levels reduced the sea level pressure on the continental shelf and steepened the
water table gradient at the coast, leading to reduced water availability in the interior,
but increased freshwater via springs on the newly-emerged coastlines. This model
26
has considerable significance for research into human settlement patterns in the Gulf
Basin both during, and immediately after, the Holocene marine transgression.
27
CHAPTER 2: REVIEW OF PREVIOUS RESEARCH AND ARCHAEOLOGICAL
POTENTIAL
2.1 An overview of Archaeology in the Arabian Peninsula
2.1.1 The Palaeolithic
Very little is known about the Palaeolithic of the Arabian Peninsula as a whole, and it
is very difficult to characterise Palaeolithic settlement (Parker et al., 2006) as there
are still very few stratified sites, most of the evidence coming from unstratified
surface finds, often in deflated landscapes with no geological context (Marks, 2009,
p.295). The largest concentrations of Palaeolithic sites are located ‘at the margins of
modern deserts in the northern and southern portions of the peninsula (Nafud Desert
in the north, the Empty Quarter in the south)’ (Petraglia, 2005, p.306). Three stratified
sites with lithic assemblages dating to the Middle Palaeolithic period (Jebel Qattar,
Jebel Katefeh and Jebel Umm Sanman) were recently found along the shores of the
Jubbah palaeolake in the Nefud desert in Northern Arabia (Petraglia et al., 2012,
p.1).
There has been a considerable amount of debate and uncertainty surrounding the
presence, or otherwise, of Palaeolithic remains in the Eastern Arabian Peninsula.
Surface finds were reported from Qatar, Bahrain, Oman and Eastern Saudi Arabia
from the 1930s onwards (Potts, 1990, p.30-31). An influential Danish survey of the
Qatar Peninsula, beginning in the 1950s and continuing into the 1960s (Kapel, 1967)
28
identified more than 120 Prehistoric sites from surface lithic scatters, which were
categorised into 4 groups, A, B, C and D. The A group was initially typologically
identified as Palaeolithic (‘Mousterian’-related), and the B, C and D groups were
typologically classified as following on chronologically from the A group (Rose, 2010).
On the basis of Kapel’s classifications, a Palaeolithic presence was also identified
elsewhere in Eastern Arabia (Kapel, in Bibby, 1973; Masry, 1974; Pullar, 1974), and
further attested to in Qatar in the early 1970s (De Cardi, 1978). In the late 1970s, a
French research programme investigated a site near Khor in southeastern Qatar and
discovered assemblages similar to Kapel’s A group present in Middle Holocene
layers (Inizan, 1988, p.62; Tixier, 1980, p.197). This research placed Kapel’s Qatar A
group material in the Neolithic, and not only led to a wholesale reassessment of all
Kapel’s groups as being Holocene in date, but also led to a general consensus that
Palaeolithic material, and by implication, Palaeolithic human activity, was absent from
the whole of Eastern Arabia (Potts, 1990, p.31).
However, this position was subsequently revised again in the light of more recent
research which has identified several new Palaeolithic sites in Eastern Arabia.
Systematic survey in Dhofar, Oman identified evidence of one stratified site, Aybut Al
Auwal, and more than 100 surface scatters, containing Middle Palaeolithic lithic
assemblages. These assemblages represent technology very similar to that from the
late Nubian Complex in northeastern Africa. The sites are all on the Nejd Plateau,
and OSL dates obtained from stratified deposits at Aybut Al Auwal produced dates of
106,000 years BP. (Rose et al., 2011, p1; Groucutt and Petraglia, 2012, p.119-120).
29
Excavations at Jebel Faya in Sharjah Emirate identified stratified Palaeolithic
material, providing clear evidence of Upper Pleistocene activity, with the earliest
stratified levels dated to c.128,000 years ago (Uerpmann and Uerpmann, 2008;
Uerpmann et al., 2009; Marks, 2009; Rose, 2010; Bretzke et al., 2013). Unstratified
surface lithic scatters containing material with similarities to the stratified material at
Jebel Faya have been identified in Abu Dhabi and Sharjah (Scott-Jackson et al.,
2009; Wahida et al. 2009; Rose, 2010).
Although these recent findings can be seen as confirmation of hominim activity in
Eastern Arabia in the Upper Pleistocene, the date of the earliest human occupation
of the Arabian Peninsula still remains uncertain. Also, there is considerable debate
about whether this occupation was continuous or intermittent (Rose, 2010), and
whether there was population continuity from the Pleistocene into the Holocene
(Uerpmann et al., 2009, p.206). Any evidence for Late Pleistocene human occupation
along the former waterway of the Ur-Shatt river corridor will now be submerged
(Parker and Goudie, 2008, p.465). Macumber (2011) argues that permanent
occupation of the area that now forms the Qatar Peninsula was unlikely in the Upper
and Middle Palaeolithic due to the lack of accessible freshwater as a result of low
groundwater tables caused by the drop in sea levels. However, it is not currently
clear how extensive surface water would have been during ameliorating climatic
conditions, or within deep sinkholes which have been subsequently infilled by
collapse and/or sediment accumulation. It is also worth noting that Lower Palaeolithic
artefacts in Qatar are most likely to have been re-deposited in secondary contexts
due to erosion or to re-working by later Neolithic groups (Cuttler, 2010, p.151).
30
2.1.2 The Neolithic
The marine transgression across the Gulf basin after the last glacial maximum took
place between 14,000 and 6,000 years ago, in the Late Pleistocene/Early Holocene,
and the environment within the exposed Gulf Basin during this period is likely to have
been highly favourable to human settlement (Teller et al., 2000). However, the
archaeological and palaeoenvironmental record for the area around the former Gulf
basin is scant and incomplete for the Early Holocene, and consists mainly of
relatively ephemeral campsites (Rose, 2010). The French investigations of the late
1970s (Tixier, 1980; Inizan, 1988) re-interpreted and re-arranged Kapel’s
classifications so that Kapel’s Qatar B group of lithics, found at a few sites around the
Oman peninsula and Qatar (Parker et al., 2006), was classed as the earliest phase of
occupation, dating to the beginning of the Holocene. Kapel’s A, C and D groups were
classed as a single, later group, dating to within the 7th millennium BP (Drechsler,
2009, p.16). Three sites from around the Gulf Basin, Wadi Wutayya in Northern
Oman, and Nad al-Thamam and Jebel Faya 1 in Sharjah Emirate, which have been
dated by radiometric dates to between 11,000 and 8,500 cal BP, have been used to
place the Qatar B sites within the same date range, on the basis of similarities
between the lithic technologies (Rose, 2010).
At c.8,000 BP, the Qatar B technology in Qatar was replaced by what was known as
the Arabian Bifacial Tradition (ABT), a technology specific to this region of the
Peninsula (Cuttler et al., 2011b, p.3). It is worth noting that use of the term ABT is
problematic, as it is now considered by flint specialists to be loosely representative of
31
a range of lithic technologies (Crassard and Dreschler, 2013, p.5-6). The northward
migration of the Indian Ocean monsoon in the Early-Mid Holocene led to a
significantly wetter period in southern parts of the Gulf region between 8,500-6,000
BP (Parker and Goudie, 2008), making conditions more favourable for human
exploitation in areas affected, although Fleitmann et al. (2004, p.20) suggest that the
effects of this wetter period reached no further north than 24° latitude. It has been
suggested that this improved climate influenced the expansion of settlement in the
8th and 7th millennia BP (Masry, 1997, p.44), a time when larger and more
permanent sites appeared in eastern Arabia (Carter, 2010, p.193). However, if we
accept Fleitmann's theory, all of the Mid-Holocene sites from the Central Gulf would
have been unaffected by the wetter conditions as they lie to the north of 24° latitude
(Cuttler, 2013, p.40). Also, there is currently little evidence for an expansion of
settlement into the interior of the Arabian Peninsula at this time.
Evidence from the 8th and 7th millennia BP for the presence of domesticated forms
of cattle, sheep and goat is increasingly appearing as a result of recent work in the
Southern Arabian Peninsula, indicating a change in food procurement. (Drechsler,
2009). However, since the evidence for domesticates is mainly confined to the
southern part of the peninsula, there is currently significant debate around their
origins (McCorriston and Martin, 2009, p.246-247; McCorriston, 2013). At this period,
semi-nomadic herding was predominant in the interior of the Arabian peninsula
(Potts, 1990; Uerpmann, 1992, cited in Parker and Goudie, 2008, p.466), and
hunting, fishing and the exploitation of marine resources such as shellfish were also
important for the coastal settlements at this time (Parker and Goudie, 2008).
32
It has long been recognised that there was a dramatic increase in settlements
appearing around the shoreline of the Gulf in the Mid-Holocene (Rose, 2010). This
occurred just after the final phase of marine incursion into the Gulf basin, when sea
levels rose to approximately 2-3m higher than present-day levels (Lambeck, 1996).
These settlements, predominantly located around the former Mid-Holocene
shoreline, are characterised by the presence of ABT lithic artefacts, and display
evidence for contact with Mesopotamia during the 8th and 7th millennia BP, in the
form of Mesopotamian ‘Ubaid pottery (Carter, 2006, p.58) (Figure 6).
These sites are generally referred to as ‘Ubaid-related sites, as they are clearly not
Mesopotamian ‘Ubaid settlements. However, the use of the term ‘Neolithic’ in relation
to these sites is the subject of some debate (Crassard and Drechsler, 2013, p.4). The
‘Ubaid-related sites appear to be relatively complex in comparison to the ephemeral
campsite-type sites that typify the Early Holocene sites in Eastern Arabia, particularly
in relation to the presence of more permanent, stone structures, sophisticated boatbuilding
technology and well-established trade networks (Carter and Crawford,
2010). However, although they do display some evidence for the domestication of
plants and animals, they are largely based around hunting, gathering and the
exploitation of coastal resources, and with the exception of Arabian Coarse Wares,
all of the pottery is imported. Rose (2010) points out that almost all of the stratified
‘Ubaid-related sites in Eastern Arabia appear to have been located in previously
unsettled areas. Only Ain Qannas in Saudi Arabia (Masry, 1997) shows evidence of
earlier occupation at the site (Rose, 2010). This relatively abrupt change in
33
settlement patterns is also reflected by unstratified surface sites, where ‘Ubaid
pottery and pre-ABT lithics have not so far been found together. (Rose, 2010)
Figure 6 Distribution of ‘Ubaid and ‘Ubaid -Related Sites around the Arabian Gulf (from Cuttler, 2013).
Marine inundation did not occur around the west coast of Qatar until after 8,000 BP
(Al Naimi et al., 2012), and occupation of sites along the Qatar coastline begins
during the second half of the 8th millennium BP (Carter, 2010). Sites in Qatar include
Ras Aburuk/Raʾs Abarāq (De Cardi, 1978, p.82), Bir Zeikrit/Biʾr Zikrīt (De Cardi,
1978, p.115), Site A4 near Dukhan, (Kapel, 1967, p.37), Al-Dasah/al-Daʿsah (De
Cardi, 1978, p.55), Al-Khor/al-Khawr (Tixier, 1980), and Wadi Debayan/Wādī al-
34
Ḍabaʿyān (Cuttler et al., 2011b; Tetlow et al., Forthcoming). Outside Qatar, major
sites include al-Markh, Bahrain (Roaf, 1976, p.146), Dosariyah/Dawsāriyyah, Saudi
Arabia (Drechsler, 2011, p.71), Abu Khamis, Saudi Arabia (Masry, 1997, p.87),
Khursaniyah, Saudi Arabia (Masry, 1997, p.84), Ain as-Sayh, Saudi Arabia (McClure
and Al-Shaikh, 1993), As-Sabiyah H3, Kuwait (Carter and Crawford, 2010) and
Bahra 1, Kuwait (Rutkowski, 2011). There are far fewer ‘Ubaid-related sites in the
Lower Gulf than in the Central Gulf, probably as a result of increasing distance from
Mesopotamia. (Carter, 2006, p.59).
So far, it has not been possible to firmly establish where these ‘Ubaid-related
communities originated from. Whilst the increase in settlement may in part be
attributed to the greater archaeological visibility of more permanent settlement, it is
also possible that the new settlers migrated into the area as a result of displacement
from the Gulf basin by relatively rapid inundation between c. 8,000BP and 7,500BP
(Rose, 2010). If this theory is accepted, then Rose proposes that we should not
necessarily view the existing, ephemeral Terminal Pleistocene/Early Holocene sites
around the Gulf as representative of the population in the region at the time, but
rather as ‘more mobile, peripheral elements of a larger core group’ (Rose, 2010,
p.850).
The archaeological evidence indicates that there was a break in human settlement in
Eastern Arabia in the early 6th Millennium BP. No ABT/’Ubaid-related sites have
been dated to later than 5,800 BP (Parker and Goudie, 2008, p.467), and there is a
complete absence of known archaeological sites throughout Eastern Arabia during
35
the following millennium (Uerpmann, 2003), with the exception of the recentlyexcavated
site at Wadi Debayan in Qatar. The fish midden at this site has been
radiocarbon dated to between 5,200 and 4,500 BP (Tetlow et al., Forthcoming). This
relatively sudden break in human settlement in Eastern Arabia, and evidence for a
major population decline in the Arabian Peninsula as a whole, coincides with the end
of the mid-Holocene wet phase, and the onset of arid conditions, following on from
the southward migration of the Indian Ocean monsoon (Parker and Goudie, 2008). It
has been suggested by Uerpmann (2003) that the break in human settlement is due
to this climatic deterioration, but there may also have been other contributory factors,
indicated by the evidence from Wadi Debayan of a high energy event affecting the
coast of Qatar soon after 4,500 BP (Tetlow et al., Forthcoming).
2.1.3 Maritime History
The potential for archaeological sites within the submerged Gulf is not simply limited
to sites from the Late Pleistocene/Early Holocene, but also includes the potential for
shipwrecks from the 8th-7th Millennium BP to the present day. The earliest direct
evidence for maritime trade in the Gulf consists of boat-related remains found at H3,
As-Sabiyah in Kuwait (Carter and Crawford, 2010). These remains took the form of
more than 50 fragments of bitumen with reed impressions and barnacles, interpreted
as the waterproof coating from sea-going reed bundle boats (Carter, 2006, p.55-56) ,
and they are the earliest sea-going boat remains yet identified anywhere in the world.
Representations of boats were also found at the site, including a ceramic model of a
reed bundle boat and a picture of a masted boat painted on a ceramic disc. These
36
boat-related remains were found in archaeological contexts dating between 7,500
and 7,000 BP (Carter, 2006, p.53-55).
There is also indirect evidence for seafaring and maritime trade in the 8th-7th
millennia BP, as represented by quantities of ‘Ubaid pottery from Mesopotamia found
at more than 60 Arabian Neolithic sites around the coast of the Eastern Arabian
Peninsula, including Saudi Arabia, Bahrain and Qatar (Burkholder, 1972; Masry,
1974; De Cardi, 1978). Recent research has led to the general acceptance that this
pottery represents an extensive maritime trading network between the Arabian Gulf
and Southern Mesopotamia (Carter, 2006; Carter and Crawford, 2010). There is no
evidence for boat-building in Qatar at this time, but presence of ‘Ubaid pottery at the
coastal Neolithic sites in Qatar clearly demonstrates that these settlements were
participating in this well-developed maritime trading relationship. The archaeological
evidence indicates that the interaction between Mesopotamia and the Gulf petered
out during the 7th millennium BP (Carter and Crawford, 2010, p.212).
In the Bronze Age, a complex maritime trade network existed between Southern
Mesopotamia, Dilmun/Bahrain, southeastern Arabia and the Indus Valley civilisations
(Al-Naimi et al., 2012, p.249) around 5,000-3,500 BP. Of particular relevance to the
Study Area, given its location, is Dilmum, which, between 4,500 and 3,500 BP,
became a very important international trading nation, channelling trade through the
Gulf. This was due in part to its strategic location between the head of the Gulf and
the Straits of Hormuz, its safe anchorage and its plentiful freshwater supplies (Killick
and Moon, 2005, p.1). There are frequent mentions in cuneiform texts from
37
Mesopotamia from the 5th Millennium BP onwards of ships sailing from Ur in
Southern Mesopotamia to Dilmun, to Magan (Southeastern Arabia) and to Meluhha
(thought to be in the Indus Valley) (Roux, p.32). Qatar was clearly a part of this
trading network, as Barbar pottery typical of Dilmum has been found at sites in Qatar,
including on the Ras Abaruk peninsula (Carter, 2003). A 4th millennium BP dye
production site on Khor island in Qatar, which produced quantities of Kassite pottery,
is also evidence of significant trade with Mesopotamia (Edens, 1999).
The Arabian Gulf continued in historical times to be a very important trade route
linking the West and the East. The Sassanid dynasty, established in Mesopotamia in
AD 225, controlled the trade of a wide range of commodities in the Gulf, and by the
early Islamic period Arab dhows were trading with India, China, Southeast Asia and
East Africa, playing a central role in trade, and in other aspects of society such as
exploration, and defence (Al-Naimi et al., 2012). The Arabian Gulf has long been
famous for its role in the historic pearl trade, and extensive pearl beds are found all
along the Arabian side of the Gulf, with the most productive located to the north and
east of Bahrain (Bowen, 1951, p.166). The importance of Bahrain as a pearl-trading
centre is apparent from the many references to the pearls of Bahrain by early
travellers to the Gulf (Bowen, 1951, p.161). It was estimated in the early 19th Century
that there were 1500 pearl fishing boats in Bahrain (Wilson, 1883, p.285), and 900 at
the turn of the 20th Century (200 in Qatar) (Bowen, 1951, p.168). The northwest
coast of Qatar was well-known to early Arab navigators (Johnson and Wilkinson,
1960, p.444), and the remains of the former pearl-fishing and trading port of AlZubārah,
in close proximity to the Study Area, are testament to the importance of the
38
northwest coast of Qatar to maritime trade in the 18th and 19th centuries. Khor
Hassan (Khuwayr), further up the coast to the north of Al-Zubārah, became the base
of the notorious pirate Rahma bin Jabir, in the early 19th Century (Johnson and
Wilkinson, 1960, p.444; Misbahuddin, 1984), again an indication of the role that this
stretch of water played in shipping and trade.
However, despite the longevity of this maritime history, there has been very little
research into wrecks or other maritime archaeological finds in the Gulf. A relatively
recent discovery of an ancient merchant ship and its cargo, lying at a depth of 70m
near the port of Siraf in Iran, was reported in 2006 by local fishermen, who
discovered it accidentally. More than 40 amphora have been recorded on the
seabed, and the wreck is believed to date from the Parthian or Sassanid periods
(NOAA, 2006, p.17). There is also an unconfirmed report of a wreck with a cargo of
amphora lying at a depth of 10m in a deep-water channel in Bahrain territorial waters.
Fragments of amphorae from this site have been retrieved on several occasions by
local divers (Al-Naimi et al., 2012, p.249). Elsewhere in the Gulf, twenty-eight stone
anchors and ring-stones were discovered from a single submerged site near Qalhat,
an important Early Islamic and Medieval port in Oman (Vosmer, 1999, p.250).
Regarding more recent times, there is documentary evidence, in the form of a series
of telegrams between various officials, of a major sea battle involving the British navy
that took place around the Bay of Al-Zubārah in Qatar in 1895. According to the
correspondence, this battle occurred as a result of a potential threat from a fleet of
dhows, assembled by the Ottomans and Sheikh Jassim bin Thani, that were
39
apparently armed and ready for an imminent attack on Bahrain (Wilson, 1895). The
British Navy became involved as a result of their relations with, and influence over,
the affairs of Bahrain. The battle took place on 6th September 1895, and
correspondence from the captain of the British warship H.M.S. Sphinx, records that
the warship shelled the fleet of dhows for hours, and then despatched boats, under
cover of firing blanks, to set fire to the dhows. The captain states that ‘The total
number of dhows destroyed as far as can be ascertained from the ship is about 44’
(Pelly, 1895. However, no traces of these boats have ever been found.
An early 18th Century naval cannon was recovered from the sea relatively recently
(the exact recovery date is unknown) near Mesaieed/MusayΚīd, off the east coast of
Qatar, reportedly during the excavation of a new pipeline (QNHER ID 355; Al-Naimi
et al., 2102, p.249). There are also documented modern wrecks in Qatari waters,
including 17 from the past century recorded within the Qatar National Historic
Environment record, which have been recorded from maritime charts. These include
wrecks off the northern tip of the Qatar peninsula, to the northeast of the Study Area,
which are depicted on an old Admiralty chart, the survey date for which is given as
1925 (UKHO 1925).
40
2.2 Submerged Landscapes Research
2.2.1 Research Outside the Arabian Gulf
There has been very little work carried out into submerged landscapes in the Arabian
Gulf basin, and relatively little has been written about its archaeological significance
or potential (Cuttler, 2014). Before detailing the small amount of work that has been
done, it is useful to place it into a wider context by providing an overview of
investigations into areas of submerged landscape elsewhere in the world.
Research on submerged coastlines has been carried out in the Red Sea, around the
Farasan Islands (Bailey et al., 2007), in the context of human occupation in the area
of the Bab al-Mandab Straits during periods of lower sea levels. Archaeological
survey and underwater exploration recorded archaeological sites, largely shell
middens formed in the last 6,000 years, located at characteristic geomorphological
locations on land, and then recorded similar locations underwater at submerged
former shorelines. Work is continuing to try and identify archaeological sites
associated with these shorelines.
In the USA, underwater geoarchaeological studies were carried out in an area of
drowned Karst landscape in Apalachee Bay, off Florida in the Gulf of Mexico. The
research used seismic profiling, vibrocoring, diver surveys and underwater
excavations to reconstruct palaeo-landscape features and to successfully locate and
investigate submerged archaeological sites (Faught and Donoghue, 1997).
41
In Europe, pioneering research was carried out into the palaeo-landscapes of the
North Sea, using an extensive 3D seismic dataset that was originally collected for oil
exploration purposes, to map Holocene features across a large area of the southern
North Sea (Fitch et al., 2005; Gaffney et al., 2007). This work demonstrated how
regional-scale studies of submerged landscapes using remotely-sensed data can
reveal highly significant information about an archaeological landscape that would
have been a major habitation area in the Mesolithic period.
Also in Europe, research has been undertaken into submerged landscapes off the
north coast of Ireland using high-resolution multibeam bathymetry to map the palaeolandscape.
The result of this work was the identification of ten areas of high
archaeological potential which will be subject to more detailed analysis in subsequent
stages of the research (Westley et al., 2011a).
A significant amount of work has been carried out on developing marine classification
and characterisation projects around the UK coast. Some of these have been based
entirely on environmental attributes, such as the Irish Sea Pilot (Golding et al., 2004),
which was not specifically an archaeological or historic classification project. This
project was aimed more at ecological aspects of marine conservation, and used
geophysical and hydrographical data to map marine landscape types. The Isles of
Scilly Rapid Coastal Zone Assessment Survey (Johns et al., 2004), used historical,
archaeological, geographical, topographical and environmental data to investigate
the submerged heritage around the Isles of Scilly. The survey methodology included
extending the existing Historic Landscape Characterisation for the Isles of Scilly
42
(Land Use Consultants, 1996) to the intertidal and maritime zones. English Heritage
undertook an extensive Historic Seascapes Programme in several areas around the
English coast for the purposes of resource management (Cornwall County Council,
2008). This programme was guided by the principles of terrestrial historic landscape
characterisation, and aimed to map the historic character of the marine environment
using maps, charts and other documents together with recent marine data sources
such as bathymetry and sediment information. However, none of these projects
utilised sidescan sonar data or acoustic classification as part of their methodology.
The research carried out in different parts of the world has shown clearly that random
surveys of submerged landscapes are unlikely to yield successful results. All of these
studies used landscape-based approaches to a greater or lesser extent as the first
stage of investigation, in order to highlight areas of potential that could then be
targeted for further, more detailed study, and all of them incorporated topography and
geomorphology to some extent. The key to the success of the study in the Gulf of
Mexico lay in using knowledge of geological and topographic markers from terrestrial
archaeological sites to try and find similar settings on the seabed. In this study, such
settings included locations close to fluvial and aquatic features, including karst
features, at river estuaries and near lithic resources. This emphasises how important
the identification of palaeo-drainage patterns, buried karst features and former
coastlines is to the study of the submerged karst landscape off Northern Qatar. The
work in the Gulf of Mexico also demonstrated the value of utilising a range of different
investigative techniques as the study became more targeted.
43
2.2.2 Research Within the Arabian Gulf
The lack of research into the archaeological and palaeoenvironmental potential of the
Arabian Gulf basin is now beginning to be addressed with the pioneering work that is
being carried out by the QNHER project around the Qatar peninsula. A limited
amount of assessment work was carried out in 2010/2011 on the sidescan sonar
dataset that was collected for the Qatar Bahrain Causeway project. This is the same
sidescan sonar dataset that was used for the research outlined in this thesis. This
assessment consisted of a rapid review of the sonar contacts originally identified by
the survey company, and 83 features of potential archaeological interest were
highlighted (Cuttler et al., 2011a). Approximately half of these anomalies were diverinspected
by a team of marine archaeologists in March 2011, but all proved to be of
modern or natural origin (Al-Naimi et al., 2012). Despite the lack of significant
archaeological finds, this work proved very successful in establishing effective
methodologies for ground-truthing seabed anomalies, and in increasing
understanding of the interpretation of acoustic signatures from the seabed around
Qatar. However, it was apparent from this work that investigating geophysical
anomalies in isolation, and using low frequency sidescan sonar data as the sole
dataset, was not an effective way of understanding the submerged landscape. It was
also equally clear that random surveys would be extremely unlikely to be successful
in locating archaeological and palaeoenvironmental remains.
The QNHER project has also been conducting further marine surveys, including a
programme of high-resolution sidescan sonar survey using a Klein Hydroscan, which
44
is capable of collecting very high resolution sidescan data over large areas, and
using this to target diver inspections (Cuttler, Forthcoming). In 2012-2013, this work
led to the confirmation of the location of seven relatively recent shipwrecks off the
north coast of Qatar, which were subsequently inspected by divers. These were all
steel-hulled vessels dating to within the last 60 years. Their concentration in the north
is probably related to the fact that the principal trade route towards Bahrain and Iraq
lies to the north of Qatar (Cuttler, 2014).
Research carried out by Cuttler (2013; 2014), using high-resolution 3D seismic
survey data that was collected from the offshore Al Shaheen oil field to the northeast
of Qatar for oil exploration purposes, has identified fluvial landscape features. This
research has confirmed the presence and character of the Ur-Shatt river (originally
postulated by Lambeck, 1996) as a wide, flat valley of marshland, lake and swamps,
although it is still not clear if this was a freshwater or saline environment.
Research into human settlement in Arabia in the Palaeolithic period undertaken by
Rose (2010) has focused on Prehistoric occupation in and around the Arabian Gulf
basin. This research uses hydrological, geoarchaelogical and archaeological
evidence to support the theory that the former Gulf basin was once an oasis which
supported a sizeable human population in the Late Pleistocene and early Holocene.
There has been a significant amount of non-archaeological survey work in the Study
Area. Geophysical surveys have been carried out by commercial companies as part
of the planning phases of the proposed Qatar-Bahrain Causeway. An acoustic survey
45
was carried out in 2002 (Marin Mätteknik AB, 2002) which captured bathymetry using
a single beam echo sounder (a geophysical technique that captures depth
information), shallow seismic data using a sub-bottom profiler (a geophysical
technique which provides profiles of the upper layers of the ocean bottom), and
sidescan sonar data. However, the acoustic data from this survey was not made
available for this research. The survey company logged 107 targets in the sidescan
sonar data, and also classified the seabed in the whole of the causeway area in both
Qatari and Bahrain waters into 12 types (Marin Mätteknik AB, 2002, p.17), but no
information is provided in the report as to how the classification was carried out. Each
seabed type is summarised in a short description (two or three sentences) and given
a single-sentence description of the area location.
Further acoustic data was captured in the causeway area in 2008 for the purposes of
sand-search survey and hydrographic survey (GEMS, 2008). This survey utilised a
single beam echo sounder and sidescan sonar, and also a sub-bottom profiler in
Qatari waters. The survey company logged a total of 601 contacts in the data from
the Qatari side of the causeway project area. The sidescan sonar data collected for
this survey is the core dataset that was used for the sediment texture classification
and geophysical anomaly identification sections of this research. Geotechnical data,
including core sampling (Fugro Peninsular, 2008; QBC Consortium, 2009a, 2009b),
and an underwater video survey (Creocean, 2008) were also carried out as part of
the causeway planning phase.
46
CHAPTER 3: OVERVIEW OF METHODOLOGY
This chapter provides an overview of the theory and usage behind historic landscape
characterisation methodology, and briefly summarises the stages of the seabed
characterisation, and the data sources and techniques to be used. Detailed
methodologies for each stage of the characterisation are provided at the beginning of
the relevant sections.
3.1 Processes of Seabed Characterisation
A major aim of this research was to test methodologies for investigating the
submerged landscape. It was decided, therefore, to explore how concepts and
methods used for terrestrial Historic Landscape Characterisation (HLC) could be
adapted in novel ways and applied to submerged landscapes in the form of ‘seabed
characterisation’. The seabed characterisation aimed to combine principles of HLC,
acoustic classification and natural environment mapping, based primarily on a range
of environmental attributes, but drawing on human exploitation and interpretational
elements wherever possible.
The basic premise of landscape characterisation is to define areas ‘on the basis of
combined shared values of dominant character attributes, with secondary attributes
recorded in a consistent and structured manner’ (Cornwall County Council, 2008:
14). Terrestrial HLC can encompass topography, habitats, natural and semi-natural
features and palaeoenvironmental deposits as well as archaeological sites. It can
47
also encompass intangible and non-visual elements such as cultural and
psychological perceptions, and historical associations (Fairclough, 2001). The core
principles of landscape characterisation require it to be systematic, comprehensive,
repeatable and interpretational.
Many different methodologies have been developed and used for terrestrial
landscape characterisation (Aldred and Fairclough, 2003). The methodology used for
this seabed characterisation follows the ‘multi-mode’ approach (Aldred and
Fairclough, 2003, p.18), a combination of attribute-led description followed by
interpretative classification. In this approach, the entire Study Area is divided into
initial landscape units on the basis of one or more core attributes. The basic
classification can then be built on and supplemented by gathering data on a range of
natural environmental and cultural attributes which can then be assigned to each
unit. Whilst environmental data can be considered to be more objective and
consistent, the human and interpretational dimension is also an important factor in
providing a holistic understanding of the landscape. When discussing attributes, it is
worth noting that the differentiation between ‘environmental’ attributes and ‘cultural’
attributes is slightly blurred, since environmental attributes can both influence and
reflect cultural attributes, and vice versa. It is the ability to accommodate the interplay
of these linked factors that is one of the strengths of the seabed characterisation
approach.
The characterisation types are defined through analysis and identification of the
trends and combinations in which certain attributes occur. Further analysis is then
48
undertaken to allocate a characterisation type to each unit, according to which type
best represents its predominant landscape character. This approach is a very flexible
and transparent method, allowing the generation of many different classifications and
interpretations, and the creation of intermediate theme mapping which may be used
as individual information sources (for example the extents of a particular type of
seabed habitat).
The factors influencing the character of the seabed can be classified as primary (or
core) attributes, which form the basis for the division of the seabed into initial
landscape units, and secondary attributes, which do not necessarily provide
seamless, continuous coverage over the landscape, but can be used as supporting
data sources to further refine and calibrate the initial landscape units. The procedure
is recursive, with the data collection stages and data analysis stages having the
potential to inform each other, and the character areas can be gradually refined with
each new stage of data input and analysis (Dingwall and Gaffney, 2007). Some
patterns will emerge in relation to Prehistoric settlement, and others in relation to
maritime exploitation and shipwrecks. This process of calibration and refinement is
the essence of the seabed characterisation. The end result of this process is the
identification of areas formerly favourable to human occupation, and areas
expressing potential for the preservation of palaeoenvironmental and archaeological
remains. This latter process takes a step beyond first-stage historic landscape
characterisation, as initial landscape characterisation does not normally seek to
provide value judgments on the landscape under investigation, the core principle
49
being that every part of the landscape is classifiable and in need of changemanagement
to some degree.
The core data sources used for the primary characterisation were low-resolution
sidescan sonar data and LiDAR bathymetry data. These datasets covered extensive
areas of the seabed and were deemed suitable for the primary classification of the
seabed. The aim was to undertake the primary classification by carrying out sediment
texture classification using the sidescan sonar data, and mapping of seabed
topography using the LiDAR bathymetry. This would be further informed by groundtruthing
data from direct sediment samples and video footage. The secondary
classification would then be undertaken, and would consist of the analysis of
individual geophysical anomalies within the sidescan sonar data, and the further
targeted investigation of selected anomalies via high-resolution geophysical survey
and diver inspection, in order to support the primary classification. The end result of
this characterisation process would be a set of contiguous polygons covering the
entire Study Area, representing broad character zones, which could be used to
assess potential and inform further research.
3.2 Primary Characterisation
3.2.1 Sediment Texture Classification
The first step of the seabed characterisation process is to establish how the seabed
can be divided up into initial landscape units in the most effective and objective way.
50
Sidescan sonar is an efficient and rapid method of creating images of large areas of
seabed based on the acoustic ‘backscatter’ reflected from the sonar beam. It can be
used to detect objects that are either raised from, or sunken into, the seabed, and
also to identify distinct changes to the material and texture type of the natural seabed
(Blondel, 2007). The sidescan sonar data available for the Study Area provides
continuous, seamless coverage, and this type of acoustic backscatter data has been
successfully used elsewhere in the world for seabed sediment classification, largely
for the purposes of natural habitat mapping. Since the seabed sediment usually
reflects the character of the underlying substrate, and is therefore a fundamental
characteristic of the seabed, sediment classes are a suitable primary attribute that
can be used for the division of the seabed into initial landscape units. The acoustic
backscatter data captured by the sidescan sonar can be used to aggregate areas
(classes) of similar acoustic signature (Penrose et al., 2005), the aim being to reflect
the variability in the nature of the seabed sediment, and highlight landscape-scale
changes.
3.2.2 Topographic Mapping
The topography of the seabed (bathymetry) is also a major influence on the character
of the seabed, although in this characterisation, the aim was not to use bathymetry
as the primary dataset for the initial determination of landscape units, but rather to
use it to inform and refine the initial units. A surface model can be created from
bathymetry that can then be visually inspected for evidence of topographic features
that formed when sea-levels were lower. Seabed topography, like sediment
51
character, is, at least partially, a reflection of the underlying geology, and provides
the ability to highlight landscape-scale features that would have been attractive for
human settlement, such as former shorelines, and/or features that are conducive to
the survival of palaeoenvironmental caches, such as palaeochannels. The shape of
the seabed evolves through factors such as erosion, deposition, the energy of the
sea and the hardness of the underlying rock. However, the present day seabed is not
going to be a direct reflection of the palaeo-land surface, as sedimentary processes
associated with marine transgression are likely to have buried the past landscape in
some areas and eroded it in others. We need to consider the processes that would
have either protected or eroded land surfaces as sea levels rose. Also, any study of
seabed mapping and classification has to take into account the mobility of features
and sediments, since seabed features such as sand waves and sand ripples are not
static features. However, the aim of this part of the research was to define broad
areas where conditions are suitable for features of high archaeological and
palaeoenvironmental potential to be created and maintained, and use these to inform
and refine the character areas.
Sea level changes are critical, since they can be used, together with the topographic
data, to locate and define former shorelines. Former shorelines are highly significant,
both in relation to patterns of human settlement, and in relation to preservation
factors. The marine incursion since the last glacial maximum was not a steady
process, with periods of both slow and rapid incursion, and also periods of stillstands.
The lack of reliable regional sea-level curves, and the debate surrounding the
effect of tectonic movement and isostatic loading movements (Lambeck, 1996;
52
Uchupi et al., 1999), means that it is not possible to be exact about the timing of
marine transgression within the Study Area. However, we have enough detail for the
purposes of broad-brush seabed characterisation.
Marine taphonomic processes are also very important in this respect, and it is critical
to determine the types of changes which might affect the form and condition of
submerged sites prior to, during and after the rise in sea level (Faught and
Donoghue, 1997). This is particularly important, since we currently have very little
information about how sedimentary regimes and taphonomic processes in the Gulf
affect potential archaeological remains. Different depositional environments and
preservation factors are likely to apply to Prehistoric sites as opposed to more recent
wrecks.
3.2.3 Ground-Truthing
Ground-truthing data is essential for characterisation undertaken via acoustic
classification, as it provides validation for the initial classes. The most common ways
to ground-truth acoustic classification data are direct sediment sampling and
photography. The aim for this characterisation was to undertake a sampling
programme over as much of the Study Area as possible, using a Van Veen grab
sampler, and then to subject the grab samples to texture analysis (granulometry) in
order to test the correlation between the acoustic classes and the sediment
characteristics. A further aim was to supplement the ground-truthing element of the
seabed characterisation by utilising geotechnical data that was collected for the
53
planning stages of the Qatar-Bahrain Causeway project (Fugro Peninsular, 2008,
QBC Consortium, 2009a; 2009b), together with underwater video survey transects
that were taken for benthic habitat assessment purposes, also for the Qatar-Bahrain
Causeway project (Creocean, 2008) .
3.3. Secondary Characterisation
3.3.1 Identification of Geophysical Anomalies
The sidescan sonar data can also be used to identify geophysical anomalies that
may represent areas of archaeological significance. The influence of Prehistoric
human exploitation of the landscape and the subsequent marine environment, as
represented by physical remains on the seabed, is a very important interpretational
aspect of the characterisation, and this is where identifying geophysical anomalies
from the sidescan sonar data plays a key role. The earliest archaeological remains
could consist of relict land surfaces and palaeochannels, as well as buried
archaeological sites. Cairns dating from 6,000 years ago, constructed out of rubble,
and measuring 1-2 m high and 5m across, are present in the terrestrial Qatari
landscape. Building in stone was being undertaken at ‘Ubaid-related sites around the
Gulf shores 6,500 years ago, as documented by sites such as Ras Abaruk 4b, Qatar
(De Cardi 1978, p182), where traces of stone structures were found, As-Sabiyah H3,
Kuwait (Carter and Crawford, 2010), where a series of stone structures were
excavated, and Marawah Island, Abu Dhabi (Beech et al., 2005), where stone walls
up to 0.7m in height survived. It is reasonable to suggest that comparable stone
54
structures would have been constructed by the populations occupying the former
Gulf basin in the period preceding the final phase of marine incursion. Such
structures as these would be large enough to be visible on sidescan sonar, but the
likelihood of this would depend on subsequent depositional and erosional processes.
Features would have to have been buried rapidly in order for them to have been
preserved, but would also require some level of subsequent erosion in order to be
detected by sidescan sonar.
After the most recent marine transgressions, archaeological remains will consist of
shipwrecks, lost cargoes and other marine debris. The aim was to try and reflect
these factors through thorough analysis of all geophysical anomalies visible in the
sidescan sonar data, and use this analysis to inform the primary characterisation.
55
3.3.2 Clarification of Geophysical Signatures
As the sidescan sonar data used for the research was low-resolution data, it was
expected that selected geophysical anomalies may need further clarification, via a
targeted programme of high-resolution geophysical survey and diver inspection, in
order to try and further clarify geophysical signature types.
3.4 Defining Character Areas and Assigning Potential
The final stage of the characterisation process is to integrate the analyses from all
the different data sources and techniques. Once all the data is collected and
integrated, it is possible to analyse and seek consistent patterning in the data, and
use this to refine the landscape units into character areas, These character areas
should reflect broad environmental trends, which will have had an influence on past
human activity in the landscape.
It is essential that part of the interpretational aspect of the seabed characterisation
comprises the study of known human settlement patterns in the coastal region of
Qatar in the Mid-Holocene, when sea levels were approximately 2-3 m higher than
today (Lambeck, 1996, p.50), and the subsequent utilisation of this knowledge to
inform pre-transgression settlement patterns in the submerged landscape. The
successful discovery of submerged archaeological sites depends on applying our
knowledge of geological and topographic markers from terrestrial archaeological
sites to find analogous settings on the seabed.
56
One of the advantages of utilising landscape characterisation approaches is that the
recursive nature of the characterisation process allows the integrated data to be
manipulated in different ways for different purposes, by combining attributes to derive
new information, and using the results to feed back into the characterisation again. In
this way, the resulting character areas can be used to generate zones of
archaeological and palaeoenvironmental potential. Preservation factors and human
settlement patterns are key to this, and certain locations score relatively highly both
in terms of being formerly favourable locations for human settlement and for having
high preservation potential in terms of the probable presence of extensive
sedimentary deposits.
57
CHAPTER 4: PRIMARY CHARACTERISATION
The following chapter describes the methodology and results of the primary
characterisation. This comprises three different strands - the sediment texture
classification using the sidescan sonar data, the topographic mapping using the
bathymetric LiDAR data, and the ground-truthing data using the direct sampling and
visual observation data. Each strand has three sections consisting of the detailed
methodology, the results and a discussion.
4.1 Sediment Texture Classification
The sediment texture classification was undertaken by applying acoustic
classification techniques to the available sidescan sonar data.
4.1.1 Methodology
4.1.1.1 Overview of Sidescan Sonar
The sidescan sonar data collected for the Qatar-Bahrain Causeway project has
provided an image of the seabed in the Study Area, which can be used as the basis
for landscape-scale investigations. The data consisted of 229 files in xtf format, each
representing a single survey line and together covering 365 square kilometres of
seabed on the Qatari side of the Causeway Project area. Sidescan sonars are either
towed by ships or hull-mounted (Figure 7), and carry sideways-looking transducers
58
which emit sound pulses, giving an image of sound (sonograph) reflected from the
sea floor beneath, and on either side of, the survey vessel (Jones, 1999, p.37).
Figure 7 Sidescan sonar operation (from Al-Naimi et al., 2012).
A range of different frequencies can be used to give different resolutions, but as with
most geophysics, there is a trade-off between resolution and coverage. Higher
frequencies give higher resolution data, but attenuate more quickly with range than
lower frequencies, so the widths of the survey lines have to be smaller than with lowresolution
surveys. The survey dataset for the Study Area was acquired using single
channel sonar at a frequency of 325 Kh in the Qatar-Bahrain Causeway’s custom
coordinate system (GEMS, 2008). This resolution is relatively low, and the data
quality is therefore not as good as could have been obtained with a higher resolution
survey, so accurate interpretation of some features could be potentially more difficult.
59
However, low-resolution surveys enable quicker data capture, thus enabling a larger
area of seabed to be surveyed within the time and resources available. In order to
ensure comprehensive coverage of the seabed, there should be sufficient overlap
between adjacent survey lines, ideally a minimum of 100%. This is because sidescan
sonar systems have two side-facing transducers that cannot ensonify the seabed
directly beneath the towfish, so an overlapping adjacent swath is necessary to
achieve complete seafloor coverage. It is also good practice to overlap survey lines
in different directions, as the appearance of certain features can vary considerably
when seen from different angles (Blondel, 2007). In the dataset for the Study Area
there is very little overlap in the survey lines, they are mainly contiguous, except for
the northern part of the Study Area (Area 1) where there are actually gaps between
the survey lines (Figure 8).
Figure 8 Coverage of sidescan sonar survey lines in the Study Area.
60
Although the low resolution and the lack of overlap in the survey lines are recognised
limitations on the data, a large-scale, low-resolution dataset such as this is suitable
for seabed characterisation when used in combination with other datasets, since
characterisation is intended as a ‘broad-brush’ study, designed to identify areas of
high potential and target them for more detailed investigation, rather than focusing on
the detail of individual, small-scale features .
4.1.1.2 Acoustic Techniques for Sediment Texture Classification
Sediment classification was undertaken using the sidescan sonar data as the base
data source. The use of acoustic techniques for large-scale seabed classification and
benthic habitat mapping was established in the mid-1990s (Anderson et al., 2008),
initially using backscatter data from sidescan sonar, and more recently multibeam
echo sounders (MBES), and their use is increasing as the technology evolves to
allow larger areas to be surveyed more cheaply at higher resolutions. Multibeam
systems produce contoured bathymetry from beneath and to the sides of the survey
vessel, and hybrid multibeam systems combine both sidescan sonar and bathymetry
(Jones, 1999, p.37).
Acoustic classification consists of classifying the area of surveyed seabed into similar
and distinct regions based on the characteristics of the acoustic backscatter. The
theory behind it is that classified acoustic data can be used as a proxy for sediment
classification, based on the fact that the character of the backscatter is influenced by
the roughness of the surface, and therefore the ‘roughness’ of the backscatter is an
61
indication of the roughness of the physical sediment. The seabed sediment, in turn,
usually reflects the character of the underlying substrate. Generally, hard substrate
such as exposed rock generates a high backscatter value, whilst soft material such
as mud or silt will produce a low backscatter value (Penrose et al., 2005).
The big advantage of this method over photographic and direct sampling methods for
mapping seabed sediments is the ability to obtain rapid coverage of large areas of
seabed, without the big spatial gaps in the data that photography and discrete
sampling inevitably leave. However, the classified acoustic data alone can only
provide preliminary baseline mapping, and needs to be analysed in conjunction with
ground-truthing data, usually obtained by direct sampling and photography. Although
ground-truthing is a necessity for a valid and robust classification, and can be timeconsuming
to collect, the attributes obtained from well-targeted ground-truthing sites
can be potentially applied to large areas of seabed, even to areas which may be
remote from the ground-truth sites (Preston, 2009). Depending on the classification
methodology, the classified dataset can be used to guide the ground truthing, and
also further, more detailed acoustic survey.
In the past, interpretation and classification of the images of the seabed obtained
from the acoustic data was usually carried out by skilled, human visual interpreters.
Although this is a relatively subjective process, it has been shown to be effective in
areas where adjacent seabed types are well-differentiated, or where specific seabed
features produce distinctive backscatter. However, in areas of heterogeneous
seabed, or where the seabed character changes gradually without clear
62
differentiation in the backscatter response, visual interpretation becomes more
difficult, and other methods have to be investigated (Collier and Brown, 2005).
Increasingly, automated processes are being explored, and although generally more
objective, these processes can provide new and different challenges. Whilst
automated classification can be done more rapidly than visual interpretation, more
intensive data cleaning and preparation is required before classification is attempted.
This is because the impact of image artefacts such as beam-pattern effects, which
are easily detectable and excluded from the classification by human interpreters,
needs to be minimised if coherent automated classification results are to be obtained,
(Preston, 2009, p.1277).
The classification carried out for the Study Area was ‘unsupervised’ classification.
This type of classification is carried out without any foreknowledge of the classes to
be created, and can be used in areas where there is little or no ground-truthing data
(such as sediment samples or underwater photographs). This is different to
‘supervised’ classification, which can be undertaken where there is sufficient groundtruthing
data available from the survey area to create training data that can be used
to classify the entire dataset (Ellingsen et al., 2002).
4.1.1.3 Trialling the Classification Methodology
Trials were carried out using different software to try and establish the best method
for classification and interpolation of the data. The first trial used the relatively newlydeveloped
seabed classification module in Chesapeake’s SonarWiz software. The
63
raw data (xtf format) from the parallel tracks covered by the survey boat were
imported into the software, batch processed and used to create a mosaic image of
the seabed. The classification module was then run on the mosaic data a number of
times using a variety of parameters, including using different gain settings, varying
window sizes, trimming the sidescan swaths by varying amounts, using different
texturers (statistical processes), setting different window sizes and applying different
filters, but no coherent results could be obtained; the classification module was
unable to detect enough differentiation in the data (Figure 9). It is not clear why the
results were so poor, given the amount of testing that was undertaken with different
parameters. It is possible that there may have been issues with the processing of the
data, and also at the time of use in 2011, the classification module had only just been
released and was still being refined. It was decided that the limited time available
could be more usefully spent trialling other software to see if better results could be
obtained.
Figure 9 Results of a trial of Sonarwiz's seabed classification module.
The second trial was undertaken using Erdas Imagine image processing software.
The mosaic image that was initially created in SonarWiz was exported as a GeoTiff
and imported into Erdas Imagine. Unsupervised classification was then carried out on
this image, but again no coherent results could be obtained, and the survey lines
were all classified as the same complex and unusable mix of pixels (Figure 10).
64
Figure 10 Results of unsupervised image classification in Erdas Imagine.
The third trial was carried out using a specialised seabed classification software
called Swathview and a specialised interpolation software called CLAMS, both
developed and supplied by Quester Tangent Ltd. These two software packages were
very kindly made freely available for a trial period by Quester Tangent, purely for the
purposes of this research. The Swathview software classifies the data using the
statistical properties of the backscatter images. It allows the selection of an optimal
number of classes relating to the composition and diversity of the seabed and
assigns records to those classes using principle component analysis and clustering
algorithms (Keller, 2011. p.16). Unlike the previous two trials, which used the mosaic
data for the classification, Swathview uses the data in image space (Preston, 2009).
The advantage of this is that analysis can be carried out with reference to the original
geometry with which the data were acquired, such as survey direction and beam
angle (Penrose et al., 2005). Fonseca et al. (2009, p.1300) consider that the
processing and choices required to create a mosaic image of backscatter data (such
as AVG correction methods and normalisation angles) are necessarily subjective,
and therefore visual or pixel-based analyses may not be the best approach to seabed
classification. They argue that the inherent angular response should be used instead.
65
The CLAMS interpolation software was specifically designed for the mapping and
visualization of seabed classification data. This software differs from other image
processing software in that it can perform categorical interpolation of the spatial data.
This means that it does not interpolate the data as though it were a continuous
variable, and it does not use intermediate class numbers, since the numbering
sequence in the seabed classification is arbitrary (Preston, 2009). This categorical
interpolation, using the mode rather than the mean of the points for the interpolation,
gives much more coherent results than that produced by more general image
processing software. After initial trials on subsets of the data, it was clear that the
results obtained using Swathview and CLAMS were far better than those obtained in
either of the trials that used the mosaic data, so it was decided to use this software
for the entire classification.
4.1.1.4 Classification using Swathview
The basic steps of classification using Swathview consist of importing the raw data,
data cleaning, image compensation, generation of image statistics, reducing
dimensionality, clustering and mapping (Preston, 2009). The Swathview software
performs most of the classification automatically, but as with all image processing, it
is necessary for the user to make decisions regarding various parameters that need
to be set, and a considerable amount of trial and error is therefore involved in order
to establish the optimum settings. It is beyond the scope of this thesis to provide a
detailed explanation of the complex statistics behind the processing that is carried
out by the software (Quester Tangent, 2012a), but an overview of the processing
66
steps is necessary in order to demonstrate that sufficient testing was carried out to
create a robust classification.
Various parameters can be set for data cleaning, including range masking and angle
masking of data. This step is crucial in order to remove along-track artefacts and
avoid striping in the classified data. The most significant decision to make is the size
of the rectangular sub-images that the software generates during the classification
process (Figure 11). The generation of image statistics (called features in Swathview)
is carried out within these sub-images or rectangles. Amplitude and texture are
analysed within each rectangle, and the centre pixel in the rectangle becomes a data
point position for that class. If more rectangles are generated, more data points are
created, leading to a higher image resolution, and consequently a longer processing
time (Personal Communication, Tony Tipple, Quester Tangent Ltd, 2012).
Figure 11 Sub-Images (rectangles) generated by Swathview for the purpose of generating image
statistics.
67
The features generated are merged into a catalogue file and principle component
analysis is used to reduce the features to the first three principal components. Once
built, the feature catalogue can be applied to whole datasets, or parts of them, to
create classified seabed files. Cluster analysis is then performed on a range of
classes, chosen by the user, who can select the optimal number of classes once the
analysis is complete. The final product is a georeferenced map of the classified data
points.
Figure 12 A three-dimensional plot of the clusters generated during the classification process in
Swathview.
The resulting georeferenced point data can then be interpolated using the CLAMS
software, to create a continuous raster dataset. Again, systematic experimentation
has to be carried out using different settings and parameters in order to obtain the
best results for the raster dataset. The most significant decision to be made is the
grid spacing for the interpolation, as this influences the resolution of the final image.
68
The raw sonar data, in xtf format, was batch-imported into Swathview. Following
initial testing of settings and parameters, a preliminary classification and interpolation
was carried out on the entire dataset, and the results analysed. Comparative testing
was then carried out on smaller areas in order to test the robustness of the
classification results. This was done by running the classification process, using the
same parameters as those used for the entire dataset, on two subsets of the data,
one from Area 3 and one in Area 4. A catalogue of features was generated from each
test area, and each one of these catalogues was used to classify each test area, and
also to classify the entire dataset (Figure 13). The settings used for this preliminary
classification, the test area files, and the interpolation, are listed in Appendix 1.
Figure 13 Results of the Swathview preliminary seabed classification using catalogues based on the
entire dataset, and using catalogues based on trial areas.
69
The results of the preliminary classification and the trial areas showed that although
there were small discrepancies, the same broad trends were apparent in all of the
classifications. Once the general validity of the method had been established by the
preliminary classification and the trial areas, further experimentation was undertaken
to identify the settings and parameters that gave the best results. Valuable feedback
was provided by the expert staff at Quester Tangent during this stage of the process.
The preliminary classification indicated that more aggressive angle and range
masking needed to be carried out to reduce the along-track striping that was visible
in the georeferenced plot. This was always going to be a problem in this particular
sidescan sonar dataset as there was no overlap in the survey lines, and in some
areas there were gaps between the survey lines. The expert interpolation across data
gaps that is provided by the CLAMS software allows for this type of aggressive
masking-out of data. Alterations were also made to the rectangle size, so that smaller
and more square-shaped sub-images were created.
The settings for the CLAMS interpolation were adjusted, so that the search radius
was increased to close up gaps in the interpolated image, and the search size was
increased, thus decreasing the amount of fine detail (Quester Tangent, 2012b). The
full classification and interpolation process was then run again on the entire dataset,
and also on the smaller test areas, using the altered parameters, to obtain the final
classification data. The parameters used for the final classification are also included
in Appendix 1.
70
After the final classification was complete, it was necessary to choose the optimal
number of classes that would be used for the final classification dataset. On the basis
of the statistics generated by Swathview, ten was selected as the optimal class
number, so the seabed was ultimately divided into a very large number of data points
(more than 2 million), each one of which was allocated to one of ten distinct classes.
The resulting track plot displays the geographical position of data associated with
each class (Figure 14).
Figure 14 Track plot from Swathview showing classified data points (left) and interpolation of data
points using Clams (right).
The software also generates confidence data for each of the classified points,
showing the probability that each point belongs to the class to which it has been
71
assigned. Using this data, it was possible to verify that for the chosen number of
classes, the majority of the points had a high level of confidence, and therefore that
the classification is robust. Out of a total of 2,033,711 data points, 85% of them had a
confidence level of 75% or more, and 51% had a confidence level of 98% or more.
The georeferenced map of the classified data points was converted to a continuous
image file using the CLAMS interpolation software (Figure 14), which was then
imported into ArcGIS, and analysed. This raster surface was used to create
contiguous polygons that formed the initial landscape units for the seabed
characterisation.
A considerable amount of testing was carried out using different parts of the dataset
and comparing the results in order to ensure that the resulting classification was as
robust as possible. Since the classification was unsupervised, the classes needed to
be validated by ground-truthing in representative sample areas, in order to more fully
understand what type of seabed sediment was represented by each class. However,
the classification carried out on the acoustic survey data provided a preliminary
dataset which could then be further refined using a range of other data sets.
4.1.2 Results
As outlined in the methodology section, following extensive trials, the acoustic data
from the sidescan sonar survey was eventually classified into ten acoustically distinct
classes, and the output of this process was a georeferenced raster image of the
72
Study Area. The colours allocated to the different classes in the raster image do not
have any significance in regard to their interpretation, they have been allocated solely
based on the best way to visualise the data. The preliminary interpretation of the
classes is summarised in Table 1 below.
Table 1: List of acoustic classes generated by Swathview, and their preliminary interpretation.
Class No. Colour in Classified
Image
Preliminary Interpretation
1 Dark Green Finer Grained Sediment (Lower Reflectivity)
3 Red Coarser Grained Sediment (Higher Reflectivity)
4 Yellow Finer Grained Sediment (Lower Reflectivity)
5 Lilac Medium Grained Sediment (Medium Reflectivity)
6 Brown Coarser Sediment (Higher Reflectivity)
8 Lime Green Coral (High Reflectivity)
9 Salmon Finer Sediment (Lower Reflectivity)
10 Blue Medium Grained Sediment (Medium Reflectivity)
The analysis is based on the premise that sandy environments tend to be good
reflectors of acoustic energy, while silty/muddy environments tend to be poor
reflectors of acoustic energy (Sutherland et al., 2007). However, at this point it is not
useful to undertake too much analysis based solely on the classified data, as unless
there is more information about what the classes actually represent, then the analysis
will have little meaning for archaeological purposes. The real strength of the
classification will be demonstrated later in this chapter when integrated analysis of
different datasets is undertaken, in order to zone the seabed on the basis of
73
character and potential. However, before carrying out the integrated analysis it is
useful to summarise the general trends displayed by the classified data, and to
examine how the classified data compares to the mosaic image of the sidescan data.
The crucial issue when analysing the classification of an extensive area of seabed is
to treat it as a broad-brush technique, viewing the image in its entirety, and not to use
the classified data for fine detailed analysis, as that is not its purpose. In this respect
the classification was successful, in that the overall image appears coherent, and
dominant trends in the data can be picked out. There are clearly large, discrete areas
where a single class is dominant, especially in the south and centre of the Study
Area. The picture is more complex in the north of the Study Area, but this was
expected, as there were gaps in the survey lines in this area, making it more difficult
to achieve a sensible automated classification. The lack of overlap in the survey lines
has led to striping in the classification data that could not be completely removed by
pre-classification processing, but it was minimised as far as possible (see
methodology section). The different regions of the Study Area are analysed below,
starting from the south and moving northwards.
74
4.1.2.1 Area 4
Figure 15 Classified data (left) and mosaic of sidescan sonar data (right) in Area 4.
75
Classes 4 (yellow) and 5 (lilac) fill the majority of Area 4 (Figure 15). The large yellow
area (Class 4) in the southwestern part can be relatively clearly seen in the sonar
mosaic, in which it appears as a darker, possibly finer sediment, demarcated from the
surrounding area by an edge of brighter reflectivity (Figure 16). Extensive trawler
scars are evident all over this area, and the acoustic classification has clearly picked
the area out as separate and distinct from the surrounding area.
Figure 16 Edge of brighter reflectivity demarcating a change in texture in Area 4.
In the very far southwest corner, brighter patches, possibly indicating coarser
sediment and/or hard, rocky substrate, are apparent (Figure 17).
76
Figure 17 Trawler scarring in Area 4.
Further north in Area 4, closer to the reef, although there is an area of predominantly
Class 9 (salmon) in the classified data, it looks very mixed at this point, probably
reflecting the complexity of the seabed in the vicinity of a coral reef, where the
acoustic classification is able to discern differences that aren't visible to the human
eye. The mosaic image here, although it has darker, less reflective areas of possibly
finer sediment (represented by Class 9 in the classified data), looks brighter in
patches, indicating rougher, harder reflectors such as coral and rocks, in a roughly
triangular-shaped area. This triangular shape also emerges as an overall pattern of
mixed classes in the classified image (Figure 18).
77
Figure 18 Pattern of mixed classes south of the reef in Area 4 corresponding with visual patterns in the
mosaic data.
The area to the south has been classified predominantly as a mix of Class 5 (lilac)
and Class 10 (blue), with a band of Class 3 (red), which is an area of brighter
reflectivity and therefore possibly coarser sediment, running up the centre. At first
view, this area looks relatively homogenous in the mosaic image, but the
classification has picked out differences that can also be seen on closer inspection of
the mosaic image, which shows an area of northwest-southeast trending sand ridges
in the red area (Figure 19).
Figure 19 Sand ridges visible in the mosaic data in Area 4 and differentiated as Class 3 in the
classified data.
78
4.1.2.2 Area 2/3
Figure 20 Classified data and mosaic of sidescan sonar data in Area 2/3.
79
Class 3 (red) dominates Area 2/3, in the centre of the Study Area (Figure 20). There
is a gap in the data in the west of Area 2/3 due to the reef, which could not be
surveyed with the sidescan sonar as the water was too shallow for the survey vessel.
The area immediately to the west of the reef is clearly picked out by the classification
as a large area of lime green (Class 8). The same class also appears as small
patches ringing the reef, as well as odd outcrops elsewhere, indicating that it is coral.
This is where the automated classification really shows its strengths, as the large
area of coral does not show up particularly clearly as a separate class in the mosaic
image, (Figure 21) although the smaller outcrops do show up as bright reflectors.
Figure 21 Mosaic of sidescan sonar data around the reef in Area 3.
80
The area to the east of the reef is predominantly Class 3 (red) and Class 5 (lilac). The
red areas in the classified data seem to correspond to brighter reflective areas of
north-south trending linear bands in the mosaic data, possibly sand ridges, whereas
the lilac areas correspond to less reflective, possibly finer sediment (Figure 22).
Figure 22 Sand ridges visible in the mosaic data in Area 2 and differentiated as Class 3 in the
classified data.
Immediately to the north of the reef, the classified image again shows a rather
complex picture. North-south linear bands here have been picked out as Class 6
(brown) (Figure 23) rather than Class 3 (red), as they were further east. The less
reflective area to the west has been classed as predominantly Class 9 (salmon),
whereas a similar-looking area further to the east was classed as Class 5 (lilac). The
automated classification here appears to be picking out differences between classes
that actually look quite similar to the human eye.
81
Figure 23 Sand ridges visible in the mosaic data in Area 3 and differentiated as Class 6 in the
classified data.
82
4.1.2.3 Area 1
Figure 24 Classified data and mosaic of sidescan sonar data in Area 1.
83
As previously mentioned, the classification in the north of the Study Area (Area 1 Figure
24) appears very mixed, particularly the area in the far north, and it is
therefore necessary to categorise this area based more on discrete areas of mixed
classes rather than areas of single classes.
The area to the north of the reef appears in the classified data as predominantly
Class 3 (red) with green patches (Class 8). Although it is quite difficult to see area
differentiation in the mosaic image where there are gaps in the survey lines, this area
does show up as a more brightly reflective area than the area further to the north,
suggesting that it could consist of coarser sediment, and/or rock and coral, which
would fit with its location close to the reef (Figure 25).
Figure 25 Brighter reflective area immediately north of the reef in Area1.
84
The most obvious feature in Area 1 is an extensive and relatively cohesive
southwest-northeast trending band of classes 4 (yellow) and 9 (salmon), with a few
patches of Class 1 (dark green) contained within it (Figure 26). It generally appears in
the mosaic image as an area of slightly lower reflectivity than the surrounding areas,
possibly because it is finer grained, and contains less rocks and coral.
Figure 26 Large band of lower reflectivity in Area 1, differentiated in the classified data mainly as
Classes 9 and 4.
This is flanked to the north by an another extensive area that, although quite mixed,
predominantly consists of classes 5 (lilac) and 10 (blue). This latter area manifests
itself in the mosaic image as containing large-scale, north-south aligned linear
features (Figure 27), which are probably sand ripples.
85
Figure 27 Sand ripples in the north of Area 1, visible in the mosaic data.
In the southeast of Area 1 is another quite mixed area that is largely composed of
Class 3 (red) , merging into patches of Class 6 (brown) in the north and mixed with
patches of Class 8 (lime) in the south. This area can be differentiated in the mosaic
image as a concentrated area of bright, reflective curvilinear features which would
appear to be sand waves and/or reef structures (Figure 28).
86
Figure 28 Sand banks in the southeast of Area 1, visible in the mosaic data and differentiated as
Classes 3 (red) and 8 (lime) in the classified data.
On initial analysis using solely the mosaic sidescan data, the classes generated by
the acoustic classification process largely seem to be coherent and logical. The red
(Class 3) and brown (Class 6) areas roughly correlate with areas of brighter
reflectivity, suggesting that these areas represent coarser sediments, often
87
containing areas of sand ripples. The lime green areas (class 8) coincide with the
reef, and with other bright reflectors dotted around the mosaic data, suggesting coral
reef and coral/rocky outcrops. The salmon (class 9) and yellow (class 4) areas
correlate with less reflective areas of seabed, suggesting finer sediments, and no
rocky substrate. The lilac (class 5) and blue (class 10) areas indicate finer-grained
sediment than that in the red and brown areas, but not as fine as that in the salmon
and yellow areas. However, it is not immediately clear, solely on the basis of the
mosaic data, as to why the automated classification process has classed the blue
and lilac areas separately, since it is difficult to distinguish much difference between
these areas in the mosaic data. The dark green areas (Class 1) are harder to
interpret as they do not appear as large cohesive areas anywhere, but rather appear
mixed in with other classes. However, they occur most frequently with classes that
have been interpreted as finer-grained sediments, so it seems logical to assume the
green areas also represent a class of finer sediment, albeit with some differences
that are not discernible by the human eye in the mosaic data.
4.1.2.4 Generating Initial Landscape Units
As part of the seabed characterisation process, the classified raster file was
polygonised to create initial landscape units, in order to carry out integrated data
analysis with other data sets. Initially these polygons were created by automatically
converting the raster image to polygon format, which resulted in an enormous and
unworkable amount of polygons (more than 39,000), many of which were very small,
and not suitable for use as base landscape units. Accordingly, the data was
88
iteratively processed to merge small polygons into neighbouring polygons. However,
although this was an objective and repeatable method that resulted in polygons of a
sensible size, it did not always produce a good representation of the dominant class
within each merged polygon in areas where the classes were very fragmentary. A
different methodology for creating the polygons was then trialled, where the raster
image file of the seabed classes was used as the basis for manually digitising
polygon boundaries, mostly based on single classes. However, in areas where the
classes were very fragmented and mixed, visual judgement was used to define
polygons based on the dominant class within a mixed area. This method proved to
be the most successful at capturing coherent and useable base landscape units.
Although by using this method the initial landscape units were created manually, they
were still based on the automated classification undertaken using Swathview, which
had picked out differences in the acoustic data that could not be seen by manual
visual examination of the unclassified sidescan data. A total of 90 individual polygons
were created as a result of this process (Figure 29 and Table 2). Also, as a result of
refining the data into polygons of a more useable size at a landscape level, two of the
classes, which only covered negligibly small parts of the Study Area, were removed
(Classes 2 and 7).
89
Figure 29 Initial landscape units based on classification of acoustic backscatter.
90
Table 2: Initial landscape units - polygon statistics.
Class No. No. Polygons
Total Class
Area
(Sq Km)
Average
Polygon Size
(Sq Km)
Max Polygon
Size
(Sq Km)
Min Polygon
Size
(Sq Km)
1 7 31.10 4.44 10.44 0.18
3 14 74.97 5.36 30.23 0.19
4 4 32.41 8.10 11.93 1.97
5 22 111.88 5.09 20.86 0.22
6 3 6.45 2.15 4.75 0.71
8 5 13.84 2.78 10.32 1.13
9 15 37.31 2.49 7.76 0.50
10 20 48.26 2.41 18.51 0.19
TOTAL 90 356.22 3.96
4.1.3 Discussion
Following on from the results of the acoustic classification, it is necessary to critically
examine how valid the acoustic classification methodology is in the context of
previous work in this field. Sidescan sonar backscatter data has been used for more
than 30 years for seabed mapping, and more recently, multibeam backscatter data
has also been used. Considerable research has been undertaken in the last decade
into assessing different seabed classification methodologies using acoustic
backscatter, although almost all of this work has been done for the purposes of
geological and/or benthic habitat mapping, rather than for exploring archaeological
potential (for example Penrose et al., 2005; Anderson et al., 2008; Brown and
91
Blondel, 2009; Van Rein et al., 2011). It should also be noted that none of this work
has been done in the Arabian Gulf.
Although there are studies that have produced inconclusive results (Kostylev et al.
2001, cited in Orpin and Kostylev, 2006), the published research generally supports
the theory that there is a correlation between acoustic backscatter and seabed
sediment characteristics (Brown and Blondel, 2009). Collier and Brown (2005)
undertook a study to examine the dependence of acoustic backscatter on sediment
grain size distribution using sidescan sonar data and sediment grab samples from
the Loch Linnhe artificial reef site on the West Coast of Scotland. Overall, the results
did support previous studies that have suggested that the highest backscatter tends
to correspond with the coarsest sediments, although they also found some issues
with coarse sediments having a disproportionate effect on the backscatter response ,
and stated that ‘it would be desirable to improve the correlations obtained between
backscatter and sediment grain size properties’ (Collier and Brown, 2005, p.446).
Brown et al. (2011) undertook sediment mapping of a 4,800km2
area of seabed off
Canada by using a semi-automated backscatter classification software (QTC
Multiview) on a large multibeam sonar dataset, and obtained promising results for the
production of geological maps. However, a study carried out in the Aleutian Islands
utilised backscatter response, seafloor rugosity and complexity data to classify the
substrate, and did not find a strong correlation between mean sediment grain size
and acoustic reflectivity. The researchers emphasised the need for a combination of
techniques for ground-truthing, a relatively large number of samples, and ground-
92
truthing of all potential substrate types represented in the classification (Rooper and
Zimmerman, 2007, p.956). Acoustic seabed classification carried out in Western
Norway (Ellingsen et al., 2002) using the QTC View system, found that there was
generally concordance between the acoustic classification and sediment grain size.
However, the study also found that other physical and/or biological factors that could
not be detected by grab sampling or coring were influencing the acoustic classes.
When comparing seabed classification from acoustic methods with direct sampling
methods or optical methods, it can be seen that each technique has its strengths and
weaknesses, for example acoustic survey provides the opportunity to examine large
areas more efficiently than direct sampling. It can be difficult to establish the spatial
scale of different seabed classes based on discrete sample stations alone, whereas
acoustic backscatter provides continuous data. It is likely that this continuous data
will result in a more diverse seabed map than would be produced using photography
or sampling at discrete locations (Sutherland et al., 2007). Also, research by Orpin
and Kostylev (2006) suggests that classification from photography is likely to result in
more seabed classes than would be identified from grab samples alone, as
photography can show a wider range of textural variability. The researchers
attributed this in part to the difficulties in quantifying coarse gravel using grab
samplers, which have a tendency to lose the coarsest grains. Rooper and
Zimmerman (2007) found that it was not possible to determine sediment grain size
from video footage, and Orpin and Kostylev (2006) found that that similar backscatter
responses could be produced by different mean grain sizes, hence the importance of
direct sediment sampling. Also, factors which may affect classification, such as ripple
93
marks and other micro-topography will not be apparent in grab samples or sediment
grain size analysis, but may be detectable in video or backscatter imagery (Ellingsen
et al., 2002). The most robust classification, therefore, is likely to result from
combining acoustic techniques with photography and grab sampling. The aim of a
successful acoustic classification system should be to bring greater powers of
analysis into the classification than would be achievable through manual
interpretation (Blondel and Sichi, 2009).
A workshop was carried out in 2006 at the University of Ulster in Northern Ireland,
which brought together several international research teams involved in developing
techniques for interpreting MBES backscatter data. The teams applied a range of
acoustic classification approaches to a single common MBES dataset collected from
Stanton Banks, in the northeast Atlantic, 120 km west of mainland Scotland, for the
purpose of geological and habitat mapping (Brown and Blondel, 2009). The dataset
covered a 7.5 x 9 km area, and ground-truthing data was provided by 90 seabed
photographs.
Blondel and Sichi (2009) undertook textural analysis via unsupervised classification
of the backscatter imagery, using a software called TexAn that was specifically
designed for textural analysis of sidescan sonar rather than multibeam data.
Although the results were not perfect, the classification worked well in relation to the
ground-truthing data, and the translation to multibeam proved to be successful.
94
Fonseca et al. (2009) used a combination of visual interpretation of the mosaic data
and angular response analysis techniques to classify the data, and the approach
resulted in a good correlation with the available ground-truthing data.
Marsh and Brown (2009) utilised an artificial neural network model called a selforganising
map (SOM) for classifying the data. This approach, based on the beam
level angular response and bathymetry rather than just on the mosaic data, does not
require any a priori specification of the number or type of classes, and is capable of
producing a range of maps with different parameters based on end-user
requirements. This work demonstrated that much better results were obtained using
the beam-level classification than the directionally-filtered, mosaiced backscatter
strength data (Marsh and Brown, 2009, p.1274-1275).
Preston (2009) undertook unsupervised classification of the data based on image
amplitudes and texture, followed by assigning attributes based on the available
ground-truthing data. Rather than being derived from a geographic mosaic, the
classification was done in image space, which means that the axes consist of sample
number and ping number, in order to avoid artefacts common in mosaics (Preston,
2009, p.1278). The classification process was carried out using specialised
multibeam classification software, (QTC Multiview), and categorical interpolation
software (QTC Clams). The resulting acoustic classes correlated well with the
ground-truthing data. This is the processing method that is the most similar to that
undertaken for the seabed classification off the coast of Qatar, albeit using multibeam
data rather than sidescan sonar data.
95
Simons and Snellen (2009) applied a Bayesian approach, using the averaged
backscatter data per beam, so that the MBES calibration was irrelevant, and also,
along-swathe seafloor type variations did not affect the classification, since no use
was made of the angular dependence of the backscatter data (Simons and Snellen,
2009, p.1267). The results showed a good correlation with the available groundtruthing
data.
It has been noted that dense concentrations of seagrass and algae contribute to the
acoustic backscatter response, even to the extent where they can mask the acoustic
signals from the actual sea floor (Penrose et al., 2005). However, this is less likely to
be a concern for low frequency data, where only the dominant return is required, and
there are certainly cases where the roughness of rocks, boulders, and bedrock have
produced very different backscatter responses, even though the benthic habitat type
is the same on all of them (Preston, 2009). Research based on human visual
interpretation of high-resolution acoustic backscatter imagery undertaken by Van
Rein et al. (2011) in Church Bay, off Rathlin Island near the North of Ireland, found
that there was no discernible difference between the imagery taken before and after
the removal of 100m2 of kelp from three sites. The authors also referenced other
studies where it was found that it was not possible to easily identify low density
seagrass beds from acoustic backscatter data (Van Rein et al., 2011, p.345).
Conversely, previous research using acoustic classification has also demonstrated
that there is at least some correlation between benthic habitat and substrate type
(Brown and Blondel, 2009).
96
Although the published literature shows varied approaches to, and results from the
classification of seabeds using acoustic backscatter data, the two issues that all of
the studies emphasise heavily are, firstly, the need for a combination of techniques to
be used for classification, as the results from each become far more informative in
combination, and secondly, the importance of the amount and quality of the groundtruthing
data. It is clear that that there is still a lot of work to be done on different
methodologies, and that some of the results from trials and experiments seem to
conflict with each other. The reality is that the results are probably very specific to the
particular seabed type that is being investigated, and it is therefore very important to
build up a regional picture and establish which methods work and which don't work
for a particular region. Concepts and ideas can be taken from other areas but they
need to be refined according to regional variances.
The results presented in section 4.1.2 show that the acoustic classification process in
the Study Area has successfully enabled the division of the seabed into a set of initial
landscape units based on sediment texture, which is a huge step towards the
process of building up knowledge about the archaeological and palaeoenvironmental
potential of zones within the submerged landscape. However, these initial polygons
are only an interim step in the characterisation process, and need further analysis
and refinement. In this form, there are too many landscape units to be useful for
landscape-scale study, and not enough is known of their character for them to be
meaningful, but they provide the basic building blocks for further characterisation
based on bathymetry, ground-truthing data and analysis of geophysical anomalies
97
(see sections 4.2, 4.3 and Chapter 5). Attributes relating to the character and
potential of the seabed in particular zones, derived from the other datasets, can be
attached to the base landscape units, and the units refined accordingly.
4.2 Topographic Mapping
4.2.1 Methodology
A surface model of the seabed in the Study Area was generated from the LiDAR
bathymetry in order to be able to identify broad zones of high archaeological and
palaeoenvironmental potential which could be used to refine the landscape units
created from the sediment texture classification.
4.2.1.1 Overview of LiDAR Bathymetry
The Bathymetry data (seabed topography) was kindly made available by the
Hydrographic Section of the Ministry of Municipality and Urban Planning in Qatar.
The bathymetry was captured by LiDAR (Light Detection and Ranging) survey, an
effective technique for measuring depth in shallow water. The technique consists of
an aircraft-mounted laser transmitting light pulses downwards. These pulses reflect
from the sea surface and the sea floor, and therefore measure the depth. This type of
airborne survey is extremely useful for coastal and shallow-water research. It allows
very rapid data collection over large areas in comparison to sonar techniques. Also, it
98
can cover very shallow areas that cannot be covered by ship-based techniques and
therefore allows seamless mapping through the land/sea interface.
Figure 30 Extent of available LiDAR bathymetry.
(m)
99
The extent of the available bathymetry did not coincide exactly with the extent of the
defined Study Area (the extent of the sidescan sonar data), but it did cover a large
part of the Study Area, and extended significantly beyond it in the south into the Gulf
of Salwa. The survey started approximately 15 km north of Al-Zubārah and continued
for some 130 km down the west coast of Qatar to approximately 15km beyond Umm
Bab (Figure 30).
4.2.1.2 Generating the Surface Model
The LiDAR data was provided as ASCII xyz files, containing the three-dimensional
coordinates of the points collected by the LiDAR survey, projected in the UTM39N
coordinate system. Vertical heights were given in metres based on the Qatar Chart
Datum, which lies 0.88m below Qatar Vertical Control Datum (Mean Sea Level at
Doha Port). It was processed and analysed using the 3D analyst module in ArcGIS.
The first step was to convert the ASCII data to 3D feature data and add the height
information (z values). The z values in the data had to be reversed as drying heights
were recorded as negative values, and depths below sea level were recorded as
positive values. This resulted in a point-based shapefile which could then be used to
interpolate a raster elevation model for the seafloor surface. This was done by using
the raster interpolation function in ArcGIS, and choosing the ‘natural neighbour’
method of interpolation. It was necessary to experiment with different cell size
settings to establish the best balance between resolution of the surface model and
feasible processing time and power. As the point data was spaced on average
between 5 and 7m apart, it was originally decided to use a cell size of 7m for the
100
interpolation, as this fitted with the resolution of the original data, and did not use too
much processing power, data storage or time. However, this made it difficult to see
some of the features when zoomed in, so after further testing, it was decided to use a
cell size of 2m for the bathymetry data in the Study Area, and 5m for the data outside
the Study Area. It was not possible to use such a high-resolution for the entire
dataset due to the processing time and data storage requirements involved. Although
the 2m resolution did not add any more detail to the features visible on the surface
due to the 7m spacing of the original LiDAR points, it did mean that features were
easier to examine when zoomed in. The resulting raster elevation surface was
visualised using stretched values along a colour ramp (Figure 30).
It became apparent at this stage, once the coastline was visible in the bathymetry
and could be compared to the coastline data that had already been obtained from the
Qatar Centre for GIS, that there appeared to be a shift in the bathymetry data to the
northwest. All possibilities were examined, including possible coordinate
system/projection issues, and the processing of the data was repeated again from
scratch. This was done first using the UTM39N projection, which still showed the shift
in data, and then using the Qatar National Grid projection in case there was an error
in the metadata provided, but the data was still not located correctly. Finally, it was
decided to use the coastline data provided by the Centre for GIS to correctly locate
the bathymetry data. Two control points were logged with a GPS at known locations
on the coast (piers) for ground-truthing purposes, to ensure that the coastline data
being used for the correction was actually in the right place itself. Then a series of
reference points were logged at identifiable locations over the extent of the
101
bathymetry and in the coastline data, and the differences in the X and Y coordinates
were compared and averaged. It was apparent that in the bathymetry data, the X
coordinates were all approximately 80m too far to the west, and the Y coordinates
were all approximately 80m too far to the north. This was a constant error noted in all
of the reference points, so it was possible to shift the data by adding 80m to the X
coordinates and subtracting 80m from the Y coordinates. Once this was done there
was a clear fit between the bathymetry and the coastline data. The coordinates of the
reference points and the values used for shifting the data are provided in Appendix 2.
The next step was to carry out a hillshade operation on the raster elevation surface in
order to create a shaded relief model of the seafloor, and therefore greatly enhance
the visualisation of the topography. This was done by using the hillshade tool in 3D
analyst, which works by considering the illumination source angle. Shadows can also
be considered in the analysis if required, but were not used in this case as a better
hillshaded surface was created without the use of shadows. A variety of models were
created using different vertical exaggeration factors to further enhance the surface
visualisation, and for most of the analysis, a vertical exaggeration factor of 20 was
deemed to provide the best results. The elevation surface was then set to be 40%
transparent, and layered on top of the hillshaded surface to create an enhanced
visualisation of the seafloor topography. The elevation surface draped over the
hillshaded surface, with a vertical exaggeration factor of 20, was the final surface
model that was used for the analysis. The surface model was also exported into
Erdas Imagine software and used to generate a 3D visualisation in order to assist
interpretation (Figure 31).
102
Figure 31 3D visualisation of the seabed in the Study Area (created in Erdas Imagine, based on the surface model generated from the bathymetry).
N
103
4.2.1.3 Mapping Features
The aim was to use the surface model to create a map of significant landscape
features by digitising polygons around any significant features or areas of seabed
that could be identified. Features or areas were deemed to be significant on the basis
of two criteria. The first criteria was whether areas had the potential to be attractive to
past humans (Westley et al., 2011a). On the basis of what we know about human
settlement in the region in the Early Holocene, attractive areas include former
coastlines, which could appear in the surface model as distinct breaks in slope, and
sheltered coastal settings such as former bays, estuaries and lagoons. Other
features known to be attractive to past humans in terrestrial Qatar in the MidHolocene
include protective Karst-related settings, for example, former sinkholes.
The second criteria was whether areas had the potential for the burial of palaeo-landsurfaces
and the presence of palaeoenvironmental caches, that is, areas
characterised by sediment deposition rather than erosion. Such areas could include
sediment traps, for example, depressions and palaeochannels, as well as smooth
areas of seabed with protruding rock outcrops. Other features such as spits and
barriers are also indicative of landscape preservation rather than erosion. Key factors
in the preservation of wrecks are not exactly the same as those for archaeological
sites, but the probable presence of extensive sedimentary deposits is common to
both. More features were mapped in the Study Area itself, as the higher resolution of
the surface model within the Study Area allowed more detailed analysis. Beyond the
Study Area, only larger, landscape-scale features were mapped.
104
It was also possible to use the surface model generated from the bathymetry in
conjunction with information about former sea levels to create models of where
shorelines would have been likely to occur in the Early Holocene. This was done by
taking minimum and maximum sea level values at specific time periods from the
global sea level curve produced by Stanford et al. (2011), and using these values to
differentially shade those areas of the surface model that would have been under
water and those that would have been dry land within a certain time period. The
same was also done with sea level values from the Qatar sea level curve produced
by Jameson and Strohmenger (2012), although with this curve there was just a single
value for any particular point in time rather than minimum and maximum values. This
process was relatively easy to do as the LiDAR bathymetry, unlike bathymetry data
generated from sea-based sonar survey, provides seamless coverage from land to
sea. The differentially shaded surface models were then compared to the
topographic features that had been mapped from the surface model.
4.2.2 Results
The surface model created from the LiDAR bathymetry has provided a significant
amount of detail about the present character and shape of the seabed. However, it is
worth noting that there are limitations to the surface model produced by the
bathymetry. The resolution of around 7m means that is possible to identify and
interpret large, landscape-scale features such as former shorelines and large
channels, but smaller, more discrete or ephemeral features will not show up at this
resolution. There are also a few small areas within the LiDAR survey area where no
105
data was collected, and these appear as completely smooth areas in the surface
model.
Also, it also has to be borne in mind that the present topography of the seabed will
not be an exact reflection of what it would have been in the Early Holocene, as it will
have been substantially affected by subsequent processes of erosion and deposition
(see Chapter 7). Generally, it is thought that sediment deposition will have had the
effect of smoothing out the seabed surface in this area (Marin Mätteknik AB, 2002,
p.17) by filling in depressions, hollows and channels.
4.2.2.1 Topography Within the Study Area
The bathymetry shows that generally the seabed within the Study Area is relatively
flat. The two deepest areas are along the southwestern edge of the Study Area,
where the seabed averages around 12 to 13m deep (the deepest point here is just
under 15m deep), and a southwest-northeast orientated channel in the north of the
Study Area, averaging 8 to 9 m deep. Apart from near-shore, the highest areas in the
Study Area are on and around the Qit’at ash Shajarah reef in the centre-west of the
Study Area, which is only just below the surface of the water at certain points. Apart
from the shallower near-shore area to the north of the Ras ‘Ushayriq peninsula, the
rest of the Study Area averages between 4-6m in depth (Figure 32).
106
Figure 32 Surface model of the Study Area created from LiDAR bathymetry.
(m)
107
Figure 33 Features mapped from the surface model.
(m)
108
The most prominent seabed feature is the Qit’at ash Shajarah drying reef in the west
of the Study Area. The other major feature is a wide, deep, southwest-northeasttrending
channel which runs across the north of the Study Area (Figure 33). This
channel is more than 5 km wide at its widest point, and it reaches depths of more
than 9m in parts.
Most significantly for the purposes of archaeological potential, the surface model
shows three coherent linear areas of relatively abrupt change in depth, which have
been tentatively interpreted as former shorelines (Figures 34 and 35) that may have
formed during periods of still-stands in the marine transgression. These putative
former shorelines are extremely significant, as the archaeological evidence from
terrestrial sites in Qatar in the Mid-Holocene demonstrates that coastlines were clear
foci for human settlement (see Chapter 7).
Two of these potential former shorelines, running roughly north-south, lie close to the
present-day coastline to the north of the Ras ‘Ushayriq peninsula (Figure 35). The
easternmost one roughly follows the -1m contour, possibly on the seaward edge of
the intertidal zone. The westernmost one does not closely follow a single contour,
instead mirroring the -5m contour in the south and rising to the -2m contour in the
north. Both of them appear to lie roughly parallel with each other, looping around a
headland and then merging.
109
Figure 34 Putative former shorelines in the Study Area.
Figure 35 Hillshaded surface model with putative former shorelines.
(m)
(m)
110
The third putative shoreline appears as a very abrupt change in depth which runs
west-east across the centre of the Study Area (Figure 36), roughly following the -7m
contour line. This edge is very clear along most of its length, but it becomes more
indistinct towards its western end. The deep, southwest-northeast orientated channel
mentioned above lies immediately to the north of this feature.
Figure 36 Hillshaded surface model with putative east-west former shoreline.
The edge of this change in depth is also clearly visible in the mosaic image of the
sidescan sonar data (Figure 37).
Figure 37 Putative east-west former shoreline with sidescan sonar data overlain.
(m)
(m)
111
Several significant topographic features lie in close association with the north-south
former shorelines, and are probably related to them (Figure 38). These features
consist of raised ridges, likely to be former promontories, spits or islands. These
types of feature are highly significant locations for potential human settlement, as are
the possible bays and headlands on these shorelines, which may have been both
attractive areas for settlement, and protective in terms of deposits.
Figure 38 Putative former shorelines with possible associated mapped features.
(m)
112
Another feature in close proximity to the former shorelines is a north-south aligned
deeper channel, more than 700m wide, which runs from the east of the Ras
‘Ushayriq peninsula northwards (Figure 39). This may represent a former river
channel, and its orientation and proximity to a present-day Wadi (Wadi Debayan)
suggest that this palaeochannel is actually a continuation of Wadi Debayan.
Figure 39 Palaeochannel possibly relating to the former course of Wadi Debayan.
A cluster of small curvilinear depressions, some interconnected, are visible in very
shallow depths approximately 1km offshore, in the intertidal zone (Figure 40). These
features, interpreted as palaeochannels, are narrow and very shallow, and the most
(m)
113
well-defined channel appears to end at the postulated former shoreline. The location
and orientation of this cluster of palaeochannels suggests that they represent a
drainage system relating to that shoreline, and are therefore of considerable
significance in terms of palaeoenvironmental potential.
Figure 40 Cluster of palaeochannels.
A few fish traps (currently undated) are just about discernible in the intertidal zone in
this area (Figure 41), and these are so ephemeral that it would have been very easy
to miss them had their locations not been mapped already as a result of the QNHER
cultural mapping project (Breeze et al., 2011). Only a few of the traps mapped by the
QNHER can be identified in the surface model, as most of them lie in 'no-data' areas
in the bathymetry.
(m)
114
Figure 41 Fishtraps visible in the surface model (top), mapped during the QNHER cultural mapping
project (middle) and an example photographed by the QNHER project (bottom).
(m)
(m)
115
A number of linear and curvilinear depressions or channels occur in the Study Area.
These are concentrated in the deep channel previously mentioned to the north of the
west-east shelf (Figure 42), and are almost all on the same southwest-northeast
orientation as the channel itself. However, there are also two north-south aligned
linear depressions or channels in the south of the Study Area, between the Bay of AlZubārah
and the reef (Figure 43).
Figure 42 Channels and depressions in the north of the Study Area.
Figure 43 Channels and depressions in the south of the Study Area.
(m)
(m)
116
The orientation of these linear and curvilinear depressions are worth highlighting.
Those lying in the deep channel in the north (Figure 42), as well as reflecting the
orientation of the channel itself, also lie roughly parallel with the southwest-northeast
alignment of the present-day and putative former shorelines to the southwest. The
linear depressions in the south, to the east of the reef (Figure 43), reflect the northsouth
orientation of the putative extension of Wadi Debayan, and it is likely that these
three features represent palaeochannels that may all be part of a larger drainage
system.
Two extensive areas of linear ridges are discernible in the north of the study data,
one north-south trending (Figure 44), and one northwest-southeast trending. These
are very large, widely-spaced features, between 100m and 200m apart, and are
interpreted as sand ripples or 'megaripples'. Another area of smaller, east-west
trending sand ripples is visible in the south of the Study Area.
Figure 44 Megaripples in the north of Area 1.
(m)
117
The central area, between the reef and the Bay of Al-Zubārah is dominated by
narrower, less prominent north-south trending linear features which are interpreted
as either sand banks or linear reef structures (Figure 45).
Figure 45 Possible sand banks between the reef and the bay of Al-Zubārah.
Several small hollows can be identified in the surface model, exclusively in the north
of the Study Area. These are concentrated into two groups, one in the extreme
northwest of the Study Area, and one to the north of the west-east potential former
shoreline. The two concentrations are separated by an extensive area of sand ripples
where no hollows are present. It seems logical to assume that the sand ripples will
have masked any hollows that may have been present in those areas. These
features appear as small, but very distinctive circular or sub-circular holes on the sea
floor, ranging from 20m to 50m in diameter. They do not appear to be very deep,
either because they are filled with sediment, or because the resolution of the LiDAR
data may not be sufficient to adequately reflect the morphology of such small
(m)
118
features. These hollows were cross-checked against the sidescan sonar data to see
if they could be further clarified, but unfortunately most of them do not lie within the
area covered by the sidescan sonar survey. One that did lie within the survey area
was not visible in the sidescan sonar data at all, at least in part because it lay at the
inner range of the swath, directly below the towfish (Figure 46). It is not clear exactly
what these features are, but they have been tentatively interpreted as former solution
hollows, as found in the terrestrial Karst landscape of present-day Northern Qatar. It
is interesting to note that one of the potential solution hollows lies very close to one of
the linear depressions, as it has been noted that sequences of Karst-related
depressions in terrestrial Qatar can form linear depressions (Sadiq and Nasir, 2002).
Figure 46 Possible solution hollow visible in the surface model but not in the sidescan sonar data.
119
There are some very low mound-like features present, mainly to the north of the
west-east former shoreline (Figure 47). These are only very slightly elevated, by
around 1m, and range in size from 50m to 150m in diameter, so could be better
described as undulations. It is possible that these features could represent outcrops
in a sediment-covered seabed.
Figure 47 Undulations in the seabed to the north of the putative east-west shoreline in Area 1.
More complex topography exists around the reef (Figure 48), making it difficult to
interpret, but it includes a raised area to the west that is likely to be part of the reef
formation. There is also an elevated area to the south, which may represent an
accumulation of sediment on the leeward side of the reef, and it includes two
prominent larger outcrops of either rock or coral. A deeper area in the southwest may
represent a basin, but this can only be a tentative interpretation as the feature lies at
the edge of the surface model.
120
Figure 48 The Reef in the west of the Study Area.
4.2.2.2 Topography Outside the Study Area
In the area to the south of the Study Area, where topographic features were not
mapped in as much detail, greater depths are present, particularly in the far south in
the Gulf of Salwa (Figure 49). Here, depths increase to more than 28m in the west,
but there are also numerous large areas of significantly higher elevation within this.
The predominant features identified were potential former shorelines, which can be
traced virtually all the way down the coast, although their occurrence is more
sporadic in the area immediately to the south of the Study Area. Due to the narrow
width of the data set around the Hawar Islands (which lie in Bahrain territorial waters
(m)
121
and therefore not included in the survey), it is not possible to tell if these putative
shorelines in the central area link up to form a continuous feature with those in the
south or if they represent different phases of shoreline. However, it is likely that the
area around the Hawar Islands may incorporate part of this shoreline.
Figure 49 Potential former shorelines to the south of the Study Area.
(m)
122
The areas of higher elevation down in the Gulf of Salwa form quite significant
protrusions from the seabed (Figure 50), with elevation changes of 7 or 8 m, and
would have been islands at some time in the past, when sea levels were more than
10m lower then present-day levels.
Figure 50 Seabed surface in the Gulf of Salwa.
(m)
123
4.2.2.3 Topography and Sediment Classification
As part of the integrated analysis of different datasets in order to refine the zoning of
the submerged landscape, the results of the acoustic classification and the
bathymetry analysis were examined together to see how each could inform and
enhance the other. The aim was to highlight those areas where the two datasets
were consistent with each other and vice versa, and in particular to see if the surface
model could provide useful additional information about the acoustic classes that
were generated by the unsupervised acoustic classification process.
The broad patterns of the classified data in the south of Area 4 seems to show a
good correlation with the broad patterns that can be discerned in the surface model
(Figure 51). The deeper area that is present along the western edge of Area 4 in the
surface model is largely reflected by the yellow area (Class 4) in the classification.
This coincides with the extensive area of trawler scarring that is visible in the mosaic
of the sidescan sonar data. These scars are not visible in the surface model,
probably because they are too superficial.
The pattern of raised areas visible in the surface model is also discernible in the
classification, including the sub-oval area in the south, and the triangular elevated
area immediately south of the reef. There is a complex pattern evident in the
topography in the elevated area to the south of the reef, and the classified data in this
area is also very complex. The patterns in the two datasets don't reflect each other in
detail, but a similar general outline shape can be traced in both datasets.
124
Figure 51 Surface model with the classified data in Area 4.
In Area 2, the deeper north-south trending channels that were identified in the
surface model show some coincidence with the lilac north-south-trending areas
(Class 5) generated by the acoustic classification (Figure 52), possibly because
these channels contain finer sediment than the surrounding area. It is also possible
that the northern part of the possible wadi extension that was identified in the surface
model to the east of the Ras ‘Ushayriq peninsula can start to be discerned in the
classified data, but as this is the only part of the channel that lay within the area
covered by the sidescan sonar survey, there is not enough data to be more positive
about this.
(m)
125
Figure 52 Surface model with the classified data in Area 2, West of the Ras Ushayriq Peninsula.
The raised area to the west of the reef that is clearly visible in the surface model, is
equally clearly visible as the lime green area (Class 8) in the classification (Figure
53). The complexity of the classified data immediately to the north of the reef is not
really reflected in the topography shown in the surface model, or in the mosaic image
of the sidescan data. The automated acoustic classification is obviously detecting
changes in the seabed sediment that are not just a factor of depth in this area, but
could be influenced by coral formation and differential sediment accumulation around
the reef.
(m)
126
Figure 53 Surface model with the classified data in Area 3, north of the reef.
In Area 1, the large, southwest-northeast trending channel that was identified in the
surface model is broadly coincident with the large band of mixed salmon and yellow
(Classes 9 and 4) in the classified data (Figure 54). The smaller linear depressions
that could be discerned within the bigger channel in the surface model are not readily
apparent in the classified data, but the gap between survey lines in this area does not
make interpretation of smaller features like this very easy. However, it is clear from
this integrated analysis that the yellow and salmon areas correlate closely with the
deeper areas within the Study Area.
(m)
127
Figure 54 Surface model with the classified data in the deep channel in Area 1.
In the north of Area 1, the classified data consists of a mixed area of blue and lilac
(class 10 and class 5 respectively). In the surface model, this area shows a very
clearly-defined area of north-south trending, large scale sand ripples. However, the
boundary of this area of sand ripples is not clearly-discernible in the classified data.
4.2.3 Discussion
The analysis of the surface model has produced some very promising results through
the identification of significant landscape features, that will form the basis of future
marine research frameworks and further investigation. Despite the limitations of the
data in terms of resolution and coverage, for the purpose of this research, which is to
identify areas of archaeological and palaeoenvironmental potential at a landscape
scale, the analysis of this dataset has been crucially important as it has enabled the
identification of previously unknown landscape features that could be highly
significant to locating past human settlement in the Late Pleistocene and Early
(m)
128
Holocene, particularly the putative former shorelines. However, it is necessary to
examine these shorelines in association with information about Holocene sea level
change. By applying reconstructions of global sea levels to the surface model
created from the bathymetry, it is possible to attempt to map the extent of dry land
that would have been exposed within certain date ranges and reconstruct shorelines.
Accurate sea level curves have to include elevation data and age dates, and also
require the interpretation of palaeo-bathymetry (Jameson and Strohmenger, 2012).
However, there are considerable uncertainties and limitations relating both to the
application of global sea level curves specifically to Qatar, and to the use of
bathymetric models with sea level curves. These issues have been discussed in
detail by Lambeck (1996), Rose (2010) and Cuttler (Cuttler and Al-Naimi, 2013;
Cuttler, 2013; Cuttler, 2014), and have been summarised in section 1.2.6. As long as
these limitations are acknowledged, applying sea level data to the surface model
created from the LiDAR bathymetry provides invaluable information about the
locations of potential former shorelines. A global sea level reconstruction curve was
produced by Stanford et al. (2011), by undertaking statistical analysis on a range of
key far-field datasets of relative sea-level change. Even though this model of sea
level change has been very carefully and substantially researched, and is applicable
for global and macro-scale studies, it is still necessary to accept that there will be
limitations in accuracy when applying the model to more detailed studies of sea
levels around Qatar. This model gives global sea level ranges (with 99% confidence)
between 15m and 11m below present day levels for 8,000 BP, and between 9 and
7.5 m below present day levels for 7,000 BP. Applying these values to the seabed
129
surface model for the Study Area produces interesting results (Figure 55). On the
basis of these values, virtually the entire Study Area apart from the deeper area at
the southwest edge, and parts of the deep channel in the north, would still have been
dry land at 7,000 BP, regardless of whether the lowest or the highest value in the
range is used. The possibility that most of the Study Area remained as dry land as
late as 7,000 BP is worth considering, given the theories of Lambeck (1996) and
Heyvaert and Baeteman (2007) that the present-day coastline was reached at 6,000
BP, and given the recent evidence dating relic beach terraces at Wadi Debayan to
6,000 BP (Tetlow et al., Forthcoming).
Figure 55 Postulated extent of dry land at 7,000 BP (based on Stanford et al,. 2011).
130
A Qatar-specific sea level curve was published by Jameson and Strohmenger
(2012), which was derived from age dating of raised beaches in Qatar. Jameson and
Strohmenger propose that the Holocene high stand was initially reached at 7,000 BP
in Qatar, which is considerably earlier than the timescale proposed by Lambeck
(1996) and Heyvaert and Baeteman (2007). According to Jameson and
Strohmenger’s curve, the sea level around Qatar at 8,000 BP is placed at 4.3m
below present day levels, dropping to 4.7m at around 7,800 BP, before rising to 2m
higher than present day levels at 7,000 BP.
Figure 56 Postulated extent of dry land at 8,000 BP (based on Jameson and Strohmenger, 2012).
131
Applying these values to the surface model of the Study Area produced quite
different results to the Stanford model. On this basis, the model shows that the
majority of the Study Area would already have been inundated by 8,000 BP (Figure
56). The western-most potential former shoreline identified to the north of the Ras
‘Ushayriq peninsula fits well with this model, roughly coinciding with the land/sea
interface at 8,000 BP. Furthermore, applying these values to the surface model
demonstrates that areas of Qatar that are currently dry-land would have been
submerged at 7,000 BP (Figure 57).
Figure 57 Postulated extent of dry land at 7,000 BP (based on Jameson and Strohmenger, 2012).
132
The interpretation of the third, west-east orientated potential former shoreline is less
clear. When the values from both of the sea level curves discussed above were
applied to the surface model, neither model highlighted this feature as a land-sea
interface between 8,000 and 7,000 BP. There is no substantial evidence in the
literature for a still stand at -7m in the Early Holocene, and this feature may represent
a shoreline that had its origins in the previous interglacial period, the Eemian.
However, even if this is the case, it would have been a shoreline again at some point
during the Early Holocene. This could equally apply to the other two putative
shorelines, since there is also no specific evidence for still stands at -2m or -1m
either.
The different results obtained for the possible extent of dry land in the Study Area in
the Early Holocene outlined above demonstrate that these attempts at shoreline
reconstructions cannot be considered to be definitive due to all the problems
previously described relating to sea level reconstructions and lack of detailed data.
However, they do show that although there is still a lot of work to be done in
establishing more detailed and robust models of sea level change around Qatar,
there is nothing to conflict with the interpretation of any of these three features as
former shorelines.
The wide, deep, southwest-northeast aligned channel in the north of the Study Area
could represent a former river bed. The two linear channels to the east of the reef,
together with the possible extension of Wadi Debayan to the east, may also be part
of a former river channel, which possibly forms a tributary of the deep channel to the
133
north. This may explain why the west-east orientated putative former shoreline is less
distinct at its western end, perhaps displaying estuary-like characteristics where it is
bisected by the channel running from the south.
As mentioned previously, the hollows identified in the surface model have been
tentatively identified as solution hollows. However, it is also possible that they could
be craters that have developed as a result of gas venting from the sea floor. Such
features, sometimes termed 'pockmarks', have been noted in the Arabian Gulf,
including off Qatar (Uchupi et al., 1996), and also elsewhere in the world, especially
in hydrocarbon-bearing areas, such as the North Sea and off Alaska (Ellis and
McGuinness, 1986).
The integrated analysis of the surface model and the acoustic classification has
proved to be extremely useful in terms of providing context and character to the
acoustic classes. One significant result is that in certain parts of the Study Area,
there is a clear correlation between depth and the reflectivity of the seabed sediment
type, and therefore by extension, there appears to be a clear correlation between
depth and sediment texture. The seabed in deeper waters appears to exhibit
evidence for finer sediments, although this requires further investigation (see Chapter
7). This is particularly obvious in the case of the deep channel in the north of the
Study Area, the broad orientation and shape of which is clearly reflected in the
classified image (see Figure 54).
134
4.3 Ground Truthing
An important component of the seabed characterisation was the validation of the
acoustic classes, and the calibration and refining of the defined landscape units by
ground-truthing via direct sampling and visual observation.
4.3.1 Methodology
4.3.1.1 Direct Sediment Sampling
The ground-truthing carried out to validate the sediment classification consisted of
direct sampling of the seafloor sediment at locations throughout the Study Area. The
locations were selected on the basis of the landscape units defined after the acoustic
classification, and 57 sediment samples were collected using a Van Veen grab
sampler, collecting on average four grabs at each sample site (Figure 58). It was not
possible to get as many samples as ideally would have been desirable from the
entire Study Area, especially from the south and east of the Study Area, due to the
limited time that the survey team had available to go out on the boat, and the bad
weather encountered during the survey. However, as far as possible, at least one
sample was taken from every landscape unit that was defined by the acoustic
classification.
135
Figure 58 Grab samples overlain on Initial Landscape Units.
136
The samples were subject to particle size analysis (granulometry) by specialists from
the Environmental Studies Centre (ESC), at Qatar University. The samples were
analysed using a laser diffraction particle size analyser (a Malvern Mastersizer). This
method requires the samples to be dried, then mixed with distilled water, and passed
through a laser beam which hits each particle in the sample. The machine measures
the intensity of the light scattered, and uses this data to calculate the size of the
particles, and therefore the percentage of each fraction (sand, silt or clay) in the
sample (Blott and Pye, 2006). Graphs showing the particle size distribution, and
tables showing the percentage volume of particle sizes were produced for each
sample, and each sample was then allocated the appropriate sediment type, such as
sand or silt, based on the statistics generated by the analysis. The complete
granulometry report as provided by the ESC is included as Appendix 3. The results of
the granulometry analysis were then used to enhance the acoustic classification
analysis, and were integrated into the overall characterisation analysis.
In addition to the targeted grab sampling programme, a significant amount of preexisting
direct sampling data was also available as a result of extensive geotechnical
investigations that were undertaken for the Qatar-Bahrain Causeway project. Several
coring programmes were carried out in 2008 in order to provide information on the
geology and the geotechnical properties of the potential causeway area. 24
exploratory boreholes were drilled along the Qatari side of the causeway route in July
2008 (Figure 59) Fugro Peninsular, 2008).
137
Figure 59 Exploratory boreholes drilled in 2008, overlain on Initial Landscape Units.
A separate coring programme, also carried out in July 2008. This focused on
potential sand traps in Qatari waters, located by earlier geophysical survey, that
could possibly be suitable for use as borrow areas for the construction of the
causeway (Figures 60 and 61), and on areas of natural elevation which had the
potential to be used as embankment areas for the causeway (Figure 62) (QBC
Consortium, 2009a; 2009b). Two of the potential embankment areas lay in Qatari
waters, and 12 boreholes were drilled on one of them, an area of coral heads close
to on the Qit’at ash Shajarah reef. A vibrocore sampling programme was undertaken
to confirm (or otherwise) the presence of the sand traps and their suitability as
borrow areas. Vibrocore sampling uses a vibrating motor to liquefy the marine
138
sediments, which can then be retrieved as a core sample. The two areas in Qatari
waters that were targeted for vibrocoring lay in the north and south of the Study Area,
with 43 vibrocores taken from the area in the north (Figure 60) and 10 from the area
in the south (Figure 61). Laboratory analysis was then carried out on the core
samples. This has provided very useful information about the seafloor and subbottom
sediment type at the coring locations, which can be used to inform the
acoustic classification of sediments in those areas covered by the coring programme.
Figure 60 Vibrocores taken in north of the Study Area in 2009, overlain on Initial Landscape Units.
139
Figure 61 Vibrocores taken in south of the Study Area in 2009, overlain on Initial Landscape Units.
Figure 62 Boreholes drilled over the reef in 2009, overlain on Initial Landscape Units.
140
4.3.1.2 Video and Photography
In addition to the grab samples, video and photographic ground-truthing data was
also available for the Study Area. The main video data available was an underwater
video-based analysis of the seabed that was carried out by a commercial company
for the Qatar-Bahrain Causeway project (Creocean, 2008). This survey was focused
on identifying, characterising and assessing the condition of the different biotopes in
the area of the proposed causeway. A series of video transects were captured, in
shallow areas (under 2m in depth) by divers and in deeper areas (over 2m) by an
ROV towed behind a boat (Creocean, 2008, p.1.) The video footage that lay within
the Qatar side of the proposed causeway area consisted of 13 transects of
approximately 1km each in length, resulting in a total of 5 hours and 12 minutes of
footage of the seabed. Although this survey was aimed at recording the seabed
habitat, and the video transects only cover the central part of the Study Area, the
video footage is very useful for providing a picture of the seabed in known locations
that can then be compared with, and used to help clarify, the other data sets in the
characterisation, including the sediment classification and the bathymetry.
Some additional photography and video footage was also captured during the grab
sampling programme, but the area of seabed covered was very small and the
visibility was generally quite poor, so this footage is not very useful either for
providing broad-brush landscape information or for providing detailed sediment
information for small areas.
141
4.3.2 Results
4.3.2.1 Direct Sediment Sampling
The distribution of the grab samples in relation to the different datasets can be seen
in Figures 58, 63 and 64. The original aim was to try and cover as much of the Study
Area as possible, and to try, within the limited resources available, to get samples
from as many of the landscape units as possible. However, as explained previously,
for practical reasons some areas were not able to be covered as well as others, for
example there is much more limited coverage in Area 2 than any of the other areas.
Figure 63 Grab sample locations overlain on the surface model.
(m)
142
Figure 64 Grab sample locations overlain on the classified data.
143
Overall, the granulometry analysis appears to indicate that the grab samples were
quite homogenous. Out of 57 samples collected, 40 were categorised as sand, and
17 as silty sand.
The samples, symbolised by sediment type (sand or silt) are shown in Figures 63 and
64, overlain over the surface model and the classified data respectively, and it can be
seen from this that the distribution of silty sand appears to be concentrated to the
northeast of the reef and in the western part of the deep channel in the north of the
Study Area. 16 of the 17 samples that were categorised as silty sand are located in
this area, with the other one (sample 49) located to the southwest of the reef. One of
the samples (43) did not retrieve any sediment at all, as it was located on the coral to
the west of the reef. Samples categorised as sand predominate in the northeast
corner of the Study Area, and to the east and south of the reef.
The distribution of sediment types as derived from cores and boreholes from the
various geotechnical surveys that were undertaken as part of the planning phase for
the proposed causeway can be analysed in conjunction with the results of the grab
sampling programme (Figures 65 and 66). It has to be borne in mind that the analysis
is not a comparison of like with like, as unlike the cores and boreholes, the grab
samples were solely surface samples, involving a relatively small amount of material.
It is also worth emphasising when analysing the core and borehole results that these
surveys were undertaken in targeted locations for specific geotechnical purposes, so
the distribution plots will display an inherent bias relating to the purpose of the
surveys. Where the marine sediment type changed with depth, the boreholes and
144
vibrocores have been symbolised in the distribution plots based on the uppermost
sediment type.
Figure 65 Vibrocores and boreholes overlain on the surface model (grab samples are visible in the
background).
(m)
145
Figure 66 Vibrocores and boreholes overlain on the classified data (grab samples are visible in the
background).
146
The 43 vibrocores in the north of Area 1 were taken specifically in an area that had
been previously identified as having high potential for sand deposits, as the aim of
the survey was to confirm locations that could be used as borrow areas for the
causeway construction. The distribution plot of these vibrocores, as expected,
therefore, shows a heavy concentration of sand in the survey area. The distribution
plot shows a band of cores categorised as sand along the north of the deep channel,
with a few classed as gravel, whereas the grab samples indicate silty sand in the
areas surrounding the vibrocores. This may be a genuine change in the sediment
type rather than just a result of the different techniques used, as the grab sample
distribution does seem to indicate that the sediments in the area along the north of
the channel are coarser. The logs from the vibrocores indicate that the sand deposits
in the survey area are thick, mainly over 1m, but extending for more than 3.5 m in
some places.
Ten vibrocores were also taken in the area to the south of the reef, for the same
purpose of confirming the location of sand deposits for borrow areas. Again, as
expected, they show thick sand deposits occurring, which is also consistent with the
distribution of the grab samples. The sand survey area coincides with two deeper
areas on either side of the submerged reef area that are visible in the surface model.
As in the north, the survey areas again correlate with areas that have been classified
as yellow and salmon (classes 4 and 9 respectively) in the acoustic classification.
Boreholes from the centre of the Study Area were taken specifically for the purpose
of studying proposed embankment areas for the causeway. They are all located on
147
the reef, as that is the highest land in the Study Area, and therefore they can't
provide much useful information about acoustic classes, as there was no sidescan
data for the reef area. Apart from one silt sample, the rest are all composed of sand
or coral.
The borehole survey carried out along the route line of the proposed causeway in the
centre of the Study Area encountered marine sediments from seabed level to a
maximum depth of 6m within the Study Area. (Fugro Peninsular, 2008, p.10-11).
Marine sediments were generally thicker further westwards, away from the coast and
towards the reef. The deposits were characterised by silty fine and medium sand or
sandy gravel, locally with thin to thick beds of sandy silt/clay, variably underlain by
shelly calcarenite (caprock). The boreholes appear to show a differentiation of
sediment texture along the route. The boreholes taken from the area stretching from
the coastline west of the Ras ‘Ushayriq peninsula to the reef are composed of sand
and gravel, whereas most of those taken from the area to the west of the reef are
composed of finer, more silty sediments. There were very few grab samples taken
from this area so it is difficult to compare them with the borehole data, but the three
grab samples that were taken to the east of the reef (46, 56 and 95) were all
categorised as sand.
148
4.3.2.2 Video Transects
Figure 67 Video transects overlain on surface model.
Figure 68 Video transects overlain on the classified data.
(m)
149
The seabed habitat video survey provided additional optical ground-truthing
information (Figures 67 and 68). This was concentrated along the route of the
causeway, and served as a useful addition to the borehole data that was obtained
along the causeway route. A summary of the seabed descriptions from the video
transects is provided in Table 3.
Table 3: Seabed descriptions summarised from the video survey information provided by Creocean
Transect
No.
Summary of seabed character Predominant
Acoustic Class
Features
visible in
Surface Model
3a Flat, sandy bottom, hard substrate visible beneath
sand layer
Class 3 (red)
3b Sand covering hard substrate, some gravels Class 3 (red)
8a Flat, sandy bottom, ripple marks Class 3 (red) Linear features
8b Flat, sandy bottom, hard substrate visible
underneath, some gravels, some sand
banks/ripples
Class 3 (red) Linear features
13a Sand and gravels, hard substrate visible
underneath, sand ripples
Class 5 (lilac) Ridges
13b Coral/rock on hard substrate, covered in sand,
sand banks
Class 5 (lilac) Ridges
18a Sand and gravels Class 3 (red)
18b Sand bank, gravels, some coral Class 3 (red)
23a Sand, rock or coral debris Class 5 (lilac)
23b Sand, calcareous debris Class 3 (red)
28a Sand bank, coral, hard calcareous bottom Class 8 (lime) Edge of reef
150
Transect
No.
Summary of seabed character Predominant
Acoustic Class
Features
visible in
Surface Model
28b Gravels, sand banks, coral, hard substrate Class 10 (blue) Edge of reef
33a Reef No data Reef
33b Reef No data Reef
38a Reef Class 8 (lime) Reef
38b Reef Class 8 (lime) Reef
103a Sand, seagrass bed Class 3 (red)
103b Sand, seagrass bed Class 3 (red)
108a Sand, coral Mixed Edge of reef
108b Sand, coral, undulating Mixed Edge of reef
113a Coral, some sand Class 8 (lime) Reef
113b Coral, some sand Class 8 (lime) Reef
118a Flat, sandy bottom, featureless Class 4 (yellow)
Class 9 (salmon)
118b Flat, sandy bottom, few gravels, featureless Class 4 (yellow)
Class 9 (salmon)
The video transects indicate that the seabed west of the Ras ‘Ushayriq peninsula
largely consists of a flat sandy bottom, with hard substrate often visible, and some
ripple marks and sand banks. Moving westwards, coral and rocky debris appear in
more and more dense concentrations until the reef itself is reached. Away from the
reef to the southwest, the seabed appears to be flat, sandy and featureless, whilst to
the north it is sandy with seagrass beds.
151
4.3.3 Discussion
The homogeneity of the grab samples has meant that the distribution of grab
samples alone is not particularly informative in terms of clarifying the classes that
were generated by the acoustic classification, as there is not enough differentiation
between the samples to define patterns. Also, it is clear that all of the direct sediment
samples display a high sand content. However, integrated analysis of all of the
ground-truthing information together with the acoustic classification and the surface
model has provided considerable illumination on the character of certain parts of the
Study Area. Figure 69 shows a percentage breakdown of sediment types for each
class, derived from the different types of ground-truthing data obtained - boreholes,
vibrocores, grab samples and video transects. The predominant sediment type within
each video transect was allocated to the centre point of each transect location, to
align the data with the direct sample locations.
In the central part of the Study Area, a comparison of the results from the boreholes
across the causeway route to the surface model and the classified acoustic data
shows that the area to the east of the reef, appears to be relatively homogenous in
terms of depth and topographic features visible in the surface model, and also in
terms of acoustic classes. The classified acoustic data in this area is predominantly
class 3 (red), which, as mentioned earlier, roughly corresponded in the mosaic data
to areas of brighter reflective material forming possible sand ridges. The borehole
logs of sand and gravel, and occasionally calcarenite (caprock) in this area, all of
which would be likely to be brighter reflectors than silty sand, would seem to support
152
this interpretation, and the comparison graph (Figure 69) shows that this class has
the highest proportion of gravel samples out of all the classes.
The identification of the vibrocore survey area in the north of the Study Area as a
potential sand trap is significant in itself, in that it shows a correlation with the area of
the deep channel identified in the surface model. In addition, both of the potential
sand trap areas in the north and south of the Study Area also correlate with areas
that have been classified as yellow and salmon (classes 4 and 9 respectively),
probably representing finer sediment, in the acoustic classification. The comparison
graph (Figure 69) shows that Class 4 has the highest proportion of silt samples (apart
from the reef), and Class 9 has the smallest proportion of gravel samples apart from
Class 1 (the latter of which was also interpreted from the original classification as
possibly representing areas of finer sediment). As has already been noted in the
integrated analysis of the surface model and the classified data, there seems to be a
clear correlation between Classes 4 and 9 and areas of deeper sea. The sediment
sampling evidence has shown that these areas can also be correlated with areas of
thick sand/silty sand deposits. The single silty sand sample in the southwest of the
Study Area, lies in an area of deeper water that is coincident with the large area
classified as Class 4 (yellow), suggesting that it could represent a sediment-filled
basin. These results show a clear relationship between sonar reflectivity and seabed
depth within the Study Area, indicating that finer sediments are occurring in the
deeper areas. This is a significant finding as both the surface model and the
sediment samples provide independent evidence for the type of sediment and
character of the seabed that is represented by classes 4 and 9.
153
Figure 69 Graph displaying sediment types for each class as derived from ground truthing data (expressed as a percentage of the total number of
samples/observations recorded for each class).
Class No.
Sediment
Type (%)
154
The gravels identified in the southwest-northeast aligned deep channel in the north of
the Study Area could be related to the coarse-grained sands and gravels identified in
the boreholes to the west of the Ras' Ushayriq Peninsula. These may be part of the
Hofuf Formation sands and gravels that are found elsewhere around the western
side of Qatar, which were continental gravels from Saudi Arabia, laid down during the
Miocene (Cavalier, 1970, p.32). Sedimentological evidence from the western region
of Abu Dhabi has demonstrated that there was a large river system in the area in the
late Miocene (Hill et al., 2012, p.28). The presence of gravels in the Study Area, both
in the deep channel in the north, and also in the area to the east of the reef, supports
the theory that a former river bed is present in these parts of the Study Area, possibly
dating from the Miocene. It is possible that this channel was interrupted by uplift from
the Bahrain anticline, which could explain why it is so much more distinct in the north
of the Study Area than in the south.
The video footage has confirmed the overall character of the seabed in the area of
the causeway, and as far as possible the observations have broadly correlated with
the classes generated by the acoustic classification procedure. The interpretation of
Class 3 as representing a relatively highly reflective seabed sediment type is
supported, in that the seabed character provided by the video transects appears to
be mostly sand with visible rock (caprock?) substrate.
However, the video footage has proved to be of limited use for distinguishing fine
detail, as the visibility is not very good in some of the transects, and they were filmed
relatively quickly, so it is not easy to see the sediment types in detail and establish
155
whether the seabed sediments are coarse or fine-grained. There is not sufficient
scope or detail in the video footage to be able to draw out some of the differences in
sediment type that have been identified through the acoustic classification. For
example, there appears to be no great difference between the observations in
Transects 8A and 8B, which are predominantly Class 3, and the observations in
Transects 13A and 13B, which are predominantly Class 5. The comparison graph
(Figure 69) shows proportionately more gravel samples from Class 3, and more silt
samples from Class 5.
Also, the transects were not as useful as hoped for in clarifying some of the largerscale
landscape features that were visible in the surface model and the mosaic of the
sidescan sonar data, as the field of vision is too narrow. However, it has been
possible to confirm from the video footage that what had been tentatively interpreted
from the bathymetry as possible linear reef structures to the west of the Ras
‘Ushayriq peninsula are actually sand ripples. More panoramic views of the seabed,
which could be stitched together to create photo mosaics of large areas would be of
great benefit for this type of landscape-scale investigation.
156
CHAPTER 5: SECONDARY CHARACTERISATION
The following chapter describes the methodology and results of the secondary
characterisation. This comprises two different strands - the identification of
geophysical anomalies within the sidescan sonar data, and the clarification of
geophysical signatures through diver inspection and high-resolution geophysical
survey. Each strand has three sections consisting of the detailed methodology, the
results and a discussion.
5.1. Identification of Geophysical Anomalies
5.1.1 Methodology
A comprehensive and systematic analysis of the entire sidescan sonar dataset for
the Qatar-Bahrain Causeway was undertaken in order to identify all geophysical
anomalies that may represent either the remains of shipwrecks, or potential locations
for archaeological sites and/or palaeoenvironmental remains. As mentioned
previously (see Chapter 2), a limited analysis of selected geophysical anomalies from
the dataset was carried out in 2010 (Cuttler et al., 2011a), but the dataset had never
been systematically assessed in its entirety for archaeological purposes. The files of
data from the parallel tracks covered by the survey boat, obtained from the survey
company in xtf format, were batch-imported into a specialist software package called
SonarWiz 5, produced by Chesapeake Technology Inc. This is a sidescan and subbottom
sonar mapping software that is particularly useful for creating mosaic images
157
of the sonar data, and for the logging of anomalies (contacts) detected in the data.
These files were imported, batch processed and stitched together to create a mosaic
image of the seabed. Thresholding was set at the import stage, and following import,
gain processing was applied to all files. The processing was designed to create the
best possible image for aiding visual interpretation of the data. Testing was carried
out on a subset of the data in order to ascertain the optimum settings for viewing the
data without losing any of the information that the data contained. The testing
established that the best results were obtained by using the Empirical Gain
Normalization (EGN) function, set to an intensity of 41. The settings and parameters
used in the final import and processing are included in Appendix 4.
Bottom tracking had already been carried out on the data, but this was checked
individually before commencing the logging of anomalies for each file, and any
necessary adjustments were made manually. Every file in the dataset was then
examined in the digitizing view in Sonarwiz and any anomalies identified were
recorded using the contact manager. Recording consisted of logging the coordinates
of each anomaly, recording the measurements (length and width, and also shadow
and scour where appropriate), and writing an objective, non-interpretational
description. Finally, two attributes were systematically logged for each anomaly, a
preliminary interpretation (e.g. debris, buried feature, depression, seabed scar) and a
confidence level for likely anthropogenic origin. All of the anomalies were graded with
a confidence level of either high, medium or unknown, apart from those that had
been ground-truthed during the diver surveys in 2011, which were graded as ‘ground-
truthed’.
158
All anomalies were initially recorded without reference to the previous work carried
out in 2010 (Cuttler et al., 2011a), or to any other datasets. This was done in order to
be as objective and consistent as possible, so that it was possible to test how useful
the recording of anomalies was as a technique for deriving meaningful data at a
landscape level. The results of the diver inspections carried out in 2011 (Al-Naimi et
al., 2012) were subsequently used to refine the interpretation and confidence levels.
The criteria for recording anomalies were not the conventional criteria normally used
for commercial surveys, where the aim is generally to record potential obstructions on
the seabed. In this study, any area that looked slightly different to the surrounding
seabed was logged as an anomaly. Anomalies were graded as high confidence
where they appeared to be similar in character to those anomalies known from the
2011 work to be anthropogenic. Anomalies were graded as medium confidence
where there were reflectors showing in the data, but the signature was less welldefined,
presumably because of the nature of the material or feature causing the
anomaly, or because the anomaly lay at the far edges of the swath range and was
therefore more difficult to resolve. Any other type of anomaly was recorded as
unknown confidence, and these included any areas of seabed that looked different
from the surrounding area. This grade covered more ephemeral, poorly-defined
anomalies.
Specifically, anomalies were logged if they could possibly indicate debris of
anthropogenic origin, if they could indicate buried structures or features such as
sediment mounds and sand waves, or indicate geological features or other natural
locations that may have archaeological or palaeoenvironmental potential, such as
159
depressions. Sometimes it was not possible to categorise an anomaly other than that
it looked different to the surrounding seabed, in which case it was logged as
'unclassified'. The list of terms used to categorise the anomalies is provided in Table
4 below.
Table 4: List of terms used for initial categorisation of anomalies.
Categorisation Term
Buried Features
Debris
Depression
Disturbed Seabed
Hole
Linear Debris
Mound
Natural Location
Partially Buried Object
Seabed Scar
Sediment Accumulation
Unclassified
It should be noted that high confidence does not necessarily mean high
archaeological potential, as the diver inspections in 2011 showed that the most
obviously anthropogenic anomalies were most likely to be of recent origin, largely
caused by dumping of cars, tyres and other objects to create artificial fish reefs (AlNaimi
et al., 2012). In fact, it was considered that the anomalies that were classified
as unknown confidence may be of greater archaeological potential. It is these types
of more ephemeral anomaly that were perceived to be good candidates for further
160
research, as so little was currently known about their character or archaeological
potential. As a result of this recording policy, a great number of anomalies were
identified - 795 in Area 1, 184 in Area 2/3, and 118 in Area 4. All of the recorded
anomalies were reviewed, and a representative range was selected to be further
clarified and validated via the analysis of other data sets, including bathymetry, and
through further survey, including high resolution geophysical survey and diver
inspection. The complete register of all recorded anomalies is provided in Appendix 5
and the results of the analysis are presented and discussed below.
5.1.2 Results
5.1.2.1 Geophysical Anomalies
As discussed in the methodology section, the criteria for recording anomalies was to
log any areas that looked slightly different to the surrounding area, and accordingly, a
complete distribution plot of all recorded anomalies is shown in Figure 70. Most of
these anomalies do not have a confirmed interpretation, but this figure shows the
initial distribution of where anomalies were visible in the data. It is immediately clear
that there is a high concentration in the northwest of the Study Area (the western part
of Area 1), and a much sparser distribution in the east of the Study Area (Area 2).
The gap in the southwest is due to the presence of the reef, where no sidescan sonar
data was available.
161
Figure 70 Distribution of all recorded anomalies within the Study Area.
Figure 71 shows the distribution of all anomalies that were categorised as being
possible debris (green dots), and therefore of potential anthropogenic origin. The
differential distribution of these is quite marked - 196 of these were in Area1, 16 were
in Area 2/3 and only 2 in Area 4. Within Area 1, there is a clear concentration in the
west of the area. All of the debris anomalies in Areas 2/3 and 4 were classed as
medium confidence. In Area 1, 112 were classed as medium confidence and 84 were
classed as high confidence. In addition to these, 13 anomalies in Area 1 (the pink
crosses) were also logged as debris, and when cross-checked against the work that
162
was done in 2011 (Cuttler et al., 2011a), were found to have been diver-inspected,
and all proved to be dumped cars and/or tyres and old fishing equipment.
Figure 71 Anomalies interpreted as potential debris, and proven to be debris from ground-truthing.
Most of the high confidence anomalies showed up very clearly as modern debris
during the logging of anomalies. When compared to those that had been groundtruthed
through diver inspection, the signatures were very characteristic. They
consisted of one or more very bright reflectors (the debris), usually quite sharply
outlined, with clear acoustic shadows, often lying in a darker area of seabed
163
representing scouring around the debris, and occasionally including brighter patches
indicating areas of coarse sediment build-up around the debris (Figure 72).
Figure 72 Examples of high-confidence anomalies in Area 1: Modern debris relating to artificial reefs
(cars, tyres etc).Clockwise starting from the top left corner: QBC_Q10126, QBC_Q10270,
QBC_Q10334, QBC_Q10902. (Each image depicts an area measuring 100m x 100m.)
164
This type of anomaly occasionally generated a slightly different signature when
located at the outer range of the swath, where the sonar signal was more attenuated.
In these cases, the bright reflectors were less sharply defined, did not always display
clear shadows, and usually appeared in a dark area of seabed. However, there were
a few survey lines running perpendicular to, and overlapping, the main body of
survey lines through Area 1, and where anomalies could be cross-referenced with
other survey lines in which they didn't lie in the outer range, it was possible to clarify
the signature type and apply it to other similar anomalies. Figure 73 shows the same
anomaly visible in different survey lines. The image on the right shows the anomaly
lying at the outer range of the swath, where it is brighter but less clearly defined than
in the image on the left, where the same anomaly lies in the centre of the swath.
Figure 73 The same anomaly (QBC_Q10815) visible in different survey lines: Modern debris relating
to artificial reefs (each image depicts an area measuring 100m x 100m).
Anomalies like these were not generally selected for further study, as it was clear
from the ground-truthing work carried out in 2011 that they represent the dumping of
165
modern debris, usually cars and tyres, probably for the creation of artificial reefs (AlNaimi
et al., 2012).
Figure 74 shows the distribution of anomalies logged as possible depressions. This
was a very tentative classification, as it was not known how depressions may show
up in sidescan sonar data in this area, as no one has ever looked for them here
before. Features were classed as possible depressions if they displayed what
appeared to be finer seabed material within areas of surrounding coarser material,
the implication being that this fine material may be sediment that has accumulated in
hollows or depressions. It is possible that these features may simply represent
differential sediment accumulation, but if they are sediment-filled depressions, they
could have the potential to contain palaeoenvironmental remains or buried structures
and features, and therefore they were recorded as possible depressions so that they
could be easily selected for further investigation. Depressions may also represent
locations that were once favourable for human settlement. Some of these anomalies
are quite large features, ranging in size from around 20m to over 100m in length (see
Figure 75).
166
Figure 745 Distribution of anomalies that are possibly depressions.
Figure 75 A selection of anomalies that were logged as possible depressions. From left to right:
QBC_Q20071, QBC_Q40064, QBC_Q40119 (each image depicts an area measuring 100m x 100m).
167
A range of different types of anomaly were selected for more detailed examination,
either by cross-referencing with other datasets, such as bathymetry, or by higher
resolution geophysics survey or diver inspection. The range was selected in order to
try and clarify the signatures of types of anomaly that appeared different to those that
had been investigated by the work carried out in 2011. As mentioned earlier,
anomalies representing obvious modern car-related debris were avoided, but some
anomalies classified as possible debris were selected for further study in case they
could represent debris from shipwrecks. Other types of anomaly selected included
locations of possible buried features/objects, and possible depressions, hollows,
mounds, sediment accumulations and other natural features that may represent
topographic locations of potential interest for human settlement. Figures 76 to 81
show representative examples of the types of anomaly that were selected for further
study, and the full list of selected anomalies is provided in Table 5.
168
Table 5: List of anomalies selected for more detailed examination.
Contact No. Latitude Longitude Confidence Level Preliminary
Interpretation
QBC_Q10001 26.1641788483 51.0675086975 High Confidence Debris
QBC_Q10008 26.1921634674 51.0260429382 Medium Confidence Buried Feature?
QBC_Q10027 26.1201820374 50.9740638733 Unknown Confidence Buried Feature?
QBC_Q10098 26.1222991943 50.9296150208 High Confidence Debris
QBC_Q10099 26.1220684052 50.9299507141 High Confidence Debris
QBC_Q10126 26.1277561188 50.9390869141 Medium Confidence Buried Feature?
QBC_Q10159 26.0871925354 51.0052795410 Medium Confidence Buried Feature?
QBC_Q10173 26.0903301239 51.0237998962 Unknown Confidence Natural Location
QBC_Q10176 26.0943851471 51.0183601379 Medium Confidence Natural Location
QBC_Q10195 26.0978145599 51.0285263062 Medium Confidence Buried Feature
QBC_Q10202 26.1259479523 50.9980239868 Medium Confidence Natural Location
QBC_Q10235 26.0509109497 50.9292297363 Unknown Confidence Depression?
QBC_Q10345 26.0795593262 50.9499397278 Unknown Confidence Natural Location
QBC_Q10362 26.0913200378 50.9708328247 High Confidence Buried Feature?
QBC_Q10376 26.0902423859 50.9799804688 Unknown Confidence Natural Location
QBC_Q10574 26.1638336182 50.9941406250 Unknown Confidence Depression?
QBC_Q10582 26.1781387329 50.9776573181 Unknown Confidence Buried Feature?
QBC_Q10634 26.1872463226 51.0019569397 Medium Confidence Partially Buried Objects?
QBC_Q10645 26.1772842407 51.0245933533 Medium Confidence Partially Buried Objects?
QBC_Q10676 26.2013893127 51.0221672058 Unknown Confidence Buried Feature?
QBC_Q10702 26.2007694244 51.0457038879 Unknown Confidence Depression?
QBC_Q10726 26.1942138672 51.0716514587 Unknown Confidence Buried Feature?
QBC_Q10729 26.0780544281 50.9166069031 Unknown Confidence Buried Feature?
QBC_Q20008 26.0396232605 50.9539299011 Unknown Confidence Natural Location
169
Contact No. Latitude Longitude Confidence Level Preliminary
Interpretation
QBC_Q20032 25.9976711273 51.0048599243 Unknown Confidence Depression?
QBC_Q20045 26.0156154633 50.9666290283 Medium Confidence Debris
QBC_Q20046 26.0135402679 50.9689636230 Unknown Confidence Natural Location
QBC_Q20060 26.0014724731 50.9691925049 Unknown Confidence Mound?
QBC_Q20066 25.9976844788 50.9691009521 Unknown Confidence Natural Location
QBC_Q20068 26.0000514984 50.9640655518 Unknown Confidence Debris?
QBC_Q20071 26.0349769592 50.9097480774 Unknown Confidence Depression?
QBC_Q20080 26.0018577576 50.9552879333 Unknown Confidence Buried Feature?
QBC_Q20088 26.0020446777 50.9487915039 Unknown Confidence Depression?
QBC_Q20093 26.0029048920 50.9438056946 Unknown Confidence Linear Debris?
QBC_Q20094 25.9825687408 50.9758758545 Unknown Confidence Hole
QBC_Q20098 25.9928531647 50.9495239258 Unknown Confidence Depression?
QBC_Q30107 25.9956703186 50.9137268066 Unknown Confidence Unclassified
QBC_Q30113 25.9930953979 50.9116210938 Unknown Confidence Linear Debris?
QBC_Q30116 25.9878578186 50.9224739075 Unknown Confidence Depression?
QBC_Q30117 25.9647445679 50.9630203247 Unknown Confidence Depression?
QBC_Q30121 25.9924755096 50.8957824707 Unknown Confidence Natural Location
QBC_Q30124 25.9860668182 50.9170112610 Unknown Confidence Unclassified
QBC_Q30132 25.9577922821 50.9669189453 Unknown Confidence Depression
QBC_Q30146 25.9918575287 50.8720436096 Unknown Confidence Linear Debris?
QBC_Q30147 25.9881267548 50.8642044067 Unknown Confidence Linear Debris?
QBC_Q30149 25.9908180237 50.8585395813 Unknown Confidence Debris?
QBC_Q30174 26.0338172913 50.8948898315 Unknown Confidence Buried Feature?
QBC_Q30175 26.0339908600 50.8949546814 Unknown Confidence Debris?
QBC_Q30176 26.0340347290 50.8953170776 Unknown Confidence Buried Feature?
QBC_Q30177 26.0348167419 50.8955154419 Unknown Confidence Linear Debris?
170
Contact No. Latitude Longitude Confidence Level Preliminary
Interpretation
QBC_Q30178 26.0456485748 50.8976631165 Unknown Confidence Depression?
QBC_Q30180 26.0327358246 50.8860626221 Unknown Confidence Debris?
QBC_Q30183 25.9920825958 50.8459053040 Unknown Confidence Unclassified
QBC_Q30189 26.0602684021 50.8891639709 Medium Confidence Debris?
QBC_Q40024 25.9772624969 50.8816719055 Unknown Confidence Natural Location
QBC_Q40026 25.9844532013 50.8379058838 Unknown Confidence Natural Location
QBC_Q40031 25.9713287354 50.8812484741 Unknown Confidence Natural Location
QBC_Q40049 25.9601192474 50.8865165710 Unknown Confidence Natural Location
QBC_Q40055 25.9574604034 50.8707466125 Unknown Confidence Sediment Accumulation?
QBC_Q40059 25.9588909149 50.8624877930 Unknown Confidence Depression?
QBC_Q40064 25.9563694000 50.8669052124 Unknown Confidence Depression?
QBC_Q40071 25.9561614990 50.8610496521 Unknown Confidence Buried Feature?
QBC_Q40072 25.9574222565 50.8355674744 Unknown Confidence Unclassified
QBC_Q40075 25.9515342712 50.8676567078 Unknown Confidence Natural Location
QBC_Q40089 25.9422588348 50.8270797729 Unknown Confidence Debris?
QBC_Q40097 25.9399833679 50.8259239197 Unknown Confidence Natural Location
QBC_Q40100 25.9359226227 50.8315277100 Unknown Confidence Debris?
QBC_Q40102 25.9289989471 50.8649368286 Unknown Confidence Depression?
QBC_Q40113 25.9325790405 50.8269119263 Unknown Confidence Unclassified
QBC_Q40118 25.9579067230 50.8681564331 Unknown Confidence Depression?
171
Figure 76 Anomalies logged as possible debris. From left to right: QBC_Q40089, QBC_Q10001 (each
image depicts an area measuring 100m x 100m).
Figure 77 Anomalies logged as areas of possible sediment accumulation. From left to right:
QBC_Q40055, QBC_Q20060 (each image depicts an area measuring 100m x 100m).
172
Figure 78 Anomalies logged as natural features of potential topographic significance. From left to right:
QBC_Q40026, QBC_Q20046, QBC_Q10195 (each image depicts an area measuring 100m x 100m).
Figure 79 Anomalies logged as potentially partially-buried objects or features. From left to right:
QBC_Q10126, QBC_Q10202 (each image depicts an area measuring 100m x 100m).
Two anomalies from this selection were of particular interest, a small, but very
pronounced hole or crater in the seabed (QBC_Q20094) and the possible linear
feature (QBC_Q10729). The former looked like it could potentially have been either a
small limestone sinkhole, or a hole/crater created as a result of gas venting from the
seabed. The character of the linear feature was difficult to determine, but it appeared
to be composed of slightly raised, bright linear reflectors lying parallel to, and
perpendicular to, each other. This anomaly was investigated by diver inspection in
173
2011 (Al-Naimi et al., 2012, p.252) but the divers reported nothing visible on the
seabed at that location that would explain the anomaly.
Figure 80 A potential seabed crater: QBC_Q20094 (Image depicts an area measuring 100m x 100m).
Figure 81 Long linear anomaly: QBC_Q10729/Bham0028 (Image depicts an area measuring 100m x
100m).
174
In addition, a high concentration of debris-related anomalies was identified in a
cluster in the west of Area 1 (Figure 82). This cluster of debris could be indicative of
the way that debris is spread around the seabed after a shipwreck, and so this area
was also considered to be of particular interest.
Figure 82 Possible debris cluster in Area 1 (the green grid lines represent 500m x 500m squares).
175
5.1.2.2 Geophysical Anomalies and Topography
It was not possible within the timescale of this research, or the resources of the
QNHER project, to carry out visual observation and/or intensive geophysical survey
of all of the anomalies selected for further study. However, as part of the integrated
analysis and testing of methodologies, the topographic surface generated from the
LiDAR bathymetry was cross-referenced with the selected anomalies to see if they
could be further clarified. The location of each one of the anomalies selected for
further study was examined in the surface model, and the results analysed.
Area 1
Out of 23 anomalies selected for further study, nine did not lie within the area
covered by the surface model, eight did not show up in the surface model at all, and
five were clearly natural features that did not differ significantly from the surrounding
seabed. The remaining anomaly from the selected set was the long linear anomaly
(QBC_Q10729) already investigated in 2011. Examination of the surface model in
the area of this anomaly does show a feature that could possibly relate to the
reflective anomaly (Figure 83), but it is not possible to establish conclusively whether
it represents a natural feature or not.
176
Figure 83 Long linear anomaly (QBC_Q10729/BHAM0028) visible in the sidescan sonar data, and the
surface model of the seabed from the same location (Both images are at same scale and orientation).
Area 2
Out of 13 anomalies selected for further study, three did not show up in the surface
model at all, and ten were clearly natural features that did not differ significantly from
the surrounding seabed.
None of the anomalies that were recorded as possible debris in Area 2 were visible in
the surface model at all (QBC_Q20012, QBC_Q20045, QBC_Q20063 and
QBC_Q20074). One of these, QBC_Q20045, although not visible itself in the surface
model, is surrounded by several other anomalies that were recorded as bright
reflectors, and all of these can be seen from the surface model to be rocky/coral
outcrops (Figure 84).
Direction of
current (swings
round with tide)
177
Figure 84 Bright reflective anomalies in Area 2, shown in the surface model to be natural outcrops.
The anomaly that looked like a possible sinkhole or gas vent on the sidescan sonar
data (QBC_Q20094), did not look particularly different to other possible holes in the
same area (see QBC_Q20096 to the southwest on Figure 85). This is probably
because the surface model is not of sufficient resolution to clarify features that are
this small.
Figure 85 Seabed crater (QBC_Q20094) visible in the sidescan sonar data (image depicts an area
measuring 100m x 100m), and the surface model of the seabed from the same location, showing
anomalies QBC_Q20094 and QBC_Q20096.
178
Area 3
Out of 18 anomalies selected for further study, all but two were clearly natural
features that did not differ significantly from the surrounding seabed. The other two,
which were initially recorded as possible debris from the sidescan data, could not be
seen in the surface model.
All of the anomalies that were logged as possible debris in Area 3, including a cluster
near the reef (Figure 86), could be seen from the surface model to be natural
features, largely coral heads.
Figure 86 Bright reflective anomalies in Area 3, shown in the surface model to be natural coral
outcrops.
179
Area 4
Out of 16 anomalies selected for further study in Area 4, four did not lie within the
area covered by the surface model, six did not show up in the surface model at all,
and six were clearly natural features that did not differ significantly from the
surrounding seabed.
All of the anomalies that were logged as possible debris in Area 4, mainly in the
southwest corner of the area, could be shown from the surface model to be rocky
outcrops (Figure 87).
Figure 87 Bright reflective anomalies in the southwest of Area 4, shown in the surface model to be
natural outcrops.
Many of the anomalies logged as potential depressions appear to be areas that are
in the acoustic shadow of sand banks, rather than actual seafloor depressions. This
includes the cluster of potential depressions in Area 4, and that in the north of Area 2.
180
5.1.3 Discussion
Overall, the identification and analysis of the geophysical anomalies in this lowresolution
sidescan sonar dataset has demonstrated that it has limited application as
a technique for investigating submerged landscapes. It has not proved a suitable
method for locating, identifying and clarifying large-scale features relevant to the use
of the former landscape by humans. However, the ability to identify recent debris on
the sea floor and produce distribution maps of the anomalies at a landscape level
does provide some useful insights into more recent human exploitation of the marine
environment, which does have a bearing on seabed characterisation.
It is clear from the distribution of probable debris-related anomalies in Area 1 (Figure
71) that these occur more densely further out from the shore, in the west of the area.
This could be due to the increased risks both of causing a shipping hazard, and of
being caught dumping material illegally, when closer to the shore. It may also be
related to the fact that areas in the south are further away from the main shipping
routes around the western Gulf. However, the concentration could also be due to
environmental factors, such as the need to create artificial reefs in deeper water, or
even the seafloor conditions in certain areas influencing the visibility of seafloor
debris in the sidescan sonar data. It may be more difficult, for example, for debris to
show up well in areas of seabed where there may be a lot of other bright reflective
material such as rocky outcrops or coral heads. It could also be that in these areas
the debris gets covered by vegetation and/or sediment more quickly.
181
It was hoped that integrated analysis of the sonar anomalies with the surface model
would be helpful in clarifying the anomalies, but this proved not to be the case with
the smaller anomalies. With a few exceptions, the anomalies that were interpreted as
debris are either not clearly visible, or not visible at all in the surface model. This
could be because the debris was not present when the LiDAR survey was carried out
in 2005 (the sidescan sonar data was collected in 2008), or more likely because the
debris is too small to show up on a surface model created from data at a resolution of
7m. This is illustrated quite clearly in Figure 88, which shows anomaly QBC_Q10051
as visible in the sidescan sonar data, the spacing of the LiDAR points in the same
area, and the surface model created from those points, which does not show the
anomaly at all.
Figure 88 Anomaly QBC_Q10051 (modern debris): sidescan sonar, LiDAR points and surface model.
182
Another notable example is QBC_Q10128, showing very clearly in the sidescan
sonar, which diver-inspections in 2011 proved to be a car reef (Bham 0017). This
anomaly is not visible in the surface model at all, whereas QBC_Q10127 to the north
(not diver-inspected) is visible in the surface model (Figure 89). The difference in
visibility of the two anomalies could be due to the difference in size of the debris, or
the distribution of the LiDAR points, or it could simply be that anomaly QBC-Q10128
represents debris that was dumped at a later date, after the LiDAR survey was
undertaken.
Figure 89 Anomaly QBC_Q10128 (car reef) visible in the sidescan sonar but not visible in the surface
model.
Although, as previously stated, the analysis of the geophysical anomalies has not
proved useful for landscape-scale investigations, the surface model used in
conjunction with the sidescan sonar data has been extremely useful for clarifying
some of the larger and more tentatively-interpreted anomalies, and has largely
confirmed that most of these are natural features such as sand ridges, reef
183
structures, and rock or coral outcrops, that do not hold particular significance for
archaeological or palaeoenvironmental potential.
Anomaly QBC_Q10376 is a good example of why it is not suitable to use geophysical
anomalies, and survey data that contains large gaps, as a method for identifying
landscape-scale features. When looking at the surface model, it is clear that this
anomaly is a part of a larger landscape feature, but that wasn't at all obvious when
analysing the survey line in isolation (Figure 90).
Figure 90 Anomaly QBC_Q10376 (topographic location of potential interest for human settlement) in
the sidescan sonar data (image depicts an area measuring 100m x 100m), and in its landscape
context in the surface model.
However, despite the limitations, this analysis has identified two anomalies of
potential interest which still merit further investigation, the long linear feature
(QBC_Q10729) and the seabed hole (QBC_Q20094).
184
5.2 Clarification of Geophysical Signatures
5.2.1 Methodology
5.2.1.1 Diver Inspections
The diver inspections carried out on selected geophysical anomalies in 2011
provided very useful ground-truthing data for the clarification of geophysical anomaly
signatures in the sidescan sonar data. These diver inspections were targeted on a
particular type of anomaly, largely the type with well-defined bright reflectors, often
with associated scour. This type of anomaly was targeted because at the time there
was no knowledge-base of geophysical signature types for the area, so it was
thought that these may represent shipwrecks, or debris from shipwrecks. In the
event, they all proved to be the result of modern dumping of cars, tyres and other
material in order to create artificial reefs, or fish traps. However, following the
systematic assessment and recording of anomalies that was undertaken as part of
this characterisation, and in the light of the information gained from the 2011 diver
inspections, a different range of anomalies were selected for ground-truthing by diver
inspection. The selection did not prioritise the well-defined, clearly anthropogenic
anomalies, instead focusing on the more ephemeral anomalies, such as potential
depressions, sand waves, areas of differential sediment accumulation and sediment
mounds. The reasoning for this was that these types of anomalies may indicate
buried objects and features, such as older shipwreck debris, ballast dumps or even,
potentially, archaeological sites. Even if these anomalies did not prove to be of
185
archaeological or palaeoenvironmental interest, the diver inspections would help to
build up the knowledge base of geophysical signature types for the area.
5.2.1.2 High-Resolution Geophysical Survey
Further intensive geophysical survey was carried out at a specific, targeted location
in order to compare the results with the low-resolution survey, and further clarify
selected anomalies. Some testing had already been carried out in the Bay of AlZubārah
in 2011 using a high-resolution sidescan sonar, and a marine
magnetometer, which is designed for the detection of ferromagnetic anomalies
(Cuttler, 2011). Further survey in 2013 was carried out using a Klein Hydroscan,
which is capable of collecting very high resolution sidescan data over large areas.
This data was collected at frequencies of 455KHz and 900KHz. The 900KHZ
frequency provided high resolution images up to 75m (total swath of 150m), while the
455KHz simultaneously collected data up to 175m (350m total swath). During the
survey the total swath width was limited to 150m to ensure a high-resolution dataset.
5.2.2 Results
5.2.2.1 Diver Inspections
Eight of the selected anomalies were subject to ground-truthing via diver inspection.
These eight were chosen for diver inspection partly because they represented good
examples of the anomaly signatures that needed to be clarified, and partly because
186
they were situated in the north of the Study Area and were therefore comparatively
accessible for the survey boat, which was based at the port at Al Ruwais. The results
can be shown in Table 6 below, and unfortunately, none of the investigated
anomalies produced useful information. The anomalies either proved to be natural
features of low archaeological and palaeoenvironmental potential, or no features
could be observed at the location (Figures 91 and 92). However, those locations
where nothing was found should not be dismissed as of no interest on the basis of
these diver inspections, since the divers reported that visibility was extremely poor,
less than 5m in most cases (see Figure 92), and it is clear that methodologies for
ground-truthing these types of anomaly need to be improved.
Table 6: Summary of results of diver inspections on selected anomalies.
Anomaly No. Longitude Latitude
Preliminary
Interpretation
Interpretation Following
Diver Inspection
QBC_Q10702 51.0457038333333 26.2013893 Depression? No features observed
QBC_Q30183 50.845905304 25.9920825958 Sediment
Accumulation
No features observed
QBC_Q40118 50.8681564331 25.957906723 Depression Linear ridges
QBC_Q10001 51.0675086666667 26.1641788333333 Debris No features observed
QBC_Q10098 50.929615 26.1222991666667 Debris Natural undulation
QBC_Q10634 51.0019568333333 26.1872461666667 Partially Buried
Objects
Rock outcrop in sediment
basin
QBC_Q10645 51.0245933333333 26.1772841666667 Partially Buried
Objects
Rock outcrop in sandy
hollow
QBC_Q10726 51.0716513333333 26.1942138333333 Buried Feature Sediment basin
187
Figure 91 Anomaly QBC_Q10634 in Area 1: Sediment Basin (Photo by QNHER marine team, 2013).
Figure 92 Anomaly QBC_Q40118 in Area 4: Linear Ridges (Photo by QNHER marine team, 2013).
188
A further group of anomalies in the vicinity of the long linear anomaly (QBC_Q10729)
were diver inspected following further intensive geophysical survey, partly to test
positional accuracy, and these all proved, as expected, to be car reefs (Figure 93).
The results can be shown in Table 7 below.
Table 7: Summary of results of diver inspections on anomalies near QBC_Q10729.
Anomaly No. Longitude Latitude
Prelim
Interpretation
Dive
Interpretation
QBC_Q10336 50.92655265 26.0960172 Modern Debris Car reef
G2/03.05.14/L9/TG1 50.9134333333333 26.0759666666667 Bright reflector Rocks
G2/20.04.13/L4/TG123 50.919195 26.07834 Modern Debris Car Reef
G3/07.05.13/L8/TG1 50.9146033333333 26.0806816666667 Modern Debris Car Reef
Figure 93 Anomaly G3/07.05.13/L8/TG1:Car reef in Area 1 (Photo by QNHER marine team, 2013).
189
5.2.2.2 High-Resolution Geophysics
High-resolution geophysical survey was carried out in the area of the long, linear
anomaly (QBC_Q10729) that was previously identified as meriting further
investigation, in order to see if this type of investigation would be more successful at
clarifying its nature than the diver inspection in 2011 was (Cuttler, 2014). Although
the diver inspection in 2011 did not find anything at the location of the anomaly
(Figure 94), it was thought possible that the anomaly was simply too big (greater than
90m long and nearly 30m wide) for divers with limited visibility to gain a wide enough
view of the seabed to be able to distinguish the feature from the surrounding seabed,
especially as it did not appear to be protruding very high from the seabed.
Figure 94 Long linear anomaly QBC_Q10729/BHAM0028 (Photo by Hampshire and Wight Trust for
Maritime Archaeology, 2011).
190
The high resolution survey was carried out using a Klein hydroscan sidescan sonar
at 900KHz. East-west transects were surveyed, at a resolution of 900KHz, in the
location of the anomaly, and also covering a significant distance to the north and
south. However, nothing relating to the long linear anomaly could be identified in the
resulting data. Initially it was thought that there may be a positioning problem, so all
coordinates and projections were double-checked, but no error was found. The
conclusion was that the anomaly may have been very ephemeral and had
disappeared, probably due to the dynamic sedimentation environment of the seabed,
in the 5 years since the low-resolution survey was carried out. Finally, it was decided
that a southwest-northeast transect would be surveyed along the exact path of the
2008 survey transect in which the anomaly was originally recorded. This time, the
anomaly was identified, as shown in Figure 95.
Figure 95 Long linear anomaly (QBC_Q10729/BHAM0028) surveyed in high resolution (each image
depicts an area measuring 100m x 100m).
191
5.2.3 Discussion
The absence of the long linear anomaly in the east-west survey lines, and its
presence in the southwest-northeast survey lines, must be due to the effects of the
direction of the sonar on the creation of acoustic shadows. This is a significant
finding, as it clearly demonstrates how important the direction of survey is, and
consequently how important it is to have overlapping survey lines, preferably
surveyed from different directions (Cuttler, 2014).
It is also interesting to note that this anomaly did not show up any better in the highresolution
sidescan data than it did in the low-resolution data. It is not clear why this
should be the case with this anomaly, as other anomalies were considerably more
clear in the high resolution data, for example anomaly QBC_Q10815 (see Figure 96).
Figure 96 Anomaly QBC_Q10815 (car reef) surveyed in low resolution (left) and high resolution (right)
(each image depicts an area measuring 100m x 100m).
192
Unfortunately, after surveying the area around the long linear anomaly with lowresolution
and high-resolution sidescan sonar, examining the seabed topography in
the surface model generated from the bathymetry, and carrying out a diver
inspection, we are still no nearer to identifying what the anomaly actually is. The
suggested next step would be to use extensive optical techniques, such as an ROV
camera, to try and get photographic images of a wide area of the seabed around the
anomaly.
Overall, the results of the clarification of geophysical signatures through diver
inspection clearly show that diver inspection of selected anomalies has not produced
good results, and that this technique is not appropriate for investigating landscapescale
anomalies. This is partly down to poor visibility and the lack of wider vision that
is necessary for the interpretation of large topographic features. It is entirely possible,
for example, that a diver could actually be in the middle of a feature such as a large
depression but not even be aware of it due to the lack of panoramic vision. Diverinspection
is more likely to be successful when targeted at very specific, small-scale
locations, as with the artificial car reefs that were successfully identified and recorded
in 2011.
However, the comparison of the low and high-resolution data sets has proved very
useful for the testing of methodologies, as it demonstrates that the low-resolution
sidescan sonar is still sufficient to enable the identification of these more ephemeral
anomalies. This, however, does not remove the need or use for targeted highresolution
geophysics once areas of potential have been identified.
193
CHAPTER 6: DEFINING CHARACTER AREAS AND ASSIGNING POTENTIAL
6.1 Refining the Initial Landscape Units into Character Areas
The end process of the characterisation was to refine the initial landscape units, and
designate useable and meaningful Character Areas based on a set of shared
attributes, which could subsequently be used to generate zones of archaeological
and palaeoenvironmental potential. The refining of the initial landscape units was
done by manual cross-correlation of the results of the sediment texture analysis with
the topographic mapping and the ground-truthing data, supplemented with
information on more recent maritime exploitation from the secondary characterisation
process (Figure 97). These factors were used to guide the re-defining of the
boundaries of the initial landscape units into character areas, simplifying and
interpreting them using the different datasets, thereby creating more cohesive,
discrete areas, in order to better represent zones with shared characteristics.
194
Figure 97 Cross-correlation of different input datasets with the Initial Landscape Units.
195
Figure 98 Final Character Areas (labelled with Character Area numbers).
196
The result of this process was a reduction in the number of landscape units, from the
initial 90 (Figures 29 and 97) down to 37 larger character areas (Figure 98). Figure
98 shows the final character areas shaded according to their predominant acoustic
class, and Table 8 provides a description of each character area. Figure 98 shows
that the overall pattern of the character areas is still heavily based on the acoustic
classification, as expected since the acoustic data was the primary dataset. An
additional character area that lay outside the main Study Areas was added to the
dataset (Character Area 37), based solely on the information gained from the surface
model, as there was no sidescan sonar data or ground-truthing data available for
these area. This was the near-shore zone northwards of the Ras ‘Ushayriq
Peninsula, which contained a concentration of highly significant topographic features,
such as possible former shorelines and palaeochannels.
197
Table 8: Summary description of Final Character Areas
Unit ID
Acoustic
Class
Sediment Type Summary
Potential for
Extensive Sediment
Deposits
Potential for
Significant
Topographic
Features
Overall
Archaeological/
Palaeoenvironmental
Potential
0 10 Sand Flat, featureless sandy area Medium Low Low
1 9 Silty Sand Slight depression containing silty material Medium Medium Medium
2 8 Coral Ridge, with coral/rocky outcrops Low Low Low
3 5 Sand Slightly elevated area, with north-south trending sand
ripples
Medium Medium Medium
4 3 Coarse Sand Slightly elevated area of coarse sand, with north-south
trending sand ripples
Medium Medium Medium
5 9 Silty Sand Slightly lower, silty area adjacent to the reef, containing
deep sand deposits
High Medium Medium
6 4 Sand Deep area with extensive trawler scars on the seabed Medium Low Low
7 9 Silty Sand Silty area adjacent to the reef Medium Medium Medium
8 10 Sand/Coral Coral area adjacent to the reef Low Low Low
198
Unit ID
Acoustic
Class
Sediment Type Summary
Potential for
Extensive Sediment
Deposits
Potential for
Significant
Topographic
Features
Overall
Archaeological/
Palaeoenvironmental
Potential
9 5 Sand Area of small, east-west trending sand ripples Medium Low Medium
10 3 Sand Potential former bay/wadi channel, with east-west trending
small sand ripples
High High High
11 6 Sand Area of coarse sand with north-south trending large sand
ripples
Medium Low Low
12 3 Sand Area of coarse sand with north-south trending large sand
ripples
Medium Medium Medium
13 8 Coral Reef Coral reef Low Low Low
14 3 Sand Area of coarse sand with north-south trending large sand
ripples
Medium Medium Medium
15 5 Sand East-west trending and north-south trending sand ripples Medium Medium Medium
16 5 Sand/Gravel Flat, featureless area of coarse sand and gravel Medium Low Low
199
Unit ID
Acoustic
Class
Sediment Type Summary
Potential for
Extensive Sediment
Deposits
Potential for
Significant
Topographic
Features
Overall
Archaeological/
Palaeoenvironmental
Potential
17 3 Sand/Gravel/Coral Area of coarse sand with north-south trending large sand
ripples
Medium Low Low
18 10 Sand North-south trending, very pronounced mega ripples Medium Medium Medium
19 3 Sand Area of coarse sand, no topographic information available
but close to area of potential former shoreline
Medium No DEM Medium
20 9 Silty Sand Deep southwest-northeast trending channel containing
depressions, possible solution hollows, and deep sand
deposits
High High High
21 3 Sand Edge of potential former shoreline, with pronounced northsouth
trending sand ripples, coarse sand
Medium High High
22 10 Silty Sand Edge of potential former shoreline, with depressions and
potential solution hollows, and a high concentration of
debris of recent anthropogenic origin
High High High
200
Unit ID
Acoustic
Class
Sediment Type Summary
Potential for
Extensive Sediment
Deposits
Potential for
Significant
Topographic
Features
Overall
Archaeological/
Palaeoenvironmental
Potential
23 1 Sand Edge of potential former shoreline, with pronounced northsouth
trending sand ripples
Medium High High
24 5 Silty Sand East-west trending and north-south trending sand ripples.
Contains anomaly of potential archaeological interest.
Medium Medium Medium
25 5 Sand North-south trending, very pronounced mega ripples, and a
high concentration of debris of recent anthropogenic origin
Medium Medium Medium
26 5 Sand North-south trending, very pronounced mega ripples Medium Medium Medium
27 9 Silty Sand Edge of potential former shoreline, with potential solution
hollows, and a high concentration of debris of recent
anthropogenic origin
Medium High High
28 4 Sand Edge of potential former shoreline, with mounds,
depressions and a high concentration of debris of recent
anthropogenic origin
High High High
201
Unit ID
Acoustic
Class
Sediment Type Summary
Potential for
Extensive Sediment
Deposits
Potential for
Significant
Topographic
Features
Overall
Archaeological/
Palaeoenvironmental
Potential
29 9 Sand Area of sand, no Topographic Information Medium No DEM Medium
30 4 Sand Northern edge of deep southwest-northeast trending
channel, with depressions and mounds
High High High
31 3 Sand Area of coarse sand, tail of reef Medium Low Low
32 5 Sand Tail of reef Medium Low Low
33 5 Sand Area of finer sand containing linear depressions Medium Medium Medium
34 5 Sand/Gravel Area of patchy, coarse sand and gravel, with rocky
substrate visible in parts, and small sand ripples
Low Low Low
35 3 Sand/Gravel Area of patchy, coarse sand and gravel, with rocky
substrate visible in parts, and north-south trending large
sand ripples
Low Low Low
36 3 Sand Area of coarse sand with north-south trending large sand
ripples
Medium Medium Medium
202
6.2 Assigning Potential
The characterisation of the seabed in the Study Area represents an attempt to zone
the seabed based on extensive, continuous data sources, using a systematic
methodology. The character zones provide a broad-brush interpretation of the
character of the seabed, incorporating every part of the Study Area into the
characterisation, in accordance with the principles of historic landscape
characterisation. A fundamental principle of terrestrial historic landscape
characterisation is that the entire landscape is considered to be significant in its
entirety, as the whole of the present-day landscape, to a greater or lesser extent, has
been influenced by, and in turn has influenced, human activity. The landscape
characterisation data is usually considered to be a means of managing change within
the landscape in a non-selective, interpretative manner (Dingwall and Gaffney, 2007,
p.17). The characterisation data, therefore, is not in itself a representation of the
relative archaeological or historic significance of a particular area of landscape. The
same can be said of the character areas created by this seabed characterisation
process. However, characterisation data can be used as the basis for assessing the
archaeological significance of a landscape. Accordingly, this research aimed to take
a step beyond initial characterisation, by using the seabed character zones to provide
a coarse-grained assessment of archaeological and palaeoenvironmental potential.
To do this, the attributes of the character zones were further analysed and used to
assign levels of potential. The level of potential was assigned on the basis of the
probability of extensive sedimentary deposits occurring within the character area, and
203
on the presence in the area of topographic features known to have been influential in
the location of human settlement in the Early-Mid Holocene.
The presence of sedimentary deposits is a fairly rough measure of the potential for
preservation of archaeological or palaeoenvironmental deposits, working simply on
the basis that the presence of sediment indicates a greater chance of burial and
preservation compared to areas of exposed bedrock (Westley et al., 2011b).
Similarly, the present-day seabed topography is also a relatively rough measure of
potential, as the seabed surface may have changed considerably since the Early
Holocene (see Chapter 7 for further discussion on this issue).
The sedimentation rates within the Arabian Gulf as a whole differ markedly
depending on morphology and sediment type, and localised relief results in complex
sedimentation patterns. Erosion in the Gulf is focused around the shallow coast and
the topographic highs, and although the Study Area directly faces the direction of the
northwesterly 'Shamal' wind, it is also protected from the full-force of the associated
waves and currents by Bahrain and the Saudi Abrabian coast (KSEPL, 1973). As a
general figure, the averages of the lowest and highest values of sedimentation rates
in the different parts of the Gulf are 0.03-2.5 mm per year (Al Ghadban et al., 1998,
p.25), but more detailed studies on sedimentation rate are required for localised
areas. Geotechnical survey work has shown that sedimentation smooths out the
seabed surface, by accumulating in topographic depressions (Marin Mätteknik AB,
2002, p.17), but according to Kassler (cited in Purser, 1973, p.31) this sediment
accumulation has not yet completely masked the underlying topography. In order to
204
obtain a more specific assessment of potential we would need more information
about the sedimentation processes, for example how much the deposits are a result
of mobile features such as sand ripples. However, despite limitations in our
knowledge about sedimentation and erosions rates in specific areas, as a baseline
assessment at a landscape-scale, these rough measures are a good starting point
from which to base further work.
In their study of submerged landscapes in the North Sea, Gaffney et al. (2007, p.116)
used a ranking system to systematically score areas of seabed in their Study Area
based on potential for preservation. In this case they created two normalised
datasets with different sets of values, one based on landscape feature potential and
one based on depth of overlying sediments, and multiplied the two datasets. Areas
with both a lack of known features and low levels of sedimentation achieved low
scores, and conversely, areas with archaeologically significant features and
substantial sediment cover scored highly.
Assigning values for potential to the seabed character areas in Qatar was done on
similar principles, except that it was done at the polygon level, and a different ranking
system was used. Each character area polygon was assigned two different scores,
one for sedimentation potential and one for topographic feature potential. The score
values were Low (Value=1), Medium (Value=2) or High (Value=3). Topographic
features that led to a designation of high topographic potential for a character area
included coastlines, palaeochannels, inlets, spits, promontories and hollows. Other
topographic features such as raised areas and slight depressions led to a
205
designation of medium topographic potential, and character areas which were
relatively featureless were assigned a low topographic potential. The combination of
values for topographic potential and sediment potential generated the overall
potential score for each character zone, either Low (Combined Value of 2 or 3),
Medium (Combined Value of 4) or High (Combined Value of 5 or 6). As a result of
this process, areas of potential were defined and displayed, as shown in Figure 99.
206
Figure 99 Character Areas (labelled with Character Area numbers) showing overall potential.
207
The distribution map of the character areas, shaded by potential, clearly shows that
the centre of Area 1, and the near-shore zone northwards of the Ras ‘Ushayriq
Peninsula, are the areas of highest potential, where further investigations should be
concentrated. The centre of Area 1, where the deep southwest-northeast channel
occurs, clearly shows a cluster of high potential character areas, based on the
accumulation of sediment (as evidenced by the vibrocore logs) together with the
occurrence of topographic features that would have been attractive for human
settlement, such as channels, hollows and a potential coastline. This combination of
possible shoreline combined with the indication of extensive sedimentation could
suggest high potential for the preservation of a palaeo- land surface.
The near-shore Character Area (Area 37) north of the Ras ‘Ushayriq Peninsula has
been assessed as being of high potential without any other primary or secondary
characterisation information, apart from Character Area 10 at the north of the
potential extension of Wadi Debayan, which was covered by the acoustic sediment
classification. The high potential for these areas is due to the presence of two
putative former shorelines and a concentration of associated significant topographic
features, including the possible wadi extension and a cluster of palaeochannels.
The gentle gradient and generally smooth nature of the seabed in the area to the
north of the deep channel supports sedimentation over that area. However, the
extensive sand ripples in the north indicate that large quantities of mobile sand
overlie the substrate, which may have an effect on the preservation potential of such
sediments. It is also possible, however, based on features visible in the surrounding
208
areas, that these ripples could be infilling and masking other palaeochannels and
Karst-related depressions.
The area between the deep channel and the reef has largely been deemed to be of
medium potential, on the basis of the north-south linear depressions in the southeast,
and the silty material obtained from grab samples in the northwest. However, less
ground-truthing data is available from this area than from anywhere else in the Study
Area, so to some extent it is a rather unknown quantity, and would benefit from
further ground-truthing.
The character areas deemed to be of low potential in the south of Area 2 have been
designated as such largely due to the paucity of significant landscape features and
the outcrops of rocky substrate identified in the ground-truthing data (boreholes and
video footage). The borehole logs interpret this as caprock, formed by submarine
lithification of the marine deposits, so it does not exclude the possibility that preinundation
archaeological and palaeoenvironmental remains could survive beneath
this caprock, but palaeoenvironmental remains would be more difficult to access, and
it would be impossible to recover archaeological deposits.
The reef area and its immediate surrounding are not close to high-significance
landscape features, and there are numerous large coral heads outcropping above
the sea floor. These areas were therefore assessed and scored as being of low
archaeological and palaeoenvironmental potential. However, south of the reef, there
are generally finer-grained silts and muds, and this area of sediment accumulation
209
(character areas 5 and 7) may have greater potential than the immediate environs of
the reef. There may be greater potential for the occurrence of shipwrecks in this area,
but any such remains would be likely to be heavily masked by coral growth, and
therefore difficult to find and survey.
Character areas 3, 4 and 5 at the south of the Study Area have been assessed as of
medium potential, due to the deep sand deposits and the slightly elevated area at the
far south. The areas to the east and west are considered to be of low potential due to
the overall lack of significant topographic features and the limited information about
sedimentation in these areas. However, it may be that further ground-truthing in
these areas could change this assessment, since the area to the west (area 6) is one
of the deepest parts of the Study Area, and the sediment classification indicates that
there may be finer-grained sediments here. Further information would also be
needed on the extent of the damage done by the trawler scarring in this character
area.
As previously mentioned, there are limitations to the measures used for assigning the
potential scores, but on the basis of the evidence that we have, they provide a
generalised starting point to begin targeting resources at the areas most likely to
yield good results. Within these generalised areas it is already possible to pinpoint
specific areas for further investigation, such as possible solution hollows and
palaeochannels, and it is expected that as more work is done and the knowledge
base grows, then more such localised areas of potential will be identified.
210
CHAPTER 7: DISCUSSION
7.1 An Evaluation of the effectiveness of the methodologies, and suggestions
for further work
It is very important to evaluate how effective this characterisation methodology has
been, since using this combination of datasets (sidescan sonar, LiDAR bathymetry,
direct sediment sampling, video footage and diver inspection) together with acoustic
classification techniques and historic landscape characterisation techniques has
never been tried before. Furthermore, apart from the grab samples, the research was
undertaken purely using available data that had not been collected specifically for the
purposes of archaeological or palaeoenvironmental research. It has clearly raised
methodological questions and issues which need further exploration, and it is
necessary to examine those aspects that worked well and those that did not, and to
assess what datasets should be prioritised for future, research-specific data capture.
Overall, it is important to bear in mind that the value of the seabed characterisation
lies in its ability to aid understanding of the submerged prehistoric landscape and
delimit areas with archaeological and palaeoenvironmental potential that can be
targeted for more intensive investigation, rather than in its ability to identify individual
archaeological sites and/or shipwrecks, although discovery of the latter still remain a
possibility. Effectively-targeted surveys can provide a wealth of information about the
topography, environments and preservation conditions of the submerged landscape
even if no archaeological sites are discovered (Bailey, 2011, p.327).
211
The use of low-resolution sidescan sonar data in the characterisation allowed
relatively rapid coverage of large areas of seabed that couldn’t easily have been
covered using photographic and direct sediment sampling methods. However, this
sidescan sonar data was never captured with the intention of using it for
archaeological purposes, and one of the aims of the research was to see how
effective this low-resolution data could be for landscape-scale studies of the seabed.
The results clearly show that for seabed classification and landscape characterisation
purposes, which are inherently coarse-grained analyses, low-resolution data is ideal.
The data was not so detailed that it provided overly-complex results, and the acoustic
classification was able to highlight large-scale trends and changes in the seabed
sediment texture, which is exactly what was required for zoning purposes. Ideally,
there would have been no gaps between the survey lines in the north of the Study
Area, but the acoustic classification methods used allowed satisfactory extrapolation
across the gaps for the purposes of broad-brush classification. Now that there are
new sonar instruments available that are able to gather high-resolution sidescan
sonar data much faster than was previously possible, it may be worth collecting some
high-resolution data from a relatively large sample of the seabed within the Study
Area - large enough to cover several character areas - and then carrying out acoustic
classification trials to compare the difference with the classification results obtained
from the low-resolution data. However, the usefulness of this exercise would have to
be weighed against the resources required to do it.
The importance of using software specifically designed for acoustic seabed
classification cannot be stressed enough. The use of Swathview, and its
212
accompanying interpolation software, Clams, made the automated unsupervised
classification of the sidescan data possible. Without this software, the classification
would have had to be done manually, as sensible results could not be obtained from
any of the other software packages that were available at the time. The software was
very kindly made freely available by the suppliers, Quester Tangent Ltd, for a specific
trial period for research purposes. If the software had been purchased and therefore
been available for longer, it would have been ideal to have carried out more testing
using different parameters. However, even with the amount of testing that was done
within the time available, very successful results were obtained, and a coherent and
logical division of the seabed was achieved, which was subsequently validated using
additional datasets and analysis. There was an element of manual classification in
the methodology used, in that the initial landscape unit polygons were manually
digitised from the classified image generated by the automated classification.
However, the core generation of classes from which the polygons were manually
created was automated, unsupervised classification, which picked out acoustic
differences in the data that could not be detected manually. One of the outcomes of
this research is that successful sediment classification has been carried out using a
combination of automated techniques and manual polygonisation, with each method
informing the other. This may not be the most objective and repeatable method, but it
has certainly been the most practically useful in this context. A possible avenue for
further research could be to undertake trials of supervised classification,
incorporating what we know about sediments in the Study Area to create training
data that can be used to classify the entire dataset, and then comparing this to the
results of the unsupervised classification.
213
The bathymetric surface model has also proved to be incredibly valuable, as it has
provided the means to create a seamless surface model of the land/sea interface
from the very important shallow zone containing significant coastal structures right
out into the deeper parts of the sea in the west of the Study Area. It has therefore, for
the first time in this part of the Arabian Gulf, enabled the identification of significant
landscape features that could have had considerable influence on the location of
human settlement in the Late Pleistocene and Early Holocene. However, there are
significant issues still to be considered, a major one of which is how far we can rely
on the bathymetry to be representative of the land surface as it was in the Early
Holocene. We know from research into submerged landscapes elsewhere in the
world, for example in the North Sea (Fitch et al., 2005, p.185), that bathymetric
models of the present-day seabed are not necessarily a good reflection of what the
palaeo-landscape would have been like, as they will, to some extent, be an
expression of current sedimentary processes. It is possible, for example, that some
erosional features could have become infilled or buried by sediment, and some
landscape features that would have been significant for human settlement in the
Early Holocene could have been eroded away. (Fitch et al., 2005, p.194). This is a
major part of the reason why significant difficulties have been experienced by
researchers when applying historic landscape characterisation techniques in areas
where early features are obscured by later sediments (Gaffney et al., 2007, p.111).
However, this has been addressed to a limited extent within the characterisation by
incorporating information about sedimentation from the acoustic classification and the
ground-truthing, and using it as a criteria to assign values for potential. Whilst this
214
cannot fill the gap in knowledge that exists regarding the palaeo-bathymetry, it can at
least provide guidance as to the best areas to target for further research.
One analytical technique based on the bathymetry data that would be worth exploring
in the future is Benthic Terrain Modelling. This technique uses bathymetry data to
generate models of slope, aspect, rugosity and bathymetric position index (a
measure of where a referenced location is relative to the locations surrounding it),
which can be used for examining and classifying seabed characteristics. Tools have
been developed specifically for undertaking Benthic Terrain Modelling within ArcGIS
(NOAA, 2014), and a comparison of such analysis with the acoustic classification of
the sidescan data could potentially produce some very useful results.
As far as can be ascertained, this seabed characterisation is the only example of
marine historic landscape characterisation that has used acoustic sediment texture
classification, bathymetry and ground-truthing to determine character and potential
for archaeological purposes. The combined analysis of the topographic surface
model with the classified sediment data was pivotal to the success of the research.
The level of correlation between the acoustic classification and the topography from
the bathymetric surface model was surprisingly high, and, true to the iterative
principles of landscape characterisation techniques, each dataset has provided
valuable information with which to inform and enhance the other, thus increasing the
value of each dataset, and enabling the refining of the initial seabed zoning into
character areas.
215
The ground-truthing data, in the form of direct sampling and video transects, whilst
not without its limitations, has largely supported the character zones generated by
the combined sediment classification and topographic analysis. However, other than
the grab samples, this data was not collected specifically for this research and there
would be clear advantages to carrying out coring and optical surveys specifically for
archaeological or palaeoenvironmental purposes. This is particularly relevant for
finding evidence for buried organic deposits in core samples. There was no evidence
for organic remains in any of the vibrocores or boreholes, probably because they
were specifically designed to gather information relevant for construction purposes,
such as rock strength and compaction. Research in other parts of the world has
shown that direct observations and coring become increasingly useful once the scale
of the investigation becomes reduced (Faught and Donaghue, 1997). This seabed
characterisation could certainly be improved by more intensive ground-truthing, but it
would be very resource-intensive to collect grab samples in sufficient densities over
the entire Study Area. However, useful results could be obtained by selecting a few
smaller areas that include the boundaries of several character zones and collecting a
high density of grab samples in these areas. The results could be further improved by
extensive photography at each grab sample location.
In terms of optical ground-truthing, rather than narrow video transects or isolated
individual photographs, a more useful technique would be to create panoramic views
of the seabed, thus enabling a wider view of the character of the seabed in a
particular area. This could be done by taking large numbers of photographs in a
character zone and stitching them together to create photo-mosaics of larger areas. If
216
this could be done within the constraints of the visibility conditions that often occur in
the Study Area, then it could be of great benefit for this type of landscape-scale
investigation, as it may provide more diverse ground-truthing data, and could help to
identify some of the larger anomalies and/or landscape features that could not be
observed clearly enough during diver inspections. Stereoscopic photography has
been used very effectively (albeit in very clear conditions) at the submerged Bronze
Age town of Pavlopetri (Henderson et al., 2013). This technique would be too timeconsuming
to undertake over large areas, but as Autonomous Underwater Vehicles
(AUVs) develop and become cheaper, the creation of underwater photo-mosaics is
likely to become more cost-effective.
Previous research into classification methods for seabed mapping has suggested
that the use of continuous data, such as acoustic data, is likely to result in a more
diverse seabed map than would have been the case if direct sampling or
photography at discrete locations had been used and extrapolated ( Sutherland et al.,
2007). This certainly appears to be the case with the seabed classification carried out
in the Study Area, since, if direct sediment sampling alone had been used, the result
would have been three classes, sand, silty sand and coral, and it would not have
been possible to generate meaningful character areas on this basis alone. This
clearly shows that the value of the integrated analysis is greater than the sum of its
parts, and again validates the use of ideas drawn from historic landscape
characterisation techniques.
217
The secondary characterisation, in the form of systematic analysis of individual
geophysical anomalies, was the least successful aspect of the characterisation
methodology. However, some useful information about future methodologies to use
was obtained from the work that was carried out. The low-resolution sidescan sonar
data is clearly perfectly adequate for the initial identification of anomalies caused by
more recent debris, as many such anomalies were identified in the low-resolution
data and confirmed by diver inspection. The high resolution sidescan sonar data
collected using the Klein Hydroscan provided an almost photo-like image of such
anomalies, so this would be a useful technique to employ where there is a need to be
more certain of what an anomaly may represent before deciding on the allocation of
further, possibly expensive, investigative resources. One part of the Study Area
where this would be of use is the area of concentrated debris in the west of Area 1
(Figure 82). High resolution sidescan sonar data from this area would probably
confirm whether this debris is the result of dumping to create artificial reefs, or
whether it represents a spread of debris from a shipwreck.
However, the more ephemeral types of anomaly that were being targeted as part of
this landscape-scale analysis did not show up with sufficient clarity in the low
resolution sidescan sonar data to be informative, and there is currently no evidence
to suggest that high-resolution data would display such anomalies any more clearly.
This is evidenced by the fact that the long linear anomaly (QBC_Q10729) did not
show up any better in the high-resolution sidescan data than it did in the lowresolution
data. However, it should be noted that this is only a single example, and
further testing by gathering high-resolution sidescan sonar data over other anomalies
218
of a similar ephemeral nature would confirm this. The further high-resolution sidescan
sonar survey work undertaken by the QNHER project has shown that the direction of
survey is obviously very important for the identification of more ephemeral anomalies
(Cuttler, 2014). The results suggest that the most cost-effective and productive
methodology for initial rapid prospection of large areas would be a low frequency,
wide swath-width sidescan survey with overlapping survey lines surveyed in a
minimum of two directions, followed by higher frequency survey in targeted areas
based on the results provided by the low frequency survey.
The most useful aspect of the secondary characterisation was the information that it
provided about more recent maritime exploitation, mainly fishing activity, in the form
of the distribution of modern anthropogenic debris and trawler scarring. A further
notable result was how useful the topographic surface model was for clarifying the
more ephemeral anomalies, even if it was largely to confirm that they were not of any
significance. This enabled the clarification of many of the anomalies that were
selected for further study without having to carry out further high-resolution
geophysical survey or diver inspection. This is especially significant as diverinspection
is a time-consuming and expensive method for exploring the submerged
landscape. A clear outcome of this research, and the work carried out by the QNHER
project, is that diver-inspection should be used for very specific, smaller scale
inspection of clearly-defined targets, but it is not suitable for ground-truthing
landscape-scale geophysical anomalies. It is clear from the experience of the
QNHER diving team of diving on known shipwreck locations and not finding anything
that there are still problems in ground-truthing small-scale anomalies by diver
219
inspection, possibly due to currents, wind and GPS accuracy (Personal
Communication, Richard Cuttler, 2014). A more robust methodology may be to drop
a marker down on an identified anomaly, then carry out a sidescan survey over the
marker before diving, to ensure that the anomaly has been correctly located.
Now that we have areas of potential that can be targeted, there is a clear need to
collect new data sets in order to improve predictive modelling, and a suite of
techniques used in combination - multibeam sonar, sub-bottom profiling and
vibrocoring - is likely to produce the most effective results. Of critical importance is
the need for sub-bottom profiling in order to try and trace buried and preserved
features. Sub-bottom profiling has been successfully used to identify palaeo-features
in the North Sea (Gaffney et al., 2007), in the Gulf of Mexico, where it was also used
extensively, in combination with bathymetric enhancement methods, for identifying
palaeo-drainage patterns, including drowned Karst features such as sinkholes
(Faught and Donaghue, 1997), and to identify submarine freshwater springs in the
Baltic Sea (Schlüter et al., 2004; Alfred Wegener Institute, 2012). In the Study Area,
sub-bottom profiling could be used to confirm, or otherwise, the interpretation of
significant features identified from the topographic surface model, and also to try and
identify features that do not have a bathymetric expression, such as buried and
preserved palaeo-land surfaces, and buried palaeo-drainage features and their
sediment fills, including Karst-related depressions and solution hollows. This
technique should be prioritised in the areas along, and inland of, the putative former
shorelines, and in the areas where the postulated solution hollows are concentrated
in the northwest of the Study Area. However, the ability to undertake sub-bottom
220
profiling inland of the former shorelines may be limited by the shallow water depths in
those areas.
In addition to the sub-bottom profiling, targeted core sampling specifically for
palaeoenvironmental purposes needs to be undertaken to try and identify organic
material, and to provide further information on the sediment in the character areas.
As with the sub-bottom profiling, the coring programme should again be concentrated
along, and inland of the putative former shorelines. In addition, specific features
identified from the topographic surface model should be core-sampled. These include
the potential solution hollows in the northwest of the Study Area, the cluster of
palaeochannels near the possible former shoreline, the possible continuation of Wadi
Debayan, and the deep southwest-northeast channel in the north of the Study Area.
If the latter does relate to a large palaeochannel, it would probably be possible to use
sub-bottom profiling to map the alignment. This information could then be used to
target a vibrocore strategy aimed at the recovery of both former sub-aerial deposits
and at deposits associated with marine transgression along the channel.
More detailed bathymetry collected from the areas along and inland of the former
shorelines, including extensive multibeam survey in the area of the possible eastwest
aligned shoreline in the north, integrated with the sub-bottom profiling and the
coring from these areas, would provide a better understanding of the changes that
have occurred in the submerged landscape since the marine transgression. This
work would be improved if more accurate sea level data could be obtained, that takes
into account regional hydrostatic adjustments. There are probably enough dates from
221
terrestrial areas such as Wadi Debayan to begin developing a more secure sea-level
curve from around 6,500 BP onwards (Tetlow et al., Forthcoming). The challenge is
to obtain information from the submerged areas, and a targeted programme of core
sampling based on this research should provide a starting point for this. More
detailed bathymetry could also be used to calculate the ‘rugosity’ of the landscape.
Rugosity can be defined as a measure of terrain complexity or the ‘bumpiness’ of the
terrain. Rugosity is an important influence on the character of the seabed, and this
information could be incorporated into the character areas to provide additional
information for further targeted work.
It would also be worth exploring the potential for using 2D and 3D seismic data to try
and identify significant buried features from the Early Holocene that do not have any
bathymetric expression, as was successfully done in the North Sea (Gaffney et al.,
2007). Seismic imaging is a geophysics technique that measures the reflectivity of
seismic waves, and is used to create sub-surface geological and geomorphological
maps. This type of survey is very expensive to undertake, but it is being undertaken
extensively in the Arabian Gulf for oil exploration purposes. Future data mining of
these datasets may provide a very cost-effective method of interrogating the former
landscape and developing research frameworks for targeted, higher-resolution
survey. As the information required for archaeological and palaeoenvironmental
purposes is limited to relatively shallow depths, it is possible to slice off the top 0.25
sec of data, thus avoiding any commercially sensitive issues such as the location of
hydrocarbons.
222
Whilst the research has demonstrated that the combined analysis of the sidescan
sonar data and the bathymetry provided a robust foundation for the initial seabed
characterisation, a purpose-designed data capture programme for investigating the
submerged landscape would give high priority to volumetric data such as sub-bottom
profiling data and/or 3D seismic data. Analysis of such data in combination with highresolution
bathymetry could provide vital comparative information between the
palaeo-topography and the present-day seafloor topography in the Study Area.
However, as has already been stated, the power of the characterisation approach is
the iterative combination of datasets and techniques. Although bathymetry and
volumetric data would be primary datasets for a specifically-designed seabed
characterisation data-capture programme around Qatar, the results of this research
clearly demonstrate that acoustic texture classification of sidescan sonar would still
play an important role in characterising and refining landscape units.
There is great potential for much more work to be undertaken using further aspects
of terrestrial historic landscape characterisation methodologies, particularly the more
cultural elements. Understanding how people worked within the marine environment
in more recent centuries can be investigated via the use of both physical and nonphysical
evidence, in order to provide another layer of interpretational information.
Evidence of more recent use of the maritime environment could provide valuable
information about the nature of the seabed, preservation potential, tidal and sediment
regimes and the location of human activity in the present and the more recent past.
The types of information that could be used, in addition to geophysics and survey
data, include maritime charts, which can provide evidence of certain types of activity,
223
such as navigation information, trade routes, shipping routes, anchoring, fishing
exclusion zones, hazards, dumping grounds, place names and other cultural
information. These information sources can also indicate the absence of certain
types of activity, which can be just as significant.
It is also worth exploring further how the characterisation could incorporate local
knowledge from people who live and work on or near the sea, or have done in the
past. They will have traditions and knowledge drawn from observing the shorelines
and working on the sea on a daily basis over extended periods, in all sorts of different
conditions. This type of work has been carried out very effectively by the Outer
Hebrides Coastal Community Marine Archaeology Pilot Project (Wessex
Archaeology, 2012), which involved talking extensively to local fishermen in order to
find out about things that people had noticed in the past but had disregarded as
unimportant. Interviews with more recent seafarers may well produce anecdotes like
that related by Cheesman (1923) of boatmen replenishing drinking water from
submarine freshwater springs off Bahrain. This could be of help in locating present
day freshwater springs, although extensive pumping since the mid 1920’s suggests
that groundwater levels have dropped significantly, by as much as 7m, in recent
times, which will have affected the output from freshwater springs (Macumber, 2011,
p.11) .
It would also be worth exploring ways of encompassing some of the more subjective
human aspects into the characterisation, in order to consider the less pragmatic
aspects of why humans chose certain areas for settlement or other activity. These
224
principles have been applied successfully in terrestrial characterisation projects, for
example the Fort Hood historic landscape characterisation project in Texas (Dingwall
and Gaffney, 2007), where aspects such as proximity to significant natural features
known to be of importance to hunter-gatherer groups, such as mesa tops and
sinkholes, were factored into the classification. Whilst it would be very difficult to
include the types of visual attributes and perceptions that terrestrial landscape
characterisations involve, it is important to avoid relying too heavily on environmental
determinism, and losing the more holistic view of how humans perceived the
landscape in which they lived.
This type of landscape-scale contextualisation of the seabed can also provide
opportunities for combined management of the historic and natural environment. The
ecological character of the seabed can both influence and reflect past and present
human exploitation, visibility of features and preservation potential. It is likely that the
datasets derived from the seabed characterisation process could be useful for both
cultural and natural habitat mapping purposes, thus providing opportunities for interdisciplinary
working with natural environment researchers. There are clear
crossovers, for example biodiversity levels and species habitats can be closely
related to the rugosity of the seabed. Also, integrated analysis of natural habitat data
with other datasets could potentially provide valuable information for the location of
submarine freshwater springs, since different flora and fauna may occur in areas
where there is a higher proportion of fresh water
225
7.2 An evaluation of the archaeological and palaeoenvironmental potential of
the submerged landscape within the Study Area
The submerged Gulf and its former coastlines have the potential to provide valuable
new information on a range of important research themes. One of the most obvious
outcomes of the seabed characterisation is that it has emphasised the critical
importance of former shorelines, which have an influence on virtually every aspect of
the significance and potential of the Study Area.
This is particularly pertinent to any study of the Neolithic period in the region,
especially in relation to the relatively sudden appearance of ʿUbaid -related sites in
previously-unsettled areas around the shoreline of the Gulf in the Mid-Holocene
(Rose, 2010). This occurred just after the final phase of marine incursion into the Gulf
basin, when sea levels rose to approximately 2-3m higher than present-day levels
(Lambeck, 1996). So far, it has not been possible to firmly establish where these
communities originated from, but it is possible that they migrated into the area as a
result of displacement from the Gulf basin by relatively rapid inundation between c.
8,000BP and 7,500BP (Rose, 2010; Cuttler, 2013).
It is reasonable to propose that there would have been aspects of continuity between
the human communities that were established in the former Gulf basin prior to the
final phase of marine transgression, and those communities represented by the
ʿUbaid-related sites occurring along the eastern coast of the Arabian Peninsula
(Cuttler, 2013). It is possible to relate the information on known settlement patterns in
226
the Mid-Holocene to pre-transgression settlement patterns, and to establish what
factors we can draw upon in the submerged landscape that will help to inform this
theme.
227
Figure 100 Distribution map of major ʿUbaid-related sites in Qatar.
228
It is clear that the sites producing ʿUbaid pottery in the Arabian Peninsula are found
almost exclusively at locations situated on the coast or on the edge of coastal
sabkhas, and on islands (Drechsler, 2011). Archaeological evidence shows that
occupation of ʿUbaid ‐related sites along the Qatar coastline (Figure 100) begins
during the second half of the 8th millennium BP (Carter, 2010). Although we lack
detailed information on the topographic locations of ʿUbaid-related sites in Qatar, on
the basis of the evidence that we do have, they appear to occur on coastal cliff areas,
overlooking the mid-Holocene shoreline, and are generally above the 2m contour
(Cuttler and Al-Naimi, 2013). The dating and topographic information for ‘Ubaidrelated
sites in Qatar is highly significant in relation to the results obtained from the
shoreline reconstructions based on Jameson and Strohmenger's sea level curve
(Jameson and Strohmenger, 2012). The two putative former shorelines that were
identified running northwards of the Ras ‘Ushayriq peninsula coincide with the
predicted extent of dry land at 8,000 BP and 7,500 BP respectively. If people were
migrating to the higher ground inland of the present-day coastline after 7,500 BP,
they may well have migrated from settlements that were on these former shorelines
in 8,000 and 7,500 BP.
Ras Aburuk (Raʾs Abarāq) lies on a low plateau to the west of, and overlooking, an
oasis depression. It is on a narrow peninsula, which is 1km from the coast on the
west (De Cardi, 1978, p.82). Bir Zeikrit (Biʾr Zikrīt) lies on a limestone plateau on the
western edge of an oasis depression and within a kilometre of the present coastline.
De Cardi (1978, p.115) noted that the site at Bir Zeikrit, unlike modern nomadic
encampments in the area, is not located directly adjacent to the water hole. Kapel’s
229
Site A4, 7km south of Dukhan, lies on a sandy level about 100 yards from the coast
(Kapel, 1967, p.37). Al-Dasah (al-Daʿsah) lies on the seaward headland of a low
terrace, probably an ancient beach, overlooking sabkha (De Cardi, 1978, p.55). De
Cardi suggested that the site location was mainly influenced by the availability of
fresh water, but also noted that the site overlooks a protected sandy bay (1978,
p.72). Al-Khor (al-Khawr) lies on top of fossil cliffs overlooking the sabkha to the north
(Tixier, 1980). Wadi Debayan (Wādī al-Ḍabaʿyān) lies on the leeward side of a
peninsula, above the 3m contour. The wadi mouth may once have been navigable,
providing direct access to the sea. This site is of particular relevance to the seabed
characterisation since the wadi discharges into the Bay of Al-Zubārah, in close
proximity to the Study Area (Cuttler et al., 2011a, Tetlow et al., Forthcoming). A
possible extension to the present-day wadi has been tentatively identified in the
topographic surface model to the east of the Ras ‘Ushayriq peninsula, again
emphasising the significance and potential of this area of the submerged landscape.
Outside Qatar, the topographic locations of important ʿUbaid-related sites on the Gulf
coast generally follow the pattern observed on the sites in Qatar, although there are
exceptions. The site at al-Markh, currently on the west coast of Bahrain, 1km from
the present-day coastline and 2m above the current sea level, was an island during
the ʿUbaid-related occupation period. There are now extensive areas of sabkha to
the north and east (Roaf, 1976, p.146). Dosariyah (Dawsāriyyah), in Saudi Arabia, is
located 1.5 km from the present-day coast, within an extensive valley-like trough, and
it opens towards the south to the sabkhat al‐Ṣumm (Drechsler, 2011, p.71).
According to Masry (1997) the original ground of the site before occupation would
230
have been an east-facing, gently-sloping hill. As-Sabiyah H3 in Kuwait lies on the
inland edge of a peninsula close to the coast, the bay between the peninsula and the
mainland now infilled with sabkha. The geomorphological evidence indicates that the
sea could be directly accessed from the site during the ʿUbaid-related occupation
period (Carter and Crawford, 2010). The recently-excavated site at Bahra 1 in Kuwait
has a less clear relationship with coastal features. It is located in a former active
dune field and is 8km from the present coastline, although the distance to the coast
would have been much shorter during the Mid-Holocene sea-level high (Rutkowski,
2011).
This brief overview of topographic characteristics of a few selected ʿUbaid-related
sites is far from comprehensive, and needs to be supplemented with detailed,
localised site topography in order to establish a comprehensive picture. However,
this emphasises how important it is that potential coastal features such as former
bays, spits, promontories and drainage channels were identified as part of the
characterisation, and the critical importance, therefore, of targeting further
investigations in the area of these putative former shorelines.
The close interaction of environmental and cultural factors in seabed characterisation
is well demonstrated by the issue of the location of submerged Karst landscapes and
their relationship to archaeological potential. The QNHER national cultural mapping
program has demonstrated that the distribution of Karst depressions was a significant
factor in the location of settlement until very recent times (Breeze et al., 2011). The
depressions and holes caused by Karst are an important feature of the present day
231
terrestrial northern Qatari landscape, tending to occur as sinkholes, simple
depressions or compound depressions, some of which can be very large – up to 3km
across and more than 25m deep (Sadiq and Nasir, 2002). The Misfer cave (Figure
101), an open cave 40km to the west of Doha, is up to 100m in depth. It is almost
certain that the submerged areas between Qatar and Bahrain will have contained
Karst features, since drowned Karst is typical along all carbonate coasts except
those where there has been rapid tectonic uplift (Ford and Williams, 2007, p.430).
Research into submerged Karst landscapes in the Gulf of Mexico has indicated that
as well as isolated depressions, many sinkholes occur in linear clusters, forming
discontinuous river channel segments (Faught and Donoghue, 1997). Most Karst
features in Qatar show northeast-southwest and northwest-southeast orientations,
governed by the joints and fractures formed during periods of uplift (Sadiq and Nasir,
2002).
Figure 101 Misfer Cave, Qatar.
232
The features typical of Karst landscapes, such as springs, sinkholes, depressions
and caverns, would have provided very attractive habitats for Prehistoric
communities. These features are likely to have had significance for both pragmatic
reasons, including access to water and to the more fertile soil (rawdha) that
accumulated in the depressions caused by the collapsed Karst features, and for
symbolic reasons such as places to bury the dead, or make ritual deposits. On land,
these same features tend to preserve and protect evidence for human activity,
making them highly significant in archaeological and palaeoenvironmental terms. By
analogy, investigating drowned and sediment filled sinkholes on the seabed should
substantially increase our chances of finding in situ archaeological and
palaeoenvironmental deposits (Faught and Donoghue, 1997).
It was considered possible that larger Karst-related features may still be visible on
the seabed in the Study Area, possibly appearing as depressions or holes,
depending on the sedimentation regime in particular areas. The identification of
potential drowned Karst features (the possible solution hollows) in the surface model
created from the bathymetry during this research is therefore a particularly exciting
development, since finding features of high archaeological and palaeoenvironmental
potential such as these was one of the primary aims of the research, and up until
now, no such features had been identified in this area of the Arabian Gulf. Whilst
further work is required to establish a firm identification of these hollows as a
submarine extension of the terrestrial Karst landscape of Northern Qatar, this is
certainly a very encouraging baseline from which to start working.
233
The location of freshwater springs in the submerged landscape is also closely related
to cultural factors and archaeological potential. The well-documented occurrence of
submarine springs in Karst environments may be associated with springs that
developed during times of lower sea levels (Ford and Williams, 2007, p.141). The
area around Qatar is the terminus of several aquifers that have a wide catchment
area across the Arabian Peninsula, emerging as freshwater springs in the Gulf.
Previous research suggests that there is a direct relationship between sea levels and
coastal freshwater springs, whereby falling sea levels caused an increase of
hydraulic pressure within underground aquifers, causing more fresh water to flow
through them and leading to the creation of springs on the exposed coastal shelf, and
hence an increase in freshwater at the coast - the 'Coastal Oasis' theory (Faure et al.,
2002; Parker and Rose, 2008). These springs can remain active once sea levels rise
if there is enough hydraulic pressure from the inland aquifers.
There is plenty of evidence for active submarine freshwater springs in the
Mediterranean Sea and the Red Sea (Faure), and in the Arabian Gulf off Bahrain.
Church (1996, p.579) reports anecdotal evidence of ancient seafarers restocking
their freshwater sources from upwelling plumes that occurred offshore in parts of the
world that have limestone terrains, including the Arabian Gulf. Cheesman (1923)
visited Bahrain in the 1920s and noted that boats were able to replenish their drinking
water from a submarine freshwater spring. As late as the 1950s, a submarine spring
located a few miles north of Jubail in Saudi Arabia, from which divers would collect
freshwater in skins, was marked by a spar buoy (Bowen, 1951, p.173-174).
Submarine freshwater springs are also known from areas of drowned Karst in other
234
parts of the world, including along the Dinaric Coast of the Adriatic Sea and in New
Zealand (Ford and Williams, 2007, p.434).
Research in the Baltic sea has utilised sub-bottom profiling and sediment analysis to
understand the formation process of features known as ‘Pockmarks’, or elongated
depressions approximately 1 to 3 m below adjacent seafloor levels, which could
potentially be associated with the location of submarine groundwater discharge from
sub-seafloor aquifers (Schlüter et al., 2004; Alfred Wegener Institute, 2012). The
location of submarine freshwater springs, whether inactive or current, is a highly
significant factor when assessing areas of archaeological potential in the submerged
landscape off Qatar, since a relatively abundant freshwater supply would have been
critical for human settlement during the Early Holocene. This makes the exploration
of former shorelines even more significant, given that the evidence suggests that
freshwater supplies would have been concentrated at the coast during sea level lowstands.
A combination of sub-bottom profiling, and information about water salinity
targeted around the putative former shorelines provides a promising methodology for
the location of seabed springs off Qatar. A CTD (conductivity, temperature, depth)
scanner could provide useful information on water salinity close to the seabed, which
may help to identify freshwater discharge from submarine aquifers, and possibly the
locations of former sub-aerial springs. However, this would be a very time-consuming
method to undertake over large areas, and it may be more useful and cost-effective
to examine satellite imagery for evidence of sea surface temperature hotspots
(Personal Communication, Richard Bates, 2015).
235
The levels of potential assigned to most of the character areas relate more to
palaeoenvironmental potential than to the potential for finding archaeological sites.
This is because the evidence suggests that the sedimentation is so thick in parts,
particularly in the deep channel in the north of the Study Area, that there would be no
prospect of being able to undertake archaeological excavations in those areas. Also,
the borehole data from the central part of the Study Area demonstrated that there are
layers of caprock within the marine sediments in this area. Submarine lithification is
common around shallow sub-tidal rock around the west coast of Qatar (Marin
Mätteknik AB, 2002, p.42), and although archaeological remains could lie preserved
underneath this caprock, they could be difficult to locate or investigate further.
Although we don't currently have any evidence for the presence of terrestrial
sediments in the submerged landscape, it is clear that more needs to be understood
about the sedimentary processes that occurred during marine transgression, in order
to ascertain whether these processes would have been likely to have preserved
rather than eroded the palaeo-landscape (Westley, 2011a). Recent terrestrial coring
work and excavations at Wadi Debayan, lying inland on a palaeo-shoreline to the
southeast of the Ras ‘Ushayriq Peninsula, revealed evidence of marine deposits that
were laid down after 4,500 BP, when sea levels were higher than present day levels,
and the site was submerged. These marine deposits were overlying earlier
archaeological and palaeoenvironmental deposits, and the nature of the marine
deposits, which included a layer of large beach cobbles, was indicative of a highenergy
marine event (Figure 102).
236
Figure 102 Evidence for archaeological deposits preserved beneath marine deposits at Wadi Debayan
(Image by Richard Cuttler/Emma Tetlow).
This site clearly demonstrates that there are some environments and conditions
where rapid inundation has actively preserved archaeological and
palaeoenvironmental deposits under marine sediments (Tetlow et al., Forthcoming).
The characterisation has demonstrated that there is significant potential within the
currently-submerged landscape for targeting analogous locations, with similar
landscape settings that would have been attractive for human settlement, and similar
formerly-sheltered situations where protective deposition could have occurred. The
general paucity of palaeoenvironmental remains from terrestrial sites in Qatar means
that the palaeoenvironmental potential of the submerged landscape is of huge
significance for filling the gap in knowledge.
237
The secondary characterisation has enabled issues of cause and effect to be
explored, for example whether certain geophysical anomalies, or the lack of them,
are a reflection of the usage of the maritime environment in a particular zone, or
whether they are simply more, or less, visible because of the seabed type in that
zone, which in turn provides significant information about archaeological potential.
The sidescan sonar data shows a very pronounced concentration of geophysical
anomalies of likely recent anthropogenic origin, probably cars, tyres etc, in the north
of the Study Area, and the reasons for this could shed useful light on the visibility and
preservation of certain types of material, and on recent human exploitation of the
marine environment.
Parallels can be drawn elsewhere in the world, for example with the Red Bird artificial
reef off the coast of Delaware in the USA. This reef is made up of over 900 New York
City subway cars, and various other vehicles, decommissioned boats and other
objects, covering 1.3 square nautical miles, and creating habitat for fish and other
marine creatures (State of Delaware, 2012). This type of exploitation of the seabed
has clearly also occurred in the sea off Qatar, although not officially, and not on the
organised and extensive scale of the Red Bird Reef. The impact of these structures
on the seabed includes changes to the sediment regime around the structures, and,
in the case of the Red Bird Reef, increased abundance of marine flora and fauna.
Conversely, an area of extensive trawler-scarring is visible in the mosaic sidescan
sonar data in the south of the Study Area, and very few geophysical anomalies are
visible in this area. This also coincides with one of the deepest areas of seabed in the
238
Study Area. It is possible that the depth and character of the seabed in this area has
influenced its use as trawler-fishing grounds. However, resolution No. 29 of 1994 of
the Council of Ministers in Qatar prohibits fishing through trawling in Qatar waters, so
either these scars pre-date 1994, or there has been illicit fishing occurring in this
area. These different types of relatively recent marine exploitation have a significant
effect on the seabed character, and possibly also on the archaeological potential of
the locality.
The seabed characterisation has not really clarified the potential of the Study Area for
locating shipwrecks, either historic or more recent. None of the geophysical
anomalies identified and selected for further study could be demonstrated to relate to
shipwreck debris other than the possibility of the concentration of debris in the west
of Area 1 (Figure 82). However, expanding the seabed characterisation to
encompass more cultural aspects, as discussed above (section 7.1), may illuminate
this aspect of the character of the Study Area.
239
CHAPTER 8: CONCLUSION: IMPLICATIONS FOR RESEARCH AND
MANAGEMENT OF THE MARINE HERITAGE RESOURCE
Given how difficult and expensive it is to investigate submerged landscapes, it is
valid to consider whether we should be doing it at all. However, given the significance
and potential of the submerged landscape of the former Arabian Gulf basin, and of
the Study Area within it, as emphasised in the discussion above, the question really
should be how we should be doing it, not whether we should be doing it. Previous
research has already highlighted the environmental and archaeological importance of
the submerged landscape around Qatar, as an area that remained free from marine
influence until relatively late, that had the potential for freshwater supplies in the form
of coastal springs, and that lay in close proximity to known archaeological sites
representing new settlements that emerged on the Mid-Holocene coast just after the
final inundation. However, this landscape was considered to be unknown and
inaccessible territory, and therefore has been largely unconsidered in the
archaeological syntheses of the Early Holocene of the region until very recently. This
seabed characterisation, together with the research carried out to the east of Qatar
using seismic data (Cuttler, 2014), has demonstrated that the situation does not have
to remain like this.
The characterisation was undertaken as a response to the need to address the major
gap in knowledge about this highly significant landscape, and to address the fact that
previous investigation methods clearly weren't working. One of the most significant
results of the research is the progress that has been made towards utilising existing
240
datasets to their maximum potential and developing robust methodologies for the
investigation of submerged landscapes in this region. It has successfully drawn on
applications of historic landscape characterisation techniques and marine remote
sensing techniques that have been used for studying submerged landscapes
elsewhere in the world, as well as trialling new combinations of methodologies and
datasets. Extensive experimentation and testing was required in order to establish
which methods worked and which did not, as this particular combination of
techniques and datasets has not been applied to submerged landscapes for the
purposes of archaeological research before.
A major outcome from the research is the identification of broad areas of potential
containing possible significant landscape features and extensive sedimentary
deposits, indicating that there is potential for finding archaeological and
palaeoenvironmental remains within the submerged landscape. Another major
outcome is the provision of a framework within which to begin more detailed
investigations, by highlighting areas of potential to target, and appropriate techniques
to use for more detailed, finer-grained investigations. However, it is important to
recognise that this research represents a very early stage in the investigation of the
submerged landscape, and it is not expected that archaeological sites will be
discovered immediately as a result of this research. However, we do know that the
possibility exists for sites to be preserved, given the right conditions, and this
research has started the process of maximising our chances of finding them. As has
been discussed previously, there are limitations to the datasets that have been used,
and there is considerable scope for further development of methodologies and
241
techniques, particularly as technology improves and more data becomes available.
The seabed characterisation, like all landscape characterisations, is not a definitive
statement of archaeological knowledge or potential of the Study Area. Rather, it
should be treated as a way of representing and mapping broad patterns in the
landscape based on the best evidence currently available, thus providing a structure
to inform and stimulate further research. The characterisation model could, for
example, be refined and improved upon with more datasets and further analysis, and
used for exploring novel methods of archaeological predictive modelling, such as that
undertaken in the North Sea by Fitch (2011). Better knowledge of the topography of
the submerged landscape in the Study Area as a result of the seabed
characterisation could also be utilised in wider regional studies, such as studies into
patterns of human migration (Cuttler et al., 2012).
A sound research basis can facilitate informed decision making relating to the
management of the marine historic environment, which can be applied beyond the
Study Area. Also, as previously mentioned, landscape characterisation approaches
lend themselves very well to combined applications for both the historic and the
natural environment. The seabed character areas, and in particular the information
about the seafloor topography, could be very useful for studies of the benthic habitat
in the Study Area, which could feed into practical conservation strategies for the
natural environment. This is very important in the submerged landscape off the coast
of Qatar, as there are a range of pressures within this environment, including
construction development such as the proposed Qatar-Bahrain Causeway, dredging
for new ports, the creation of new land for tourism developments, and oil and gas
242
exploration. In this respect, the aims and achievements of the seabed
characterisation fit within the wider context of the pioneering marine research work
being carried out on behalf of the Qatar Museums Authority by the University of
Birmingham’s QNHER project.
Further research could lead ultimately to a point where it is realistic to search for
submerged sites and palaeoenvironmental remains with a reasonable expectation of
success. This seabed characterisation represents good progress towards this point,
and it is hoped that it proves to be only the beginning of research into a more
comprehensive understanding of the submerged landscape of the Arabian Gulf.
243
APPENDIX 1: PROCESS AND PARAMETERS USED FOR THE ACOUSTIC
CLASSIFICATION OF SIDESCAN SONAR DATA
Reference system parameters for original survey data (from GEMS Qatar LLC
2008)Geodetic parameters
Working Spheroid Working Projection
Spheroid WGS ‘84 Grid Projection QBC2001
Datum WGS ‘84 Projection Type Transverse Mercator
Semi-major Axis (a) 6 378 137.000 Central Meridian 050°49’ 00.000” E
Semi-minor Axis (b) 6 356 752.314 Latitude of Origin 25°00’ 00.000” N
Inv. Flattening (1/f) 298.257 223 56 False Easting 400000.00
First Eccentricity (e2) 0.0066943800 False Northing 500000.00
Scale Factor on CM 0.999996
Datum Transformation Parameters from WGS84 To working spheroid
WGS84 Spheroid WGS84 to Working Spheroid
Spheroid WGS84 X shift (dX): N/A
Datum WGS84 Y shift (dY): N/A
Semi-major Axis (a) 6 378 137.000 Z shift (dZ): N/A
Semi-minor Axis (b) 6 356 752.314 X rotation (Rx): N/A
Inv. Flattening (1/f) 298.257 223 56 Y rotation (Ry): N/A
First Eccentricity (e2) 0.006 694 38 Z rotation (Rz): N/A
Second Eccentricity (e’2) 0.006 739 50 Scale correction (ppm): N/A
Coordinate system used for data analysis
WGS84 UTM39N
Software used: Swathview and CLAMS (Quester Tangent Corporation)
TRIAL CLASSIFICATION
Swathview (classification software)
Batch Processing
 CLEAN - Sidescan Masking
Pick Bottom – None, left as original
Mask Water column – offset 5m
Angle - Max 90 degrees
Range - Relative 80%
Preserve Border edits – checked
 GENERATE RECTANGLES
Vessel speed - 4.4 knots
Ping rate - 5
Width * height (m)
Width (samples) - 129
244
Height (pings) - 17
Sample rate - 9600
 GENERATE FEATURES
Process file id 1
Features
Merge FFV files
Base FFV channel 1
Merge channel 1
Create Catalogue
FFV file id 1
Auto Cluster
Select merged .dat.utm file
Number of classes from 5 to 20
5 iterations
15,000 records per iteration
Saved the relevant number of classes as .dat.utm
Clams (Interpolation software)
Interpolation
Open the relevant .dat.utm file
Check the coordinate system
Node spacing = 5 for X and Y (makes it less pixellated the smaller the value)
Search radius = 50 (joins up gaps if bigger)
Search size = 40 (bigger produces less polygons)
Create
Save as: surfer
Xyz
Geotiff
ARGIS
Open geotiff in ArcGIS (open the _classes.tiff)
NB transformation problem – image is flipped and shifted on NW point
Resolved by doing a flip to get it on the right orientation and then a shift on the y
axis i.e. deducted values from the y coordinates (used data properties - min and max
extents - of geotiff and of sonarwiz coverage files to establish what the shift needed to
be. Shifted by -33573.69324 in y axis)
Go to layer properties, symbology, unique values, build attribute table, alter
background colour to clear, use field ‘value’
FINAL CLASSIFICATION
Swathview (Classification software)
Batch Processing
 CLEAN - Sidescan Masking
Pick Bottom – None, left as original
Mask Water column – offset 5m
245
Angle – 2 degrees min Max 70 degrees
Range - Relative 60%
Preserve Border edits – checked
 GENERATE RECTANGLES
Vessel speed - 0 knots
Ping rate - 0
Width * height (m)
Width (samples) - 65
Height (pings) - 33
Sample rate - 9600
 GENERATE FEATURES
Process file id 1
Features
Merge FFV files
Base FFV channel 1
Merge channel 1
Create Catalogue
FFV file id 1
Auto Cluster
Select merged .dat.utm file
Number of classes from 5 to 20
5 iterations
15,000 records per iteration
Saved the relevant number of classes as .dat.utm
Clams (interpolation software)
Interpolation
Open the relevant .dat.utm file
Check the coordinate system
Node spacing = 5 for X and Y (makes it less pixellated the smaller the value)
Search radius = 100 (joins up gaps if bigger)
Search size = 100 (bigger produces less polygons)
Create
Save as: Geotiff
Files used for test areas for both trials
Area3
20080724030952.xtf
20080724042616.xtf
20080724055140.xtf
20080724072234.xtf
20080724084848.xtf
20080724101218.xtf
20080724113257.xtf
20080724124624.xtf
246
Area4
20080726100849.xtf
20080726043129.xtf
20080726051534.xtf
20080726055735.xtf
20080726071558.xtf
20080726073821.xtf
20080726082638.xtf
20080726090147.xtf
20080726092620.xtf
20080726105338.xtf
20080726113249.xtf
20080726121658.xtf
20080726130112.xtf
20080726063909.xtf
Data Issues
Files excluded from classification due to problems in loading into Swathview:
20080709064910.xtf
Files excluded from classification due to problems in Swathview processing:
20080703042933.xtf
20080703121350.xtf
20080709093527.xtf
20080709194757.xtf
247
APPENDIX 2: PROCESS AND PARAMETERS USED FOR PROCESSING THE
LIDAR BATHYMETRY DATA
Datums used for the LiDAR Survey (information provided with the data supply)
Horizontal: WGS84
Vertical: QCD which lies 0.88m below QVD (MSL at Doha Port)
Coordinate system used for data analysis
WGS84 UTM39N
Software used: ArcGIS (ESRI Inc.)
Step 1: convert ASCII to 3D feature.
3D analyst/Conversion/From file
Output to point
Use -1 to reverse the z values (this is because in xyz data, drying heights are
negative and depths below sea level are positive, so both had to be reversed)
Projection set as UTM39N
Step 2: append z info
3D analyst/3D features/Add Z information. Check output property Z
Step 3: Append Point files together
Data management tools/general/append
Step 4: Interpolate surface
3D analyst/Raster interpolation/Natural neighbour
Select z value field and cell size (Both 7m and 2m resolution)
Visualise using Stretch and Invert values
Step 5: Hillshade
3D analyst/Raster surface/Hillshade
Leave defaults unchanged (model shadows unchecked, azimuth 315 altitude 45)
apart from set z factor to required exaggeration value
248
Points Used to Shift Bathymetry Data (based on coast shapefile )
Area Bathy X Coast X Shift X Bathy Y Coast Y Shift Y
North 499228.175 499307.51 79.335 2873894.052 2873813.575 80.477
North 499788.509 499867.091 78.582 2873443.491 2873364.116 79.375
North 508438.557 508523.245 84.688 2883591.301 2883499.971 91.33
North 503941.48 504021.084 79.604 2880717.915 2880638.314 79.601
Average
Shift
North
Area +80.55 -82.69
Middle 496773.595 496837.095 63.5 2861453.369 2861381.931 71.438
Middle 495551.764 495635.511 83.747 2857387.52 2857296.296 91.224
Middle 479798.61 479879.044 80.434 2823655.11 2823581.026 74.084
Average
Shift
Middle
Area +75.89 -78.92
South 476984.466 477062.041 77.575 2822519.302 2822441.727 77.575
South 475589.916 475673.102 83.186 2810301.592 2810223.525 78.067
South 475824.075 475907.737 83.662 2803880.432 2803805.67 74.762
South 478892.997 478981.107 88.11 2774132.514 2774048.409 84.105
Average
Shift
South
Area +83.06 -78.62
249
APPENDIX 3: GRAB SAMPLE GRAIN SIZE ANALYSIS REPORT
By Ismail Mahmoud Al-Shaikh, Abdul Rahman Al-Obaidly and Abdel Rahman
Sorour, Environmental Studies Centre, Qatar University.
250
Particle Size Range and Sediment Type
Particle Size Range (µm) Sediment Type
0.02 – 4.0 Clay
4.0 – 63.0 Silt
63.0 – 125.0 Very fine sand
125.0 – 250.0 Fine sand
250.0 – 500.0 Medium sand
500.0 – 1000.0 Coarse sand
1000.0 – 2000.0 Very coarse sand
Sample No. 1
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
Volume(%)
KSB 1, Wednesday, May 01, 2013 9:37:11 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 1, Wednesday, May 01, 2013 9:37:11 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
3.10
Size (µm)
63.000
125.000
250.000
Volume In %
2.95
16.56
Size (µm)
250.000
500.000
1000.000
Volume In %
31.04
31.54
Size (µm)
1000.000
2000.000
Volume In %
14.81
Size (µm) V
251
Sample No. 2
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 2, Wednesday, May 01, 2013 9:40:48 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 2, Wednesday, May 01, 2013 9:40:48 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
1.42
Size (µm)
63.000
125.000
250.000
Volume In %
0.36
8.01
Size (µm)
250.000
500.000
1000.000
Volume In %
33.19
39.73
Size (µm)
1000.000
2000.000
Volume In %
17.29
Size (µm) V
252
Sample No. 3
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 3, Wednesday, May 01, 2013 9:45:43 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 3, Wednesday, May 01, 2013 9:45:43 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.07
3.43
Size (µm)
63.000
125.000
250.000
Volume In %
4.65
30.90
Size (µm)
250.000
500.000
1000.000
Volume In %
35.80
17.96
Size (µm)
1000.000
2000.000
Volume In %
7.20
Size (µm) V
253
Sample No. 4
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
Volume(%)
KSB 4, Wednesday, May 01, 2013 9:47:59 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 4, Wednesday, May 01, 2013 9:47:59 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
3.45
Size (µm)
63.000
125.000
250.000
Volume In %
1.03
12.28
Size (µm)
250.000
500.000
1000.000
Volume In %
27.93
34.33
Size (µm)
1000.000
2000.000
Volume In %
20.98
Size (µm) V
254
Sample No. 5
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 5, Wednesday, May 01, 2013 9:50:32 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 5, Wednesday, May 01, 2013 9:50:32 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
1.32
Size (µm)
63.000
125.000
250.000
Volume In %
0.45
15.05
Size (µm)
250.000
500.000
1000.000
Volume In %
42.53
32.98
Size (µm)
1000.000
2000.000
Volume In %
7.67
Size (µm)
255
Sample No. 6
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 6, Wednesday, May 01, 2013 9:53:33 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 6, Wednesday, May 01, 2013 9:53:33 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
1.85
Size (µm)
63.000
125.000
250.000
Volume In %
1.57
8.71
Size (µm)
250.000
500.000
1000.000
Volume In %
28.60
40.38
Size (µm)
1000.000
2000.000
Volume In %
18.89
Size (µm)
256
Sample No. 7
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 7, Wednesday, May 01, 2013 9:58:26 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 7, Wednesday, May 01, 2013 9:58:26 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
0.31
Size (µm)
63.000
125.000
250.000
Volume In %
0.40
6.51
Size (µm)
250.000
500.000
1000.000
Volume In %
33.20
42.87
Size (µm)
1000.000
2000.000
Volume In %
16.70
Size (µm) V
257
Sample No. 8
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
9
Volume(%)
KSB 8, Wednesday, May 01, 2013 10:01:02 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 8, Wednesday, May 01, 2013 10:01:02 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.06
3.80
Size (µm)
63.000
125.000
250.000
Volume In %
3.19
12.82
Size (µm)
250.000
500.000
1000.000
Volume In %
29.82
34.63
Size (µm)
1000.000
2000.000
Volume In %
15.67
Size (µm)
258
Sample No. 9
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 9, Wednesday, May 01, 2013 10:03:03 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 9, Wednesday, May 01, 2013 10:03:03 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
1.81
Size (µm)
63.000
125.000
250.000
Volume In %
0.76
9.27
Size (µm)
250.000
500.000
1000.000
Volume In %
32.50
38.99
Size (µm)
1000.000
2000.000
Volume In %
16.68
Size (µm)
259
Sample No. 10
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
9
Volume(%)
KSB 10, Wednesday, May 01, 2013 10:05:23 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 10, Wednesday, May 01, 2013 10:05:23 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.07
4.71
Size (µm)
63.000
125.000
250.000
Volume In %
2.64
10.86
Size (µm)
250.000
500.000
1000.000
Volume In %
26.77
36.00
Size (µm)
1000.000
2000.000
Volume In %
18.94
Size (µm)
260
Sample No. 11
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
9
Volume(%)
KSB 11, Wednesday, May 01, 2013 10:08:49 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 11, Wednesday, May 01, 2013 10:08:49 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
1.82
Size (µm)
63.000
125.000
250.000
Volume In %
0.98
10.19
Size (µm)
250.000
500.000
1000.000
Volume In %
31.20
37.94
Size (µm)
1000.000
2000.000
Volume In %
17.86
Size (µm)
261
Sample No. 12
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
Volume(%)
KSB 12, Wednesday, May 01, 2013 10:11:12 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 12, Wednesday, May 01, 2013 10:11:12 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
1.32
Size (µm)
63.000
125.000
250.000
Volume In %
2.57
23.00
Size (µm)
250.000
500.000
1000.000
Volume In %
35.96
27.01
Size (µm)
1000.000
2000.000
Volume In %
10.13
Size (µm) V
262
Sample No. 13
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 13, Wednesday, May 01, 2013 10:14:03 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 13, Wednesday, May 01, 2013 10:14:03 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
2.55
Size (µm)
63.000
125.000
250.000
Volume In %
1.35
8.25
Size (µm)
250.000
500.000
1000.000
Volume In %
23.49
39.80
Size (µm)
1000.000
2000.000
Volume In %
24.57
Size (µm)
263
Sample No. 14
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
9
Volume(%)
KSB 14, Wednesday, May 01, 2013 10:16:54 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 14, Wednesday, May 01, 2013 10:16:54 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.06
6.23
Size (µm)
63.000
125.000
250.000
Volume In %
2.44
10.92
Size (µm)
250.000
500.000
1000.000
Volume In %
31.18
35.74
Size (µm)
1000.000
2000.000
Volume In %
13.43
Size (µm)
264
Sample No. 15
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 15, Wednesday, May 01, 2013 10:19:42 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 15, Wednesday, May 01, 2013 10:19:42 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
4.09
Size (µm)
63.000
125.000
250.000
Volume In %
1.49
10.07
Size (µm)
250.000
500.000
1000.000
Volume In %
34.47
38.01
Size (µm)
1000.000
2000.000
Volume In %
11.87
Size (µm)
265
Sample No. 16
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
Volume(%)
KSB 16, Wednesday, May 01, 2013 10:22:04 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 16, Wednesday, May 01, 2013 10:22:04 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.29
9.97
Size (µm)
63.000
125.000
250.000
Volume In %
4.06
16.22
Size (µm)
250.000
500.000
1000.000
Volume In %
31.33
27.58
Size (µm)
1000.000
2000.000
Volume In %
10.54
Size (µm)
266
Sample No. 17
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
Volume(%)
KSB 17, Wednesday, May 01, 2013 10:23:50 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 17, Wednesday, May 01, 2013 10:23:50 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.16
5.23
Size (µm)
63.000
125.000
250.000
Volume In %
3.79
14.44
Size (µm)
250.000
500.000
1000.000
Volume In %
26.70
31.86
Size (µm)
1000.000
2000.000
Volume In %
17.83
Size (µm) V
267
Sample No. 18
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
Volume(%)
KSB 18, Wednesday, May 01, 2013 10:25:45 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 18, Wednesday, May 01, 2013 10:25:45 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.16
6.17
Size (µm)
63.000
125.000
250.000
Volume In %
4.44
17.18
Size (µm)
250.000
500.000
1000.000
Volume In %
28.97
29.79
Size (µm)
1000.000
2000.000
Volume In %
13.27
Size (µm)
268
Sample No. 19
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
Volume(%)
KSB 19, Wednesday, May 01, 2013 10:28:56 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 19, Wednesday, May 01, 2013 10:28:56 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.45
12.17
Size (µm)
63.000
125.000
250.000
Volume In %
7.65
21.25
Size (µm)
250.000
500.000
1000.000
Volume In %
27.44
23.41
Size (µm)
1000.000
2000.000
Volume In %
7.62
Size (µm)
269
Sample No. 20
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
Volume(%)
KSB 20, Wednesday, May 01, 2013 10:31:13 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 20, Wednesday, May 01, 2013 10:31:13 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.79
16.76
Size (µm)
63.000
125.000
250.000
Volume In %
5.92
14.92
Size (µm)
250.000
500.000
1000.000
Volume In %
22.33
25.19
Size (µm)
1000.000
2000.000
Volume In %
14.09
Size (µm) V
270
Sample No. 21
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
Volume(%)
KSB 21, Wednesday, May 01, 2013 10:33:18 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 21, Wednesday, May 01, 2013 10:33:18 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.42
9.60
Size (µm)
63.000
125.000
250.000
Volume In %
8.24
17.79
Size (µm)
250.000
500.000
1000.000
Volume In %
22.39
26.28
Size (µm)
1000.000
2000.000
Volume In %
15.28
Size (µm) V
271
Sample No. 22
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
Volume(%)
KSB 22, Wednesday, May 01, 2013 10:34:55 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 22, Wednesday, May 01, 2013 10:34:55 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.16
5.69
Size (µm)
63.000
125.000
250.000
Volume In %
3.99
16.95
Size (µm)
250.000
500.000
1000.000
Volume In %
28.13
29.38
Size (µm)
1000.000
2000.000
Volume In %
15.70
Size (µm)
272
Sample No. 23
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
Volume(%)
KSB 23, Wednesday, May 01, 2013 10:37:34 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 23, Wednesday, May 01, 2013 10:37:34 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.29
8.28
Size (µm)
63.000
125.000
250.000
Volume In %
5.15
14.55
Size (µm)
250.000
500.000
1000.000
Volume In %
23.99
28.89
Size (µm)
1000.000
2000.000
Volume In %
18.84
Size (µm) V
273
Sample No. 24
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 24, Wednesday, May 01, 2013 10:39:41 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 24, Wednesday, May 01, 2013 10:39:41 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.07
4.55
Size (µm)
63.000
125.000
250.000
Volume In %
2.32
7.60
Size (µm)
250.000
500.000
1000.000
Volume In %
24.65
39.85
Size (µm)
1000.000
2000.000
Volume In %
20.96
Size (µm) V
274
Sample No. 25
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
Volume(%)
KSB 25, Wednesday, May 01, 2013 10:41:44 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 25, Wednesday, May 01, 2013 10:41:44 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.40
9.54
Size (µm)
63.000
125.000
250.000
Volume In %
8.46
19.10
Size (µm)
250.000
500.000
1000.000
Volume In %
21.00
24.80
Size (µm)
1000.000
2000.000
Volume In %
16.72
Size (µm) V
275
Sample No. 26
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
Volume(%)
KSB 26, Wednesday, May 01, 2013 10:43:42 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
Volume(%)
KSB 26, Wednesday, May 01, 2013 10:43:42 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.46
9.51
Size (µm)
63.000
125.000
250.000
Volume In %
11.45
22.50
Size (µm)
250.000
500.000
1000.000
Volume In %
19.83
22.31
Size (µm)
1000.000
2000.000
Volume In %
13.93
Size (µm)
276
Sample No. 27
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
Volume(%)
KSB 27, Wednesday, May 01, 2013 10:46:06 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 27, Wednesday, May 01, 2013 10:46:06 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.27
7.78
Size (µm)
63.000
125.000
250.000
Volume In %
9.44
22.53
Size (µm)
250.000
500.000
1000.000
Volume In %
23.93
23.35
Size (µm)
1000.000
2000.000
Volume In %
12.69
Size (µm) V
277
Sample No. 28
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Volume(%)
KSB 28, Wednesday, May 01, 2013 10:48:01 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 28, Wednesday, May 01, 2013 10:48:01 AM
Size (µm)
0.010
4.000
63.000
Volume In %
1.08
26.94
Size (µm)
63.000
125.000
250.000
Volume In %
13.95
18.23
Size (µm)
250.000
500.000
1000.000
Volume In %
17.40
15.74
Size (µm)
1000.000
2000.000
Volume In %
6.66
Size (µm) V
278
Sample No. 29
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
0.5
1
1.5
2
2.5
3
3.5
4
Volume(%)
KSB 29, Wednesday, May 01, 2013 10:50:36 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 29, Wednesday, May 01, 2013 10:50:36 AM
Size (µm)
0.010
4.000
63.000
Volume In %
1.26
26.79
Size (µm)
63.000
125.000
250.000
Volume In %
12.95
17.48
Size (µm)
250.000
500.000
1000.000
Volume In %
15.08
16.19
Size (µm)
1000.000
2000.000
Volume In %
10.24
Size (µm) V
279
Sample No. 30
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Volume(%)
KSB 30, Wednesday, May 01, 2013 10:52:20 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 30, Wednesday, May 01, 2013 10:52:20 AM
Size (µm)
0.010
4.000
63.000
Volume In %
1.02
25.96
Size (µm)
63.000
125.000
250.000
Volume In %
12.76
18.12
Size (µm)
250.000
500.000
1000.000
Volume In %
18.66
16.32
Size (µm)
1000.000
2000.000
Volume In %
7.16
Size (µm) V
280
Sample No. 31
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
Volume(%)
KSB 31, Wednesday, May 01, 2013 10:54:27 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
Volume(%)
KSB 31, Wednesday, May 01, 2013 10:54:27 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.72
20.18
Size (µm)
63.000
125.000
250.000
Volume In %
11.46
16.04
Size (µm)
250.000
500.000
1000.000
Volume In %
16.01
21.12
Size (µm)
1000.000
2000.000
Volume In %
14.45
Size (µm)
281
Sample No. 32A
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
Volume(%)
KSB 32 A, Wednesday, May 01, 2013 10:57:04 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 32 A, Wednesday, May 01, 2013 10:57:04 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.65
15.56
Size (µm)
63.000
125.000
250.000
Volume In %
7.26
17.01
Size (µm)
250.000
500.000
1000.000
Volume In %
23.32
23.79
Size (µm)
1000.000
2000.000
Volume In %
12.42
Size (µm) V
282
Sample No. 32B
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
Volume(%)
KSB 32 B, Wednesday, May 01, 2013 11:00:14 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 32 B, Wednesday, May 01, 2013 11:00:14 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.69
13.13
Size (µm)
63.000
125.000
250.000
Volume In %
6.70
13.02
Size (µm)
250.000
500.000
1000.000
Volume In %
23.46
29.20
Size (µm)
1000.000
2000.000
Volume In %
13.80
Size (µm)
283
Sample No. 33
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
Volume(%)
KSB 33, Wednesday, May 01, 2013 11:02:21 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 33, Wednesday, May 01, 2013 11:02:21 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.38
11.15
Size (µm)
63.000
125.000
250.000
Volume In %
4.66
12.54
Size (µm)
250.000
500.000
1000.000
Volume In %
24.28
30.23
Size (µm)
1000.000
2000.000
Volume In %
16.75
Size (µm)
284
Sample No. 34
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
Volume(%)
KSB 34, Wednesday, May 01, 2013 11:04:41 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 34, Wednesday, May 01, 2013 11:04:41 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.15
7.05
Size (µm)
63.000
125.000
250.000
Volume In %
3.01
14.91
Size (µm)
250.000
500.000
1000.000
Volume In %
30.54
29.99
Size (µm)
1000.000
2000.000
Volume In %
14.35
Size (µm) V
285
Sample No. 36
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
Volume(%)
KSB 36, Wednesday, May 01, 2013 11:07:07 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 36, Wednesday, May 01, 2013 11:07:07 AM
Size (µm)
0.010
4.000
63.000
Volume In %
1.33
21.69
Size (µm)
63.000
125.000
250.000
Volume In %
6.80
15.33
Size (µm)
250.000
500.000
1000.000
Volume In %
24.75
22.06
Size (µm)
1000.000
2000.000
Volume In %
8.04
Size (µm)
286
Sample No. 37
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
Volume(%)
KSB 37, Wednesday, May 01, 2013 11:10:15 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 37, Wednesday, May 01, 2013 11:10:15 AM
Size (µm)
0.010
4.000
63.000
Volume In %
1.04
15.30
Size (µm)
63.000
125.000
250.000
Volume In %
5.68
14.96
Size (µm)
250.000
500.000
1000.000
Volume In %
25.83
25.29
Size (µm)
1000.000
2000.000
Volume In %
11.90
Size (µm) V
287
Sample No. 38
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
Volume(%)
KSB 38, Wednesday, May 01, 2013 11:12:35 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 38, Wednesday, May 01, 2013 11:12:35 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.13
5.02
Size (µm)
63.000
125.000
250.000
Volume In %
2.56
19.12
Size (µm)
250.000
500.000
1000.000
Volume In %
34.28
27.10
Size (µm)
1000.000
2000.000
Volume In %
11.78
Size (µm)
288
Sample No. 39
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 39, Wednesday, May 01, 2013 11:14:16 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 39, Wednesday, May 01, 2013 11:14:16 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
0.62
Size (µm)
63.000
125.000
250.000
Volume In %
0.10
5.88
Size (µm)
250.000
500.000
1000.000
Volume In %
33.30
41.49
Size (µm)
1000.000
2000.000
Volume In %
18.60
Size (µm) V
289
Sample No. 40
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
Volume(%)
KSB 40, Wednesday, May 01, 2013 11:17:24 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 40, Wednesday, May 01, 2013 11:17:24 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.16
5.12
Size (µm)
63.000
125.000
250.000
Volume In %
1.56
11.76
Size (µm)
250.000
500.000
1000.000
Volume In %
29.75
34.74
Size (µm)
1000.000
2000.000
Volume In %
16.90
Size (µm) V
290
Sample No. 41
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
Volume(%)
KSB 41, Wednesday, May 01, 2013 11:20:05 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 41, Wednesday, May 01, 2013 11:20:05 AM
Size (µm)
0.010
4.000
63.000
Volume In %
1.89
18.88
Size (µm)
63.000
125.000
250.000
Volume In %
2.82
8.75
Size (µm)
250.000
500.000
1000.000
Volume In %
23.24
29.96
Size (µm)
1000.000
2000.000
Volume In %
14.47
Size (µm)
291
Sample No. 42
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
0.5
1
1.5
2
2.5
3
3.5
Volume(%)
KSB 42, Wednesday, May 01, 2013 11:22:11 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 42, Wednesday, May 01, 2013 11:22:11 AM
Size (µm)
0.010
4.000
63.000
Volume In %
3.54
32.64
Size (µm)
63.000
125.000
250.000
Volume In %
10.55
13.50
Size (µm)
250.000
500.000
1000.000
Volume In %
13.90
15.45
Size (µm)
1000.000
2000.000
Volume In %
10.42
Size (µm) V
292
Sample No. 46
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 46, Wednesday, May 01, 2013 11:25:35 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 46, Wednesday, May 01, 2013 11:25:35 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
2.66
Size (µm)
63.000
125.000
250.000
Volume In %
0.89
7.30
Size (µm)
250.000
500.000
1000.000
Volume In %
32.11
40.95
Size (µm)
1000.000
2000.000
Volume In %
16.08
Size (µm) V
293
Sample No. 47
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
9
Volume(%)
KSB 47, Monday, July 08, 2013 12:58:04 PM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 47, Monday, July 08, 2013 12:58:04 PM
Size (µm)
0.010
4.000
63.000
Volume In %
0.26
6.57
Size (µm)
63.000
125.000
250.000
Volume In %
4.60
13.87
Size (µm)
250.000
500.000
1000.000
Volume In %
31.04
32.61
Size (µm)
1000.000
2000.000
Volume In %
11.06
Size (µm)
294
Sample No. 48
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
Volume(%)
KSB 48, Monday, July 08, 2013 1:01:15 PM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 48, Monday, July 08, 2013 1:01:15 PM
Size (µm)
0.010
4.000
63.000
Volume In %
0.72
14.74
Size (µm)
63.000
125.000
250.000
Volume In %
6.94
16.16
Size (µm)
250.000
500.000
1000.000
Volume In %
27.29
25.55
Size (µm)
1000.000
2000.000
Volume In %
8.60
Size (µm)
295
Sample No. 49
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
0.5
1
1.5
2
2.5
3
3.5
4
Volume(%)
KSB 49, Monday, July 08, 2013 1:04:33 PM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 49, Monday, July 08, 2013 1:04:33 PM
Size (µm)
0.010
4.000
63.000
Volume In %
10.41
43.01
Size (µm)
63.000
125.000
250.000
Volume In %
5.83
11.06
Size (µm)
250.000
500.000
1000.000
Volume In %
16.20
10.43
Size (µm)
1000.000
2000.000
Volume In %
3.06
Size (µm)
296
Sample No. 50
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
9
Volume(%)
KSB 50, Monday, July 08, 2013 1:14:04 PM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
35
Volume(%)
KSB 50, Monday, July 08, 2013 1:14:04 PM
Size (µm)
0.010
4.000
63.000
Volume In %
0.07
4.93
Size (µm)
63.000
125.000
250.000
Volume In %
1.12
13.96
Size (µm)
250.000
500.000
1000.000
Volume In %
35.77
31.63
Size (µm)
1000.000
2000.000
Volume In %
12.52
Size (µm)
297
Sample No. 51
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
Volume(%)
KSB 51, Monday, July 08, 2013 1:17:42 PM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 51, Monday, July 08, 2013 1:17:42 PM
Size (µm)
0.010
4.000
63.000
Volume In %
0.75
13.91
Size (µm)
63.000
125.000
250.000
Volume In %
5.57
15.41
Size (µm)
250.000
500.000
1000.000
Volume In %
29.29
27.12
Size (µm)
1000.000
2000.000
Volume In %
7.95
Size (µm)
298
Sample No. 52
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
Volume(%)
KSB 52, Monday, July 08, 2013 1:20:25 PM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 52, Monday, July 08, 2013 1:20:25 PM
Size (µm)
0.010
4.000
63.000
Volume In %
1.28
24.05
Size (µm)
63.000
125.000
250.000
Volume In %
10.20
17.03
Size (µm)
250.000
500.000
1000.000
Volume In %
20.78
18.08
Size (µm)
1000.000
2000.000
Volume In %
8.59
Size (µm)
299
Sample No. 53
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
9
Volume(%)
KSB 53, Monday, July 08, 2013 1:25:14 PM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 53, Monday, July 08, 2013 1:25:14 PM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
0.62
Size (µm)
63.000
125.000
250.000
Volume In %
0.80
15.79
Size (µm)
250.000
500.000
1000.000
Volume In %
37.64
33.01
Size (µm)
1000.000
2000.000
Volume In %
12.14
Size (µm)
300
Sample No. 54
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
Volume(%)
KSB 54, Monday, July 08, 2013 1:29:30 PM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 54, Monday, July 08, 2013 1:29:30 PM
Size (µm)
0.010
4.000
63.000
Volume In %
2.07
19.96
Size (µm)
63.000
125.000
250.000
Volume In %
5.35
13.12
Size (µm)
250.000
500.000
1000.000
Volume In %
23.45
24.66
Size (µm)
1000.000
2000.000
Volume In %
11.39
Size (µm)
301
Sample No. 55
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
Volume(%)
KSB 55, Monday, July 08, 2013 1:35:40 PM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
Volume(%)
KSB 55, Monday, July 08, 2013 1:35:40 PM
Size (µm)
0.010
4.000
63.000
Volume In %
2.13
25.16
Size (µm)
63.000
125.000
250.000
Volume In %
9.23
18.12
Size (µm)
250.000
500.000
1000.000
Volume In %
22.73
17.49
Size (µm)
1000.000
2000.000
Volume In %
5.14
Size (µm)
302
Sample No. 56
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
1
2
3
4
5
6
7
8
9
Volume(%)
KSB 56, Monday, July 08, 2013 1:40:31 PM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 56, Monday, July 08, 2013 1:40:31 PM
Size (µm)
0.010
4.000
63.000
Volume In %
0.06
4.34
Size (µm)
63.000
125.000
250.000
Volume In %
1.77
7.90
Size (µm)
250.000
500.000
1000.000
Volume In %
24.59
39.15
Size (µm)
1000.000
2000.000
Volume In %
22.20
Size (µm)
303
Sample No. 94
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 94, Wednesday, May 01, 2013 11:28:10 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 94, Wednesday, May 01, 2013 11:28:10 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.07
3.30
Size (µm)
63.000
125.000
250.000
Volume In %
0.96
4.73
Size (µm)
250.000
500.000
1000.000
Volume In %
19.26
41.02
Size (µm)
1000.000
2000.000
Volume In %
30.66
Size (µm)
304
Sample No. 95
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 95, Wednesday, May 01, 2013 11:30:24 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 95, Wednesday, May 01, 2013 11:30:24 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
4.14
Size (µm)
63.000
125.000
250.000
Volume In %
1.43
5.35
Size (µm)
250.000
500.000
1000.000
Volume In %
24.66
41.56
Size (µm)
1000.000
2000.000
Volume In %
22.86
Size (µm)
305
Sample No. 96
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
0.5
1
1.5
2
2.5
3
3.5
4
Volume(%)
KSB 96, Wednesday, May 01, 2013 11:32:55 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
5
10
15
20
25
30
Volume(%)
KSB 96, Wednesday, May 01, 2013 11:32:55 AM
Size (µm)
0.010
4.000
63.000
Volume In %
2.15
30.01
Size (µm)
63.000
125.000
250.000
Volume In %
14.01
15.47
Size (µm)
250.000
500.000
1000.000
Volume In %
13.89
16.82
Size (µm)
1000.000
2000.000
Volume In %
7.64
Size (µm)
306
Sample No. 97
Particle Size Distribution
0.020161 300.02 500.02 700.02 900.02 1100 1300 1500 1700 1900
Particle Size (µm)
0
2
4
6
8
10
Volume(%)
KSB 97, Wednesday, May 01, 2013 11:36:02 AM
Particle Size Distribution
0.01 200.01 400.01 600.01 800.01 1000 1200 1400 1600 1800
Particle Size (µm)
0
10
20
30
40
Volume(%)
KSB 97, Wednesday, May 01, 2013 11:36:02 AM
Size (µm)
0.010
4.000
63.000
Volume In %
0.00
0.77
Size (µm)
63.000
125.000
250.000
Volume In %
0.14
8.88
Size (µm)
250.000
500.000
1000.000
Volume In %
34.60
40.24
Size (µm)
1000.000
2000.000
Volume In %
15.37
Size (µm) V
307
Grain Size Distribution
Station No. Sand Silt Clay Sediment type
KSB1 96.9 3.1 0.00 Sand
KSB2 98.58 1.42 0.00 Sand
KSB3 96.5 3.43 0.07 Sand
KSB4 96.55 3.45 0.00 Sand
KSB5 98.68 1.32 0.00 Sand
KSB6 98.15 1.85 0.00 Sand
KSB7 99.69 0.31 0.00 Sand
KSB8 96.14 3.8 0.06 Sand
KSB9 98.19 1.81 0.00 Sand
KSB10 95.22 4.71 0.07 Sand
KSB11 98.18 1.82 0.00 Sand
KSB12 98.68 1.32 0.00 Sand
KSB13 97.45 2.55 0.00 Sand
KSB14 93.71 6.23 0.06 Sand
KSB15 95.91 4.09 0.00 Sand
KSB16 89.74 9.97 0.29 Silty sand
KSB17 94.61 5.23 0.16 Sand
KSB18 93.67 6.17 0.16 Sand
KSB19 87.38 12.17 0.45 Silty sand
KSB20 82.45 16.76 0.79 Silty sand
KSB21 89.98 9.6 0.42 Silty sand
KSB22 94.15 5.69 0.16 Sand
KSB23 91.43 8.28 0.29 Sand
KSB24 95.38 4.55 0.07 Sand
KSB25 90.06 9.54 0.40 Sand
KSB26 90.03 9.51 0.46 Sand
KSB27 91.95 7.78 0.27 Sand
KSB28 71.98 26.94 1.08 Silty sand
KSB29 71.95 26.79 1.26 Silty sand
KSB30 73.02 25.96 1.02 Silty sand
308
Station No. Sand Silt Clay Sediment type
KSB31 79.1 20.18 0.72 Silty sand
KSB32-A 83.79 15.56 0.65 Silty sand
KSB32-B 86.18 13.13 0.69 Silty sand
KSB33 88.47 11.15 0.38 Silty sand
KSB34 92.8 7.05 0.15 Sand
KSB36 76.98 21.69 1.33 Silty sand
KSB37 83.66 15.3 1.04 Silty sand
KSB38 94.85 5.02 0.13 Sand
KSB39 99.38 0.62 0.00 Sand
KSB40 94.72 5.12 0.16 Sand
KSB41 79.23 18.88 1.89 Silty sand
KSB42 63.82 32.64 3.54 Silty sand
KSB46 97.34 2.66 0.00 Sand
KSB47 99.74 6.57 0.26 Sand
KSB48 99.28 14.74 0.72 Sand
KSB49 89.59 43.01 10.41 Silty sand
KSB50 99.93 4.93 0.07 Sand
KSB51 99.25 13.91 0.75 Sand
KSB52 98.72 24.05 1.28 Sand
KSB53 100 0.62 0.00 Sand
KSB54 97.93 19.96 2.07 Sand
KSB55 97.87 25.16 2.13 Sand
KSB56 99.94 4.34 0.06 Sand
KSB94 96.63 3.3 0.07 Sand
KSB95 95.86 4.14 0.00 Sand
KSB96 67.84 30.01 2.15 Silty sand
KSB97 99.23 0.77 0.00 Sand
309
APPENDIX 4: PARAMETERS USED FOR PROCESSING AND MOSAICING
SIDESCAN SONAR DATA
Software used: Sonarwiz (Chesapeake Technology Inc.)
Coordinate system used for data analysis
WGS84 UTM39N
Logging Anomalies
Set up Project
Coordinate source – auto (use fish if valid, otherwise ship)
Time constant for course smoothing = 300 pings
Coordinate system – UTM84-39N
Batch import of files
Course made good - checked
No gain applied
Percentage of sonar range to map - 100%
Channels 3 and 4
Thresholding enabled - sniff carried out on 6 files to establish thresholding value of 2231
Batch Processing of files
EGN applied – intensity=41
Seabed Classification Test
The same settings were used as for logging anomalies except that the files were trimmed
and split where they overlapped, and the nadir transparency was set to 4. Perpendicular
survey lines were not included in the mosaic.
310
APPENDIX 5: LIST OF ALL ANOMALIES LOGGED FROM THE SIDESCAN SONAR DATA
Area 1
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10001 45.58 7.91 1.89 Debris? 26.164178 51.067508
QBC_Q10002 3.39 3.73 0.21 Debris? 26.172531 51.053955
QBC_Q10003 17.09 1.65 0.17 Seabed Scar? 26.188127 51.033427
QBC_Q10004 0 0 0 Natural? 26.190223 51.030078
QBC_Q10005 13.52 1.9 0.14 Debris? 26.190361 51.028873
QBC_Q10006 10.94 1.54 0.11 Debris? 26.190689 51.028266
QBC_Q10007 11.62 5.98 0.38 Natural? 26.191166 51.02786
QBC_Q10008 22.98 13.1 0.15 Buried Feature? 26.192163 51.026041
QBC_Q10009 4.44 5.14 1.01 Debris? - Modern? 26.198301 51.018376
QBC_Q10010 21.65 17.38 0.94 Buried Feature? 26.198905 51.016678
QBC_Q10011 28.72 2.96 0.07 Natural? 26.202091 51.012029
QBC_Q10012 16.67 3.27 0 Debris? 26.209287 51.004147
QBC_Q10013 14.95 4.82 0.06 Natural? 26.217675 50.989785
QBC_Q10014 24.01 8.2 0.34 Debris? 26.140912 50.946437
QBC_Q10016 53.44 28.49 0.78 Debris? - Modern? 26.132907 50.957824
QBC_Q10018 1.57 1.57 0.28 Debris? 26.124179 50.968431
QBC_Q10020 6.53 2.49 0.12 Natural? 26.121128 50.972238
QBC_Q10021 3.74 4.02 0.29 Unclassified 26.119927 50.973677
QBC_Q10023 52.26 11.04 0.9 Debris? 26.115726 50.980105
QBC_Q10024 14.06 2.11 0.1 Natural? 26.113622 50.983122
QBC_Q10025 11.39 4.22 0.45 Debris - Modern? 26.110508 50.98751
311
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10026 2.43 3.58 1.49 Natural? 26.066991 50.903244
QBC_Q10027 17.9 7.71 0.23 Buried Feature?- 26.120183 50.974065
QBC_Q10030 105.23 5.09 1.2 Seabed Scar? 26.05359 50.922498
QBC_Q10035 44.6 16.38 1.04 Buried Feature? 26.049929 50.927743
QBC_Q10037 5.75 3.8 0.56 Buried Feature? 26.10989 50.915878
QBC_Q10038 20.24 14.35 0.54 Buried Feature? 26.109964 50.916072
QBC_Q10039 4.19 3.48 0.25 Debris? 26.109416 50.916106
QBC_Q10044 7.58 7.63 0.51 Debris? 26.105752 50.922107
QBC_Q10045 4.59 3.04 0.89 Debris? 26.104725 50.923774
QBC_Q10046 9.64 4.98 0.93 Debris? 26.103701 50.925436
QBC_Q10047 7.95 1.67 0 Debris? 26.096653 50.93603
QBC_Q10048 8.47 5.01 0.75 Debris? 26.095658 50.936373
QBC_Q10049 2.92 2.34 0.43 Natural? 26.092541 50.941048
QBC_Q10050 9.96 4.09 0.81 Debris? 26.090631 50.944056
QBC_Q10900 18.32 3.14 0 Duplicate of QBC_Q10050 26.090739 50.944173
QBC_Q10051 21.16 11.32 0.77 Debris? - Modern? 26.089546 50.943997
QBC_Q10052 45.47 1.67 0.9 Debris? 26.089965 50.944877
QBC_Q10053 4.69 3.3 0 Natural? 26.091397 50.943247
QBC_Q10056 0 0 0 Debris? 26.089551 50.943385
QBC_Q10057 3.03 1.08 0.3 Debris? 26.088984 50.944446
QBC_Q10058 8.74 1.92 0 Debris? 26.088443 50.944708
QBC_Q10059 6.5 5.15 3.48 Debris? 26.083818 50.951767
QBC_Q10060 7.27 1.28 0.23 Natural? 26.075834 50.964598
QBC_Q10061 25.94 3.43 0.32 Natural? 26.068163 50.973555
QBC_Q10062 7.32 2.12 1.14 Debris? 26.111185 50.923062
QBC_Q10063 5.59 1.77 1 Debris? 26.108357 50.926696
312
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10064 3.42 1.82 0.2 Natural? 26.107479 50.926505
QBC_Q10065 4.83 2.32 0.48 Debris? 26.108373 50.927508
QBC_Q10066 6.05 2.89 0.51 Buried Feature? 26.107364 50.927361
QBC_Q10067 5.08 4.41 0.09 Buried Feature? 26.107701 50.926639
QBC_Q10068 3.54 1.12 0.09 Buried Feature? 26.100057 50.939206
QBC_Q10069a 2.62 2.55 0.99 Unclassified 26.094479 50.94567
QBC_Q10069b 4.7 2.72 0.41 Buried Feature? 26.094703 50.945614
QBC_Q10071 3.44 1.99 0.43 Debris? 26.091898 50.948385
QBC_Q10072 9.34 2.37 0.76 Debris? 26.091842 50.950222
QBC_Q10073 2.77 0.74 0.49 Debris? 26.089504 50.953108
QBC_Q10074 2.41 0.74 0.16 Debris? 26.08955 50.953583
QBC_Q10075 26.41 12.41 0.97 Debris? 26.08889 50.953501
QBC_Q10076 5.51 2.39 0.76 Buried Feature? 26.082374 50.962397
QBC_Q10077 15.57 1.55 0 Natural? 26.080693 50.965024
QBC_Q10078 13.09 21.76 0.41 Natural? 26.084755 50.959102
QBC_Q10079 4.28 2.2 0.36 Natural? 26.113025 50.927079
QBC_Q10080 4.21 1.84 0.81 Debris? 26.113302 50.927475
QBC_Q10081 7.17 5.55 0.72 Debris? 26.112738 50.92822
QBC_Q10082 23.98 10.99 0 Location? 26.11209 50.928692
QBC_Q10083 7.52 3.61 1.01 Debris? 26.110706 50.930578
QBC_Q10084 19.52 2.98 0 Location? 26.104293 50.93928
QBC_Q10085 17.58 13.28 0.92 Debris? 26.098616 50.947458
QBC_Q10086 10.05 7.91 0.09 Buried Feature? 26.09642 50.952379
QBC_Q10087 3.33 0.73 0.13 Debris? 26.095703 50.95399
QBC_Q10088 1.66 0.94 0.64 Natural? 26.092932 50.956692
QBC_Q10089 22.08 12.26 0 Unclassified 26.092969 50.957561
313
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10090 19.63 19.47 0 Natural? 26.083392 50.969561
QBC_Q10091 34.48 12.42 0.52 Buried Feature? 26.077678 50.977986
QBC_Q10092 6.84 6.32 0 Buried Feature? 26.115737 50.931746
QBC_Q10093 8.57 1.95 0.31 Debris? 26.114192 50.93375
QBC_Q10094 6.77 4.35 0.41 Debris? 26.108821 50.941569
QBC_Q10095 6.01 2.99 0 Debris? 26.092568 50.962845
QBC_Q10096 22.95 4.17 0.11 Buried Feature? 26.092101 50.965949
QBC_Q10097 98.26 2.28 0 Linear Debris? 26.087507 50.971216
QBC_Q10098 23.27 6.76 0 Debris? 26.122298 50.929615
QBC_Q10099 14.99 6.1 0 Debris? 26.122068 50.929949
QBC_Q10100 3.84 2.77 0 Debris? 26.121916 50.930422
QBC_Q10101 1.26 0.5 0.16 Debris? 26.120506 50.933141
QBC_Q10102 1.62 0.75 0.25 Debris? 26.12021 50.933567
QBC_Q10103 4.42 2.17 0.44 Buried Feature? 26.117515 50.936816
QBC_Q10104 11.6 5.15 0.69 Debris? 26.116663 50.939105
QBC_Q10105 4.66 1.89 1.02 Debris? 26.114032 50.942135
QBC_Q10107 6.12 6.17 1.33 Buried Feature? - Duplicate 26.111764 50.945494
QBC_Q10106 0 0 0 Duplicate of QBC_Q10107 26.111778 50.945571
QBC_Q10108 9.08 6.57 0.8 Debris? 26.10403 50.95733
QBC_Q10109 1.56 1.25 0.37 Debris? 26.103142 50.95729
QBC_Q10110 16.81 8.48 0.87 Debris? 26.097583 50.9661
QBC_Q10111 10.64 3.62 0.47 Debris? 26.093648 50.970599
QBC_Q10112 15.62 10.02 0.73 Debris? 26.092174 50.973257
QBC_Q10113 16.51 0.85 0 Natural? 26.085384 50.983023
QBC_Q10114 3.49 1.68 0.27 Natural? 26.125137 50.934866
QBC_Q10115 6.9 0.68 0.37 Debris? 26.124633 50.936281
314
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10116 11.24 1.57 0 Buried Feature? 26.12002 50.941669
QBC_Q10117 34.54 1.34 0 Natural? 26.11693 50.94535
QBC_Q10118 1.54 1.27 1.18 Natural? 26.111446 50.953896
QBC_Q10119 26.06 8.07 0 Buried Feature? 26.109767 50.956038
QBC_Q10120 12.25 3.84 0.58 Buried Feature? 26.102049 50.968652
QBC_Q10121 1.6 1.02 0.35 Natural? 26.09922 50.971846
QBC_Q10122 1.22 1.13 0.34 Natural? 26.099352 50.971723
QBC_Q10123 0.71 0.95 0.43 Natural? 26.099284 50.971785
QBC_Q10124 7.98 2.54 1.11 Debris? 26.09884 50.972745
QBC_Q10125 34.3 5.02 0.74 Buried feature? 26.098345 50.974332
QBC_Q10126 3.98 2.74 1.28 Buried Feature? 26.127756 50.939085
QBC_Q10127 6 2.75 0.25 Debris? 26.128502 50.939809
QBC_Q10128 16.21 4.9 0.66 Bham0074 - Cars 26.127721 50.9399
QBC_Q10129 0 0 0 Debris? 26.127713 50.940492
QBC_Q10130 6.55 5.48 0 Debris? 26.127253 50.941361
QBC_Q10131 23.06 1.42 0 Seabed Scar? 26.127141 50.941242
QBC_Q10132 6.64 5.41 0.89 Debris? 26.126667 50.941273
QBC_Q10133 0 0 0 Hole? 26.126209 50.94185
QBC_Q10134 19.17 2.89 0 Debris? 26.1226 50.945164
QBC_Q10135 24.74 4.12 0 Debris? 26.112494 50.959673
QBC_Q10136 6.64 4.07 0 Debris? 26.109849 50.962975
QBC_Q10137 5.22 3.97 0 Debris? 26.109132 50.967272
QBC_Q10138 7.46 1.37 0.69 Debris? 26.098517 50.979507
QBC_Q10139 50.05 1.78 0 Natural? 26.095911 50.983161
QBC_Q10140 1.41 1.08 0.58 Natural? 26.096149 50.983603
QBC_Q10141 16.18 7.11 0 Debris? 26.130633 50.944689
315
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10142 3.55 1.79 0.61 Debris? 26.12952 50.944773
QBC_Q10143 2.54 2.54 0.53 Debris? 26.128734 50.944708
QBC_Q10144 0 0 0 Debris? 26.128266 50.94669
QBC_Q10145 3.38 2.34 0.45 Buried Feature? 26.128192 50.947362
QBC_Q10146 16.76 7.05 0 Debris? 26.126096 50.948485
QBC_Q10147 2.92 1.71 0 Natural? 26.126761 50.949553
QBC_Q10148 1.76 0.58 0.23 Natural? 26.12299 50.954576
QBC_Q10149 3.11 1.8 0 Debris? 26.119636 50.960689
QBC_Q10150 33.33 3.41 0 Natural? 26.11807 50.961883
QBC_Q10151 10.15 2.76 0.5 Debris? 26.117805 50.962641
QBC_Q10152 2.25 1.14 0.62 Debris? 26.116527 50.963562
QBC_Q10153 31.79 4.59 0 Debris? 26.109352 50.974734
QBC_Q10155 8.78 3.94 0.52 Debris? 26.104284 50.981682
QBC_Q10156 3.69 2.14 0 Buried feature? 26.09757 50.989577
QBC_Q10157 12.02 1.49 0 Natural? 26.091728 50.999054
QBC_Q10158 3.01 2.77 0.1 Natural? 26.090182 51.000899
QBC_Q10159 12.43 18.1 0.15 Buried Feature? 26.087193 51.005279
QBC_Q10160 6.06 1.62 0.99 Debris? - Modern? 26.135037 50.945099
QBC_Q10161 44.17 13.51 0.88 Debris? - Modern? 26.132914 50.948831
QBC_Q10162 67.45 2.03 0 Natural? 26.122775 50.961895
QBC_Q10163 10.03 0.65 0.16 Natural? 26.121548 50.963778
QBC_Q10164 5.83 2.51 0.91 Debris? 26.12081 50.96621
QBC_Q10165 1.76 1.47 0 Debris? 26.117534 50.968357
QBC_Q10166 19.26 11.17 0.67 Debris? 26.117197 50.97072
QBC_Q10167 0 0 0 Object in Water Column? 26.116877 50.970623
QBC_Q10168 8.94 2.9 0.79 Debris? - Modern? 26.112267 50.976146
316
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10169 49.67 1.59 0 Seabed Scar? 26.112176 50.978016
QBC_Q10170 14.58 2.96 0.78 Buried Feature? 26.105195 50.986914
QBC_Q10172 58.46 4.21 0 Location? 26.098402 50.996512
QBC_Q10173 5.4 2.08 0.18 Location? 26.090331 51.023801
QBC_Q10175 4.92 3.52 0.38 Natural? 26.092546 51.022249
QBC_Q10176 0 0 0 Location? 26.094385 51.018362
QBC_Q10177 2.28 2.27 0 Hole? 26.115219 50.989037
QBC_Q10178 9.81 4.7 0.5 Debris? 26.116643 50.986132
QBC_Q10179 1.17 1.57 0.44 Debris? 26.116962 50.986285
QBC_Q10180 30.6 10.05 0 Location? 26.119999 50.983597
QBC_Q10181 5.11 1.51 0.19 Natural? 26.137381 50.957887
QBC_Q10182 9.83 3.37 0.52 Debris? - Modern? 26.140096 50.95542
QBC_Q10184 8.07 4.68 0.82 Debris? 26.122287 50.988372
QBC_Q10185 7.22 4.84 0.71 Debris? 26.125707 50.981833
QBC_Q10186 0 0 0 Debris? - Modern? 26.130007 50.976833
QBC_Q10187 5.59 3.6 0.73 Debris? 26.136482 50.966842
QBC_Q10188 20.52 6.76 0.3 Debris? - Modern? 26.140766 50.96262
QBC_Q10189 3.75 2.95 0 Hole? 26.147616 50.952451
QBC_Q10190 5.1 2.14 0 Debris? - Modern? 26.149583 50.955742
QBC_Q10191 10.91 7.98 0 Debris? 26.143373 50.967286
QBC_Q10192 6.99 2.63 0.27 Debris? - Modern? 26.140544 50.968642
QBC_Q10193 6.85 4.42 0.62 Debris? 26.138977 50.972897
QBC_Q10194 35.86 2.36 0 Seabed Scar? 26.103934 51.022202
QBC_Q10195 6.94 3.15 0 Buried Feature? 26.097815 51.028528
QBC_Q10196 11.03 3.42 0.73 Debris? - Modern? 26.089698 51.039936
QBC_Q10197 39.71 10.8 0 Natural? 26.098603 51.035768
317
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10198 87.39 28.38 0 Location? 26.116708 51.012945
QBC_Q10199 7.28 5.99 1.01 Debris? - Modern? 26.122426 51.002891
QBC_Q10200 23.65 2.32 0 Natural? 26.123063 51.003561
QBC_Q10201 0 0 0.8 Unclassified 26.124505 50.999772
QBC_Q10202 51.84 30.17 0 Location? 26.125948 50.998024
QBC_Q10203 5.2 2.44 0.68 Debris - Modern? 26.130334 50.993269
QBC_Q10204 1.69 0.7 0.27 Debris? 26.129516 50.993047
QBC_Q10205 10.56 5.94 0.13 Buried Feature? 26.140785 50.978639
QBC_Q10206 2.79 2.37 0.72 Debris? 26.140742 50.978097
QBC_Q10207 6.04 2.23 0.53 Debris? - Modern? 26.143998 50.972776
QBC_Q10208 11.67 8.95 0.53 Debris? 26.14765 50.967655
QBC_Q10209 16.49 6.85 0.81 Debris? 26.15206 50.961337
QBC_Q10210 11.33 3.38 0 Debris? 26.11286 50.922721
QBC_Q10211 0 0 0 Location? 26.112947 50.924671
QBC_Q10213 4.9 5.04 0.53 Debris? 26.103412 50.936359
QBC_Q10214 5.67 2.52 0.42 Debris? 26.102886 50.936858
QBC_Q10215 6.05 1.4 0.46 Debris? 26.102337 50.938217
QBC_Q10216 8.58 4.45 0.75 Bham0073 - Cars 26.101064 50.939624
QBC_Q10217 15.94 1.05 0 Natural? 26.097927 50.944155
QBC_Q10218 4.06 2.44 1.25 Debris? 26.095919 50.947739
QBC_Q10219 0 0 0 Unclassified 26.092693 50.950983
QBC_Q10220 14.13 8.38 0.95 Debris? 26.090578 50.956323
QBC_Q10221 0 0 0 Location? 26.089928 50.957252
QBC_Q10222 2.44 0.83 0.35 Buried Feature? 26.089615 50.956595
QBC_Q10223 2.86 1.79 0 Debris? 26.088618 50.959718
QBC_Q10224 52.84 7.75 0 Natural? 26.062097 50.906542
318
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10225 101.93 12.77 0 Natural? 26.059907 50.909682
QBC_Q10226 24.58 12.78 0 Natural? 26.068826 50.905008
QBC_Q10227 30.58 18.43 0 Location? 26.066123 50.909461
QBC_Q10228 9.65 4.97 1.4 Debris? 26.065381 50.909654
QBC_Q10229 97.17 4.84 0 Seabed Scar? 26.061457 50.914405
QBC_Q10230 92.29 6.12 0 Data problem? 26.060353 50.916225
QBC_Q10231 83.7 2.17 0 Linear Debris? 26.060518 50.91724
QBC_Q10232 28.89 8.22 0 Depression? 26.060296 50.917675
QBC_Q10233 84.05 3.61 0 Seabed Scar? 26.055883 50.922871
QBC_Q10234 0 12.3 0 Depression? 26.052369 50.927506
QBC_Q10235 0 0 0 Depression? 26.050911 50.929229
QBC_Q10236 2.87 1.68 0.98 Debris? - Modern? 26.077354 50.901243
QBC_Q10237 10.11 5.08 1.27 Debris? - Modern? 26.075569 50.903497
QBC_Q10238 9.08 2.32 0.76 Debris? 26.074429 50.90462
QBC_Q10239 38.06 21.79 0 Location? 26.070272 50.911597
QBC_Q10240 10.03 1.9 0.41 Natural? 26.062519 50.922783
QBC_Q10241 0 0 0.47 Debris? 26.043976 50.948712
QBC_Q10242 3.83 2.9 1.35 Debris? 26.043286 50.949088
QBC_Q10243 0 0 0.3 Debris? 26.043442 50.949323
QBC_Q10244 7.06 5.38 0.96 Natural? 26.054666 50.901281
QBC_Q10245 2.75 1.86 0.64 Natural? 26.051646 50.899967
QBC_Q10246 4.23 2.36 0.96 Natural? 26.067749 50.895292
QBC_Q10247 23.56 15.17 0 Location? 26.066965 50.896625
QBC_Q10248 55.37 1.44 0 Natural? 26.05887 50.907371
QBC_Q10249 4.92 2.04 0.71 Debris? 26.063726 50.916268
QBC_Q10250 2.07 2.07 1.04 Debris? 26.063292 50.917283
319
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10251 31.51 1.85 0 Natural? 26.06366 50.915651
QBC_Q10252 16 6.06 0 Location? 26.078566 50.903123
QBC_Q10253 34.11 30.74 0 Location? 26.069383 50.916591
QBC_Q10254 23.94 2.24 0 Seabed Scar? 26.068958 50.918444
QBC_Q10255 99.32 3.45 0 Data problem? 26.06761 50.917947
QBC_Q10256 5.93 1.84 0 Natural? 26.067568 50.919924
QBC_Q10257 4.8 1.48 0.71 Natural? 26.045402 50.950653
QBC_Q10258 8.12 6.24 0 Natural? 26.042201 50.95442
QBC_Q10259 21.93 12.4 0 Location? 26.069299 50.915386
QBC_Q10260 49.4 14.2 0 Depression? 26.082551 50.907172
QBC_Q10261 21.12 4.05 0 Debris? 26.07874 50.912878
QBC_Q10262 4.65 1.75 0.54 Debris? - Modern? 26.077936 50.912727
QBC_Q10263 38.51 10.29 0 Depression? 26.078133 50.913414
QBC_Q10264 12.09 11.9 0 Location? 26.073948 50.917897
QBC_Q10265 10.44 3.6 0.91 Bham0020 - cars 26.07089 50.922166
QBC_Q10266 7.75 3.46 0.38 Natural? 26.046325 50.956833
QBC_Q10267 4.6 2.06 0.26 Natural? 26.042798 50.962266
QBC_Q10268 7.43 4.27 0.33 Natural? 26.041907 50.96437
QBC_Q10269 4.65 1.6 0 Natural? 26.044931 50.960365
QBC_Q10270 10.63 3.41 0.74 Debris? - Modern? 26.085082 50.910245
QBC_Q10271 37.77 1.79 0 Natural? 26.083989 50.91146
QBC_Q10272 3.71 0.91 0.57 Natural? 26.08364 50.912842
QBC_Q10273 10.08 14.29 0.49 Bham0019 - Cars 26.078984 50.918596
QBC_Q10274 37.24 12.83 0 Location? 26.075106 50.923526
QBC_Q10275 55.93 45.92 0 Location? 26.064642 50.939217
QBC_Q10276 5.96 2.28 0.89 Debris? 26.060076 50.94573
320
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10277 63.21 3.6 0 Depression? 26.060305 50.945978
QBC_Q10278 7.28 4.72 0.23 Natural? 26.057685 50.950464
QBC_Q10279 7.35 1.92 0.74 Buried Feature? 26.055975 50.952156
QBC_Q10280 31.94 1.71 0 Natural? 26.053293 50.955595
QBC_Q10281 8.35 2.95 0.23 Natural? 26.047329 50.96471
QBC_Q10282 9.36 0.93 0.2 Natural? 26.046697 50.964255
QBC_Q10283 35.11 5.12 0.42 Buried Feature? 26.044793 50.966205
QBC_Q10284 44.75 2.26 0 Natural? 26.08994 50.911351
QBC_Q10285 4.79 1.94 0.4 Debris? - Modern? 26.086224 50.916698
QBC_Q10286 12.15 3.36 0.54 Bham0017 - Cars 26.086726 50.917386
QBC_Q10287 8.29 3.39 0.71 Debris? - Modern? 26.084214 50.918716
QBC_Q10288 2.61 0.87 0.43 Debris? 26.082951 50.922121
QBC_Q10289 6.82 1.58 0.11 Natural? 26.080507 50.924306
QBC_Q10290 8.33 4.56 0 Natural? 26.078858 50.92888
QBC_Q10291 0 0 0 Location? 26.07881 50.928273
QBC_Q10292 54.39 32.54 0.74 Natural? 26.076672 50.931454
QBC_Q10293 3.9 2.29 0.69 Debris? - Modern? 26.074413 50.933972
QBC_Q10294 12.72 1.97 0.55 Debris? 26.072967 50.934841
QBC_Q10295 19.62 9.98 0 Buried Feature? 26.070205 50.939847
QBC_Q10297 0 0 0 Duplicate of QBC_Q10295 26.070232 50.939827
QBC_Q10298 4.2 2.34 0.79 Debris? - Modern? 26.06983 50.940326
QBC_Q10299 48.04 1.62 0 Natural? 26.068463 50.941823
QBC_Q10300 10.49 8.3 0 Location? 26.0641 50.947909
QBC_Q10301 14.73 10.39 0.36 Natural ? 26.055367 50.959361
QBC_Q10302 3.02 2.42 0.7 Debris? - Modern? 26.096733 50.910892
QBC_Q10303 6.4 2.68 0 Debris? 26.092732 50.917014
321
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10304 23.89 2.62 0.81 Bham0016 - Tyres 26.091109 50.917744
QBC_Q10305 6.69 3.08 0.53 Debris? 26.089936 50.919292
QBC_Q10306 4.64 3.57 0 Debris? 26.089514 50.919463
QBC_Q10307 5.24 1.82 0.72 Debris? - Modern? 26.087683 50.922878
QBC_Q10308 11.56 3.47 0 Debris? - Modern? 26.087451 50.92484
QBC_Q10309 4.01 2.67 0.46 Natural? 26.080841 50.933292
QBC_Q10310 35.32 2.38 0 Natural? 26.076909 50.937072
QBC_Q10311 4.56 3.65 0.69 Debris? 26.076836 50.938696
QBC_Q10312 5.59 2.57 0 Natural? 26.071291 50.947673
QBC_Q10313 41.42 14.3 0.47 Natural? 26.067197 50.951187
QBC_Q10314 74.78 2.91 0 Seabed Scar? 26.065338 50.954863
QBC_Q10315 18.32 1.11 0 Natural? 26.062645 50.958109
QBC_Q10316 4.89 3.06 0.35 Natural? 26.057918 50.964623
QBC_Q10317 27.29 6.41 0 Buried Feature? 26.056922 50.965199
QBC_Q10318 3.53 1.13 0 Natural ? 26.101537 50.912217
QBC_Q10319 3.58 2.22 0 Natural? 26.097268 50.916536
QBC_Q10320 24.95 10.81 0 Seabed Scar? 26.097572 50.917563
QBC_Q10321 4.99 2.1 1.1 Debris? 26.093511 50.923065
QBC_Q10322 13.95 4.36 0.66 Bham0013 - Tyres 26.092083 50.924327
QBC_Q10323 4.08 1.3 0.58 Debris? 26.091531 50.925976
QBC_Q10324 23.37 10.23 0 Natural? 26.09102 50.925282
QBC_Q10325 5.73 3.24 0 Debris? 26.088973 50.927951
QBC_Q10326 13.12 4.8 0.89 Bham0012 - Cars 26.088844 50.929978
QBC_Q10327 57.44 19.08 0.63 Buried Feature? 26.083173 50.936775
QBC_Q10328 6.32 1.71 0 Debris? 26.083721 50.93771
QBC_Q10329 7.8 2.18 0.87 Debris? - Modern? 26.082836 50.93811
322
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10330 6.15 3.01 0 Natural? 26.080197 50.940258
QBC_Q10331 6.44 2.71 0.98 Debris? 26.103053 50.918189
QBC_Q10332 4.45 2.62 0.83 Debris? - Modern? 26.101257 50.92039
QBC_Q10333 11.7 8.91 0.17 Natural ? 26.100566 50.920102
QBC_Q10334 78.8 4.18 0.68 Debris? - Modern? 26.099281 50.921995
QBC_Q10335 4.09 2.19 0.94 Debris? 26.099227 50.92284
QBC_Q10336 57.23 34.68 0.62 Debris? - Modern? 26.096039 50.926544
QBC_Q10337 6.41 2.16 0.94 Debris? 26.095839 50.927719
QBC_Q10338 17.77 3.66 1.42 Debris? - Modern? 26.092555 50.931337
QBC_Q10339 5.27 2.1 0.7 Debris? 26.090511 50.93551
QBC_Q10340 0 0 0 Data problem? 26.090887 50.935667
QBC_Q10341 5.06 1.86 1.02 Debris? - Modern? 26.088323 50.938225
QBC_Q10342 7.62 2.53 0.71 Debris? - Modern? 26.086813 50.939104
QBC_Q10343 0 0 0 Duplicate of QBC_Q10342 26.086859 50.939092
QBC_Q10344 5.45 1.71 0.47 Debris? 26.087336 50.940866
QBC_Q10345 51.55 15.54 0 Location? 26.079559 50.949938
QBC_Q10346 15.94 15.63 0 Natural? 26.075275 50.955642
QBC_Q10347 15.43 6.64 0 Natural? 26.072676 50.959533
QBC_Q10349 10.28 1.71 0 Natural ? 26.062442 50.975204
QBC_Q10350 3.78 3.33 0.65 Debris? 26.114509 50.930767
QBC_Q10352 66.08 1.53 0 Natural? 26.10597 50.940842
QBC_Q10353 12.94 3.21 0.37 Debris? 26.10262 50.945964
QBC_Q10354 34.81 3.58 0 Data problem? 26.10227 50.94849
QBC_Q10355 5.32 2.3 0.68 Bham0040 - Cars 26.097135 50.954447
QBC_Q10356 4.62 2.1 0 Natural? 26.092559 50.959382
QBC_Q10357 12.56 2.52 0.85 Debris? 26.092271 50.962169
323
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10358 18.5 8.3 0 Buried feature? 26.091498 50.960999
QBC_Q10359 6.33 3.28 0 Debris? - Modern? 26.114074 50.939762
QBC_Q10360 6.94 4.88 0.86 Debris? 26.109001 50.946489
QBC_Q10361 7.83 3.63 0 Debris? 26.104551 50.950737
QBC_Q10362 14.54 2.91 0.84 Buried Feature? 26.091321 50.970833
QBC_Q10363 3.44 3.19 1.16 Debris? 26.124418 50.931742
QBC_Q10364 29.94 1.5 0 Net? 26.118995 50.938643
QBC_Q10365 1.51 2.23 0.49 Debris? 26.119324 50.93896
QBC_Q10366 4.2 3.33 1.17 Debris? 26.117135 50.94257
QBC_Q10367 7.03 3.33 0.95 Debris - Modern? 26.112085 50.948764
QBC_Q10368 5.47 1.91 0.96 Debris? - Modern? 26.107309 50.957101
QBC_Q10369 17.69 3.41 0.91 Debris? - Modern? 26.104606 50.959414
QBC_Q10370 3.53 1.23 0.61 Debris? 26.099705 50.96755
QBC_Q10371 16.69 3.95 0.96 Debris? - Modern? 26.0983 50.969533
QBC_Q10372 8.04 4.55 0 Debris? 26.098361 50.969887
QBC_Q10373 8.61 6.31 0.38 Buried Feature? 26.095926 50.971923
QBC_Q10374 100.7 1.13 0 Seabed Scar? 26.093658 50.97629
QBC_Q10375 4.46 2.17 0 Bham0038 - Cars 26.093331 50.976766
QBC_Q10376 29.13 1.52 0 Location? 26.090243 50.979979
QBC_Q10377 2.83 3.04 0 Buried Feature? 26.12209 50.944676
QBC_Q10378 15.37 12.25 0.51 Buried Feature? 26.117737 50.948732
QBC_Q10379 11.26 4.67 0.31 Natural? 26.115013 50.952333
QBC_Q10380 12.77 9.72 0.59 Buried Feature? 26.114713 50.952828
QBC_Q10381 6.54 3.28 0.7 Debris? - Modern? 26.113137 50.955331
QBC_Q10382 6.61 2.24 0.79 Debris? - Modern? 26.109259 50.960807
QBC_Q10383 3.51 1.9 0.96 Buried Feature? 26.108103 50.962087
324
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10384 7.4 2.97 0 Debris? 26.099536 50.973515
QBC_Q10385 80.89 2.34 0 Natural ? 26.099046 50.976504
QBC_Q10386 10.38 2.83 0.82 Buried feature? 26.098194 50.977084
QBC_Q10387 5.4 2.33 0.74 Debris? - Modern? 26.096483 50.979281
QBC_Q10388 48.43 1.69 0 Natural? 26.092071 50.986189
QBC_Q10389 29.51 10.77 0 Location? 26.090276 50.989229
QBC_Q10390 5.75 1.07 0.24 Location? 26.088617 50.990707
QBC_Q10391 7.26 4.01 0.13 Natural ? 26.088166 50.992154
QBC_Q10392 38.71 17.8 0 Depression? 26.081654 50.965677
QBC_Q10393 16.84 7.15 0.42 Debris? 26.082116 50.964915
QBC_Q10394 25.28 11.11 0 Depression? 26.081151 50.965134
QBC_Q10395 3.49 1.75 0.48 Natural ? 26.074486 50.959364
QBC_Q10396 29.57 4.91 0.54 Duplicate of QBC_Q10313 26.067142 50.951194
QBC_Q10397 7.75 3.57 0.96 Duplicate of QBC_Q10276 26.060116 50.945687
QBC_Q10398 21.23 3.96 0.28 Debris? 26.132888 50.937619
QBC_Q10399 8.99 6.81 0 Depression? 26.132258 50.936895
QBC_Q10400 16.74 3.35 1.11 Debris? 26.131783 50.937512
QBC_Q10401 10.8 5.03 1.18 Debris? 26.13177 50.938311
QBC_Q10402 17.03 3.44 0 Debris? 26.130814 50.937899
QBC_Q10403 6.42 2.2 1.05 Debris? 26.130915 50.939438
QBC_Q10407 8.55 5.93 0 Debris? 26.130206 50.938566
QBC_Q10408 24.87 3.09 0.93 Debris? 26.129648 50.9402
QBC_Q10409 10.53 3.08 1.85 Debris? 26.129444 50.940623
QBC_Q10410 6.53 3.49 0 Debris? 26.129233 50.940292
QBC_Q10411 6.08 4.32 1.17 Debris? 26.130272 50.940832
QBC_Q10412 3.22 3.27 0 Natural? 26.130334 50.939137
325
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10413 3.28 2.06 0.51 Debris? 26.129181 50.941226
QBC_Q10414 74.61 2.1 0 Seabed Scar? 26.128383 50.941777
QBC_Q10415 29.82 1.47 0 Natural? 26.128076 50.942728
QBC_Q10416 0 0 0 Debris? 26.12765 50.942574
QBC_Q10417 4.71 1.85 0 Natural? 26.131683 50.936887
QBC_Q10418 3.61 1.82 0.84 Debris? 26.127657 50.943502
QBC_Q10419 12.05 7.85 0.94 Debris? 26.119333 50.955536
QBC_Q10420 66.94 11.25 0.74 Debris? 26.118021 50.956311
QBC_Q10421 5.94 2.77 0 Natural? 26.112134 50.966925
QBC_Q10422 4.73 1.94 0.45 Natural? 26.10829 50.971582
QBC_Q10423 7.27 2.59 0 Natural? 26.097671 50.984475
QBC_Q10424 7.85 2.53 0.65 Natural? 26.096792 50.987606
QBC_Q10425 9.63 4.64 0 Natural? 26.08853 50.997432
QBC_Q10426 2.2 1.47 0.67 Natural? 26.132898 50.944483
QBC_Q10427 11.02 2.02 0.62 Buried Feature? 26.130293 50.94753
QBC_Q10428 15.59 5.21 0.57 Debris? 26.129287 50.94973
QBC_Q10429 12.98 1.83 0 Debris? 26.12179 50.959458
QBC_Q10430 6.82 6.23 0.8 Debris? 26.121587 50.960683
QBC_Q10431 12.99 3.59 0 Natural? 26.121122 50.959972
QBC_Q10432 6.19 2.77 0.77 Natural? 26.120058 50.96336
QBC_Q10433 8.74 5.66 0 Natural? 26.117114 50.967657
QBC_Q10434 4.89 3.38 0.83 Debris? 26.113673 50.971041
QBC_Q10435 5.08 2.5 0.86 Debris - Modern? 26.110855 50.974373
QBC_Q10436 15.4 8.21 0.34 Natural? 26.088292 51.015315
QBC_Q10437 49 2.38 0 Natural? 26.093642 51.00693
QBC_Q10438 13.33 7.38 0 Object in Water Column? 26.107673 50.987568
326
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10439 14.11 4.58 0 Object in Water Column? 26.109215 50.98577
QBC_Q10440 10.65 2.94 0 Natural? 26.112253 50.982443
QBC_Q10441 6.15 2.29 0 Natural? 26.115561 50.9784
QBC_Q10442 3.85 2.26 0 Debris? 26.119473 50.973045
QBC_Q10443 37.07 19.91 0 Natural? 26.118104 50.972227
QBC_Q10444 12.03 9.41 0 Object in Water Column? 26.119496 50.971046
QBC_Q10445 11.19 3.23 0.94 Debris? 26.119187 50.970769
QBC_Q10446 7.53 5.78 0.7 Natural? 26.120371 50.96879
QBC_Q10447 15.25 9 0 Object in Water Column? 26.122053 50.967423
QBC_Q10448 4.4 1.87 0.24 Natural? 26.133035 50.953097
QBC_Q10449 8.92 2.97 0 Debris? 26.134125 50.957602
QBC_Q10450 5.17 1.89 0.23 Natural? 26.128833 50.965904
QBC_Q10451 13.78 5.18 0.76 Debris? 26.127867 50.96907
QBC_Q10452 43.53 3.92 0 Natural? 26.122308 50.974889
QBC_Q10453 11.89 1.8 1.23 Debris? 26.122296 50.975878
QBC_Q10454 13.81 7.03 0 Debris? 26.120713 50.976605
QBC_Q10455 27.69 15.84 0 Unclassified 26.12092 50.97602
QBC_Q10456 21.23 11.37 0 Mound? 26.101249 51.006154
QBC_Q10457 0 0 0 Depression? 26.0986 51.009047
QBC_Q10458 25.54 3.82 0 Unclassified 26.090337 51.019905
QBC_Q10459 11.72 7.85 0.63 Unclassified 26.140022 50.957661
QBC_Q10460 13.45 15.23 0 Unclassified 26.140349 50.959474
QBC_Q10461 5.79 1.6 0.9 Buried Feature? 26.137013 50.963788
QBC_Q10462 1.17 0.57 0 Buried Feature? 26.137314 50.963271
QBC_Q10463 9.03 2.7 0 Debris? 26.124795 50.978596
QBC_Q10464 8.43 5.9 0.33 Debris? 26.116303 50.990638
327
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10465 116.46 35.15 0 Location? 26.098932 51.016057
QBC_Q10466 14.57 5.64 0.67 Debris? 26.141276 50.965515
QBC_Q10467 19.38 8.29 0 Depression? 26.122558 50.992131
QBC_Q10468 16.43 5.65 0.25 Natural? 26.117767 50.997205
QBC_Q10469 17.16 5.24 0.58 Natural? 26.116921 50.998238
QBC_Q10470 37.55 1.46 0.27 Natural? 26.095451 51.02946
QBC_Q10471 30.07 9.78 0 Data problem? 26.142646 50.972823
QBC_Q10472 54.31 26.26 0 Data problem? 26.142073 50.973382
QBC_Q10473 2.98 2.1 0.78 Debris? 26.1388 50.977506
QBC_Q10474 4.25 2.58 0 Debris? 26.138399 50.976924
QBC_Q10475 8.35 1.52 0.4 Natural? 26.13651 50.980969
QBC_Q10476 54.89 2.44 0.66 Bham0032 - Cars 26.126558 50.993165
QBC_Q10477 3.09 1.45 0.15 Natural? 26.12219 51.001116
QBC_Q10478 8.42 1.85 1.27 Duplicate 26.122387 51.002858
QBC_Q10479 109.15 31.53 0 Location? 26.100921 51.030072
QBC_Q10480 7.38 3.77 0.88 Debris - Modern? 26.150078 50.968992
QBC_Q10481 15.49 5.94 0.55 Debris? 26.145429 50.975783
QBC_Q10482 9.61 9.62 0 Data problem? 26.141467 50.981648
QBC_Q10483 6.62 2.9 0 Natural? 26.141977 50.981637
QBC_Q10484 17.52 6.99 0 Debris? 26.141425 50.982197
QBC_Q10485 8.03 2.52 0.14 Debris? 26.140669 50.982893
QBC_Q10486 0 0 0 Location? 26.129037 50.998825
QBC_Q10487 5.08 4.15 0.62 Debris - Modern? 26.155279 50.964161
QBC_Q10488 8.02 2.38 0.24 Natural? 26.146583 50.978489
QBC_Q10489 3.87 3.11 0 Object in Water Column? 26.144162 50.98089
QBC_Q10490 50.58 1.71 0 Natural? 26.144382 50.981644
328
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10491 24.74 12.95 0.22 Buried Feature? 26.142245 50.982567
QBC_Q10492 40.21 2.9 0 Natural? 26.129952 51.000018
QBC_Q10493 4.99 2.17 0.24 Debris? 26.155025 50.970307
QBC_Q10494 8.34 4.94 0 Natural? 26.152498 50.974397
QBC_Q10495 5.29 2.12 0.64 Debris - Modern? 26.150855 50.975168
QBC_Q10496 11.65 11.06 0 Debris? 26.146534 50.980692
QBC_Q10497 0.32 1.35 0.75 Natural? 26.146553 50.982844
QBC_Q10498 39.75 1.78 0 Natural? 26.136876 50.996083
QBC_Q10499 6.16 6.16 0 Natural? 26.124703 51.013823
QBC_Q10500 36.82 19.71 0 Depression? 26.114017 51.027796
QBC_Q10501 2.5 1.5 0 Buried Feature? 26.111836 51.030317
QBC_Q10502 8.16 3.6 0.59 Debris? 26.166536 50.957756
QBC_Q10503 9.45 3.69 0 Debris? 26.166218 50.959665
QBC_Q10504 2.81 2.1 0.65 Debris - Modern? 26.158738 50.968206
QBC_Q10505 6 4.86 0.39 Debris? 26.147941 50.983211
QBC_Q10506 22.79 12.87 0.31 Debris? 26.14809 50.983509
QBC_Q10507 17.52 2.46 0.78 Debris? 26.145884 50.987333
QBC_Q10508 47.22 2.16 0 Natural? 26.132228 51.005321
QBC_Q10509 38.78 2.62 0 Depression? 26.121144 51.020555
QBC_Q10510 8 7.81 0.85 Debris - Modern? 26.166047 50.961688
QBC_Q10511 5.02 2.36 0.97 Debris? 26.16495 50.964133
QBC_Q10512 16.62 1.71 0 Natural? 26.159574 50.972543
QBC_Q10513 64.28 26.03 0 Data problem? 26.11668 51.031117
QBC_Q10514 1.17 1.03 1.1 Linear Debris? 26.167868 50.964272
QBC_Q10515 4.13 1.79 0.34 Debris? 26.152482 50.985184
QBC_Q10516 23.66 15.91 0 Location? 26.150527 50.988769
329
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10517 11.15 7.75 0.18 Buried Feature? 26.149864 50.988512
QBC_Q10518 12.56 2.76 0 Natural? 26.148256 50.992184
QBC_Q10519 37.03 19.75 0 Natural? 26.147807 50.991207
QBC_Q10520 2.42 2.46 0.5 Debris? 26.140752 51.001409
QBC_Q10521 72.52 35.64 0 Location? 26.129018 51.019557
QBC_Q10522 9.98 8.04 0.48 Buried Feature? 26.113749 51.039725
QBC_Q10523 4.6 3 0 Debris? 26.162758 50.976398
QBC_Q10524 3.97 1.32 0 Natural? 26.155817 50.985357
QBC_Q10525 8.06 2.12 0.2 Natural? 26.152347 50.990761
QBC_Q10526 18.11 8.04 0 Natural? 26.146921 50.996365
QBC_Q10527 13.65 4.49 0.77 Debris? 26.146352 50.998449
QBC_Q10528 31.79 3.78 0 Data problem? 26.140886 51.007253
QBC_Q10529 4.39 1.97 0.25 Debris? 26.131362 51.020367
QBC_Q10530 15.03 6.97 0.45 Mound? 26.116336 51.043928
QBC_Q10531 55.84 4 0 Seabed Scar? 26.120435 51.039476
QBC_Q10532 21.15 7.12 0.31 Mound? 26.123701 51.034668
QBC_Q10533 0 0 0 Location? 26.1307 51.023848
QBC_Q10534 41.5 2.24 0 Seabed Scar? 26.132132 51.022748
QBC_Q10536 5.5 3.92 0 Natural? 26.148023 50.999444
QBC_Q10537 5.97 3.66 0 Debris? 26.159526 50.982383
QBC_Q10538 32.13 3.73 0.4 Debris? 26.161349 50.980513
QBC_Q10539 10.28 4.22 0.32 Debris? 26.162772 50.978272
QBC_Q10540 9.45 2.46 0 Debris? 26.163896 50.976284
QBC_Q10541 0 0 0 Seabed Scar? 26.16652 50.975039
QBC_Q10542 7.74 5.14 0.38 Debris? 26.166842 50.972396
QBC_Q10543 12.58 3.1 0.46 Debris? 26.171508 50.966158
330
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10544 28.58 4.57 0 Data problem? 26.174761 50.967491
QBC_Q10545 7.08 2.15 0.79 Debris? - Modern? 26.166958 50.977564
QBC_Q10546 2.65 2.51 1.03 Debris? 26.163965 50.981973
QBC_Q10547 16.97 3.77 0 Debris? 26.159758 50.98872
QBC_Q10548 6.61 2.02 0.88 Debris? - Modern? 26.170851 50.976167
QBC_Q10549 15.57 16.41 0 Natural? 26.168964 50.979454
QBC_Q10550 9.34 4.68 0.52 Debris? 26.167225 50.981809
QBC_Q10551 10.53 2.82 0.3 Debris? 26.165512 50.981963
QBC_Q10553 13.34 5.71 0 Depression? 26.1587 50.993389
QBC_Q10554 27.51 10.63 0 Natural? 26.156295 50.997212
QBC_Q10555 13.63 3.58 0.23 Natural? 26.154688 50.999428
QBC_Q10556 20.02 4.54 0.85 Debris? 26.149203 51.005646
QBC_Q10557 12.46 5.48 0.53 Natural? 26.146558 51.010588
QBC_Q10558 6.52 3.68 0 Natural? 26.14617 51.011798
QBC_Q10559 11.92 6.12 0.67 Debris? 26.145515 51.012185
QBC_Q10560 6.33 4.85 0 Natural? 26.145584 51.010885
QBC_Q10561 15.57 12.1 0.78 Natural? 26.145229 51.012155
QBC_Q10562 23.42 4.12 0 Natural? 26.135378 51.024372
QBC_Q10563 46.61 2.49 0 Location? 26.13448 51.027722
QBC_Q10564 5.63 3.31 0.42 Natural? 26.170553 50.981335
QBC_Q10565 8.23 5.33 0 Debris? 26.167984 50.985344
QBC_Q10566 6.01 1.76 0.89 Debris? 26.166546 50.985764
QBC_Q10567 114.51 59.16 0 Location? 26.160261 50.995137
QBC_Q10568 6.24 3.23 0.83 Debris? 26.159135 50.99692
QBC_Q10569 8.33 4.06 0.51 Debris? 26.147722 51.011473
QBC_Q10570 1.34 1.04 0.7 Debris? 26.178741 50.972303
331
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10571 6.01 2.63 0.63 Debris? - Modern? 26.177788 50.974457
QBC_Q10572 11.51 4.71 0.76 Buried Feature? 26.16657 50.990345
QBC_Q10573 7.1 5.57 0 Debris? 26.165421 50.990076
QBC_Q10574 10.28 8.97 0 Depressions? 26.163834 50.994139
QBC_Q10575 56.51 23.73 0 Location? 26.160125 50.998018
QBC_Q10576 13.72 7.52 0 Depression? 26.159465 50.999357
QBC_Q10577 8.7 6.93 1.73 Debris? 26.149734 51.012472
QBC_Q10578 2.55 0.75 0.31 Debris? 26.141918 51.025165
QBC_Q10579 11.53 7.56 0 Natural? 26.137986 51.028675
QBC_Q10580 30.78 11.58 0 Depression? 26.128764 51.043474
QBC_Q10581 13.38 12.62 0 Hole? 26.184715 50.967615
QBC_Q10582 0.7 0.7 0.28 Buried Feature? 26.178139 50.977658
QBC_Q10583 5.85 3.21 0.88 Buried Feature? 26.163234 50.998661
QBC_Q10584 87.5 37.91 0 Location? 26.137295 51.035618
QBC_Q10585 6.22 2.84 0.44 Buried Feature? 26.170374 50.991912
QBC_Q10586 24.98 10.32 0 Depression? 26.168026 50.996343
QBC_Q10587 23.34 10.27 0 Depression? 26.167003 50.996646
QBC_Q10588 35.91 28.56 0 Location? 26.159944 51.007962
QBC_Q10590 5.77 3.69 1.07 Debris? - Modern? 26.182407 50.979411
QBC_Q10591 25.01 19.16 0 Depression? 26.178901 50.985306
QBC_Q10592 4.78 2.57 0.58 Debris? - Modern? 26.17788 50.986311
QBC_Q10593 10.19 2.68 0 Natural? 26.165607 51.002126
QBC_Q10594 7.85 3.8 0 Natural? 26.131552 51.050247
QBC_Q10595 5.54 4.69 0.4 Buried Feature? 26.189683 50.974796
QBC_Q10596 4.98 2.99 1.16 Debris? 26.186551 50.978421
QBC_Q10597 74.85 16.43 0 Location? 26.186509 50.977432
332
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10598 5.58 1.58 0 Buried Feature? 26.184185 50.980576
QBC_Q10599 10.78 3.86 0.87 Buried Feature? 26.17903 50.988005
QBC_Q10600 6.86 1.59 0.2 Natural? 26.176783 50.990843
QBC_Q10601 1.64 1.02 0.62 Debris? 26.175218 50.993276
QBC_Q10602 9.09 5.43 0.22 Natural? 26.152577 51.026579
QBC_Q10603 12.85 9.62 0.45 Partially Buried Objects? 26.190087 50.979844
QBC_Q10604 6.61 1.6 0 Debris? 26.18898 50.981179
QBC_Q10605 3.38 3.14 1 Debris? 26.18466 50.988254
QBC_Q10607 6.08 5.84 1.01 Buried Feature? 26.183321 50.991329
QBC_Q10608 9.41 5.47 0 Natural? 26.141162 51.051057
QBC_Q10609 40.17 3.46 0 Buried Feature? 26.140807 51.050964
QBC_Q10610 102.16 9.88 0 Location? 26.13782 51.05379
QBC_Q10611 5.16 2.11 0 Natural? 26.201705 50.971514
QBC_Q10612 6.91 2.24 0.77 Partially Buried Objects? 26.20069 50.97364
QBC_Q10613 18.72 1.57 0 Natural? 26.201114 50.972691
QBC_Q10614 8 2.46 0 Natural? 26.19561 50.982688
QBC_Q10616 79.28 1.22 0 Seabed Scar? 26.190535 50.989366
QBC_Q10617 9.95 5.72 0 Buried Feature? 26.186639 50.993278
QBC_Q10618 4.22 1.05 0.39 Debris? 26.186522 50.995093
QBC_Q10619 4.24 2.58 0 Debris? 26.180739 51.00083
QBC_Q10620 8.92 3.44 0 Debris? 26.177484 51.005494
QBC_Q10621 3.72 1.62 0 Partially Buried Objects? 26.176614 51.007299
QBC_Q10622 6.26 0.61 0.45 Partially Buried Objects? 26.170585 51.016889
QBC_Q10623 6.77 1.75 0 Natural? 26.151562 51.042322
QBC_Q10624 61.39 23.74 0 Data problem? 26.174451 51.019493
QBC_Q10625 33.99 15.43 0 Natural? 26.161599 51.036097
333
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10626 3.45 1.13 0.29 Natural? 26.149057 51.05548
QBC_Q10627 4.87 1.19 0.38 Natural? 26.20637 50.975389
QBC_Q10628 79.21 2.7 0 Natural? 26.200845 50.983092
QBC_Q10629 74.16 2.33 0 Seabed Scar? 26.196759 50.988643
QBC_Q10630 1.99 0.92 0.41 Partially Buried Objects? 26.196139 50.989419
QBC_Q10631 1.27 0.63 0.4 Debris? 26.190244 50.996395
QBC_Q10632 4.52 3.91 0 Debris? 26.188869 50.998967
QBC_Q10633 82.51 13.26 0 Location? 26.186787 51.001022
QBC_Q10634 11.1 5.94 1.24 Partially Buried Objects? 26.187246 51.001958
QBC_Q10635 11.01 1.21 0 Natural? 26.184534 51.00623
QBC_Q10636 5.51 3.26 0.96 Partially Buried Objects? 26.200751 50.990649
QBC_Q10637 9.59 4.37 0.8 Debris? 26.20047 50.99143
QBC_Q10638 11.62 9.5 0 Natural? 26.199125 50.9938
QBC_Q10639 6.57 3.86 0.46 Partially Buried Objects? 26.195732 50.99836
QBC_Q10640 16.82 9.4 0 Debris? 26.196356 50.997651
QBC_Q10641 9.02 4.2 0.97 Debris? 26.189589 51.005185
QBC_Q10642 7.9 2.42 0.56 Debris? 26.185853 51.010229
QBC_Q10643 7.68 4.57 0.72 BHAM0022 - coral head 26.183558 51.015647
QBC_Q10644 15.42 4.65 2.44 Debris? 26.177719 51.023168
QBC_Q10645 7.3 2.66 0 Partially Buried Objects? 26.177284 51.024593
QBC_Q10646 6.78 4.53 0 Mound? 26.157792 51.050043
QBC_Q10647 3.09 2.55 0 Natural? 26.153292 51.055598
QBC_Q10648 16.28 7.55 0 Natural? 26.212976 50.980034
QBC_Q10649 5.97 2.49 0.57 Natural? 26.213094 50.98174
QBC_Q10650 6.39 2.66 0.94 Partially Buried Objects? - Asses 26.196144 51.005155
QBC_Q10651 33.44 7.24 0 Debris? 26.189869 51.012283
334
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10652 11.53 9.4 0 Debris? 26.18925 51.013595
QBC_Q10653 7.02 6.09 0 Natural? 26.186561 51.017296
QBC_Q10654 15.42 10.51 0 Depression? 26.183959 51.021857
QBC_Q10655 13.32 8.76 0 Depression? 26.16408 51.049672
QBC_Q10656 5.68 3.03 0 Debris? 26.161197 51.055817
QBC_Q10657 36.08 2.78 0 Unclassified 26.154432 51.064211
QBC_Q10658 6.43 4.17 0.59 Natural? 26.219383 50.979669
QBC_Q10659 9.22 3.21 0 Natural? 26.218145 50.983593
QBC_Q10660 4.14 2.26 0.79 Debris? 26.21736 50.984428
QBC_Q10661 7.19 3.27 0.84 Debris? 26.216642 50.98503
QBC_Q10662 77.74 2.2 0 Seabed Scar? 26.216053 50.985567
QBC_Q10663 2.01 2.45 0.78 Debris? 26.214963 50.986993
QBC_Q10664 7.59 3.14 0 Debris? 26.210859 50.990777
QBC_Q10665 34.55 12.87 0 Debris? 26.210029 50.994777
QBC_Q10666 5.1 1.47 0.96 Partially Buried Objects? - Asses 26.20775 50.997027
QBC_Q10667 48.7 2.84 0 Seabed Scar? 26.203179 51.00384
QBC_Q10668 10.47 7.75 0 Object in Water Column? 26.170889 51.047996
QBC_Q10669 3.77 3.31 0.27 Partially Buried Objects? 26.171655 51.04787
QBC_Q10670 88.05 2.73 0 Linear Debris? 26.200506 51.014543
QBC_Q10671 7.1 8.35 0.73 Bham0003 - 2 cars 26.222486 50.991231
QBC_Q10672 6.98 4.33 1.08 Partially Buried Objects? 26.217146 50.998854
QBC_Q10673 4.5 3.06 0.47 Partially Buried Objects? 26.217635 50.999494
QBC_Q10674 2.65 2.42 1.04 Debris? - Modern? 26.205136 51.017049
QBC_Q10675 1.81 1.19 0.76 Debris? 26.201617 51.020594
QBC_Q10676 23.03 17.08 0 Buried Feature? 26.20139 51.022168
QBC_Q10677 2.33 1.84 0.81 Debris? 26.199145 51.024446
335
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10678 5.98 2.3 0.85 Partially Buried Objects? 26.196575 51.0279
QBC_Q10679 1.66 0.67 0.45 Debris? 26.187724 51.040428
QBC_Q10680 1.8 1.07 0 Debris? 26.18188 51.049147
QBC_Q10681 4.82 1.92 0.16 Natural? 26.172046 51.062369
QBC_Q10682 1.18 0.88 0.5 Debris? 26.217385 51.007933
QBC_Q10683 0.91 0.62 0.34 Debris? 26.215904 51.009766
QBC_Q10684 0 0 0 Unclassified 26.206748 51.02219
QBC_Q10685 78.19 1.17 0 Seabed Scar? 26.203884 51.027346
QBC_Q10686 30.87 4.86 0 Natural? 26.184855 51.05447
QBC_Q10687 13.34 9.84 0 Depression? 26.175814 51.064984
QBC_Q10688 0.77 0.78 0.3 Natural? 26.211105 51.024764
QBC_Q10689 5.69 2.32 1.22 Partially Buried Objects? - Duplicate 26.207801 51.029506
QBC_Q10690 6.51 4.41 0.52 Natural? 26.204605 51.03247
QBC_Q10691 0 0 0 Unclassified 26.19526 51.046282
QBC_Q10692 7.49 5.08 0 Debris? 26.181437 51.067004
QBC_Q10693 9.18 6.63 0 Location? 26.178017 51.071213
QBC_Q10694 6.53 2.04 0.57 BHAM0001 - metal angle iron 26.210442 51.032916
QBC_Q10695 1.38 0.8 0.35 Natural? 26.21056 51.033607
QBC_Q10696 60.47 14.24 0 Location? 26.20892 51.034508
QBC_Q10697 62.84 7.46 0 Location? 26.208273 51.035456
QBC_Q10698 2.79 1.19 0.45 Natural? 26.207628 51.035773
QBC_Q10699 1.58 0.58 0.25 Natural? 26.206796 51.038223
QBC_Q10700 24.85 5.03 0 Natural? 26.204538 51.042405
QBC_Q10701 25.62 1.72 0 Location? 26.204608 51.041937
QBC_Q10702 101.59 13.77 0 Depression? 26.20077 51.045704
QBC_Q10703 49.74 2.6 0 Natural? 26.19924 51.048303
336
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10704 1.28 1.56 0.64 Debris? 26.1932 51.056634
QBC_Q10705 75.29 3.97 0 Buried Feature? 26.193103 51.057825
QBC_Q10706 0.84 0.28 0.69 Debris? 26.192831 51.05833
QBC_Q10707 13.27 1.98 0.25 Natural? 26.192357 51.059489
QBC_Q10708 1.32 0.64 0.63 Natural? 26.182613 51.072906
QBC_Q10709 1.12 0.93 0.66 Natural? 26.20862 51.043284
QBC_Q10710 2.6 1.12 0.53 Natural? 26.208763 51.044473
QBC_Q10711 1.59 0.85 0.24 Natural? 26.207304 51.046158
QBC_Q10712 2.26 0.79 0.4 Natural? 26.205436 51.048515
QBC_Q10713 1.72 1.25 0.47 Natural? 26.204865 51.048351
QBC_Q10714 0.35 0.35 0.27 Natural? 26.205088 51.048356
QBC_Q10715 1.29 0.97 0.59 Natural? 26.204389 51.050096
QBC_Q10716 10.63 8.12 0 Unclassified 26.204007 51.050221
QBC_Q10717 0 0 0.79 Debris? 26.19546 51.062176
QBC_Q10718 1.29 1.08 0.29 Natural? 26.19307 51.066298
QBC_Q10719 6.34 3.35 0 Mound? 26.19078 51.068742
QBC_Q10720 30.67 24.82 0 Location? 26.201079 51.062081
QBC_Q10721 0.63 0.38 0.48 Natural? 26.200876 51.063127
QBC_Q10722 0.88 0.38 0.28 Natural? 26.200962 51.063092
QBC_Q10723 1.38 0.75 0.39 Natural? 26.200706 51.063513
QBC_Q10724 1.13 0.92 0.21 Natural? 26.200986 51.06344
QBC_Q10725 43.15 5.26 0 Buried Feature? 26.194527 51.071913
QBC_Q10726 19.88 11.89 0 Buried Feature? 26.194214 51.071651
QBC_Q10727 29.49 2.04 0 Seabed Scar? 26.193718 51.073781
QBC_Q10728 7.94 3.98 0.21 Buried Feature? 26.192236 51.075747
QBC_Q10729 86.3 32.31 0 BHAM0028 - NOT PROPERLY LOCATED ? 26.078054 50.916605
337
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10800 0 0 0 Partially Buried Objects? - Duplicate of QBC_Q100689 26.207795 51.029454
QBC_Q10801 4.97 2.37 0 Debris? 26.198023 51.021605
QBC_Q10802 26.77 4.03 0 Debris? - Duplicate of QBC_Q10651 26.189848 51.012473
QBC_Q10803 10.41 3.4 0 Debris? - Duplicate 26.145484 50.975717
QBC_Q10804 5.63 2.63 0.68 Debris - Modern? - Duplicate of QBC_Q10207 26.143949 50.972733
QBC_Q10805 22.25 6.6 0 Debris? 26.140839 50.971974
QBC_Q10806 16.93 7.01 0 Duplicate of QBC_Q10187 26.136399 50.966768
QBC_Q10807 19.95 2.69 0 Debris? 26.117499 50.951097
QBC_Q10809 10.5 6.32 0 Debris? 26.102439 50.938059
QBC_Q10810 18.23 3.56 0 26.102819 50.936521
QBC_Q10812 21.17 14.72 1.05 Partially Buried Objects? 26.100444 50.934951
QBC_Q10813 5.82 2.29 0.51 Partially Buried Objects? 26.089422 50.925223
QBC_Q10814 8.24 3.33 0.9 Debris - Modern? - duplicate of QBC_Q10308 26.087427 50.924932
QBC_Q10815 26.34 14.15 0.91 Debris - Modern? 26.085005 50.92127
QBC_Q10816 0 0 0 Disturbed seabed ? 26.073977 50.913471
QBC_Q10817 57.5 9.44 0 26.074453 50.91239
QBC_Q10818 12.8 5.17 0 Unclassified 26.073308 50.910517
QBC_Q10819 15.65 6.53 0 Unclassified 26.07261 50.909795
QBC_Q10820 30.8 16.32 0 Depression? 26.066651 50.905448
QBC_Q10821 0 0 0 Duplicate of QBC_Q10729 26.077841 50.916601
QBC_Q10822 11.24 6.24 0.96 Partially Buried Objects? 26.084648 50.921714
QBC_Q10824 7.42 2.78 0.85 Duplicate of QBC_Q1807 26.117655 50.951166
QBC_Q10825 0 0 0 Debris - Modern? Duplicate of QBC_Q10806 and QBC_Q10187 26.136519 50.966809
QBC_Q10826 0 0 0 Debris - Modern? Duplicate of QBC_Q10804 26.143991 50.972805
QBC_Q10827 0 0 0 Duplicate of QBC_Q10803 26.145584 50.975927
QBC_Q10829 6.76 2.48 0.7 Debris - Modern? 26.148292 50.977067
338
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10830 19.74 4.38 0 Depression? 26.169129 50.996058
QBC_Q10831 0 0 0 Duplicate of QBC_Q10651 26.189858 51.012523
QBC_Q10832 6.52 5.22 0 Debris? 26.19204 51.0143
QBC_Q10833 0 0 0 Duplicate of QBC_Q100689 26.207859 51.029534
QBC_Q10834 0 0 0 Duplicate of QBC_Q10816 26.07404 50.913514
QBC_Q10835 0 0 0 Duplicate of QBC_Q10815 26.08498 50.921268
QBC_Q10836 0 0 0 duplicate of QBC_Q10308 26.087481 50.925022
QBC_Q10837 0 0 0 Duplicate of QBC_Q10813 26.089259 50.925087
QBC_Q10838 0 0 0 Duplicate of QBC_Q10812 26.100483 50.934926
QBC_Q10839 0 0 0 Duplicate of QBC_Q10107 26.111848 50.945664
QBC_Q10840 3.89 2.8 0.86 Debris - Modern? 26.064826 50.936662
QBC_Q10841 2.47 1.05 0.38 Partially Buried Objects? 26.068584 50.940235
QBC_Q10842 0 0 0 Duplicate of QBC_Q10298 26.069907 50.940324
QBC_Q10843 6.83 2.52 0.43 Debris - Modern? 26.088594 50.956422
QBC_Q10844 1.01 0.6 0.45 Debris? 26.089272 50.95716
QBC_Q10845 6.21 2.36 1.01 Debris - Modern? 26.090141 50.958149
QBC_Q10846 1.43 0.41 0.54 Debris? 26.090735 50.959415
QBC_Q10847 0 0 0 Duplicate of QBC_Q10110 26.097505 50.96587
QBC_Q10848 3.63 1.17 0.53 Debris? 26.181361 51.037826
QBC_Q10849 5.33 1.66 0.56 Debris? 26.146677 51.00794
QBC_Q10850 0 0 0 Duplicate of QBC_Q10203 26.130408 50.993318
QBC_Q10851 0 0 0 Duplicate of QBC_Q10023 26.115736 50.979872
QBC_Q10852 0 0 0 Duplicate of QBC_Q10168 26.112219 50.976101
QBC_Q10853 0 0 0 Duplicate of QBC_Q10120 26.102052 50.968747
QBC_Q10854 10.85 1.42 0.76 Duplicate of QBC_Q10840 26.064851 50.936623
QBC_Q10855 4.41 2.3 0.76 Duplicate of QBC_Q10298 26.069825 50.940309
339
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q10856 11.42 3.45 0.68 Duplicate of QBC_Q10843 26.088514 50.956385
QBC_Q10857 4.73 2.02 0 Duplicate of QBC_Q10221 26.089954 50.957033
QBC_Q10858 9.3 5.48 0.69 Duplicate of QBC_Q10845 26.09017 50.958092
QBC_Q10859 0.37 0.18 0 Debris? 26.090517 50.958006
QBC_Q10860 3.81 1.44 0.41 Duplicate of QBC_Q10846 26.0908 50.959455
QBC_Q10861 39.99 1.31 0 Linear Debris? 26.099693 50.966635
QBC_Q10862 4.42 2.36 0.55 Debris - Modern? 26.102055 50.968746
QBC_Q10863 4.93 3.01 0.49 Duplicate of QBC_Q10023 26.115733 50.979878
QBC_Q10864 0 0 0 Duplicate of QBC_Q10203 26.130369 50.993332
QBC_Q10865 0 0 0 26.097435 50.964845
QBC_Q10866 0 0 0 Duplicate of QBC_Q10849 26.146739 51.007851
QBC_Q10867 0 0 0 Duplicate of QBC_Q10848 26.181363 51.037869
QBC_Q10869 6.04 5.96 1.28 Natural? 26.049892 50.892192
QBC_Q10901 6.6 3.55 0.72 Debris - Modern? 26.089642 50.942611
QBC_Q10902 5.96 3.54 1.05 Debris - Modern? 26.089445 50.941131
QBC_Q10903 0 0 0 Duplicate of QBC_Q10344 26.087449 50.940918
QBC_Q10904 7.42 4.05 0 Debris? 26.08548 50.939336
QBC_Q10905 0 0 0 Duplicate of QBC_Q10327 26.083135 50.936817
QBC_Q10906 27.85 15.69 0 Disturbed seabed ? 26.077989 50.931212
QBC_Q10907 8.85 5.52 0.48 Debris? 26.072828 50.928273
QBC_Q10908 0 0 0 Duplicate of QBC_Q10231 26.060548 50.917317
QBC_Q10909 3.98 1.87 0.98 Debris? 26.059662 50.916193
340
Area 2 and 3
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q20001 17.45 7.88 0.35 Natural? 26.034871 50.969744
QBC_Q20002 3.78 2.09 0.26 Natural? 26.022824 50.989369
QBC_Q20003 28.81 14.45 0 Location? 26.015682 51.00121
QBC_Q20004 11.15 4.29 0 Natural? 26.029393 50.975943
QBC_Q20005 18.62 1.64 0 Debris? 26.021609 50.985721
QBC_Q20006 5.74 1.49 0.38 Natural? 26.037301 50.961285
QBC_Q20007 5.41 2.96 1.32 Debris? 26.043259 50.949102
QBC_Q20008 32.26 5.33 0 Location? 26.039622 50.953929
QBC_Q20009 7.58 4.17 0 Natural? 26.031682 50.966108
QBC_Q20010 15.52 5.42 0.59 Location? 26.023692 50.980616
QBC_Q20011 30.22 20.9 0 Depression? 26.021348 50.985627
QBC_Q20012 3.79 2.23 0.61 Debris? 26.039721 50.949273
QBC_Q20013 10.91 6.17 0 Hole? 26.028365 50.969804
QBC_Q20014 10.83 3.15 0.74 Natural? 26.038261 50.95344
QBC_Q20015 89.29 1.93 0 Linear Debris? 26.035609 50.953278
QBC_Q20016 3.41 1.81 0.62 Natural? 26.028178 50.966301
QBC_Q20017 0.79 19.91 0 Location? 26.018519 50.982515
QBC_Q20018 16.84 2.11 0.53 Natural? 26.012999 50.991011
QBC_Q20019 3.25 0.96 0.32 Natural? 26.037128 50.946973
QBC_Q20020 47.18 7.05 0 Natural? 26.031579 50.957676
QBC_Q20021 2.87 1.73 0.64 Debris? 26.03056 50.958093
QBC_Q20022 47.36 4.22 0.37 Unclassified 26.01596 50.983347
QBC_Q20023 13.55 3.34 0.88 Natural? 26.014797 50.983851
QBC_Q20024 49.06 33.56 0 Location? 26.047384 50.92857
341
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q20025 13.9 2.44 0 Debris? 26.040639 50.937346
QBC_Q20027 6.8 4.71 0 Natural? 26.032283 50.950897
QBC_Q20028 35.31 1.64 0 Natural? 26.012014 50.98432
QBC_Q20029 27.21 5.3 0 Unclassified 26.008609 50.989855
QBC_Q20030 0 0 0 Unclassified 25.999482 51.005706
QBC_Q20031 27.53 7.47 0 Depression? 25.997185 51.006401
QBC_Q20032 30.63 22.52 0 Depression? 25.997672 51.004862
QBC_Q20033 16.23 11.15 0 Hole? 26.025869 50.960361
QBC_Q20034 96.04 52.54 0 Location? 26.026234 50.957138
QBC_Q20035 8.44 1.56 0.3 Natural? 26.030321 50.94946
QBC_Q20036 12.37 2.67 0.61 Debris? 26.023908 50.95936
QBC_Q20037 41.82 4.7 0 Seabed scar? 26.008555 50.983086
QBC_Q20038 18.73 0.9 0 Natural? 26.015007 50.970753
QBC_Q20039 74.1 41.85 0 natural 26.018131 50.963395
QBC_Q20040 52.45 34.27 0 natural 26.019713 50.962363
QBC_Q20041 8.53 5.19 1.3 Debris? 26.032559 50.938596
QBC_Q20043 37.41 15.23 0.58 location 26.017383 50.963429
QBC_Q20044 12.12 1.04 0 hole? 26.016424 50.962491
QBC_Q20045 16.75 2.18 0 Debris? 26.015616 50.966629
QBC_Q20046 65.8 20.3 0 location 26.013541 50.968965
QBC_Q20047 25.21 4.29 0 Seabed scar? 26.012459 50.969711
QBC_Q20048 113.87 53.67 0 location 26.011772 50.967575
QBC_Q20049 14.69 3.27 0.34 Natural? 26.012287 50.968259
QBC_Q20050 17.63 4.01 0.49 Natural? 26.024893 50.943524
QBC_Q20051 13.86 8.25 0 Hole? 26.023127 50.945239
QBC_Q20052 96.86 71.72 0 Natural? 26.011805 50.964874
342
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q20053 33.67 12.93 0 Depression? 26.005168 50.970761
QBC_Q20054 23.18 3.95 1.56 Natural? 26.006324 50.970422
QBC_Q20055 12.64 4.84 0.74 Natural? 26.00827 50.967464
QBC_Q20056 9.27 5.01 0.61 Natural? 26.008231 50.96518
QBC_Q20057 29.83 4.27 0 Natural? 26.008954 50.963154
QBC_Q20058 18.91 5.53 0 Depression? 26.003936 50.968802
QBC_Q20059 3.18 1.41 0.73 Debris? 25.993794 50.985838
QBC_Q20060 61.18 37.13 0 Mound? 26.001472 50.969192
QBC_Q20061 15.49 7.51 0 Hole? 26.007525 50.961633
QBC_Q20062 48.69 21.63 0 Sediment accumulation? 26.000014 50.970078
QBC_Q20063 8.69 7.28 2.16 Debris? 25.998462 50.971358
QBC_Q20064 5.64 3.33 1.31 Debris? 25.987718 50.986676
QBC_Q20065 40.74 12.06 0 Depression? 25.98895 50.983912
QBC_Q20066 8.35 3.79 1.35 Natural? 25.997685 50.969099
QBC_Q20067 31.05 6.36 1.03 Natural? 25.998971 50.968183
QBC_Q20068 19.12 4.65 0.59 Debris? 26.000052 50.964064
QBC_Q20069 32.09 3.03 0 Natural? 26.003802 50.960695
QBC_Q20070 2.73 2.49 0.64 Debris? 26.010633 50.949374
QBC_Q20071 38.32 12.2 0 Depression? 26.034977 50.909748
QBC_Q20072 25.21 2.98 0 Natural? 26.036274 50.907219
QBC_Q20073 11.59 3.98 0.31 Debris? 25.99973 50.963871
QBC_Q20074 4.68 2.29 1.19 Debris? 25.985496 50.986134
QBC_Q20075 0 0 0 Debris? 25.985699 50.986791
QBC_Q20076 11.24 4.75 0 Debris? 25.98548 50.987437
QBC_Q20077 8.89 2.96 1.24 Debris? 25.990019 50.974952
QBC_Q20078 8.85 2.56 0.37 Debris? 25.998364 50.959999
343
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q20079 10.35 3.01 0.58 Debris? 26.000616 50.957843
QBC_Q20080 25.93 31.87 1.21 Buried features? 26.001858 50.955288
QBC_Q20081 5.82 1.45 0.34 Debris? 26.026197 50.916223
QBC_Q20082 7.22 3.47 1.5 Debris? 26.032657 50.906264
QBC_Q20083 57.88 9.44 0 Depression? 26.009534 50.938401
QBC_Q20084 11.87 2.69 0.55 Debris? 26.007582 50.941897
QBC_Q20085 12.36 6.1 0.29 Debris? 26.008765 50.942045
QBC_Q20086 82.15 25.16 0 Depression? 25.999753 50.956708
QBC_Q20087 10.92 9 0.65 Debris? 25.985574 50.977561
QBC_Q20088 24.61 12.44 0 Depression ? 26.002045 50.94879
QBC_Q20089 3.01 2.26 1.16 Buried features? 26.005229 50.943802
QBC_Q20090 21.7 6.03 0 hole? 26.026321 50.903746
QBC_Q20091 20.52 7.04 0 Depression? 26.025408 50.9074
QBC_Q20092 29.71 1.61 0 Hole? 26.005373 50.940511
QBC_Q20093 43.41 3.94 0.5 Linear Debris? 26.002905 50.943807
QBC_Q20094 7.71 4.32 0 Hole? 25.982569 50.975877
QBC_Q20096 14.9 12.33 1.22 Debris? 25.981652 50.973519
QBC_Q20097 4.06 2.65 0.66 Debris? 25.974627 50.978978
QBC_Q20098 31.37 17.29 0 Depression? 25.992853 50.949522
QBC_Q20099 19.87 3.64 0 Natural? 26.01989 50.904247
QBC_Q20100 68.85 2.53 0 Natural? 26.010368 50.913935
QBC_Q20101 31.91 3.37 0 Location? 26.007956 50.912479
QBC_Q20102 17.88 5.05 0.56 Debris? 25.993403 50.932861
QBC_Q20103 6.38 3.6 1.1 Debris? 26.003767 50.913513
QBC_Q20104 24.32 1.36 0 Seabed scar? 26.003197 50.913515
QBC_Q30105 11.35 4.4 0 Debris? 25.977768 50.960233
344
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q30106 11.83 10.31 0.87 Debris? 25.977577 50.960399
QBC_Q30107 29.3 4.84 0 Unclassified 25.995671 50.913725
QBC_Q30108 3.88 1.51 1.15 Debris? 25.958976 50.986071
QBC_Q30109 10.7 6.9 0 Natural? 25.986144 50.934308
QBC_Q30110 55.23 4.81 0.98 Natural? 25.975427 50.955838
QBC_Q30111 34.99 3.32 0 Seabed scar? 25.988777 50.923005
QBC_Q30112 8.24 3.96 0.46 Debris? 25.992048 50.91507
QBC_Q30113 47.15 3.1 0 Linear Debris? 25.993095 50.911622
QBC_Q30114 15.49 4.05 0.46 Natural? 25.998981 50.897624
QBC_Q30115 20.51 8.26 0 Hole? 25.988612 50.917231
QBC_Q30116 33.68 12.91 0 Depression? 25.987858 50.922473
QBC_Q30117 46.89 23.3 0 Depression? 25.964744 50.963022
QBC_Q30118 23.42 9.08 0 Depression? 25.992698 50.904102
QBC_Q30120 19.72 2.98 0 Linear Debris? 25.992489 50.902197
QBC_Q30121 103.01 3 0 Natural? 25.992475 50.895784
QBC_Q30122 5.1 3.82 0 Hole? 25.990982 50.902169
QBC_Q30123 88.19 7.71 0 Unclassified 25.987257 50.915365
QBC_Q30124 96.05 4.07 0 Unclassified 25.986066 50.917011
QBC_Q30125 11.56 3.37 0 Debris? 25.977564 50.93689
QBC_Q30126 33.12 3.09 0 Seabed scar? 25.977464 50.937645
QBC_Q30127 17.39 3.98 0 Depression? 25.960588 50.966248
QBC_Q30128 93.87 4.71 0 Natural? 25.99261 50.890159
QBC_Q30129 98.22 2.98 0 Seabed scar? 25.979662 50.927369
QBC_Q30130 12.46 4.36 0.46 Debris? 25.975852 50.936445
QBC_Q30131 34.51 1.81 0 Natural? 25.960882 50.96314
QBC_Q30132 33.38 8.95 0 Depression? 25.957793 50.96692
345
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q30133 63.54 2.97 0 Seabed scar? 25.978715 50.926276
QBC_Q30134 5.47 1.47 0.22 Debris? 25.986519 50.892238
QBC_Q30135 8.73 2.17 0.57 Debris? 25.962336 50.950633
QBC_Q30136 96.79 1.7 0 Natural? 25.985557 50.893222
QBC_Q30137 20.36 3.14 0 Depression? 25.948145 50.969633
QBC_Q30138 37.3 7.62 0 Depression? 25.969652 50.930996
QBC_Q30139 80.15 26.76 0 Location? 25.955473 50.956545
QBC_Q30140 34.58 27.58 0 Natural? 25.975749 50.91299
QBC_Q30141 95.65 6.02 0.86 Natural? 25.975314 50.916018
QBC_Q30142 6.77 3.28 0.4 Debris? 25.990113 50.866632
QBC_Q30143 3.48 3.01 0 Debris? 25.990316 50.866171
QBC_Q30144 14.79 3.79 0 Unclassified 25.989013 50.86602
QBC_Q30145 36.85 8.08 0 Natural? 25.991626 50.867082
QBC_Q30146 24.46 1.22 0 Linear Debris? 25.991857 50.872043
QBC_Q30147 31.23 1.05 0 Linear Debris? 25.988126 50.864203
QBC_Q30148 30.6 3.16 0 Natural? 25.991997 50.86839
QBC_Q30149 6.09 3.72 0.6 Debris? 25.990818 50.858539
QBC_Q30150 5.3 3.32 1.37 Debris? 25.990016 50.860044
QBC_Q30151 7.02 4.33 0 Natural? 25.995585 50.863937
QBC_Q30152 9.04 3.89 0 Unclassified 25.991102 50.856636
QBC_Q30153 5.16 5 0 Unclassified 25.989567 50.856227
QBC_Q30154 9.38 2.91 0 Unclassified 25.990474 50.856215
QBC_Q30155 7.88 2.85 0 Unclassified 25.990071 50.857458
QBC_Q30156 6.12 4.87 1.1 Unclassified 25.988864 50.856032
QBC_Q30157 5.05 2.64 0 Unclassified 25.99027 50.855844
QBC_Q30158 10.77 3.02 0 Natural? 25.993669 50.856892
346
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q30159 18.33 5.78 1.16 Debris? 25.997508 50.860406
QBC_Q30160 11.21 1.55 0 Linear Debris? 25.992489 50.850839
QBC_Q30161 20.77 10.97 0.93 Debris? 25.993421 50.851087
QBC_Q30162 4.4 3.46 1.67 Debris? 26.026144 50.891788
QBC_Q30163 5.67 2.27 2.22 Debris? 26.025856 50.89145
QBC_Q30164 6.33 3.85 0.59 Debris? 26.025013 50.891097
QBC_Q30165 5.21 5.2 1.13 Debris? 26.025462 50.890914
QBC_Q30166 5.18 3.5 0.88 Debris? 26.025673 50.890524
QBC_Q30167 4.16 3.35 0.71 Debris? 26.025872 50.890714
QBC_Q30168 6 5.38 0.88 Debris? 26.026389 50.891085
QBC_Q30169 3.24 1.36 0.37 Debris? 26.025976 50.891257
QBC_Q30170 6.81 2.48 0.53 Debris? 26.028128 50.89273
QBC_Q30171 5.31 4.04 0.47 Debris? 26.030331 50.894548
QBC_Q30172 5.17 2.79 0.91 Debris? 26.032058 50.895138
QBC_Q30174 38.24 3.32 0.44 Buried features? 26.033817 50.894891
QBC_Q30175 5.59 2.57 0.55 Debris? 26.033991 50.894954
QBC_Q30176 33.97 3.82 0.21 Buried features? 26.034035 50.895316
QBC_Q30177 47.63 1.4 0.49 Linear Debris? 26.034817 50.895515
QBC_Q30178 42.05 10.3 0 Depression? 26.045648 50.897662
QBC_Q30179 7.8 4.83 2.18 Debris? 26.044424 50.895195
QBC_Q30180 6.12 1.43 1.02 Debris? 26.032736 50.886064
QBC_Q30181 8.29 7.15 1.49 Debris? 26.044475 50.895084
QBC_Q30182 7.69 2.49 0 Debris? 26.04388 50.894803
QBC_Q30183 45.21 40.42 0 Unclassified 25.992082 50.845904
QBC_Q30184 41.06 19.18 0 Unclassified 25.992179 50.845926
QBC_Q30185 8.07 7.23 0 Unclassified 25.993254 50.843139
347
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q30186 5.97 2.61 0 Natural? 25.996373 50.843966
QBC_Q30187 28.63 2.77 0 Linear Debris? 25.992991 50.841089
QBC_Q30188 23.14 7.49 2.36 Natural? 26.053965 50.884765
QBC_Q30189 7.86 3.06 1.89 Debris? 26.060268 50.889165
Area 4
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q40001 9.17 6.3 0.32 Buried feature? 25.932107 50.861681
QBC_Q40002 94.47 1.96 0 Seabed scar? 25.924947 50.858498
QBC_Q40003 8.64 6.75 0 Natural? 25.926054 50.858424
QBC_Q40004 99.51 2.43 0 Seabed scar? 25.931503 50.861094
QBC_Q40005 33.62 2.36 0 Seabed scar? 25.938567 50.862092
QBC_Q40006 22.84 5.17 0 Disturbed Seabed 25.940958 50.862832
QBC_Q40008 13.98 5.53 0.3 Natural? 25.986061 50.873693
QBC_Q40009 66.13 2.08 0 Seabed scar? 25.987743 50.869673
QBC_Q40010 3.58 1.53 1.53 Debris? 25.988274 50.8648
QBC_Q40011 2.66 1.92 0.64 Debris? 25.987836 50.86551
QBC_Q40012 4.28 1.54 0.22 Natural? 25.989284 50.85452
QBC_Q40013 23.67 1.01 0 Natural? 25.991374 50.839758
QBC_Q40014 2.77 1.67 0.91 Buried feature? 25.990097 50.837026
QBC_Q40015 7.69 2.02 0 Natural? 25.990882 50.840948
QBC_Q40016 8.94 3.36 0 Natural? 25.987067 50.857546
QBC_Q40017 71.48 1.94 0.29 Linear Debris? 25.980771 50.88926
QBC_Q40018 74.73 3.13 0 Seabed scar? 25.986048 50.849994
348
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q40019 61.72 1.26 0 Seabed scar? 25.987655 50.846209
QBC_Q40020 70.44 0.55 0 Unclassified 25.987127 50.842702
QBC_Q40021 31.21 1.23 0 Seabed scar? 25.988522 50.841997
QBC_Q40022 3.95 1.99 0 Natural? 25.987938 50.838971
QBC_Q40023 49.01 26.34 0 Location? 25.977381 50.884258
QBC_Q40024 97.43 0 0 Location? 25.977262 50.88167
QBC_Q40025 4.47 3.12 0 Unclassified 25.981833 50.854398
QBC_Q40026 104.92 0 0 Location? 25.984453 50.837905
QBC_Q40027 99.93 0 0 Location? 25.981649 50.85249
QBC_Q40028 67.49 2.02 0 Seabed scar? 25.977321 50.865097
QBC_Q40029 6.26 4.05 0 Natural? 25.978028 50.851914
QBC_Q40030 4.76 3.8 0 Natural? 25.977766 50.852149
QBC_Q40031 77.92 7.31 0.84 Location? 25.971329 50.88125
QBC_Q40032 9.23 5.44 1.07 Natural? 25.973929 50.863919
QBC_Q40033 0 0.74 0 Linear Debris? 25.97656 50.847923
QBC_Q40034 8.29 5.27 1.2 Debris - modern? 25.972184 50.860145
QBC_Q40035 0 0.76 0 Linear Debris? 25.973636 50.83635
QBC_Q40036 9.03 3.45 0 Unclassified 25.972746 50.838404
QBC_Q40037 7.31 2.76 0 Unclassified 25.969904 50.848385
QBC_Q40038 126.85 17.28 0 Depression? 25.964237 50.887264
QBC_Q40039 48.53 12.33 0 Depression? 25.962862 50.887205
QBC_Q40040 0 0 0 Unclassified 25.961924 50.887342
QBC_Q40041 7.92 3.96 0 Natural? 25.963152 50.880036
QBC_Q40042 6.94 4.41 0 Unclassified 25.964352 50.875995
QBC_Q40044 92.28 0.73 0 Seabed scar? 25.966476 50.85934
QBC_Q40045 4.48 1.59 0 Debris? 25.969016 50.839139
349
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q40046 7.15 1.12 0.1 Natural? 25.967363 50.850407
QBC_Q40047 4.65 2.85 0 Unclassified 25.965044 50.857902
QBC_Q40048 32.07 1.21 0 Seabed scar? 25.960408 50.886675
QBC_Q40049 49.03 12.43 0 Natural? 25.960118 50.886517
QBC_Q40050 19.89 16.94 0 Depression? 25.961303 50.867656
QBC_Q40051 4 2 0 Unclassified 25.966443 50.843104
QBC_Q40052 23.52 6.94 0 Depression? 25.960415 50.868114
QBC_Q40053 32.63 13.84 0 Depression? 25.960072 50.871546
QBC_Q40054 8.62 5.25 0 Hole? 25.957434 50.875804
QBC_Q40055 40.62 19.22 0 Sediment accumulation? 25.957461 50.870748
QBC_Q40056 44.65 13.59 0 Depression? - Duplicate 25.958747 50.869402
QBC_Q40057 45.78 15.97 0 Depression? - Duplicate 25.957867 50.86819
QBC_Q40058 5.99 3.25 0.32 Debris? 25.958019 50.866742
QBC_Q40059 97.61 20.33 0 Depression? 25.95889 50.862487
QBC_Q40060 4.52 3.19 0.85 Debris? 25.964182 50.834745
QBC_Q40061 11.63 3.42 0 Unclassified 25.963187 50.832818
QBC_Q40062 30.29 1.39 0 Seabed scar? 25.963085 50.829337
QBC_Q40063 7.75 2.89 0 Unclassified 25.96245 50.832142
QBC_Q40064 47.34 13.77 0 Depression? 25.956369 50.866906
QBC_Q40065 32.36 10.82 0 Depression? 25.956751 50.869363
QBC_Q40066 20.95 12.78 0 Sediment accumulation? 25.955581 50.876598
QBC_Q40067 64.96 1.67 0 Seabed scar? 25.953895 50.882226
QBC_Q40068 30.47 2.92 0 Hole? 25.955256 50.867615
QBC_Q40069 22.05 6.03 0 Hole? 25.955473 50.867196
QBC_Q40070 4.2 2.42 0 Unclassified 25.95574 50.861931
QBC_Q40071 79.69 20.53 0 Buried features? 25.956161 50.86105
350
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q40072 9.8 6.67 0.42 Unclassified 25.957421 50.835568
QBC_Q40073 11.25 8.02 0 Debris? 25.950385 50.878201
QBC_Q40074 19.91 15.89 0 Sediment accumulation? 25.951354 50.879572
QBC_Q40075 23.81 6.86 0 Natural? 25.951534 50.867655
QBC_Q40076 19.47 3.33 0 Natural? 25.951729 50.86673
QBC_Q40077 78.28 1.86 0 Seabed scar? 25.953003 50.848071
QBC_Q40078 50.42 3.32 0 Seabed scar? 25.948526 50.866514
QBC_Q40079 11.63 12.86 0 Unclassified 25.948783 50.87525
QBC_Q40080 0 0 0 Location? 25.947379 50.878228
QBC_Q40081 11.1 7.62 0 Unclassified 25.945731 50.874101
QBC_Q40082 98.21 1.23 0 Seabed scar? 25.950966 50.840941
QBC_Q40083 10.14 5.53 0 Buried features? 25.949666 50.837692
QBC_Q40084 9.75 5.07 0 Debris? 25.941409 50.874538
QBC_Q40085 71.09 1.34 0 Seabed scar? 25.94 50.863017
QBC_Q40086 8.37 4.83 0 Debris? 25.944913 50.841861
QBC_Q40087 6.43 4.66 0 Debris? 25.945882 50.825839
QBC_Q40088 7.21 2.61 0 Debris? 25.937896 50.87336
QBC_Q40089 44.6 6.21 0 Debris? 25.942259 50.82708
QBC_Q40090 37.83 3.31 0 Seabed scar? 25.94043 50.827707
QBC_Q40091 10.14 9.68 1.35 Debris? 25.93982 50.830971
QBC_Q40092 6.63 2.66 1.37 Debris? 25.939852 50.832139
QBC_Q40093 46.63 1.72 0 Seabed scar? 25.936133 50.861157
QBC_Q40094 7.19 3.42 0 Natural? 25.932808 50.874257
QBC_Q40095 5.18 2.73 0 Debris? 25.939733 50.82807
QBC_Q40096 5.71 3.23 0 Debris? 25.94 50.826521
QBC_Q40097 29.24 8.5 0 Natural? 25.939984 50.825925
351
Contact No Length(ft) Width(ft) Height(ft) Preliminary Interpretation Lat(WGS84) Lon(WGS84)
QBC_Q40098 88.18 2.99 0 Unclassified 25.938848 50.826256
QBC_Q40099 38.71 24.66 0 Unclassified 25.936521 50.828418
QBC_Q40100 14.26 7.67 0 Debris? 25.935923 50.831528
QBC_Q40101 92.46 1.46 0 Seabed scar? 25.935402 50.836563
QBC_Q40102 104.76 17.6 0 Depression? 25.928998 50.864938
QBC_Q40103 116.89 0.71 0 Seabed scar? 25.93095 50.85438
QBC_Q40104 8.05 4.27 0.9 Natural? 25.934183 50.83233
QBC_Q40105 16.52 5.05 1.35 Natural? 25.934876 50.827527
QBC_Q40106 5.72 4.98 0.38 Natural? 25.932611 50.829857
QBC_Q40107 6.86 2.26 0.75 Natural? 25.933705 50.830691
QBC_Q40108 26.86 3.4 0 Natural? 25.932396 50.831335
QBC_Q40109 10.55 3.43 0.45 Natural? 25.932459 50.832626
QBC_Q40110 62.63 2.12 0 Seabed scar? 25.924592 50.869542
QBC_Q40111 107.23 1.55 0 Seabed scar? 25.925479 50.869238
QBC_Q40112 14.75 10.57 1.02 Natural? 25.930967 50.827903
QBC_Q40113 99.32 26.23 0 Unclassified 25.93258 50.826912
QBC_Q40114 6.68 3.82 0.21 Natural? 25.98773 50.857431
QBC_Q40116 12.88 7.83 0 Unclassified 25.963887 50.850258
QBC_Q40117 40.44 8.35 0 Sediment accumulation? 25.952159 50.846663
QBC_Q40118 59.29 15.25 0 Depression? - Duplicate 25.957907 50.868155
QBC_Q40119 44.47 14.35 0 Depression? - Duplicate 25.958824 50.869361
QBC_Q40120 3.8 1.79 0.26 Natural? 25.967127 50.871502
QBC_Q40121 28.05 4.68 0.32 Natural? 25.985138 50.875706
352
APPENDIX 6: DERIVED DATA INCLUDED ON ACCOMPANYING CD
The data created during the research is included on an accompanying data CD. The data
included is purely derived data, and no original survey data has been included due to data
ownership/copyright reasons. The exception to this is the grab sample locations and
granulometry report, which was undertaken specifically for this research.
The data can either be opened as a map package or a map document in ArcGIS, or the data
files can be imported into other compatible GIS packages.
ArcGIS Map Package
DataCD.mpk
ArcGIS Project
DataCD.mxd
Shapefiles (In folder called DataForCD)
(NB layer files (.lyr) for each shapefile/grid are also included)
Background Mapping:
Coast_UTM39N.shp
SurveyArea.shp
Qatar_UbaidSites.shp
Acoustic Classification
Classes (Raster grid File)
Topographic Mapping
TopographicFeatures.shp
Ground Truthing
GrabSamples.shp (shapefile contains hyperlinks to granulometry report:
../hyperlinks/GranulometryReport.pdf)
Geophysical Anomalies:
Area1AllContacts.shp
Area3AllContacts.shp
Area4AllContacts.shp
Area1Debris.shp
Area3Debris.shp
Area4Debris.shp
Area1Depressions.shp
Area3Depressions.shp
Area4Depressions.shp
Area1GroundTruthed.shp
353
Area1LongLinearFeature.shp
Seabed Characterisation
InitialLandscapeUnits.shp
FinalCharacterAreas.shp (by acoustic class)
FinalCharacterAreas.shp (by overall potential)
354
BIBLIOGRAPHY
Abdel Monim-Mubarak, W. and Kubryakov, A.I. 2000 Hydrological Structure of
Waters of the Persian Gulf According to the Data of Observations in 1992. Physical
Oceanography, 11 (5): 459-471.
Abu-Zeid, M.M., Abd El Hameed, A.E.T., Al Kuwari , A.J. (2001) Diagenesis in the
Coastal Quaternary Carbonates in Qatar, Arabian Gulf. Carbonates and
Evaporites, 16 (1): 26-36.
Al-Ghadban, A.N., Abdali, F. and Massoud, M.S. (1998) Sedimentation Rate and
Bioturbation in the Arabian Gulf. Environment International, 24 (1/2): 23-31.
Al-Hinai, K.G., Moore, J.M. and Bush, P.R. (1987) LANDSAT Image Enhancement
Study of Possible Submerged Sand-Dunes in the Arabian Gulf. International
Journal of Remote Sensing, 8 (2): 251-258.
Al-Naimi, F., Cuttler, R., Alhaidous, I.I., Dingwall, L., Momber, G., Al-Naimi, S.,
Breeze, P. and Al-Kawari, A.A. (2012) Landscape Signatures and Seabed
Characterisation in the Marine Environment of North-west Qatar. Proceedings of the
Seminar for Arabian Studies, 42: 245-260.
Al-Saad, H. (2005) Lithostratigraphy of the Middle Eocene Dammam Formation in
Qatar, Arabian Gulf: Effects of Sea-Level Fluctuations Along a Tidal Environment.
Journal of Asian Earth Sciences, 25: 781–789.
Aldred, O. and Fairclough, G. (2003) Historic Landscape Characterisation: Taking
Stock of the Method. London: English Heritage/Somerset County Council.
Alfred Wegener Institute (2012) Case Study “Eckernfoerde Bay” – The SubGate
Project [online]. Available from :
www.awi.de/en/research/research_divisions/geosciences/marine_geochemistry/sub
marine_groundwater_discharge_in_the_coastal_zone/eckernfoerde_bay_the_subgat
e_project/ [Accessed 31 August 2013].
Alsharhan, A.S. and Kendall, C.G.St.C. (2003) Holocene Coastal Carbonates and
Evaporites of the Southern Arabian Gulf and their Ancient Analogues. Earth-Science
Reviews, 61: 191–243
Anderson, J.T., Holliday, D.V., Kloser, R., Reid, D.G., and Simard, Y. (2008) Acoustic
Seabed Classification: Current Practice and Future Directions. ICES Journal of
Marine Science, 65: 1004–1011.
Bailey, G., Al Sharekh, A., Flemming, N., Lambeck, K., Momber, G., Sinclair, A. and
Vita-Finzi, C. (2007) Coastal Prehistory in the Southern Red Sea Basin, Underwater
Archaeology and the Farasan Islands. Proceedings of the Seminar for Arabian
Studies, 37: 1-16.
355
Bailey, G.N. (2009) “The Red Sea, Coastal Landscapes, and Hominin Dispersals.” In
Petraglia, M.D. and Rose, J.I. (eds.) The Evolution of Human Populations in
Arabia: Paleoenvironments, Prehistory and Genetics. Dordrecht: Springer. pp.15-
37.
Bailey, G.N. (2011) “Continental Shelf Archaeology: Where Next ?” In Benjamin, J.,
Bonsall, C., Pickard, C. and Fischer, A. (eds.) Submerged Prehistory. Oxford:
Oxbow Books.
Beech, M., Cuttler, R., Moscrop, D., Kallweit, H. and Martin, J. (2005) New Evidence
for the Neolithic Settlement of Marawah Island, Abu Dhabi, United Arab Emirates.
Proceedings of the Seminar for Arabian Studies, 35: 37-56.
Bibby, T.G. (1973) Preliminary Survey in East Arabia, 1968. Copenhagen:
Gyldendal.
Bird, M.I., Austin, W.E.N., Wurster, C.M., Fifield, L.K., Mojtahid, M. and Sargeant, C.
(2010) Punctuated Eustatic Sea-Level Rise in the Early Mid-Holocene. Geology, 38
(9): 803-806.
Blondel, P. (2007) The Handbook of Sidescan Sonar. Chichester: Springer-Praxis.
Blondel, P. and Sichi, O.G. (2009) Textural Analyses of Multibeam Sonar Imagery
from Stanton Banks, Northern Ireland Continental Shelf. Applied Acoustics, 70:
1288–1297.
Blott, S.J. and Pye, K. (2006) Particle Size Distribution Analysis of Sand-Sized
Particles by Laser Diffraction: an Experimental Investigation of Instrument Sensitivity
and the Effects of Particle Shape. Sedimentology, 53: 671–685.
Bowen, R.L. (1951) The Pearl Fisheries of the Persian Gulf. Middle East Journal, 5,
(2): 161-180.
Breeze, P., Cuttler, R. and Collins, P. (2011) Archaeological Landscape
Characterisation in Qatar through Satellite and Aerial Photographic Analysis, 2009 to
2010. Proceedings of the Seminar for Arabian Studies, 41: 13-26.
Bretzke, K., Armitage, S.J., Parker, A.G., Walkington, H. and Uerpmann, H-P. (2013)
The Environmental Context of Palaeolithic Settlement at Jebel Faya, Emirate
Sharjah, UAE. Quaternary International, 300: 83–93.
Brown, C.J. and Blondel, P. (2009) Developments in the Application of Multibeam
Sonar Backscatter for Seafloor Habitat Mapping. Applied Acoustics, 70: 1242–
1247.
Brown, C.J., Todd, B.J., Kostylev, V.E. and Pickrill, R.A. (2011) Image-Based
Classification of Multibeam Sonar Backscatter Data for Objective Surficial Sediment
Mapping of Georges Bank, Canada. Continental Shelf Research, 31: S110–S119.
356
Burkholder, G. (1972) ‘Ubaid Sites and Pottery in Saudi Arabia. Archaeology, 25:
264-269.
Carter, R. (2003) “Tracing Bronze Age Trade in the Arabian Gulf. Evidence for WayStations
of the Merchants of Dilmum between Bahrain and the Northern Emirates.” In
Potts, D., Al-Naboodah, H. and Hellyer, P. (eds.) Archaeology of the United Arab
Emirates. Proceedings of the First International Conference on the
Archaeology of the U.A.E (Abu Dhabi, 15-18 April 2001). London: Trident Press.
pp.124–131.
Carter, R. (2006) Boat Remains and Maritime Trade in the Persian Gulf During the
Sixth and Fifth Millennia BC. Antiquity, 80: 52–63.
Carter, R. (2010) “The Social and Environmental Context of Neolithic Seafaring in the
Persian Gulf.” In Anderson, A., Barrett, J.H. and Boyle, K.V. (eds.) The Global
Origins and Development of Seafaring. Cambridge: McDonald Institute for
Archaeological Research, University of Cambridge.
Carter, R. and Crawford, H. (eds.) (2010) Maritime Interactions in the Arabian
Neolithic: Evidence from H3, As-Sabiyah, an ‘Ubaid-Related Site in Kuwait.
American School of Prehistoric Research Monograph Series. Leiden: Brill.
Cavelier, C. 1970 Geological Description of the Qatar Peninsula (Arabian Gulf).
Government of Qatar Department of Petroleum Affairs Unpublished Report
Cheesman, R.E. (1923) From Oqair to the Ruins of Salwa. The Geographical
Journal 62 (5): 321-335.
Church, T.M. (1996) An Underground Route for the Water Cycle. Nature, 380: 579-
580.
Collier, J.S. and Brown, C.J. (2005) Correlation of Sidescan Backscatter with Grain
Size Distribution of Surficial Seabed Sediments. Marine Geology, 214: 431–449.
Cornwall County Council (2008) England’s Historic Seascapes: Historic
Seascape Characterisation (HSC), National HSC Method Statement. Report No.
2008R024. Cornwall: Cornwall: County Council
Crassard, R. and Drechsler (2013) Towards New Paradigms: Multiple Pathways for
the Arabian Neolithic. Arabian Archaeology and Epigraphy, 24: 3-8.
Creocean (2008) Qatar Bahrain Causeway Project. Underwater Video Survey:
Causeway Layout and Shipping Channels. Final Report. Report No. 1080055.
Unpublished.
Cuttler R.T.H. (ed.) (2011) Reconstruction of the Late Pleistocene and Holocene
Depositional Systems and Palaeogeography of Qatar. Stage 2 Data Analysis,
357
Assessment and Potential. Qatar National Historic Environment Record Project.
Doha: Qatar Museums Authority/University of Birmingham.
Cuttler, R.T.H. (2013) Considering Marine Transgression as a Mechanism for
Enforced Migration and the Coastal Gulf ‘Ubaid Phenomenon. The Neolithic of
Arabia: New Paradigms and Future Perspectives, Lyon (France). Arabian
Archaeology and Epigraphy, 24: 37-43.
Cuttler, R.T.H. (2014) Human Populations and Former Sub-aerial Landscapes of
the Arabian Gulf: Research and conservation. PhD Thesis, University of
Birmingham, UK.
Cuttler R.T.H. (ed.) (Forthcoming) Qatar National Historic Environment Record
Project. Historic Landscapes and Palaeoenvironments: Terrestrial and Marine
Analysis. Unpublished. Doha: Qatar Museums Authority/University of Birmingham.
Cuttler, R.T.H., Fitch, S. and Breeze, P. (2011a) An Archaeological Impact
Assessment of Sidescan Sonar Data from the Qatar Areas of the Qatar-Bahrain
Causeway. IBM VISTA Centre, University of Birmingham for the Qatar Museums
Authority. Unpublished report.
Cuttler, R.T.H., Tetlow, E. and Al-Naimi, F. (2011b) Assessing the Value of
Palaeoenvironmental Data and Geomorphological Processes for Understanding Late
Quaternary Population Dynamics in Qatar. Proceedings of the Seminar for
Arabian Studies, 41: 47-60.
Cuttler, R.T.H., Fitch, S., and Al-Naimi, F.A. (2012) “Considering the ‘Terra Incognita’
and the Implications for the Cultural Resource Management of the Arabian Gulf
Palaeolandscape.” In Hellyer, P., Al Naboodah, H. and Potts, D. (eds.) Archaeology
of the United Arab Emirates: Proceedings of the Second International
Conference on the Archaeology of the U.A.E. London: Trident Press. pp.235-245.
Cuttler, R.T.H. and Al-Naimi, F.A. (2013). From Land-Locked Desert to Maritime
Nation: Understanding Taphonomic Processes and Landscape Evolution in Qatar
and the Arabian Peninsula from 14,000BP. Adumatu, 28
De Cardi, B. (ed.) (1978) Qatar Archaeological Report: Excavations 1973. Oxford:
Oxford University Press.
Dingwall, L. and Gaffney, V. (eds.) (2007) Heritage Management at Fort Hood,
Texas: Experiments in Historic Landscape Characterisation. Oxford:
Archaeopress.
Drechsler, P. (2009) The Dispersal of the Neolithic Over the Arabian Peninsula.
BAR International Series 1969. Oxford: Archaeopress.
Drechsler, P. (2011) Places of Contact, Spheres of Interaction. The ‘Ubaid
Phenomenon in the Central Gulf Area as Seen from a First Season of
358
Reinvestigations at Dosariyah (Dawsāriyyah), Eastern Province, Saudi Arabia.
Proceedings of the Seminar for Arabian Studies, 41: 69-82.
Edens, C. (1999) Khor Ile-Sud, Qatar: The Archaeology of Late Bronze Age PurpleDye
Production in the Arabian Gulf. Iraq, 61: 71-88
Ellingsen, K.E., Gray, J.S., and Bjørnbom, E. (2002) Acoustic Classification of
Seabed Habitats Using the QTC VIEWTM
System. ICES Journal of Marine Science,
59: 825–835.
Ellis, J.P. and McGuinness, W.T. (1986) Pockmarks of the Northwestern Arabian
Gulf. Advances in Underwater Technology, Ocean Science and Offshore
Engineering, 6: 353-367.
Evans, G. 1966 The Recent Sedimentary Facies of the Persian Gulf Region.
Philosophical Transactions of the Royal Society of London. Series A,
Mathematical and Physical Sciences, 259 (1099): 291-298.
Fairclough, G.J. (2001) Boundless Horizons – Historic Landscape Characterisation.
English Heritage Conservation Bulletin, 40: 21-26.
Faught, M.K. and Donoghue, J.F. (1997) Marine Inundated Archaeological Sites and
Palaeofluvial Systems: Examples from a Karst-Controlled Continental Shelf Setting in
Apalachee Bay, Northeastern Gulf of Mexico. Geoarchaeology, 12 (5): 417–458.
Faure, H., Walter, R.C. and Grant, D.R. (2002) The Coastal Oasis: Ice Age Springs
on Emerged Continental Shelves. Global and Planetary Change, 33: 47–56.
Fleitmann, D., Matter, A., Pint, J.J. and Al-Shanti, M.A. (2004) The Speleothem
Record of Climate Change in Saudi Arabia. Saudi Geological Survey Open-File
Report SGS-OF-2004-8. Unpublished.
Fitch, S., Thomson, K. and Gaffney, V. (2005) Late Pleistocene and Holocene
Depositional Systems and the Palaeogeography of the Dogger Bank, North Sea.
Quaternary Research, 64: 185-196.
Fitch, S. (2011) The Mesolithic Landscape of the Southern North Sea. PhD
Thesis, University of Birmingham, UK.
Fonseca, L., Brown, C., Calder, B., Mayer, L. and Rzhanov, Y. (2009) Angular Range
Analysis of Acoustic Themes from Stanton Banks Ireland: A Link Between Visual
Interpretation and Multibeam Echosounder Angular Signatures. Applied Acoustics,
70: 1298–1304.
Ford, D. And Williams, P. (2007) Karst Hydrogeology and Geomorphology.
England: Wiley.
359
Fugro Peninsular (2008) Qatar Bahrain Causeway Project Offshore Geotechnical
Investigation. Qatar-Bahrain Volume 1. Report No. PQ/3145/08/01. Unpublished.
Gaffney, V., Thomson, K. and Fitch, S. (eds.) (2007) Mapping Doggerland: The
Mesolithic Landscapes of the Southern North Sea. Oxford: Archaeopress.
GEMS (2008) Geophysical Survey Report, Qatar Site, Qatar-Bahrain Causeway.
Report No. GME08030-GPH-001. Unpublished.
Golding, N., Vincent, M.A. and Connor, D.W. (2004) The Irish Sea Pilot. Report on
the Development of a Marine Landscape classification for the Irish Sea. JNCC
Report No. 346. Unpublished.
Groucutt, H.S. and Petraglia, M.D. (2012) The Prehistory of the Arabian Peninsula:
Deserts, Dispersals, and Demography. Evolutionary Anthropology, 21: 113–125.
Henderson, J., Pizarro, O., Johnson-Roberson,,
M. and Mahon, I. (2013) Mapping
Submerged Archaeological Sites using Stereo-Vision Photogrammetry. International
Journal of Nautical Archaeology, 42 (2): 243–256.
Herman, J.J.H.C. (1957) Surface Sediments of the Persian Gulf Near the Qatar
Peninsula. PhD Thesis, University of Utrecht, Netherlands.
Heyvaert , V.M.A. and Baeteman , C. (2007) Holocene Sedimentary Evolution and
Palaeocoastlines of the Lower Khuzestan Plain (Southwest Iran). Marine Geology,
242: 83–108.
Hill, A., Bibi, F., Beech, M. and Yasin Al-Tikriti, W. (2012) “Before Archaeology: Life
and Environments in the Miocene of Abu Dhabi.” In Potts, D.T. and Hellyer, P. (eds.)
Fifty Years of Emirates Archaeology. Abu Dhabi/Dubai: Motivate Publishing. pp21-
33.
Inizan, M-L. (1988) Préhistoire à Qatar: Mission Archéologique Française à
Qatar: Volume. 2. Paris: CNRS.
Jameson, J. and Strohmenger, C.J. (2012) Relative Sea-Level Changes During the
Late Pleistocene to Holocene of Qatar: Implications for Eustasy and Tectonics.
Adapted from Oral Presentation at AAPG Annual Convention and Exhibition,
Long Beach, California, April 22-25, 2012 [online]. Available from:
http://www.searchanddiscovery.com/pdfz/documents/2012/50704jameson/ndx_jame
son.pdf.html [Accessed 25 March 2014]
Jones, E.J.W. (1999) Marine Geophysics. Chichester: Wiley.
Johns, C., Larn, R. and Tapper, B. (2004) Rapid Coastal Zone Assessment for the
Isles of Scilly. Cornwall County Council Report No. 2004R030. Cornwall: Cornwall:
County Council
360
Johnstone, T.M. and Wilkinson,J.C. (1960) Some Geographical Aspects of Qatar.
The Geographical Journal, 126 (4): 442-450.
Kapel, H. (1967) Atlas of the Stone-Age Cultures of Qatar. Reports of the Danish
Archaeological Expedition to the Arabian Gulf. Aarhus: Aarhus University Press.
Keller, B.M. (2011) Imaging the Twilight Zone: The Morphology and Distribution
of Mesophotic Zone Features, A Case Study from Bonaire, Dutch Caribbean.
Master of Science Thesis, University of Delaware, USA.
Killick, R. and Moon, J. (eds.) (2005) The Early Dilmun Settlement at Saar. Saar
Excavation Report Volume 3. Ludlow: Archaeology International Ltd.
KSEPL Carbonate Group of 1965-1969 (1973) Holocene Carbonate Sediments of
the Southern Arabian Gulf, Parts I-VIII. Netherlands: Koninklijke/Shell.
Lambeck, K. (1996) Shoreline Reconstructions for the Persian Gulf Since the Last
Glacial Maximum. Earth and Planetary Science Letters, 142: 43‐57.
Land Use Consultants (1996) Isles of Scilly Historic Landscape Assessment and
Management Strategy. Unpublished.
LeBlanc, J. (2008) A Fossil Hunting Guide To the Tertiary Formations of Qatar,
Middle East. First Edition [online]. Available from:
https://drive.google.com/folderview?id=0B_FRE6vSR2MeMGwyaXdsNDVsdnM&usp
=sharing [accessed December 2013].
Macumber, P.G. (2011) A Geomorphological and Hydrological Underpinning
for Archaeological Research in Northern Qatar. Proceedings of the Seminar for
Arabian Studies, 41: 1-14.
Marin Mätteknik AB (2002) Qatar-Bahrain Causeway. Acoustic Survey Report.
Preliminary Environmental and Engineering Studies. Phase 1 – Studies and
Surveys. Report No. 54085-PR-009 Issue No. 1. Unpublished.
Marks, A.E. (2009) “The Palaeolithic of Arabia in an Inter-Regional Context.” In
Petraglia, M.D. and Rose, J.I. (eds.) The Evolution of Human Populations in
Arabia: Paleoenvironments, Prehistory and Genetics. Dordrecht: Springer.
pp.295-308.
Marsh, I and Brown, C. (2009) Neural Network Classification of Multibeam
Backscatter and Bathymetry Data from Stanton Bank (Area IV). Applied Acoustics,
70: 1269–1276.
Masry, A.H. (1997) Prehistory in Northeastern Arabia: The Problem of
Interregional Interaction. London: Kegan Paul International.
361
McClure, H.A. (1976) Radiocarbon Chronology of Late Quaternary Lakes in the
Arabian Desert. Nature, 263: 755-756.
McClure, H.A. and Al-Shaikh, N.Y. (1993) Palaeogeography of an 'Ubaid
Archaeological Site, Saudi Arabia. Arabian Archaeology and Epigraphy, 4: 107–
125.
McCorriston, J. and Martin, L. (2009) “Southern Arabia’s Early Pastoral Population
History: Some Recent Evidence.” In Petraglia, M.D. and Rose, J.I. (eds.) The
Evolution of Human Populations in Arabia: Paleoenvironments, Prehistory and
Genetics. Dordrecht: Springer. pp.237-250.
McCorriston, J. (2013) The Neolithic in Arabia: a View from the South. Arabian
Archaeology and Epigraphy, 24: 68–72.
McCorriston, J. and Martin, L. (2009) “Southern Arabia’s Early Pastoral Population
History: Some Recent Evidence.” In Petraglia, M.D. and Rose, J.I. (eds.) The
Evolution of Human Populations in Arabia: Paleoenvironments, Prehistory and
Genetics. Dordrecht: Springer. pp.237-250.
Misbahuddin, K. (1984) The Role of Raḥmah bin Jabīr bin ʿAdhbī in the History
of Eastern Arabia (1783-1826). Master of Arts Thesis, Institute of Islamic Studies,
McGill University, Canada.
NOAA (2006) National Marine Protected Areas Center Marine Cultural and
Historic Newsletter. Volume 3 (12), 1st December 2006.
NOAA (2014) Digital Coast. Benthic Terrain Modeler [online]. Available from:
http://coast.noaa.gov/digitalcoast/tools/btm [accessed 3rd February 2015].
Orpin, A.R. and Kostylev, V.E. (2006) Towards a Statistically Valid Method of
Textural Sea Floor Characterization of Benthic Habitats. Marine Geology, 225: 209–
222.
Parker, A.G., Goudie, A.S., Stokes, S., White, K., Hodson, M.J., Manning, M. and
Kennet, D. (2006) A Record of Holocene Climate Change from Lake Geochemical
Analyses in Southeastern Arabia. Quaternary Research, 66: 465–476.
Parker, A.G. and Goudie, A.S. (2008) Geomorphological and Palaeoenvironmental
Investigations in the Southeastern Arabian Gulf Region and the Implication for the
Archaeology of the Region. Geomorphology, 101: 458-470
Parker, A.G. and Rose, J.I. (2008) Climate Change and Human Origins in Southern
Arabia. Proceedings of the Seminar for Arabian Studies, 38: 25-42.
Pelly, J.H. (1895) Telegram No. 360. H.M.S. “Sphinx” at Zobara, the 7th September
1895. From Captain J. H. Pelly, Commander and Senior Naval Officer to Colonel F.
362
A. Wilson, Political Resident, Persian Gulf. British Library Indian Office Reports
Collections, IOR R/15/1/314.
Penrose, J.D., Siwabessy, P.J.W., Gavrilov, A., Parnum, I., Hamilton, L.J., Bickers,
A., Brooke, B., Ryan, D.A. and Kennedy, P. (2005) Acoustic Techniques for
Seabed Classification. Cooperative Research Centre for Coastal Zone Estuary and
Waterway Management. Technical Report 32, September 2005.
Petraglia, M.D. (2005) Hominin Responses to Pleistocene Environmental Change in
Arabia and South Asia. Geological Society, London, Special Publications,
January 1, 2005, 247: 305-319
Petraglia, M.D. and Rose, J.I. (eds.) (2009) The Evolution of Human Populations
in Arabia: Paleoenvironments, Prehistory and Genetics. Dordrecht: Springer.
Petraglia, M.D., Alsharekh, A., Breeze, P., Clarkson, C., Crassard, R., Drake, N.A.,
Groucutt, H.S., Jennings, R., Parker, A.G., Parton, A., Roberts, R.G., Shipton, C.,
Matheson, C., Al-Omari, A. and Veall, M.A. (2012) Hominin Dispersal into the Nefud
Desert and Middle Palaeolithic Settlement along the Jubbah Palaeolake, Northern
Arabia. PLoS ONE [online], 7 (11): e49840. doi:10.1371/journal.pone.0049840.
Potts, D.W. (1990) The Arabian Gulf in Antiquity. Volume I. From Prehistory to
the Fall of the Achaemenid Empire. Oxford: Clarendon Press.
Preston, J. (2009) Automated Acoustic Seabed Classification of Multibeam Images of
Stanton Banks. Applied Acoustics, 70: 1277–1287.
Pullar, J. (1974) Harvard Archaeological Survey in Oman, 1973: Flint Sites in Oman.
Proceedings of the Seminar for Arabian Studies, 4: 33-48.
Purser, B.H. (ed.) (1973) The Persian Gulf. Holocene Carbonate Sedimentation
and Diagenesis in a Shallow Epicontinental Sea.
QBC Consortium (2009a) Geotechnical Factual Report. Qatar-Bahrain Causeway
Project Embankment Data. Report No. QBC-ME-P2GT-ALL-GEN-TRP-001 Rev.
0.1. Unpublished.
QBC Consortium (2009b) Geotechnical Factual Report. Qatar-Bahrain Causeway
Project Borrow Areas Data. Report No. QBC-ME-P2GT-ALL-GEN-TRP-002 Rev.
1.0. Unpublished.
Quester Tangent (2012a) QTC SWATHVIEW 12.1 User Manual and Reference:
DMNSWVW-0000-R06. Unpublished. Canada: Quester Tangent.
Quester Tangent (2012b) QTC CLAMS™ 12.1 User Guide and Reference: DMNCLAM-0000-R03.
Unpublished. Canada: Quester Tangent.
363
Roaf, M. (1976) Excavations at Al Markh, Bahrain. Proceedings of the Seminar for
Arabian Studies, 6: 144-160.
Rooper, C.N. and Zimmermann, M. (2007) A Bottom-Up Methodology for Integrating
Underwater Video and Acoustic Mapping for Seafloor Substrate Classification.
Continental Shelf Research, 27: 947–957.
Rose, J.I. (2010) New Light on Human Prehistory in the Arabo-Persian Gulf Oasis.
Current Anthropology, 51 (6): 849-883.
Rose, J.I., Usik, V.I., Marks, A.E., Hilbert, Y.H., Galletti, S.C., Parton, A., Geiling,
J.M., Černý, V., Morley, M.W & Roberts, R.G. (2011) The Nubian Complex of Dhofar,
Oman: An African Middle Stone Age Industry in Southern Arabia. PLoS ONE
[online], 6(11): e28239. doi:10.1371/journal.pone.0028239
Roux, G. (1964) Ancient Iraq. UK: Penguin.
Rutkowski, L. (ed.) (2011) Kuwaiti-Polish Investigations in Northern Kuwait. AsSabbiya
2007-2010. Unpublished. Al-Jahra/Warsaw: National Council for Culture,
Arts and Letters, Kuwait, and Polish Centre of Mediterranean Archaeology, University
of Warsaw.
Sadiq, A.M. and Nasir, S.J. (2002) Middle Pleistocene Karst Evolution in the State of
Qatar, Arabian Gulf. Journal of Cave and Karst Studies, 64 (2): 132-139.
Sarnthein, M. (1972) Sediments and History of the Postglacial Transgression in the
Persian Gulf and Northwest Gulf of Oman. Marine Geology, 12: 245-266.
Schlüter, M.J., Sauter, E.J., Andersen, C.E., Dahlgaard, H. and Dando, P. (2004)
Spatial Distribution and Budget for Submarine Groundwater Discharge in
Eckernförde Bay (Western Baltic Sea). Limnology and Oceanography, 49 (1): 157–
167
Scott-Jackson, J.E., Scott-Jackson, W.B. and Rose, J.I. (2009) “Palaeolithic Stone
Tool Assemblages from Sharjah and Ras al Khaimah in the United Arab Emirates.” In
Petraglia, M.D. and Rose, J.I. (eds.) The Evolution of Human Populations in
Arabia: Paleoenvironments, Prehistory and Genetics. Dordrecht: Springer.
pp.125-238.
Shinn, E.A. (1969) Submarine Lithification of Holocene Carbonate Sediments in the
Persian Gulf. Sedimentology, 12: 109-144.
Simons, D.G. and Snellen, M. (2009) A Bayesian Approach to Seafloor Classification
Using Multi-Beam Echo-Sounder Backscatter Data. Applied Acoustics, 70: 1258–
1268.
Spoor, R.H. (1997) Human Population Groups and the Distribution of Lithic
Arrowheads in the Arabian Gulf. Arabian Archaeology and Epigraphy, 8: 143-160.
364
Stanford, J.D., Hemmingway, R., Rohling, E.J., Challenor, P.G., Medina-Elizalde, M.
and Lester, A.J. (2011) Sea-level Probability for the Last Deglaciation: A Statistical
Analysis of Far-Field Records. Global and Planetary Change, 79: 193-203.
State of Delaware (2012) Division of Fish and Wildlife. Red Bird Reef Sinkings
[online]. Available from:
www.dnrec.delaware.gov/fw/Fisheries/Pages/Red%20Bird%20Reef%20sinkings.asp
x [accessed 27th August 2013].
Stoffers, P. and Ross, D.A. (1979) Late Pleistocene and Holocene Sedimentation in
the Persian Gulf/Gulf of Oman. Geology, 23: 181-208.
Sutherland, T.F., Galloway, J., Loschiavo, R., Levings, C.D. and Hare, R. (2007)
Calibration Techniques and Sampling Resolution Requirements for Groundtruthing
Multibeam Acoustic Backscatter (EM3000) and QTC VIEWTM
Classification
Technology. Estuarine, Coastal and Shelf Science, 75: 447-458.
Taylor, J.C.M. and Illing, L.V. (1969) Holocene Intertidal Calcium Carbonate
Cementation, Qatar, Persian Gulf. Sedimentology, 12: 69-107.
Teller, J.T., Glennie, K.W., Lancaster, N. and Singhvi, A.K. (2000) Calcareous Dunes
of the United Arab Emirates and Noah's Flood: the Postglacial Reflooding of the
Persian (Arabian) Gulf. Quaternary International, 68-71: 297-308.
Tetlow, E.A, Cuttler, R.T.H and Al Naimi, F.A (Forthcoming) Wadi Debayan: A
Neolithic Landscape.
Tixier, J. (1980) Mission Archéologique Française à Qatar: 1976-77, 1977-78.
Paris: CNRS Recherches Anthropologiques.
Uchupi, E., Swift S.A. and Ross D.A. (1996) Gas Venting and Late Quaternary
Sedimentation in the Persian (Arabian) Gulf. Marine Geology, 129: 237-269.
Uchupi, E., Swift S.A. and Ross D.A. (1999) Late Quaternary Stratigraphy,
Palaeoclimate and Neotectonism of the Persian (Arabian) Gulf Region. Marine
Geology, 160: 1–23.
Uerpmann, M. (2003) “The Dark Millennium — Remarks on the Final Stone Age in
the Emirates and Oman.” In Potts, D., Al-Naboodah, H. and Hellyer, P. (eds.)
Archaeology of the United Arab Emirates. Proceedings of the First International
Conference on the Archaeology of the U.A.E (Abu Dhabi, 15-18 April 2001).
London: Trident Press. pp.74–81.
Uerpmann, M. and Uerpmann, H-P. (2008) “Neolithic Faunal Remains from al-Buhais
18 (Sharjah, UAE).” In Uerpmann, H-P., Uerpmann, M. and Jasim, S.A. (eds.) The
Natural Environment of Jebel al-Buhais: Past and Present, The Archaeology of
Jebel al-Buhais, Volume 2. Tübingen: Kerns Verlag. pp.97-132.
365
Uerpmann, H-P., Potts, D.T. and Uerpmann, M. (2009) “Holocene (Re-)Occupation of
Eastern Arabia.” In Petraglia, M.D. and Rose, J.I. (eds.) The Evolution of Human
Populations in Arabia: Paleoenvironments, Prehistory and Genetics. Dordrecht:
Springer. pp.205-214.
UKHO 1925 Maritime Chart, Persian Gulf.
Van Rein, H., Brown, C.J., Quinn, R., Breen, J. and Schoeman, D. (2011) An
Evaluation of Acoustic Seabed Classification Techniques for Marine Biotope
Monitoring Over Broad-Scales (>1 km2
) and Meso-Scales (10 m2-
1 km2
). Estuarine,
Coastal and Shelf Science, 93: 336-349.
Vosmer, T. (1999) Indo-Arabian Stone Anchors in the Western Indian Ocean and
Arabian Sea. Arabian Archaeology and Epigraphy, 10: 248-263.
Wahida, G., Al-Tikriti, W.Y., Beech, M. and Al Meqbali, A. (2009) “A Middle
Palaeolithic Assemblage from Jebel Barakah, Coastal Abu Dhabi Emirate.” In
Petraglia, M.D. and Rose, J.I. (eds.) The Evolution of Human Populations in
Arabia: Paleoenvironments, Prehistory and Genetics. Dordrecht: Springer.
pp.117-124.
Wessex Archaeology (2012) Outer Hebrides Coastal Community Marine
Archaeology Pilot Project [online]. Available from:
http://blogs.wessexarch.co.uk/ohccmapp/ [accessed 31st August 2013].
Westley, K., Quinn, R., Forsythe, W., Plets, R., Bell, T., Benetti, S., McGrath, F. And
Robinson, R. (2011a) Mapping Submerged Landscapes Using Multibeam
Bathymetric Data: a Case Study from the North Coast of Ireland. The International
Journal of Nautical Archaeology, 40 (1): 99–112.
Westley, K., Bell, T., Plets, R. and Quinn, R. (2011b) “Investigating Submerged
Archaeological Landscapes: a Research Strategy Illustrated with Case Studies from
Ireland and Newfoundland, Canada.” In Benjamin, J., Bonsall, C., Pickard, C. and
Fischer, A. (eds.) Submerged Prehistory. Oxford: Oxbow Books.
Williams, A.H. and Walkden, G.M. (2002) “Late Quaternary Highstand Deposits of the
Southern Arabian Gulf: a Record of Sea-Level and Climate Change.” In Clift, P.D.,
Kroon, D., Gaedicke, C. and Craig, J. (eds.) The Tectonic and Climatic Evolution
of the Arabian Sea Region. London: Geological Society. pp371-386.
Wilson, D. (1833) Memorandum Respecting the Pearl Fisheries in the Persian Gulf.
Journal of the Royal Geographical Society of London, (3): 283-286.
Wilson, Col. F.A. (1895) Telegram No. 98, Dated Bushire, the 15th September 1895.
From Colonel F. A. Wilson, Political Resident, Persian Gulf To The Secretary to the
Government of India, Foreign Department. British Library Indian Office Reports
Collections, IOR R/15/1/314.