Morphology and origin of modern seabed features in the central
Transcription
Morphology and origin of modern seabed features in the central
AGH University of Science and Technology in Kraków, Poland Faculty of Geology, Geophysics and Environmental Protection Department of Energetic Resources Doctoral dissertation Morphology and origin of modern seabed features in the central basin of the Gulf of Thailand Radosław Puchała 1,2 Supervisor: Prof. Dr. Szczepan Porębski3 Kraków, 2014 1 2 3 Fugro Survey B.V., 2260 AC Leidschendam, The Netherlands Fugro Survey Africa Pte Ltd, Cape Town, South Africa AGH University of Science and Technology, Poland Akademia Górniczo-Hutnicza im. Stanisława Staszica w Krakowie Wydział Geologii Geofizyki i Ochrony Środowiska Katedra Surowców Energetycznych Rozprawa doktorska Morfologia i pochodzenie współczesnych form dna w centralnej części Zatoki Tajlandzkiej Radosław Puchała1,2 Promotor: prof. dr hab. Szczepan Porębski 3 Kraków, 2014 _______________________ 1 2 3 Fugro Survey B.V., 2260 AC Leidschendam, Holandia Fugro Survey Africa Pte Ltd, Kapsztad, Republika Południowej Afryki Akademia Górniczo-Hutnicza im. Stanisława Staszica w Krakowie, Polska ACKNOWLEDGEMENTS I acknowledge Fugro Group of Companies for making available all of the geophysical datasets, provision of sediment cores, and financial patronage. My special thanks go to Fugro directors: Wilhard Kreijkes, Peter Aarts and Mark Heine. I wish to thank thesis supervisor Prof. Szczepan Porębski, AGH University of Science and Technology in Kraków, for full support, shared knowledge and valuable input to the research. I would like to extend my gratitude to the staff of the Institute of Geological Sciences, University of Wrocław, especially to Krzysztof Moskwa for managing the laboratory tests, Dr. Czesław August for performing X-ray analyses, Dr. Wojciech Śliwiński for scientific and logistical support, and to Prof. Andrzej Solecki for supervision during my pre-doctoral studies. I am also in debt to Joachim Ariel Kloskowski for photographing the microfossils and to my colleagues from Fugro, Simon Ayers and Dr. Marcin Jankowski for revising the text and many useful suggestions. I am grateful to the Institute of Geological Sciences, Polish Academy of Sciences, for providing scientific supervision during the years 2009-2010, and to the Gliwice Radiocarbon Laboratory for performing radiocarbon dating. Finally, I must thank my wife Anna for patience and a lot of sacrifice, which helped me to finalise the work. 3 ABSTRACT Three main stratigraphic units have been identified in the shallowest deposits of the central part of the Gulf of Thailand. Unit C represents an upper Pleistocene transgressiveregressive sequence. Unit B comprises marine sediments which were subjected to lateritization during the last glacial period and then flooding by the sea. Unit A consists of Holocene marine muds and clays (A1), which are in places underlain by muds (A2) that originated during an early transgressive stage of the Pleistocene/Holocene transition period. This tri-partite stratigraphy is consistent across the whole basin. The erosional unconformity between the two uppermost soft marine muds of Unit A and underlying stiff silt of Unit B is a ravinement surface (R1) associated with early Holocene marine transgression. Radiocarbon ages date this surface at 10.4-10.6 cal kyr BP for the centralsouthern part of the basin and more than 6.5 cal kyr BP for the northwestern area, which reflects a diachronous spread of the transgression in the landward (northward) direction. In water depths beyond 50 m, the Gulf of Thailand displays unique seafloor morphology, which is typified by the occurrence of elongated soft-mud mounds and depressions, as well as numerous pockmarks. These features result from a combination of fluid seepage, sediment dehydration processes and erosion by marine currents. Sediment dewatering has led to the formation of numerous, widely distributed, small pits and pockmarks within the unconsolidated muds of Unit A. The superimposed, long-term erosive activity of bottom currents modify gradually these features into small elongated pockmarks, long runnels and depressions, and ultimately into large fields of elongated mounds and ridges, as well as residual fragments of as yet un-eroded mud and clay sheets. Mud mounds represent a half-way stage in the spectrum of erosionally-controlled seabed morphology, encompassing continuous mud sheets at one end and isolated mud residues at the other, and all intermediate morphological forms have been found on the modern seafloor of the Gulf of Thailand. The occurrence of these features below the 50 m isobath reflects water stratification at a thermohalocline separating two water layers with different physical properties. The unidirectional bottom circulation operating within the lower water layer is interpreted to reflect a combination of tidal and density-driven currents associated with water exchange between the South China Sea and Gulf of Thailand. 4 TABLE OF CONTENTS ACKNOWLEDGEMENTS .............................................................................................................................. 3 ABSTRACT ............................................................................................................................................. 4 TABLE OF CONTENTS ................................................................................................................................ 5 LIST OF FIGURES .................................................................................................................................... 7 LIST OF TABLES .................................................................................................................................... 11 1 INTRODUCTION AND PURPOSE OF RESEARCH ....................................................................... 12 2 GULF OF THAILAND – BASIN MORPHOLOGY AND STRUCTURAL, OCEANOGRAPHIC AND CLIMATIC SETTINGS ......................................................................................................................... 14 2.1 2.2 2.3 2.4 2.5 2.6 2.7 3 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 Location and Topography ....................................................................................................... 14 Geological Background .......................................................................................................... 16 Structural Setting ................................................................................................................. 17 2.3.1 Pre-Quaternary Stratigraphy ..................................................................................... 17 2.3.2 Upper-Pleistocene and Holocene Sediments................................................................. 18 Climate ................................................................................................................................ 20 2.4.1 Monsoons ............................................................................................................... 22 Water Circulation .................................................................................................................. 22 2.5.1 Tides ...................................................................................................................... 23 2.5.2 Water Temperature and Salinity ................................................................................ 24 Water Currents ..................................................................................................................... 25 Seabed Morphology, Sedimentation and Environment ................................................................ 26 2.7.1 Accumulation Rates and Sediment Mixing ................................................................... 28 2.7.2 River Deltas ............................................................................................................ 28 2.7.3 Mangroves .............................................................................................................. 31 2.7.4 Coral Reefs ............................................................................................................. 32 METHODOLOGY ...................................................................................................................... 33 Survey Vessels ..................................................................................................................... 33 Positioning and Navigation ..................................................................................................... 34 3.2.1 Fugro Starfix HP DGPS ............................................................................................. 34 3.2.2 Fugro Starfix MRDGPS .............................................................................................. 34 3.2.3 Dynamic Heading Reference System .......................................................................... 34 3.2.4 Underwater Positioning ............................................................................................. 35 3.2.5 Navigation .............................................................................................................. 36 Sound Velocity Measurements ................................................................................................ 36 Single Beam Echo Sounder (SBES).......................................................................................... 36 Multi Beam Echo Sounder (MBES) ........................................................................................... 38 3.5.1 Simrad EM1002 MBES .............................................................................................. 38 3.5.2 Reson SeaBat 8101 MBES ......................................................................................... 38 3.5.3 MBES System Calibration and Processing .................................................................... 39 3.5.4 Tidal Reduction ........................................................................................................ 39 3.5.5 Refraction Reduction ................................................................................................ 39 3.5.6 Post-processing Analysis and Data QC ........................................................................ 39 Motion Sensors ..................................................................................................................... 40 Side Scan Sonar (SSS) .......................................................................................................... 41 3.7.1 System Description and Data Acquisition .................................................................... 41 3.7.2 Side Scan Sonar Data Processing ............................................................................... 41 Sub-bottom Profiler (SBP) ...................................................................................................... 43 3.8.1 Acquisition Parameters ............................................................................................. 44 3.8.2 Data QC and Processing ............................................................................................ 44 Gravity Coring ...................................................................................................................... 45 Laboratory Analyses .............................................................................................................. 46 Borehole Log Analyses ........................................................................................................... 47 Survey Coverage and Methods for Each Area ............................................................................ 48 5 4 4.1 4.2 4.3 4.4 4.5 4.6 4.7 5 5.1 5.2 5.3 5.4 6 6.1 7 RESULTS ................................................................................................................................. 49 General Overview of the Whole Area ....................................................................................... 49 4.1.1 Seabed Morphology of the Gulf of Thailand.................................................................. 49 4.1.2 Stratigraphy of Shallow Sediments in the Gulf of Thailand ............................................. 49 Central-South Basin .............................................................................................................. 52 4.2.1 Bathymetry and Seabed Morphology .......................................................................... 52 4.2.2 Lithology and Sub-bottom Features ............................................................................ 53 4.2.3 Seabed Features ...................................................................................................... 59 Central-West Basin ............................................................................................................... 66 4.3.1 Bathymetry and Seabed Morphology .......................................................................... 66 4.3.2 Lithology and Sub-bottom Features ............................................................................ 66 4.3.3 Seabed Features ...................................................................................................... 89 South-West Margin (Offshore Songkhla) .................................................................................. 90 4.4.1 Bathymetry and Seabed Morphology .......................................................................... 90 4.4.2 Seabed Features ...................................................................................................... 90 4.4.3 Lithology................................................................................................................. 90 4.4.4 Sub-bottom Features................................................................................................ 92 North-South Profile ............................................................................................................... 92 4.5.1 Bathymetry and Seabed Morphology .......................................................................... 92 4.5.2 Lithology................................................................................................................. 92 4.5.3 Seabed Features ...................................................................................................... 94 4.5.4 Sub-bottom Features................................................................................................ 94 Mouth of the Gulf (Offshore Vietnam) ...................................................................................... 97 4.6.1 Bathymetry and Seabed Morphology .......................................................................... 97 4.6.2 Seabed Features ...................................................................................................... 97 4.6.3 Lithology and Sub-bottom Features ............................................................................ 97 Central Basin (Offshore Cambodia) ......................................................................................... 97 4.7.1 Bathymetry and Seabed Morphology .......................................................................... 97 4.7.2 Seabed Features ...................................................................................................... 97 4.7.3 Lithology and Sub-bottom Features ............................................................................ 98 INTEPRETATION ................................................................................................................... 105 Stratigraphy of Upper Pleistocene and Holocene of the Gulf of Thailand ..................................... 105 5.1.1 Unit C - Upper Pleistocene sedimentary sequence ...................................................... 105 5.1.2 Unit B – Uppermost Pleistocene lateritic palaeosoil ..................................................... 106 5.1.3 Reflector R1 – Lower Holocene transgressive surface .................................................. 107 5.1.4 Unit A – Transgressive systems tract and Holocene marine muds ................................. 109 The origin, shape and distribution of pockmarks of the Gulf of Thailand ..................................... 110 5.2.1 Pockmarks morphology and fluid seeps..................................................................... 110 5.2.2 Pockmarks morphology and marine currents ............................................................. 115 5.2.3 Processes driving to further changes in pockmarks morphology ................................... 117 Holocene marine mud mounds and ridges .............................................................................. 118 5.3.1 Soft mud mounds and ridges – morphology and spatial distribution ............................. 118 5.3.2 Orientation of mud mounds and ridges ..................................................................... 119 5.3.3 Age of the mud mounds and ridges .......................................................................... 122 Evolution of seabed morphology ........................................................................................... 122 5.4.1 Pockmark cluster evolution ..................................................................................... 122 5.4.2 Evolution of large pockmarks................................................................................... 123 5.4.3 Unit pockmark evolution and further development of the seabed morphology ................ 123 DISCUSSION OF RESULTS .................................................................................................... 130 Oceanographic conditions and formation of elongated pockmarks ............................................. 130 6.1.1 Bottom currents and wind direction .......................................................................... 130 6.1.2 Tidal currents – theory and definitions ...................................................................... 131 6.1.3 Tidal currents and seabed morphology...................................................................... 132 6.1.4 Density-driven currents and seabed morphology ........................................................ 134 6.1.5 Seabed morphology and thermohalocline .................................................................. 135 6.1.6 Seabed morphology and internal waves .................................................................... 135 6.1.7 Bottom currents and the nature of oceanographic data ............................................... 135 6.1.8 Summary .............................................................................................................. 136 CONCLUSIONS ...................................................................................................................... 137 REFERENCES ...................................................................................................................................... 140 6 LIST OF FIGURES Figure 2.1. Location of study area. The Gulf of Thailand. ....................................................................... 15 Figure 2.2. Regional tectonic map of western Southeast Asia (after Watcharanantakul and Morley, 2000, modified from: Packham, 1996; Leloup et al., 1995; Oudom-Ugsorn et al., 1986, Polchan and Sattayarak, 1989). ..................................................................................................... 16 Figure 2.3. Stratigraphy of the Cainozoic sequences in the Gulf of Thailand (after Watcharanantakul and Morley, 2000, modified from Lian and Bradley, 1986). .......................................................... 17 Figure 2.4. Correlation of Quaternary deposits in the coastal parts of Thailand (modified from Dheeradilok, 1995). ......................................................................................................... 19 Figure 2.5. Composite stratigraphic succession of unconsolidated sediments in the Lower Central Plain (not to scale) (after Sinsakul, 2000). .................................................................................. 20 Figure 2.6. Sea-level curve for the Gulf of Thailand and Sunda Shelf during the past 140 kyr based on oxygen isotope (Shackleton, 1987) and coral reef records (Chappell et al., 1996). Compiled by Hanebuth et al., 2003. ................................................................................................. 20 Figure 2.7. Climate of the Gulf of Thailand (after Encarta® 2006a). ....................................................... 21 Figure 2.8. Sketch of the water circulation at 5 m below the surface in the Gulf of Thailand, 1993-1994, deduced from the oceanographic data. The numbers indicate the distance in km travelled by a particle of water in 30 days (after Wattayakorn et al., 1998). ............................................. 23 Figure 2.9. Seasonal variations of vertical profiles of water temperature, salinity and density at the mouth of Gulf of Thailand (after Yanagi et al., 2001). ........................................................... 24 Figure 2.10. Schematic representation of seasonal variation in wind, heat flux through the sea surface, river discharge, stratification, density-driven currents and wind driven currents in the Gulf of Thailand (after Yanagi et al., 2001). ................................................................................... 25 Figure 2.11. Surface winds and surface sea currents during July and November in the South China Sea region. Thickness of arrows indicates the constancy of the predominant surface current directions (after Koompans, 1972). .................................................................................... 26 Figure 2.12. Graphical presentation of the coastal morphology of the south-east coast of Thailand (modified from Dheeradilok, 1995)..................................................................................... 27 Figure 2.13. (A) Geomorphology and sediment distribution of the Chao Phraya delta plain and the adjacent region. (B) Index map of Chao Phraya delta (after Tanabe et al., 2003)..................... 29 Figure 2.14. Geomorphology and Late Holocene evolution of the Mekong delta. The dashed lines indicate estimated location and age (in years from present) of palaeo-offshore break. (after Ta et al., 2002a). .......................................................................................................................... 31 Figure 3.1. Summary of the analogue system setup on-board MV Geo Surveyor (after Fugro, 2007). ......... 33 Figure 3.2. Starfix Reference Stations Coverage in Southeast Asia (after Fugro, 2007). ............................ 35 Figure 3.3. Generalised example of SBES system configuration. ............................................................. 37 Figure 3.4. Example of SBES data. Horizontal lines spacing: 2.5 m, Vertical lines spacing: 125 m (1 fix-25 m). Top channel: 200 kHz, Bottom channel: 38 kHz, Delay: 45 m/55 m. ................... 37 Figure 3.5. Example of processed MBES data. Approximate size of area 4x3 km. ..................................... 40 Figure 3.6. Example of SSS system configuration. ................................................................................ 42 Figure 3.7. Example of slant corrected SSS data. ................................................................................. 42 Figure 3.8. Simplified diagram of Pinger (SBP) system configuration....................................................... 43 Figure 3.9. Example of pinger (SBP) data. Horizontal lines spacing: 10 ms, ............................................. 45 Figure 3.10. Gravity corer model (after Fugro NV, 2001). ....................................................................... 46 Figure 4.1. Sub-bottom Profiler (SBP) data showing major units and reflectors of the central part of the Gulf of Thailand. Water depth about 70 m below MSL. .......................................................... 50 7 Figure 4.2. MBES Image, showing morphology of seabed in SW part of study area, seabed consists of very soft clay, including minor depressions and large-scale isolated pockmark “P1”. ................. 52 Figure 4.3. MBES image, showing seabed lithology at NE part of study area. The brown ridges on the right represent soft mud mounds, while the blue areas show isolated eyed pockmarks within the stiff silty sediment, covered by thin layer of soft clay. ..................................................... 53 Figure 4.4. Photography of sample DC 40 (Unit B) showing bioturbation structures developed over lateritic soil features. ........................................................................................................ 54 Figure 4.5. Grain size distribution of sample DC40 (Unit B).................................................................... 55 Figure 4.6. XRD patterns of the clay fraction from sample DC 40 (Unit B) (after Puchała et al., 2011). ....... 55 Figure 4.7. Grain-size distribution of sample DC 66B representing lower part of Unit A. ............................ 58 Figure 4.8. XRD patterns of the clay fraction from sample DC 66B representing lower part of Unit A (after Puchała et al., 2011). .............................................................................................. 58 Figure 4.9. MBES Image, showing pockmark cluster and other seabed features at and around coring location 37A, central Gulf of Thailand. ................................................................................ 60 Figure 4.10. Sub-bottom Profiler (SBP) image, showing shallow geology at and around pockmark cluster (37A), Gulf of Thailand. Red rectangles represent gravity corer sampling locations. .................. 61 Figure 4.11. Side Scan Sonar image showing pockmark cluster at and around coring location 37A. SSS range: 200 m per channel, frequency: 100 kHz. .................................................................. 62 Figure 4.12. Microscope images in cross-polarized light showing internal structure of ferruginous concretions: A, C, D, E & F - sample DC 37A (pockmark cluster 37A); B – sample DC 40 (Unit B). ......................................................................................................................... 63 Figure 4.13. Sub-bottom Profiler image, showing cross-section through the large-scale pockmark P1, Gulf Of Thailand. .................................................................................................................... 64 Figure 4.14. MBES image showing double pockmark P2, central Gulf of Thailand........................................ 65 Figure 4.15. Single channel seismic (SBP) image showing a cross-section through twin pockmark P2. The red rectangle indicates the location of gravity core DC 66B. .................................................. 65 Figure 4.16. Model of shallow geology of Gulf of Thailand central west basin, based on SBP data. ................ 66 Figure 4.17. Gravity core logs oriented in reference to sea level, showing detailed location of XRD sections and lab analysed samples. Cores GC 3 and GC 4 represent upper parts of the silty mounds, while cores GC 1 and GC 2 represents areas between the mounds, and core GC 5 represents the pockmark cluster area. ................................................................................ 68 Figure 4.18. X-radiograms and photographs of selected GC 3 core slabs (Unit A): RTG 5 (left), RTG 2 (middle), RTG 1 (right). .................................................................................................... 69 Figure 4.19. X-radiograms and photographs of selected GC 4 core slabs (Unit A): RTG 3 (left), RTG 4 (middle), RTG 6 (right). .................................................................................................... 70 Figure 4.20. X-radiogram and photograph of GC 1 core slab (Unit A/Unit B): RTG 7. .................................. 71 Figure 4.21. Grain size distribution curve (left) and table (right) of samples G3 and G5 from core GC 3, showing vertical variation of sediments within the upper part of a mud mound. Sediment type defined in the table is based on BS 5930:1999, Folk (1974), and Shepard (1954) classifications, respectively. .............................................................................................. 72 Figure 4.22. Grain size distribution curve (left) and table (right) of samples G7, G8, and G9 from core GC 4, showing vertical variation of sediments within the upper part of a mud mound. Sediment type defined in the table is based on BS 5930:1999, Folk (1974), and Shepard (1954) classifications respectively. ............................................................................................... 73 Figure 4.23. Grain size distribution curve (left) and table (right) of samples G10a, G10b, G14 and G12 from core GC 1, showing vertical variation of sediments on the boundary between Units A and B. Sediment type defined in the table is based on BS 5930:1999, Folk (1974), and Shepard (1954) classifications, respectively. ....................................................................... 74 8 Figure 4.24. Grain size distribution curve (left) and table (right) of sample G15 (core GC 2), and samples G16 and G19 (GC 5 core), showing sediment types of Unit B (G15, G19) and muddy matrix from pockmark cluster (R1). Sediment type defined in the table is based on BS 5930:1999, Folk (1974), and Shepard (1954) classifications, respectively. ............................................... 75 Figure 4.25. Grain size distribution triangle of all analysed samples assigned to corresponding units/seabed forms (based on Udden-Wentworth grain scale). ............................................... 75 Figure 4.26. Microscope images showing internal structure of carbonate concretions: A & D - sample G21b (Unit B); B – sample G18a (pockmark cluster); C – sample G20 (R1 surface). ................ 78 Figure 4.27. Plate showing foraminifers found within sediment samples. Identification of these fossils is presented in Table 4.6. ..................................................................................................... 83 Figure 4.28. Plate showing gastropods found within sediment samples. Identification of these fossils is presented in Table 4.6. ..................................................................................................... 84 Figure 4.29. Plate showing bivalves found within sediment samples. Identification of these fossils is presented in Table 4.6, ..................................................................................................... 85 Figure 4.30. Plate showing micro-molluscs found within sediment samples. Identification of these fossils is presented in Table 4.6. .................................................................................................. 86 Figure 4.31. Gravity core logs oriented with reference to sea level, showing locations and ages of radiocarbon dated samples................................................................................................ 88 Figure 4.32. SSS record showing pockmark cluster within central west basin. ............................................ 89 Figure 4.33. Model of shallow geology of south-west margin of Gulf of Thailand, based on SBP data and drilling log. ...................................................................................................................... 91 Figure 4.34. Shallow seismic cross-section 170 km from northern coastline. Central basin of Gulf of Thailand. ........................................................................................................................ 93 Figure 4.35. Presence of mud mounds and ridges along North-South profile across Gulf of Thailand. ............ 95 Figure 4.36. Shallow seismic cross-section 450 km from northern coastline showing a growth fault within Unit C. Central basin of Gulf of Thailand. ............................................................................. 96 Figure 4.37. Shallow seismic cross-section 600 km from northern coastline showing a gas chimney below a palaeochannel displaying symmetrical fill. Central south basin of Gulf of Thailand. ................. 96 Figure 4.38. Borehole logs of the central basin, Gulf of Thailand – part I. .................................................. 99 Figure 4.39. Borehole logs of the central basin, Gulf of Thailand – part II. ............................................... 100 Figure 4.40. Shallow geological model for the central basin, Gulf of Thailand, based on SBP data (line SL-06) and drilling logs. The inclined stratification of Unit C4 represents deltaic clinoforms. .................................................................................................................... 101 Figure 4.41. Shallow geological model for the central basin, Gulf of Thailand, based on SBP data (line SL-09) and drilling log. ............................................................................................ 102 Figure 5.1. Distribution pattern based on Average Nearest Neighbour method (top) and Standard Directional Ellipse (bottom) of unit pockmarks, calculated in ArcGIS software. ....................... 111 Figure 5.2. Distribution pattern based on Average Nearest Neighbour method (top) and Standard Directional Ellipse (bottom) of eyed pockmarks, calculated in ArcGIS software. ..................... 112 Figure 5.3. Model of pockmark cluster 37A. Central Basin, Gulf of Thailand. .......................................... 114 Figure 5.4. Orientation of longest axis of unit pockmarks (violet roses), seabed depressions (red roses) and mud ridges (green roses) indicating dominant direction of bottom current within the Gulf of Thailand, against predominant surface wind direction (The wind directions based on Koompans 1972). .......................................................................................................... 116 Figure 5.5. Diagram showing how fluidization velocity depends on grain size and consolidation of sediment (from Hovland and Judd 1988, redrawn from Lowe 1975). .................................... 117 9 Figure 5.6. Spatial distribution of mud mounds and ridges in the Gulf of Thailand. The magenta hatch indicates areas outside survey corridors where occurrence of mud mounds and ridges is expected. ...................................................................................................................... 120 Figure 5.7. Evolutionary stages of mud mounds and ridges formation in the Gulf of Thailand, central basin. A - elongated pockmarks within soft clay - stage 2; B - elongated depression within soft clay - stage 3; C - mud mounds and ridges - stage 4; D – flay stiff seabed with eyed pockmarks– stage 5). ..................................................................................................... 126 Figure 5.8. Model of seabed morphology development in the Gulf of Thailand, central basin. ................... 127 10 LIST OF TABLES Table 2.1. Temperature values for selected locations around Gulf of Thailand (after Encarta®, 2006b). ..... 21 Table 2.2. Precipitation values for selected locations around Gulf of Thailand (after Encarta® 2006b). ....... 21 Table 3.1. List of utilised methods and survey coverage of each study area. ........................................... 48 Table 4.1. Shallow stratigraphy of the Gulf of Thailand and correlation with the adjacent areas. ............... 51 Table 4.2. Radiocarbon ages of plant matter and shelly fauna (after Puchała et al., 2011)........................ 56 Table 4.3. Lithology of gravity core samples at and around pockmark cluster 37A. Detailed coring locations are presented on Figure 4.10. .............................................................................. 61 Table 4.4. Statistical parameters of the grain size of analysed sediment samples. Methodology of calculations after Folk (1974). ........................................................................................... 76 Table 4.5. Interpretation of XRD analyses of carbonate-ferruginous concretions. Detailed locations of the samples are presented on Figure 4.17. .......................................................................... 77 Table 4.6. List of fossils identified within the sediment samples. ........................................................... 80 Table 4.7. Radiocarbon ages of shells and microfauna of the central-west basin. ..................................... 87 Table 5.8. Orientation of elongated pockmarks and seabed depressions indicating dominant direction of bottom current, southern central basin, Gulf of Thailand. .................................................... 117 Table 5.9. Summarised results of mud mounds/ridges and depressions/subaqueous channels orientation for three different areas of the Gulf of Thailand. ................................................ 121 11 1 INTRODUCTION AND PURPOSE OF RESEARCH The Gulf of Thailand is a shallow epicontinental sea. The location of the gulf within the monsoon winds zone, lack of major fluvial sediment supply, dominant muddy sedimentation, and the unique pattern of marine currents, have resulted in a specific seabed morphology. Seabed sediments consist of Holocene silts and clays, which form a regular continuous mantle within shallower parts of the basin, but in deeper waters these youngest sediments form elongated mounds arranged parallel to the axis of the basin. This morphology, which is globally unique, is widespread within the Gulf of Thailand in water depths in excess of 50 m. It covers an area of tens of thousands of square kilometres. The presence of these features indicates that there are unusual geological and oceanographic conditions within the basin. The morphology, distribution and origin of mud mounds have not been examined in detail so far and this forms one of the main objectives of this study. Other interesting features common on the Gulf of Thailand continental shelf are pockmarks. They are recognised as cone-shaped circular or elliptical depressions related to processes of dehydration and fluid escape from the sediments. Although pockmarks themselves are rather well known, the author’s interest focused on the sediments, which occur in close proximity to major clusters of pockmarks. These sediments are a mixture of mud, carbonate shells and goethite concretions. This portion of research has been largely centred on erosive processes associated with early Holocene marine transgression, which resulted in reworking of the bottom sediments in coarse-grained lags, and subsequent processes that prevented any significant deposition of marine mud within these areas. The study also aims to characterise the Late Quaternary seabed succession within the Gulf of Thailand with particular attention given to palaeoenvironmental reconstruction of the Upper Pleistocene and Holocene basin-fill increments. The three main hypotheses tested in this dissertation are as follows: An erosive unconformity between the two uppermost stratigraphic horizons of the Gulf of Thailand, i.e., uppermost soft marine muds (Bangkok Clay Formation) and underlying stiff silt (Stiff Clay Member) is a ravinement surface associated with early Holocene marine transgression. The unique seabed morphology within the deeper part of sublittoral zone of the Gulf of Thailand, characterised by pattern of elongated soft mud mounds and depressions as well as numerous pockmarks, is a result of a combination of fluid seepage, sediment dehydration processes, and marine current influence. The mud mounds and ridges are an intermediate stage of seabed morphological development in a specific erosive marine current regime, where the original stage 12 is a non-eroded soft mud mantle and the final stage is a flat seabed with all soft mud cover washed away. A wide spectrum of methods has been utilised to achieve the goals of the study. These methods mainly included geophysical data acquired during marine acoustic surveys such as: side scan sonar imagery, sub-bottom profiling and swathe bathymetry. The geoacoustic records have been ground-truthed with numerous gravity cores and supported by geotechnical drilling logs. Selected samples were subject to laboratory tests including X-radiography of sediment slabs, radiocarbon dating, sieve-pipette analyses, XRay diffraction and thermal (TGA, DTA and DTGA) techniques, analyses of microscopic images of thin sections of iron-oxide concretions, and identification of extracted fossils. All of these methods are described in detail within Chapter 3. The continental shelf of the Gulf of Thailand has been chosen as an investigation area. Desk studies cover the whole Gulf of Thailand area including islands, river deltas and coastal areas. The field investigation covered selected parts of the basin, mostly within central part of the Gulf of Thailand and was limited to the continental shelf only. 13 2 GULF OF THAILAND – BASIN MORPHOLOGY AND STRUCTURAL, OCEANOGRAPHIC AND CLIMATIC SETTINGS 2.1 Location and Topography The Gulf of Thailand, also known as the Gulf of Siam, is an inlet of the South China Sea (Figure 2.1). It is located between the Malay Peninsula and the Southeast Asian mainland. Geographically, the gulf is located between 5° and 14° latitude (N), and from 99° to 105° longitude (E). The most northern fragment of the basin, located near the Chao Phraya River delta, is called the Bight of Bangkok. The Chao Phraya River is the largest river flowing directly into the gulf. The Mekong River, the biggest river in the region, flows into the South China Sea, east of the Gulf of Thailand; however, due to the area’s water circulation patterns, it is an important source of fresh water for the Gulf of Thailand (Stansfield and Garrett, 1997). The Gulf of Thailand is a slightly elongated basin, with an axis oriented SE-NW. The length of the inlet along the axis is around 750 km and the width across its mouth is approximately 360 km (Encarta, 2006a). It is a shallow shelf reservoir with a maximum water depth of 86 m (Srisuksawad et al., 1997). Two ridges separate the deeper central Gulf from the South China Sea. The south-western ridge, located about 100 km from Cape Camau, is less than 25 m deep. The north-east ridge is approximately 150 km offshore Kota Bahru, water depth over this feature is less than 50 m. The water depth within the narrow channel between these two ridges reaches 67 m (Robinson, 1974). Six separate survey areas, located across the Gulf of Thailand basin, were subject to the detailed analyses presented in this dissertation. These areas, presented on Figure 2.1, are considered to be representative of seafloor conditions in each section of the basin. The central south basin exhibits the greatest water depths within the Gulf of Thailand as well as an area of palaeo-prodelta related to the Kelantan River. The west basin is located approximately 100 km north of Ko Samui Island within the central-west Inner Gulf. The western margin of the Gulf of Thailand is represented by a site situated approximately 35 km north-east of the city of Songkhla. This area is characterised by a generally flat part of the basin slope, approximately 20 km off the west coast. A north-south cross-section is provided by an almost 600 km long and 150 m wide survey corridor passing across the Gulf of Thailand. This route starts at the northern coastal town of Sattahip and runs south to the central south part of the Gulf. The route shows a north-south profile through the seafloor of the northern and central part of the 14 basin. At its northern extreme, the corridor begins in water depths of around 15 m and ends within the deepest part of the basin at a water depth beyond 80 m. Figure 2.1. Location of study area. The Gulf of Thailand. The survey area within the gulf mouth is situated offshore Vietnam on the most southern part of the Gulf of Thailand, close to the boundary with the South China Sea. 15 The last analysed area is situated offshore Cambodia within the central part of the Gulf of Thailand. The area represents a deep water section of the basin near its axis and is located around 200 km from the nearest shoreline. 2.2 Geological Background The Gulf of Thailand is located between the Thai-Malay Peninsula, the Indochina Massif and the eastern part of the South China Sea. It is a typical shallow epicontinental sea, which is underlain by twelve sedimentary Cainozoic sub-basins separated by northtrending linear ridges (Pigott and Sattayarak, 1993). This succession rests unconformably on a pre-Palaeogene basement (Jardine, 1997). The Quaternary history of the Gulf of Thailand was largely controlled by frequent sea level changes related to glacial and interglacial periods. Upper Pleistocene to Holocene marine sediments known from the Lower Central Plain and Chao Phraya delta areas are the Holocene Bangkok Clay Formation (Cox, 1968, Moh et al., 1969, fide Tanabe et al., 2003; Sinsakul, 2000) unconformably overlying the Upper Pleistocene Stiff Clay Member (Rau and Nutalaya, 1983; Tanabe et al., 2003). A map illustrating the regional geology and tectonics of the study area is shown on Figure 2.2. Figure 2.2. Regional tectonic map of western Southeast Asia (after Watcharanantakul and Morley, 2000, modified from: Packham, 1996; Leloup et al., 1995; Oudom-Ugsorn et al., 1986, Polchan and Sattayarak, 1989). 16 2.3 Structural Setting 2.3.1 Pre-Quaternary Stratigraphy The geology of the pre-Tertiary basement has been established mostly on the basis of onshore studies (Suensilpong et al., 1982, fide Watcharanantakul and Morley, 2000). The basement of the Pattani Basin, which is located in the central part of the Gulf of Thailand, is most likely composed of the following elements: Precambrian crystalline basement; Lower Palaeozoic highly folded and foliated green schist grade metasediments (Suensilpong et al., 1982, fide Watcharanantakul and Morley, 2000) including preDevonian mafic-ultramafic intrusions (Tan 1996); Permo-Triassic granitic belt. The Carboniferous, Permo-Triassic and Mesozoic clastics and carbonates (Ratburi) have been noted in the southern part of Thailand Peninsula and Chumpon Basin in the western part of the Gulf (Heward et al., 2000, fide Watcharanantakul and Morley 2000); Late Cretaceous and Early Tertiary granites, which flank and source the sediments of the Pattani Basin (Hutchison, 1983; Watcharanantakul and Morley, 2000). The stratigraphy of the Cainozoic sequences in the Gulf of Thailand is presented on Figure 2.3. Figure 2.3. Stratigraphy of the Cainozoic sequences in the Gulf of Thailand (after Watcharanantakul and Morley, 2000, modified from Lian and Bradley, 1986). 17 The thickness of Cainozoic sediments in the Gulf of Thailand varies from approximately 8.5 km in the deepest parts of the Pattani Basin to less than 300 m in areas where shallow subcropping, pre-Tertiary highs occur (Watcharanantakul and Morley, 2000). 2.3.2 Upper-Pleistocene and Holocene Sediments The Pleistocene sequence on the eastern coast of the Thai Peninsula consists of sands, clays, gravels, and several lateritic layers, interpreted as colluvial and fluvial in origin (Dheeradilok, 1995). The sediments of the Lower Central Plain are composed of an Upper Pleistocene Stiff Clay Member (Rau and Nutalaya, 1983; Dheeradilok, 1995; Sinsakul, 2000) and Holocene marine clays, called the Bangkok Clay Formation (Cox, 1968, Moh et al., 1969, fide Tanabe et al., 2003; Dheeradilok, 1995; Sinsakul, 2000). An erosional unconformity between these units marks the Pleistocene/Holocene boundary. The Holocene sediments of Chao Phraya Delta are divided into lower transgressive peaty (mangrove swamp) sediments and upper regressive deltaic sediments (Somboon, 1988, Somboon and Thiramongkol, 1992, Songtham et al., 2000, Woodroffe, 2000, fide Tanabe et al., 2003; Sinsakul, 2000). The composite stratigraphic succession of the Lower Central Plain is presented on Figure 2.5. The correlation of Quaternary deposits in the coastal part of Thailand is shown in Figure 2.4. The Late Pleistocene and Holocene history of the Gulf of Thailand is determined by climate changes and marine transgression, related to the close of the glacial epoch. The sea-level curve for the Gulf of Thailand during the past 140 000 years is shown on Figure 2.6. The Pleistocene/Holocene transition is marked by expansion of arboreal vegetation in response to increased precipitation and climate warming (Kealhofer, 2002; White et al., 2004). The Holocene marine transgression resulted in a gradual flooding of the gulf area. The Stiff Clay Member representing Pleistocene deposits, is unconformably covered by several peat lenses, intertidal and marine clay horizons (Sinsakul, 2000). The peat lenses originated from flooded forests and mangrove swamps. Intertidal and marine clays mark respectively the middle and final stages of marine transgression. The Middle and Late Holocene period, contrary to rapid post-glacial sea level changes, is characterised by the relatively low amplitude variations. Water level within Gulf of Thailand remained quite steady, with the highstand maximum at 8000-7000 years BP (Tanabe et al., 2003). Slight variations in vegetation occurred due to monsoon flow changes and human agricultural activity (White et al., 2004). 18 Central Plain Lower Central Top soil alluvial HOLOCENE Meander belt Subtidal shell & peat (4-5 kyr) Soft Clay Member Flood plain Terrace I Intertidal Southern Thai Peninsula (Songkhla Lake Basin) (marine) Bangkok Clay Fm Upper Central Fluvial/Recent beach deposit Channel/Lacustrine Old beach ridge/Tidalflat peat (4.3-6.6 kyr) Flood plain Estuarine Stiff Clay Member Alluvial fan Fm MIDDLE LOW PLEISTOCENE ? Kam Phoeng Phet Fm ? Redsoil Fm Lacustrine marl Fm ? Phra Pradaeng Member Fluviatile coarse sand and gravel with remains of (Terrace III) Ping Fm Fluviatile deposit Phra Nokhan Member Chao Phraya Fm (non-marine) UPPER Deltaic sand/silt Laterite Pediments / gravel beds (Terraces) Samut Prakan Member Laterite ? PLIOCENE Plio-Miocene Weathered Older rocks Figure 2.4. Correlation of Quaternary deposits in the coastal parts of Thailand (modified from Dheeradilok, 1995). 19 Figure 2.5. Composite stratigraphic succession of unconsolidated sediments in the Lower Central Plain (not to scale) (after Sinsakul, 2000). Figure 2.6. Sea-level curve for the Gulf of Thailand and Sunda Shelf during the past 140 kyr based on oxygen isotope (Shackleton, 1987) and coral reef records (Chappell et al., 1996). Compiled by Hanebuth et al., 2003. 2.4 Climate The Gulf of Thailand has a moist, tropical climate with two monsoonal winds: the north-east during mid October to March and the south-west during May to September (Thampanya et al., 2006). In the southern part of the Gulf of Thailand, the climate is more humid, with average precipitation around 2000 mm (Table 2.2, Songkhla). The northern part is more affected by seasonal changes due to monsoons; thus, dry and wet 20 seasons occur. The winds from the south-west produce a rainy season from mid April to mid October (Encarta, 2006c). The average precipitation in the northern part of the inlet is around 1500 mm (Table 2.2, Bangkok). There are no major seasonal temperature changes within Gulf of Thailand. Slightly higher temperatures are observed when area is under the influence of the south-west winds (Encarta, 2006b). The average annual air temperature ranges between 27°C and 28°C (Table 2.1). A map showing climatic zones in South East Asia is presented in Figure 2.7. Figure 2.7. Climate of the Gulf of Thailand (after Encarta® 2006a). Table 2.1. 2006b). Temperature values for selected locations around Gulf of Thailand (after Encarta®, Bangkok Songkhla Ho Chi Minh City 28 28 27 January [°C] 21-32 24-30 21-32 July [°C] 25-33 24-33 24-31 Annual average [°C] Table 2.2. 2006b). Precipitation values for selected locations around Gulf of Thailand (after Encarta® Annual average [mm] No of days with precipitation Bangkok Songkhla Ho Chi Minh City 1498 2035 1861 100 124 134 21 Two main rivers deliver fresh water into Gulf of Thailand; the Chao Phraya and Mekong. The annual peaks of water runoff occur between August and November. The Chao Phraya River mouth is located in northern part of the gulf. The area drained by this river is relatively small, thus, the influence of Chao Phraya River is rather local and dilution of sea water is restricted to northern coastal areas of the basin. The Mekong River, despite its delta being located outside the Gulf of Thailand, influences coastal waters offshore Vietnam. The Mekong is one of the world’s largest rivers carrying large amounts of fresh water derived both from rainfall and Tibetan snow melt. The river flow is subject to strong seasonal variation. The peak rate of flow in August (rainy season) is 16-19 times greater than during the dry season (Robinson, 1974). The climate of the Gulf of Thailand area has changed since the Last Glacial Maximum. During the Late Pleistocene, the climate was drier and cooler than that at present, which resulted in a wide range of vegetation types from dense forests (pine/oak dominated) to savannas (White et al., 2004). The Pleistocene to Holocene transition coincided with a change in the climate towards moist and hot. 2.4.1 Monsoons The Monsoon is the wind that changes direction with the change of the season (Encarta® 2006d). It affects mostly the interior of Asia and the Indian Ocean. The wind usually blows from the south-west from May to September and from the north-east from November to February (Figure 2.11). The south-west wind, also called the “summer monsoon” blows across the Indian Ocean and Bay of Bengal carrying warm and moist air. It causes a rainy season in the Gulf of Thailand between July and October. The dry season is caused by the north-east “winter monsoon” and occurs when cold and dry air flows from Asian interior (China) towards the Indian Ocean. This period is one of variable moderate winds over the Gulf and mild temperatures on land. The direction of the north-east monsoon is relatively steady over the South China Sea; however, it may be variable within Gulf of Thailand (Robinson, 1974). These two seasons are divided by two periods of transition between the opposing monsoons, one of two months duration in March-April and the second one in October (Robinson, 1974). 2.5 Water Circulation Based on oceanographic data collected during the NAGA Expedition (Robinson, 1974), the Gulf of Thailand is a classical two-layered, shallow-water estuary. The upper layer is composed of low salinity water, diluted due to precipitation and fresh water runoff from rivers. The lower layer comprises high-salinity, relatively cold water delivered 22 from the South China Sea. This high salinity layer usually occurs within the deepest areas of the gulf in water depths greater than 50 m below sea level. Superimposed on this twolayered system is water circulation generated by wind-driven currents related to monsoons and tides. Interactions between the variable winds, tidal currents, fresh water runoff and excessive precipitation may locally cause anomalies, which result in upwelling of cold salty water and sinking of warm, low-salinity and highly oxygenated waters. The general circulation and physical properties of the Gulf’s water undergo large seasonal as well as short-period variations. The water circulation in the Outer Gulf is principally related to the South China Sea; consequently, the circulation in the Inner Gulf is related to the Outer Gulf (Figure 2.8). Figure 2.8. Sketch of the water circulation at 5 m below the surface in the Gulf of Thailand, 19931994, deduced from the oceanographic data. The numbers indicate the distance in km travelled by a particle of water in 30 days (after Wattayakorn et al., 1998). 2.5.1 Tides Tidal motion in the South China Sea and Gulf of Thailand is largely maintained by the energy flux from the Pacific Ocean through the Luzon Strait (Fang et al., 1999). The phase of semi-diurnal tides (M2 and S2) propagates clockwise in the central part of Gulf of Thailand, opposite to the phase of diurnal tides (K1, O1, P1), which are counter clockwise (Yanagi and Takao, 1998). The tidal cycle in the Gulf of Thailand consists of irregular tides with average amplitude of 2.7 m. Regular daily tides are observed on the eastern coast of the gulf. The amplitude of tides increases from 1.5 to 3.5 m towards the central part of the gulf, where irregular daily tides with amplitude of about 4 m can be observed (Loi, 1965; Gorshkov et al., 1974; Latypov 2003). Tidal currents in the Gulf of Thailand may exceed one knot (~0.5 m/s) (Srisuksawad el al., 1997). 23 2.5.2 Water Temperature and Salinity The average salinity of the Gulf of Thailand’s water varies between 30.06 and 31.26‰. Water salinity changes according to the season. In the rainy season, salinity within the inner part of the basin may drop down to 28‰ (Emery and Niino, 1963; Latypov, 1995, fide Latypov, 2003). The saline water migrates from the South China Sea, while the main sources of fresh water are river discharges. Lower seawater salinities were noted on eastern side of the Lower and Middle Gulf of Thailand (Srisuksawad et al., 1997), most probably related to Mekong River or/and Cambodian river input. The temperature of the Gulf of Thailand surficial waters varies from 24 to 30°C, being at the maximum during May–August and minimum in November–February (Pham, 1985 fide Latypov, 2003). During seasons when winds are relatively light, stratification develops due to significant heating of the sea surface, (Yanagi et al., 2001). The stratification is additionally influenced by salinity changes caused by seasonal river discharge variation (Figure 2.9). Figure 2.9. Seasonal variations of vertical profiles of water temperature, salinity and density at the mouth of Gulf of Thailand (after Yanagi et al., 2001). A schematic representation of seasonal variation in wind, heat flux through the sea surface, river discharge, stratification, density-driven currents and wind driven currents in the Gulf of Thailand is presented on Figure 2.10. The greatest stratification and estuarine circulation in the Gulf of Thailand takes place in April. The stratification develops due to the occurrence of strong surface heating and relatively light winds, which results in formation of density driven currents. The estuarine circulation is intensified by surface Ekman transport related to the south-west monsoon. Moderate stratification remains until September and is maintained by high levels of river discharge and moderate heat flux. Cold and relatively undisturbed water masses remain in the central part of the Gulf of Thailand, where the water is deep and amplitudes of tidal currents are 24 low. In December-January, the stratification is absent due to the disruptive influence of the cold, strong, north-east monsoon, (Yanagi et al., 2001). Figure 2.10. Schematic representation of seasonal variation in wind, heat flux through the sea surface, river discharge, stratification, density-driven currents and wind driven currents in the Gulf of Thailand (after Yanagi et al., 2001). 2.6 Water Currents Studies performed by Wattayakorn and others (1998) illustrate the domination of strong, predicable, generally shore-parallel tidal currents within the Gulf of Thailand. The mean currents, monthly averaged, are usually less than 0.07 m/s in speed (max 0.12 m/s). The strongest net currents occur near the centre of the Gulf during the peak of monsoon season, while weak and variable currents predominate during the rest of the year. The north-east monsoon generally sets in during November and affects the Gulf of Thailand at full force during December and January. In May, the prevailing wind direction changes towards the south and in subsequent months the wind blows mainly from south to south-west (Koompans, 1972). The graphical presentation of winds, and related surface currents, is shown on Figure 2.11. Monsoons have a significant influence on surface currents. During the south-west monsoon season, the surface current moves clockwise and during the north-east monsoon season it moves counter clockwise. According to Robinson (1974), the windinduced motion appears to be a major component of the circulation in the Gulf of Thailand. The monsoon winds over the Gulf, however, are not simply north-east or south-west, but vary widely around these primary directions making interpretation of the results of in situ winds more difficult. The vertical wind-induced water motions that affect the density structure of the water column are costal upwelling, coastal sinking, open sea convergence and divergence (Robinson, 1974). 25 Figure 2.11. Surface winds and surface sea currents during July and November in the South China Sea region. Thickness of arrows indicates the constancy of the predominant surface current directions (after Koompans, 1972). 2.7 Seabed Morphology, Sedimentation and Environment The south-east coast of Thailand is characterised by a broad coastal plain with long and wide mainland beaches of sand and dunes. Tidal flats with large lagoons and sand spits are also common. Close the mouth of the Tiger River (Surat Thani Province), prodeltaic deposition occurs. High sediment supply resulted in Holocene coastal progradation. The sandy material delivered by rivers is distributed by a northerly-directed 26 longshore current (Figure 2.12). On the other hand, large stretches of the coastline have been found to be subject to erosion (Dheeradilok, 1995; Thampanya el al., 2006). The northern coast of the Gulf of Thailand is characterised by a coastal floodplain landscape. The floodplain, called the Lower Central Plain, has been formed by fluvial and deltaic deposition by the Chao Phraya, Mae Klong, Mae Nam Thachin and Bang Pakong rivers. Coastal sediments of the northern gulf are a mixture of mostly fluvial, muddominated material and minor amounts of sand transported from the south by littoral currents. Landscape features also include tidal flats and beaches (Dheeradilok, 1995). Ko Samui Island (coral reefs) Tiger River Delta (mangrove swamps) Figure 2.12. Graphical presentation of the coastal morphology of the south-east coast of Thailand (modified from Dheeradilok, 1995) 27 In the central part of the Gulf of Thailand, seabed morphology is dominated by a flat central depression. The maximum depth of this depression is 86 m below mean sea level. Buried palaeo-river valleys radiating into the central depression may be observed in the seabed morphology (Srisuksawad et al., 1997). The bottom sediments within the Gulf of Thailand are dominated by clay and sandy clay. Locally, especially in nearshore areas, sands and clayey sands occur. A clay deposition was reported also inshore near major river mouths in the upper Gulf of Thailand (Emery and Niino, 1963, fide Srisuksawad et al., 1997). 2.7.1 Accumulation Rates and Sediment Mixing Sediment accumulation rates measured by Srisuksawad and others (1997) amount 270-490 mg/cm2/year in the upper Gulf of Thailand and 64-190 mg/cm2/year in the central basin. The influence of sediment mixing and re-suspension by storms and bottom currents seems to be more significant in the upper Gulf of Thailand. Biological mixing and trawling activity may also cause higher mixing coefficients. Mixing from storms and currents are probably the most significant factors in the southern Gulf. 2.7.2 River Deltas Chao Phraya Delta The Chao Phraya is the biggest river flowing into Gulf of Thailand. It is located on the northern coast of the Gulf. The mouth of the Chao Phraya together with its tributaries, Mae Klong, Tha Chin and Bang Pakong rivers forms the Chao Phraya delta system. Detailed studies of the structure and evolution of the delta have been published by Sinsakul (2000) and Tanabe and others (2003). The geomorphology and sediment distribution of the Chao Phraya delta is shown on Figure 2.13. The coast on the Chao Phraya delta is a low-energy depositional environment. From the north towards the Gulf of Thailand, the entire system has been subdivided into delta plain, tidal flat, river mouth flat, delta front, and pro-delta portions (Tanabe et al., 2003). Clay is the dominant sediment on the floodplain. Along river channels mainly silt or sandy clay occurs. Well-sorted medium sand forms a 30 km-long beach ridge, which is located within the delta plain about 3 m above Mean Sea Level (Rau and Nutalaya 1983; Somboon and Thiramongkol 1992, fide Tanabe et al., 2003). A significant part of the delta plain is covered by mangrove forests, which form a 2030 km wide belt along the coast (Somboon 1988, Woodroffe, 2000, fide Tanabe et al., 2003). A 1-5 km wide zone of tidal (mud) flats occurs between the 28 mangroves and the delta front. This area lies approximately 1 m below mean sea level (MSL) and is exposed during low tide (Royal Thai Navy, 1995, 1996, fide Tanabe et al., 2003). Figure 2.13. (A) Geomorphology and sediment distribution of the Chao Phraya delta plain and the adjacent region. (B) Index map of Chao Phraya delta (after Tanabe et al., 2003). The boundary between the delta front and pro-delta is defined by the slope gradient break point, which is located approximately along the 11-13 m below MSL isobath (Tanabe et al., 2003). The delta front and pro-delta consists mostly of clay (Srisuksawad et al., 1997). 29 Kelantan River Delta The Kelantan is 355 km long and is the third largest river in the Malay Peninsula. It is located on the south-western coast of the Gulf of Thailand near the entrance of the gulf (Figure 2.1). The Kalantan Delta was described by Koompans (1972). The delta debouches into the sea through two main channels. The mouth of the Kelantan River has gradually shifted to the west due to the influence of westward orientated beach drift generated by the north-east monsoon. Coastal erosion and river sedimentation within the delta are in a delicate balance. Retreat of the coastline is counterbalanced by accretion of land elsewhere. The river sedimentation within its mouth is mostly composed of clay with a silt fraction. A sand fraction is delivered occasionally during flooding events. The river’s sediments are mostly laid down within the coastal plain, causing its rapid seaward growth. Mekong River Delta The Mekong is one of the largest rivers in the world with a length of 4200 km, drainage area of 0.79 x 106 km2 and an annual water discharge of 470 km 3 (Wolanski et al., 1996). The Mekong River forms the delta that is located in the most southern part of Vietnam near the entrance of the Gulf of Thailand into the South China Sea. The present Mekong Delta system has two major distributary channels, both discharging directly into the South China Sea. However, presently no major channels supply the Gulf of Thailand, the Holocene history of the Mekong Delta described by Ta et al., 2002a shows delta progradation of about 200 km during the last 6 kyr. This means that during the Middle Holocene the Mekong River was discharging waters into both the South China Sea and the Gulf of Thailand. The water entering the Gulf of Thailand was flowing via a palaeochannel located within the western part of the delta; north of the Camau Peninsula (Figure 2.14). The present Mekong Delta Plain can be divided into two parts: an upper delta plain influenced by fluvial processes and a lower delta plain where marine processes dominate (Ta et al., 2002a). The lower delta plain is characterised by numerous beach ridges and inter-ridge swamps. The most common environments within the sub-aerial part of the delta are mangroves, beach ridges and tidal flats. The Late Holocene evolution of the Mekong River Delta shows the change of the delta system from tide-dominated to mixed, tide and wave influenced (Ta et al., 2002b). The dominant type of suspended sediment carried by Mekong River is fine silt, while the clay fractions do not exceed 30% by volume (Wolanski et al., 1996). At least 95% of that sediment is exported to the sea and deposited within 20 km of the coast (Anikiyev et al., 1986, fide Wolanski et al., 1996). 30 Figure 2.14. Geomorphology and Late Holocene evolution of the Mekong delta. The dashed lines indicate estimated location and age (in years from present) of palaeo-offshore break. (after Ta et al., 2002a). 2.7.3 Mangroves Mangroves are trees or shrubs that grow in shallow and muddy salt or brackish water, especially along quiet shorelines, deltas or estuaries in tropical regions, where they collectively form mangrove swamps (Encarta, 2006e). A significant concentration of mangroves occurs on the Chao Phraya delta plain (Figure 2.13). Mangroves occupy approximately 10% of the south-eastern coast of Thailand (Thampanya et al., 2006). The development of mangroves is related mostly to the positioning of river mouths and sheltered bays. The spread of mangrove forests can be fast and has been measured to be as high as 140-1200 m over the period of 31 years around some river mouths (Thampanya et al., 2006). 31 Mangrove forests, which provide significant protection against coastal erosion and form an important habitat for countless aquatic species, have been seriously reduced in spread during last decades (Cheevaporn and Menasveta, 2003). The existing mangrove forest area in Thailand has decreased more than 50% in the past 32 years (Kongsangchai, 1995, fide Cheevaporn and Menasveta, 2003). Historically, mangroves occupied large bands of Southeast Asia’s coasts (Rao, 1986, Aksornkoae, 1993, fide Thampanya et al., 2006). 2.7.4 Coral Reefs The reefs of the Gulf of Thailand develop essentially around archipelagos and single islands. Most of the islands are situated in the eastern part of the Gulf. The islands are relatively high mountainous plateaus with steep, marine-cut rockfall slopes. The underwater slopes consist of boulder–block pavements in the shallowest parts, stony and gravel-prone alluvial bedrocks in the middle zone, and sandy and coral deposits with a high content of organogenic detritus in the deepest parts (Latypov, 2003). 32 3 METHODOLOGY Results presented in this dissertation are mainly based on a combination of geophysical and geological data acquired by Fugro’s designated survey vessels from the Gulf of Thailand area during the period 2006-2007. Apart from these marine geophysical records, additional utilised data include borehole logs, results of laboratory analyses of gravity core samples, and Fugro’s geophysical survey reports prepared before 2006. The author of this paper has been involved in data acquisition and interpretation as a field geophysicist. Selected sediment samples, collected by gravity corer, have been subjected to laboratory analyses carried out or assigned by the author, who is also responsible for the final data interpretation. 3.1 Survey Vessels Two Fugro survey vessels, Geo Surveyor and Geo Eastern operated by Fugro Survey Pte Ltd have been used during geophysical data acquisition and gravity coring. Both vessels were equipped with a Differential Global Positioning System (DGPS), Single Beam Echo sounder (SBES), Multi Beam Echo sounder (MBES), Pinger Sub-bottom Profiler (SBP), Side Scan Sonar (SSS), Ultra-short baseline (USBL) system for SSS fish positioning, Sound Velocity Probe (SVP), and a 3 m barrel Gravity Corer. A brief summary of the geophysical and geotechnical systems used during survey operations is given below (Figure 3.1). Figure 3.1. Summary of the analogue system setup on-board MV Geo Surveyor (after Fugro, 2007). 33 3.2 Positioning and Navigation A Fugro Starfix HP DGPS with corrections received via Starfix Spot Multiple Reference Station DGPS and the APSat satellite system was used as the primary positioning system. Fugro’s Starfix MRDGPS system was used as a secondary system and for DGPS data QC. Position of GPS antennas as well as all mounted survey sensors were calculated based on measured offsets in relation to the vessel Central Reference Point (CRP). DGPS verifications have been performed at the start of every project. 3.2.1 Fugro Starfix HP DGPS The Starfix HP DGPS is a dual frequency GPS augmentation service that provides positioning for marine users. Accuracy for the Starfix-HP service is 20 cm, 95% for the North and East components with a vertical accuracy of 30 cm, 95% (Fugro/Starfix.HP, 2003). By using dual frequency GPS receivers Starfix-HP can measure the true ionosphere at the reference and user locations, substantially eliminating this error. Using these iono-free measurements with information contained in the receiver carrier phase data, it is possible to create wide area positioning results of unmatched accuracy and performance. 3.2.2 Fugro Starfix MRDGPS A Starfix is a multiple reference station Differential GPS (DGPS) system that uses the Inmarsat communication satellites as the downlinks for the correction data from each reference station and a Multi Reference Differential Global Positioning System (MRDGPS) position solution. The primary function of MRDGPS is to use all available DGPS data collected and computed at the mobile unit to provide the user with real-time indication of the position performance. Prior to the final position computation, all pseudo ranges are statistically tested for gross errors using the w-test method. Each observation is carefully corrected and weighted considering satellite elevation, age of differential corrections, distances from the reference station (Figure 3.2) and rate of change correction. 3.2.3 Dynamic Heading Reference System Two Brown Meridian gyrocompasses were installed to provide vessel heading input to the Starfix.Seis systems and to enable real time calculation of vessel offsets. Gyro calibrations have been performed at the start of every project. These portable survey gyros are designed specifically for survey type operations allowing easy interfacing to navigation computers and have a digital output of the bearing to 0.17 of a degree. 34 Figure 3.2. Starfix Reference Stations Coverage in Southeast Asia (after Fugro, 2007). 3.2.4 Underwater Positioning A Sonardyne Ultra Short Baseline (USBL) system was used to accurately track the towed side scan sonar fish position. The Sonardyne USBL system consisted of a pole mounted transceiver, pitch and roll sensor and transponder beacons. The system measured ranges and bearings between the transducer and the transponder beacon attached to the sonar fish. Data were transmitted via an RS232 line to the navigation computer that calculated and logged the position of the tow fish at every fix point. During survey operations, the USBL beacon was mounted on the side-scan sonar cable close by the side-scan sonar tow fish. 35 3.2.5 Navigation A Starfix Navigation Suite, a recently developed real-time navigation and datalogging package designed by Fugro, was used as the navigation and logging system. Starfix.Seis module has been designed to fulfil the specific requirements of Fugro’s seismic ships, as well as for general hydrographic survey needs. The Starfix.Seis is an advanced vessel positioning software application that enables the navigation system to be configured and modified in real time. It makes use of high accuracy external time sources, such as GPS timer cards or timer boards connected to GPS (1PPS), to provide microsecond precision and synchronisation of all data. Interfacing to the various systems is achieved either directly via the PC's serial ports and add on multi-port serial cards or through a Qubit Q2780 series interface box. Data from echo sounders and gyros can also be input via the interface box. 3.3 Sound Velocity Measurements For the purpose of measuring the velocity profile of sound in seawater, an AML model SV Plus Sound Velocity Probe (SVP) was used. The AML SV Plus is an independent probe, which was lowered from the sea-surface to the seabed using the coring winch or rope. A mean velocity of derived from SVP profiling has been used for the single- and multi-beam echo sounders, side scan sonar, USBL system and sub-bottom profilers. 3.4 Single Beam Echo Sounder (SBES) The bathymetric data was acquired using the Simrad EA500 or Odom Echotrac Echo Sounders. Both types had hull-mounted transducers operating at dual frequencies of 38 and 200 kHz. A mean acoustic velocity, obtained by averaging the velocities logged by SVP, was applied into the echo sounder. The echo sounder was interfaced to a DMS2-05 or TSS 320D Heave Compensator. Sounding values compensated for motion were logged by the Starfix.Seis software and displayed on a paper record. A generalised example of the SBES system configuration is presented on Figure 3.3. An example of the SBES data plot-out is shown on Figure 3.4. Data processing was performed using Fugro’s Starfix Processing Suite and included: transducer offset correction, vessel draught correction, tidal correction, bad fixes rejection, despiking, and smoothing. 36 Figure 3.3. Generalised example of SBES system configuration. Figure 3.4. Example of SBES data. Horizontal lines spacing: 2.5 m, Vertical lines spacing: 125 m (1 fix-25 m). Top channel: 200 kHz, Bottom channel: 38 kHz, Delay: 45 m/55 m. 37 3.5 Multi Beam Echo Sounder (MBES) Swathe bathymetry data where acquired using two types of Multi Beam Echo Sounder, depending on the vessel. During surveys performed by MV Geo Surveyor a Kongsberg Simrad EM1002 was utilized, whereas a Reson Seabat 8101 was mounted onboard MV Geo Eastern. Characteristics of both models and procedure of data acquisition and processing is described below. 3.5.1 Simrad EM1002 MBES A Kongsberg Simrad EM1002 system with a hull-mounted mechanically pitch stabilized transducer array generates 111 narrow beams in a fan-shaped geometry and operates at a frequency of 95 kHz. The array has hemispherical dimensions of 0.4 m x 0.8 m. The Simrad EM1002 is a high performance Multi Beam Echo Sounder system that may operate from the shoreline and down to a depth of 1000 m. The across track coverage is up to about 1500 m in deeper waters, and in shallow waters up to 7.4 times depth beneath the transducer. The normal angular coverage is adjustable up to 150°, with the option of increasing the angular coverage to one or both sides up to 5° above the horizontal. The EM1002 swathe system maps the seafloor through 111 narrow beams in fanshaped geometry, each beam producing one sounding. The beams are spaced equidistant horizontally at the bottom, which means tighter angular spacing in the outer parts of the coverage sector. In deeper waters, the coverage sector is usually reduced and beam spacing automatically changes so that 111 equidistant beams are always available. 3.5.2 Reson SeaBat 8101 MBES The Reson SeaBat 8101 system utilises a hull-mounted transducer, has a total of 101 beams and operates at a frequency of 250 kHz (Reson, 2002). The large number of beams results in a wide swathe that forms a continuous profile of the seafloor. To enable the accurate measurement of the wide swathe, the SeaBat 8101 uses either the amplitude, or phase bottom detection method (or a combination of these methods) for every beam within the swathe. The SeaBat 8101 may measure a complete 150 degree swathe in one ping/update, which gives a 370 m across track profile in 50 m of water. 38 3.5.3 MBES System Calibration and Processing The MBES system was calibrated for roll, pitch, latency and yaw. A post survey calibration was carried out as well to re-evaluate the survey settings. Fugro’s Starfix.Proc software module was implemented to process all raw *.FBF files as logged by Starfix.Seis software using an automated script sequence. These scripts were used to tie-in MBES raw data, heading data, GPS position data, motion sensor data, heave compensator data, vessel draft, speed of sound in the water and applied the required corrections. The MBES *.pos files generated by Starfix.Proc were reviewed and edited for spurious data in Starfix.Swathedit and Starfix.Surface. Gridding was performed using Starfix.Dtm. MBES Digital Terrain Models were generated using Fugro’s in-house Starfix.Dtm and Starfix.Surface software modules. Various processing parameters were used, depending on the project characteristics. Standard parameters used during typical surveys are as follows: DTM method: Parabolic W Q Grid cell size: 1mx1m Grid search radius: 10 m Smoothing method: None Filtering method: GRAD, STD 3.5.4 Tidal Reduction Bathymetric data were reduced to mean sea level (MSL), with predicted tides calculated from the tidal constants. 3.5.5 Refraction Reduction For the purpose of determining accurate ranges and acoustic refraction, SVP profile data was used. The speed of sound was measured through the entire water column normally every 24 hours. Sound velocity profiles were applied to MBES data during processing. 3.5.6 Post-processing Analysis and Data QC The following Quality Control checks were performed on the MBES data during processing: comparison of depths from the MBES nadir beams with depths from the SBES, comparison of depths from overlapping lines, examination of sun illuminated 39 images for motion artefacts, examination of swathe profiles for velocity artefacts. A data example showing an MBES image after final processing is shown on Figure 3.5. Figure 3.5. Example of processed MBES data. Approximate size of area 4x3 km. 3.6 Motion Sensors Two TSS 320B Heave Compensators were used to remove the effects of vessel motion from the single-beam bathymetric data and from the Pinger data respectively. The output from the TSS 320B processing unit was applied to the SBES, so depth sent to the navigation computer for logging, was heave corrected. In the case of the Pinger data, the trigger from the graphic recorder was delayed by the 320B-processing unit in proportion to the heave, before being sent on to the GeoPulse 5430A transmitter. A Heave Compensator is a system for measuring vertical motion when no stationary reference is available. It is based on a gimbal-mounted vertical accelerometer. It consists of a model 321 heave sensor and a model 320B processing unit. 40 A TSS 335 system was used to provide roll and pitch data to the Sonardyne USBL system. Heave, roll and pitch data used by the Multibeam system, was provided by TSS DMS02 sensor or/and Seatex MRU-5 motion compensator. 3.7 Side Scan Sonar (SSS) 3.7.1 System Description and Data Acquisition A GeoAcoustics Side Scan Sonar System was used for seabed feature mapping. The system consisted of a dual frequency 100/500 kHz transducer mounted in a GeoAcoustics 159D tow fish, a GeoAcoustics Side Scan Transceiver Module, an Alden Model 9315 CTP Thermal Graphic Plotter and a GLOG Acquisition System. The GeoAcoustics 159D fish was the part of the system towed behind the vessel. Positioning of the sonar fish was achieved by a USBL system. The fish was normally towed at a target altitude of 10 to 15% of the selected sonar range above the seabed. A remotely controlled electric winch, with the tow-cable running through a cable counter block, allowed the tow-fish to be kept at an optimal altitude. Fugro’s GLOG/GPLOT system was used for SSS digital recording and hard copy output to the Alden plotter. TVG and white/black level adjustments were applied to the raw signal in GPLOT for display, recording to (*.xtf) and simultaneous output to hard copies on Alden film paper. A schematic diagram of the Side Scan Sonar system configuration is presented on Figure 3.6. Figure 3.7 shows a slant corrected SSS data example. 3.7.2 Side Scan Sonar Data Processing The basic processing sequence of side scan sonar data is as follows: Process and edit/de-spike of USBL position data Process position offsets for SSS datum Import XTF files to Starfix.Workbench Integrate/append processed positions with XTF data through Starfix.Workbench Generate side scan sonar mosaic image to perform interpretation. 41 Figure 3.6. Example of SSS system configuration. Figure 3.7. Example of slant corrected SSS data. 42 3.8 Sub-bottom Profiler (SBP) Sub-bottom profiling was conducted using an ORE Pinger 4x4 transducer array mounted in a ballast tank at the bottom of the vessel. High frequency (3.5 kHz) transmitted signals were typically used. The average penetration depth of the 4x4 Pinger system depends on the seabed conditions. Within typical unconsolidated clastic sediment, it exceeds 30 m below seabed. The combination of 16 transducers and a switching connector junction box allows a multiple beam pattern and source power level selection. The array is powered using a GeoPulse Model 5430A transmitter. The trigger from a Seamap Delphlink 1 unit relayed via the 5430A transmitter was sent through a TSS 320 heave compensation unit to eliminate heave components associated with the transmitter/receiver array. A diagram showing the basic configuration of the Pinger system is shown on Figure 3.8. Both raw and processed pinger data were recorded to GLOG and GPLOT PCs respectively. Substantial initial gain was applied to weak raw signals by a GeoAcoustics 5210A Receiver. The amplified signal was then passed through a FEAM Unit where more gain was applied before the signal was sent to the GLOG PC. This amplified raw signal was recorded in *.glog format on the GLOG hard disk and also passed on to the GPLOT PC where filters and TVG were applied for recording to disk as *.segy. Figure 3.8. Simplified diagram of Pinger (SBP) system configuration. 43 3.8.1 Acquisition Parameters The following acquisition parameters were used during Hull Mounted 4 x 4 Pinger sub-bottom profiling: Transmitter Type: ORE GeoAcoustics 5430A Tx frequency: 3.5 kHz Tx power: 5 kWatts max Tx Pulse Cycles: 2 Firing rate: 250 ms Recording length: 80-120 ms Recording delay: Various (Depending on water depth) Filter Low cut/High cut: 2500 Hz - 4500 Hz Digital recorder: GLOG (Version 3.01) / GPLOT (Version 5.2) Logged data and media: *.glog (raw) and *.segy (filtered) to hard disk Plotter: Alden 9315 CTP 3.8.2 Data QC and Processing Data QC, processing and interpretation of the acquired pinger data was carried out initially onboard the survey vessel using Fugro’s in-house interpretation software Starfix.Interp. The processing sequences involved for the pinger data were as follows: Process and edit positions for Pinger transducer Integrate/append processed positions with SGY data through Starfix.Interp. Interpret features and horizons in Strafix.Interp Example of pinger data plot-out is shown on Figure 3.9. 44 Figure 3.9. Example of pinger (SBP) data. Horizontal lines spacing: 10 ms, Vertical lines spacing: 50 m (1 fix – 25 m). D75 means: Delay 75 ms. 3.9 Gravity Coring Gravity coring was the method used to collect soil samples directly from the seabed. The gravity corer consists of 3 m long and 80 mm diameter core barrel, 400 Kg weight assembly, cutter, catcher and 80 mm core liner (Figure 3.10). The surveyor relayed sounding information from the SBES to facilitate placement of the corer at the required 20 m – 50 m above the seabed. Corer depth was assessed by markings placed every 10 m on the cable. As the vessel stern passed over the intended target, the bridge instructed the coring crew to drop the corer to the seabed. The corer was then rapidly lowered to the seabed in a controlled freefall descent. A very slight slackening in cable subsequently indicated contact with the seabed, and a navigation fix was taken. The corer was recovered and the core sample removed. Cores were logged on individual sample data sheets and sampled. To establish approximate shear strengths in kPa (in cohesive soils only) the Sheer Vane test and Pocket Penetrometer test were carried out (if possible). Sediment was described using British Soil Classification System - BS 5930:1999 (1999). 45 Figure 3.10. Gravity corer model (after Fugro NV, 2001). 3.10 Laboratory Analyses Four samples were fractionated by the author using sieve and pipette analysis methods and results were shown on histograms and cumulative arithmetic curves. Thirteen samples were fractionated and grain size analysed based on the Laser Diffraction Method (LDM). These analyses were conducted by Soil Lab of the Department of Physical Geography, University of Wrocław, using a diffractometer Masterseizer 2000 for the particles smaller than 1 mm, and traditional sieve analysis for larger particles. The analyses were presented graphically as frequency curves, and also in the tables as percentage of each fraction in the sample. The fractions have been classified using both the EN ISO-14688-1 norm and the Udden-Wentworth (Wentworth, 1922) grain scale. Based on these scales and taking into account specifics of the LDM method, where percentage of clay fraction is usually decreased by few per cent in favour of increased silt fraction, the sediment type was defined in the view of three separate classifications: (1) BS 5930:1999, (2) Folk and Ward, 1957; Folk, 1974, and (3) Shepard, 1954, respectively. Clay (<2 μm) and silt (<20 μm) fractions were analysed by X-Ray diffraction and thermal (TGA, DTA and DTGA) techniques. XRD patterns were generated using Co-Kα radiation and Fe-filter on a Siemens D5005 diffractometer. They were measured at the 46 interval 5-77°2θ for 1s per 0.02°2θ step. Thermal analyses were performed using Derivatograph 1500Q to determine the composition and amount of clay and Fe-hydroxide minerals. The samples were heated from 25-96 °C increasing at 10 °C/min in an air atmosphere. The X-Ray diffraction and thermal analyses were performed by Dr. Czeslaw August in the Laboratory of the Institute of Geological Sciences, University of Wrocław. Organic matter from twenty samples was dated in the Gliwice Radiocarbon Laboratory. Two samples (DC 37AtO and DC 22BG) were dated using Gas Proportional Counting Method (GPC) and the remaining eighteen using Accelerator Mass Spectrometry (AMS). The calibration of radiocarbon data was performed using the Marine04 curve (Hughen et al., 2004) for sample DC 37AtO and the IntCal04 atmospheric curve (Reimer et al, 2004) for other samples from the central south area. Dates were calculated using OxCal v4.05 software. The samples from the central west area were calibrated using the IntCal04 atmospheric curve (Reimer et al., 2013) and OxCal v4.2.3 software. Eight thin sections of iron-oxide and carbonate concretions were been made in the Institute of Geological Sciences, University of Wrocław. These concretions were also subject to X-Ray diffraction in order to determine mineral composition. The fossils extracted from sediment samples, including microfauna, have in general been identified and assigned to relevant class, order or family. Selected cores were cut into 1 cm slabs, photographed and subjected to X-Ray radiography for documenting the sediment structure. The X-radiography was done using digital RTG system with 44 kV head voltage output and radiation intensity of 10 mAs. RTG images were recorded digitally and displayed on computer using DIXRay software. 3.11 Borehole Log Analyses Geotechnical borehole logs, used to determine sediment type over the top 100 m sub-seabed have been provided by Fugro Singapore Pte Ltd. The borehole logs include a sediment type description and geotechnical properties such as water content, submerged unit weight, undrained shear strength, all referenced to depth below seabed. The drilling depth of provided logs was typically 30 m or 100 m. 47 3.12 Survey Coverage and Methods for Each Area The detailed listing of utilised methods and the extent of acoustic survey coverage of each particular study area is presented in Table 3.1. Table 3.1. Area List of utilised methods and survey coverage of each study area. Vessel Central Geo Eastern south basin Central west basin Geo Surveyor West margin (Songkhla offshore) Geo Surveyor Acoustic Survey Coverage Acoustic Sensors Over 200 km SBES, corridors MBES, 1000-1500 SSS, PGR m wide 6x11 km site (SBES only), 4x1 km site (all sensors) 4x3 km site 1 core per 1 km of survey corridor Laboratory Analyses Geotechnical Borehole Logs Radiocarbon dating - 9 samples, Granulometry (sieve+pipette) 5 borehole 4 samples, logs of 100 XRD analyses m depth 5 samples, Thin-sections 3 samples Radiocarbon dating - 11 samples, Granulometry (LDM) - 13 samples, Microfauna analysis - 13 samples XRD analyses of concretions 5 samples, Thin-sections 5 samples, X-radiography 7 slabs SBES, SSS, PGR 5 cores SBES, MBES, SSS, PGR - - 1 borehole log of 99.8 m depth - - - 3 cores - 1 borehole log of 70 m depth - - 16 borehole logs of 30 m depth Geo Over 600 km SBES, Surveyor / corridors MBES, Osam 1000 m wide SSS, PGR Dragon Two Gulf mouth 1.5x1.5 km Geo SBES, SSS, (offshore sites, Surveyor PGR Vietnam) One 1x1 km site Central Sixteen SBES, basin Geo 1x1 km sites, MBES, (Cambodia Surveyor Seventeen SSS, PGR offshore) tie-lines N-S crosssection Gravity Coring 48 4 RESULTS 4.1 General Overview of the Whole Area 4.1.1 Seabed Morphology of the Gulf of Thailand The seabed morphology of the Gulf of Thailand can be divided into three zones. These comprise: (1) coastal sections with water depths up to 15 m below mean sea level (bmsl), (2) basin margins with water depths between 15 and 50 m, and (3) the central basin where water depths range from 50 to 86 m below mean sea level (MSL). The morphology of the basin margins (Sections 4.4 and 4.5) is relatively uniform, with a generally flat seabed sloping very gently towards the axis of the gulf. The central portion of the Gulf of Thailand (Sections 4.2, 4.3, 4.5, and 4.7) is characterised by mud mounds and ridges, this topography trending parallel to the axis of the gulf. The mouth of the Gulf of Thailand (Section 4.6) is characterised by shallower waters than the centre of the basin due to the presence of submarine ridges separating the gulf from the South China Sea. Water depths along these ridges are less than 50 m, except within the narrow intervening channel. The seabed morphology in this area is mostly flat, analogous to the margins of the basin. Mud mounds vs carbonate mud mounds The term “mud mounds” was used in this paper, as these positive structures consist of a mixture of clay and silt particles. The author would like to emphasize that these features should not be confused with the “carbonate mud mounds”, commonly called “mud mounds”, which are compositionally and genetically entirely different features. 4.1.2 Stratigraphy of Shallow Sediments in the Gulf of Thailand Geoacoustic data, combined with the results of sediment sample analyses, allowed the subdivision of the shallow geological sediment section into three main stratigraphic units; A, B and C (Figure 4.1). They differ from each other in terms of lithology, geophysical properties, age, and facies characteristics. Unit C, representing Pleistocene clastic sediments, is defined as a stack of high- and low-reflectivity channel-fill and horizontally bedded strata that extends upwards to reflector R2, which marks a major channelized surface. Unit B, defined as Upper Pleistocene - Lower Holocene Stiff Clays, extends between reflectors R1 and R2. This stratum consists mainly of stiff and very stiff clays/silts, and is the primary high 49 reflectivity horizon, which underlies unconsolidated Holocene muds, here referred to as Unit A. The boundary between Unit A and Unit B runs along reflector R1, which represents an Early Holocene unconformity and is marked by downlapping stratal terminations within its overburden (Figure 4.1). The above stratigraphic sequence that has been identified in the central Gulf area (Table 4.1) is, in general, traceable over the entire basin. All detailed information regarding the characteristics of each unit in particular areas is provided below. The radiocarbon dates, which identify ages of mentioned strata, are presented in Sections 4.2 and 4.3. Figure 4.1. Sub-bottom Profiler (SBP) data showing major units and reflectors of the central part of the Gulf of Thailand. Water depth about 70 m below MSL. 50 Table 4.1. Shallow stratigraphy of the Gulf of Thailand and correlation with the adjacent areas. 51 4.2 Central-South Basin 4.2.1 Bathymetry and Seabed Morphology The water depth within the survey area ranges from 55 m to 82 m bmsl. The seabed topography within this area can be divided into two parts. The south-western part, where water depth ranges between 55 m and 70 m, is generally flat with slight undulations. Local irregularities consist of small seabed depressions and larger isolated pockmarks (Figure 4.2). The depressions tend to be elongated in a north-west to south-east direction. Isolated large scale pockmark (~250m diameter, ~10 m deep) PGR profile line (Figure 4.13) Small elongated unit pockmarks Figure 4.2. MBES Image, showing morphology of seabed in SW part of study area, seabed consists of very soft clay, including minor depressions and large-scale isolated pockmark “P1”. The north-eastern part of the survey area, with the water depths between 70 m and 82 m, is characterised by more diverse seabed terrains (Figure 4.3). The seabed morphology in this zone is controlled by variation in the thickness of the soft clay horizon 52 covering the flat stiff substratum. In areas where thickness of soft clay is less than 0.5 m, the seabed is relatively flat, with irregularities caused by randomly distributed small depressions. An increase in the thickness of soft clay results in the development of NW-SE elongated soft mud mounds, which became dominant seabed features. The maximum observed height of these mounds is 3 m. The number of soft mud mounds decreases with increasing thickness of the soft clay cover. The areas with well-developed clay cover, typically over 5 m thick, also feature a flat and uniform seabed. Soft mud mounds Isolated eyed pockmarks Figure 4.3. MBES image, showing seabed lithology at NE part of study area. The brown ridges on the right represent soft mud mounds, while the blue areas show isolated eyed pockmarks within the stiff silty sediment, covered by thin layer of soft clay. 4.2.2 Lithology and Sub-bottom Features Pleistocene Clastic Sediments (Unit C) Unit C has been defined on the single channel seismic records as a stack of high- and low-reflectivity channel-fill and horizontally bedded strata located below reflector R2. Numerous channels, showing complex internal stratal patterns, are recognizable on 53 geoacoustic sections (Figure 2b-d in: Puchała et al., 2011). The borehole logs through Unit C (Figure 5 in: Puchała et al., 2011) document the presence of terrigenous sediments, mostly clays interbedded with sands, silts, and peat layers. The clay layers are stiff to very stiff and may contain silt partings, ferruginous stains, shell fragments, and inclusions of organic matter. Although the contact between Unit B and Unit C is a major erosional boundary that is clearly recognizable on the sub-bottom profiler sections, it is poorly defined in borehole logs. This indicates that the lithology of these units is similar, but they vary in acoustic properties. No radiocarbon dating is available for these sediments. Upper Pleistocene Stiff Clays (Unit B) Unit B is defined as extending from reflector R1 down to a major channelized surface, referred to here as reflector R2. Unit B is high reflectivity horizon below the overlying deposits of Unit A. The average thickness of Unit B across the survey area is ca. 5 m with the maximum observed within the palaeochannels. However, any precise estimation of the depth of these channels is often difficult due to acoustic attenuation. Unit B thins slightly towards basin margins, where it locally occurs only as channel infill. Unit B consists of a stiff to very stiff sediments, which vary from silty clay to sandy silt. The sample DC 40 (Figure 4.4 and Figure 4.5) best characterises this unit. It returned patches composed of yellowish silt and very fine sand, mixed with mottles of greenish clay. The sample also contains angular iron-oxide concretions up to 1 cm in diameter. Silt-sand mixture Clay Micas Figure 4.4. Photography of sample DC 40 (Unit B) showing bioturbation structures developed over lateritic soil features. 54 Figure 4.5. Grain size distribution of sample DC40 (Unit B). The XRD analysis of sample DC 40 (Figure 4.6) shows the presence of kaolinite, illite, I/S mixed layer, and chlorite (possible hydrocalumnite). The concretions are of lepidocrocite and goethite in composition. Figure 4.6. XRD patterns of the clay fraction from sample DC 40 (Unit B) (after Puchała et al., 2011). The sediment samples DC 40 and DC 64 contain also microfaunal fossils, identified as foraminifers. Several samples representing Unit B reveal irregular dark brown clusters of organic matter, comprised of a highly decomposed plant material. 55 The radiocarbon ages derived from samples DC 64 Bb and DC 40b (Table 4.2) date Unit B as Late Pleistocene to Holocene (Puchała et al., 2011). The age of a plant material from sample DC 64Bb was determined as 19 962±90 cal yr BP and the age of shell detritus from samples DC 64Bb and DC 40b indicate 8 290±75 cal yr BP and 6 585±50 cal yr BP, respectively. However, both samples DC 40b and DC 64Bb represent the uppermost part of Unit B close to the R1 unconformity. It is therefore possible that these sediments could have been mixed with material containing younger marine fauna derived from Unit A. Furthermore, the plant remains from sample DC 64Bb could be redeposited and much older than microfauna from the same sample. Table 4.2. Sample Radiocarbon ages of plant matter and shelly fauna (after Puchała et al., 2011). Unit Material dated Calibrated 14C age ± error 1Σ (years BP) Water depth (m) Depth in core (m) DC-14t 62.7 0.1 Topmost A1 Snail shell 2945±55 DC-66b 78.5 1.8 A1 Plant matter 31820±220 DC-22B G 76.0 2.2 A1 Sapropelic peat 14445±325 DC-37At O 78.2 0.2 A1, shell lag on composite erosion surface R1- R1A Ostrea 10593±164 DC-37At A 78.2 0.2 A1, shell lag on composite erosion surface R1- R1A Arca 10344±73 DC-37At V 78.2 0.2 A1, shell lag on composite erosion surface R1- R1A Venus 10302±71 DC-40b 78.3 1.5 Topmost B or lowermost A, contact R1 unclear DC-64Bb S 78.4 1.8 Topmost B or lowermost A, contact R1 unclear DC-64Bb P 78.4 1.8 Topmost B or lowermost A, contact R1 unclear Foraminifers and bivalve shell detritus Foraminifers and bivalve shell detritus Plant matter 6585±50 8290±75 19962±90 Early Holocene Unconformity (Reflector R1) A strong and coherent acoustic reflector, R1, marks the boundary between Unit A and Unit B. This unconformity is characterized by a predominantly flat, horizontal surface, except over some palaeovalleys, where it assumes convex-down geometry over the stiff clays. Reflector R1 lithologically corresponds to the soft clay/stiff clay interface, which is locally associated with coarse-grained lag deposits composed of either sparse, broken shell fragments, or greater concentrations of thick-shelled Ostrea mixed with sub-angular 56 goethite concretions. The ages obtained from these carbonate fossils (Table 4.2) are 10 593±164 cal yr BP (Ostrea), 10 344±73 cal yr BP (Arca) and 10 302±71 cal yr BP (Venus). These locally preserved gravel lag horizons record a reduced sedimentation rate within a high-energy marine environment, and are interpreted to be underlain by a wavecut ravinement surface associated with the Holocene transgression. This, together with the downlapping geometry of silt-clay strata above (Figure 4.1), suggests that the surface merges upwards into a maximum flooding zone around R1. The interface between Unit A and Unit B is not always associated with coarse-grained lag sediments, but it can be represented by soft marine clays lying directly on the older stiff silt substratum representing Unit B. These sediments, including shell detritus and foraminifers (samples DC 64Bb and DC 40b), date to 8 290±75 cal yr BP and 6 585±50 cal yr BP, respectively (Table 4.2). Marine Muds and Clays (Unit A2 and A1) The surficial sediments blanketing almost the entire survey area, are clearly visible on geoacoustic profiles as the topmost, transparent layer referred here to as Unit A (Figure 4.1). The gravity cores and boring logs confirm that this unit consists of very soft to soft silty clay, gradually transforming downwards into firm silty clay. The thickness of Unit A within the study area varies from null to 32 m and generally decreases towards the centre of the basin. The observed layer thickness was less than 2 m at water depths of 80 m and 15-25 m at 60 m water depth. The maximum thickness of Unit A was observed within a palaeochannel that may represent the Pleistocene incised valley of the Kelantan River (Puchała et al., 2011). The single channel seismic records show subunit A1 as a transparent layer. Subunit A2 is characterized as a low-reflectivity strata with locally regular sub-horizontal bedding recognizable mostly within shallower (south-eastern) parts of the study area (Figure 1 in: Puchała et al., 2011). In channelized areas located within the south-west part of the study area, bedding follows the geometry of the channel base. These channels form linear patterns oriented in a generally SE-NW direction. The main observed channel is incised within an underlying palaeovalley. Subunit A2 thins towards the north-west, where it downlaps onto the R1 unconformity. It appears to pinch out at a water depth of ca. 70 m. Gravity cores from subunit A1 consist of greenish grey, very soft to soft silty clays. These highly water-saturated sediments contain numerous marine fossils, such as thinshelled gastropods, bivalves, foraminifers, and sponge needles. The lower parts of subunit A1 and subunit A2 are more consolidated and yielded fewer fossils. Subunit A2 consists of soft to firm silty clays to clayey silts with sub-horizontal bedding. The 57 sediments of subunit A1 form a continuous cover through the entire survey area, while the subunit A2 occurs only in shallower parts of basin. Sample DC 66B taken from 1.8 m below seabed at water depth of 78.5 m, was considered representative of the consolidated part of subunit A1. The grain-size distribution of this sample (presented in Figure 4.7) shows 55.5% clay, 41% silt, and 3.5% sand. The XRD analysis reveals that kaolinite is the main clay mineral; illite and mixed-layer illite-smectite occur in small quantities (Figure 4.8). Figure 4.7. Grain-size distribution of sample DC 66B representing lower part of Unit A. Figure 4.8. XRD patterns of the clay fraction from sample DC 66B representing lower part of Unit A (after Puchała et al., 2011). 58 The carbonated plant fragments obtained from sample DC 66b (Table 4.2) yielded the age of 31 820±220 cal yr BP. However, it is highly probable that these plant fragments are not in-situ remains, but were redeposited. The uppermost part of Unit A has been dated based on a snail shell (DC 14t), which provided the age of 2 945±55 cal yr BP. A few gravity cores proved the localized presence of brown organic clay, transitional to dark brown peat below the soft clays of subunit A1. The thickness of the peat body was estimated to be less than 3 m. The microscopic analysis of a peat sample showed the presence of highly decomposed plant material (mainly sapropelic peat). The boundary between the soft clay and underlying peat is gradational. The radiocarbon dating (sample DC 22B, Table 4.2) yielded an age of 14 445±325 cal yr BP for the peat. However again, the geological position of this layer (above R1), as well as the very gradual boundary with the overlying recent marine clays, indicates that this organic matter could not be in-situ material. It is likely that these deposits represent remains of sapropelic peat deposits, which were formed in a lake environment during the period preceding the Holocene marine transgression. 4.2.3 Seabed Features The main seabed features found in the survey area are pockmarks and mounds composed of very soft clay with marine shell fragments. The soft mud mounds overlie a flat, stiff clay/silt substratum. The surface of both types of sediment, the stiff substratum and the unconsolidated soft clay, are characterised by the presence of numerous pockmarks. The pockmarks within the soft clay are elongated in a NW-SE orientation, while pockmarks within the stiff silty substratum are in generally circular in shape. Other minor features observed within the survey area are seabed surface channels, pitted seabed areas, trawl scars and coarse sediment patches. Pockmark Clusters Two types of pockmark clusters, or “pockmark fields”, occur in the survey area. The first type, called pitted seabed, occurs within soft clay sediments (Unit A) and is represented by a series of small depressions, with each single pockmark usually less than 10 m in diameter. These clusters are mostly oval-shaped, 50-200 m in diameter, and appear to be not associated with particularly high sonar reflectivity. Overall, more than ten pitted seabed areas have been observed within the whole study area. The single channel seismic data and samples acquired from gravity coring show neither significant geoacoustic, nor lithological anomalies related to these areas. The second type of 59 pockmark cluster is represented by one large pockmark field, observed on the stiff silt seabed (Unit B) at DC 37A (coring location 37A). MBES, SBP and SSS data examples showing this field are presented in Figure 4.9, Figure 4.10, and Figure 4.11, respectively. The lithology of the selected gravity cores along the single channel seismic (SBP) profile presented in Figure 4.10 are summarised in Table 4.3. The discussed pockmark field is 300 m in diameter, with a surficial cover of coarse sediment indicated by the high acoustic reflectivity of geophysical records (Figure 4.11). The single channel seismic records (Figure 4.10) show attenuation and masking of the acoustic signal about 20 m below pockmark field 37A. The areas of acoustic blanking can typically be linked to biogas accumulations in sediments (Lekkerkerk et al., 2006), but there is no clear proof of shallow gas presence in this pockmark field. Stiff Silt (top) Unit B SBP profile line (Figure 4.10) Pockmark cluster (37A) Soft mud mounds/ridges Unit A Figure 4.9. MBES Image, showing pockmark cluster and other seabed features at and around coring location 37A, central Gulf of Thailand. 60 Figure 4.10. Sub-bottom Profiler (SBP) image, showing shallow geology at and around pockmark cluster (37A), Gulf of Thailand. Red rectangles represent gravity corer sampling locations. Table 4.3. Lithology of gravity core samples at and around pockmark cluster 37A. Detailed coring locations are presented on Figure 4.10. Sample No. Surficial lithology Lithology base of core Recovery DC 34 Very soft, greenish grey clay Stiff, grey with yellow and brown patches, sandy silt-clay 2.4 m DC 35 Very soft, greenish grey clay Stiff, dark grey, clayey silt 2.7 m DC 36 Very soft, greenish grey clay Stiff, greenish grey, sandy silt with shell fragments 0.8 m Medium dense, greenish grey, clayey gravel, composed of shell fragments and limonite concretions Firm, yellowish grey, sandy silt 0.5 m DC 38 Very soft, greenish grey clay Firm, light grey clay 2.6 m DC 39 Very soft, greenish grey clay Stiff, yellowish-greenish grey, sandy silt-clay 1.2 m DC 37A (pockmark cluster) The sediment recovered from the high sonar reflectivity pockmark cluster (sample DC 37A) is a conglomerate consisting of gravel-size shells, shell fragments, goethite concretions and a very soft greenish-grey clay matrix. Two fossil groups have been distinguished. One group consists of epifaunal sessile species, represented mainly by thick-shelled Ostrea often colonised by Serpula and bryozoans, as well as solitary corals. These fossils typically inhabit high-energy shallow marine environments, mainly at rocky coasts. The second group of fossils consists of the infaunal, thin-shelled bivalves 61 (Arca, Pecten, and Venus) and gastropods, which reside in quiet marine settings, located typically below the storm wave base. Pockmarks/depressions 50 m Pockmark cluster. High reflectivity sonar backscatter caused by shells and limonite concretions DC 37A coring location Figure 4.11. Side Scan Sonar image showing pockmark cluster at and around coring location 37A. SSS range: 200 m per channel, frequency: 100 kHz. The concretions from the pockmark cluster (DC 37A sample) are up to 3 cm in diameter and are noticeably larger than those from the underlying substratum (sample DC 40), which are up to 1.5 cm in diameter. No significant differences in internal structure have been noted; however, concretions from DC 37A tend to be more rounded. The concretions are irregular in shape and frequently porous. Main detrital components are subangular grains of quartz. Quartz silt falls in to two fractions: coarse, approximately 0.1 mm in diameter (Figure 4.12: B, D & F) and fine-grained averaging 0.01 mm in diameter (Figure 4.12, A). Other constituents within concretions are single mica flakes, biogenic calcareous elements, and well-preserved sea urchin plates (Figure 4.12, C). The concretions are cemented by iron oxides admixed with clay minerals. X-Ray analyses of sample DC 37A specimens have shown goethite as a main cement mineral (Figure 3c in: Puchała et al., 2011), while in sample DC 40 goethite is mixed with lepidocrocite. These ferruginous minerals mostly occur as uniform, adhesive matrix, 62 although locally re-crystallisation structures may be found as well. The presence of primitive proto-ooids, or ferruginous pellets has also been noted (Figure 4.12, E). The concentration of iron oxides inside the concretions ranges from 30% to 80% of the total volume. Thin-sections analyses have shown that larger concretions tend to be composed of smaller forms, which were subsequently cemented together. Figure 4.12. Microscope images in cross-polarized light showing internal structure of ferruginous concretions: A, C, D, E & F - sample DC 37A (pockmark cluster 37A); B – sample DC 40 (Unit B). A. Thinner (0.01 mm) quartz silt fraction cemented by goethite; B. Quartz silt grain enclosed by mixture of goethite and lepidocrocite; C. Sea urchin plates preserved within concretion; D. Weathered clay mineral grains; E. Primitive proto-ooids or ferruginous pellets preserved within concretion; F. Thicker (0.1 mm) quartz silt fraction cemented by goethite. 63 Large Isolated Pockmarks Two isolated large-scale pockmarks have been detected during the study. A pockmark named P1 occurs in the southern part of the study area within soft clay sediments (Unit A). A MBES image of P1 is presented in Figure 4.2. The depression is about 10 m deep and 250 m wide. The outline crown of the feature is slightly elongated in a SE-NW direction. A SBP data example, illustrating a cross-section through P1 (Figure 4.13), shows a significant geoacoustic anomaly beneath the pockmark. The anomaly may result from acoustic signal interference due to shallow gas presence, or alternatively, it can be interpreted as disturbed bedding formed by long-term fluid migration. A second alternative, which could reflect a signal scattering due to presence of coarse sediments, is unlikely as the acoustic disturbance starts within very regular fine grained horizon. The high reflectivity SSS character inside P1 could indicate the presence of coarser sediments and/or an accumulation of shells on the bottom of the depression. Another plausible explanation for this high backscatter involves fluid venting. The second large-scale pockmark, named P2, has been observed in the northern part of the study area, within Units A and B. The MBES image of pockmark P2 is presented on Figure 4.14. The depression is about 5 m deep, 100 m wide and 200 m long. At the bottom of the depression are two crater-shaped features, which appear to be central points of one twin pockmark. The cross-section through pockmark P2 (Figure 4.15) shows an acoustic signal diffraction and a disturbance of the sedimentary sequence right below the pockmark. Those geophysical indicators can be linked to fluid migration, while the observed crater-shaped features (Figure 4.14) could represent fluid vents. Figure 4.13. Sub-bottom Profiler image, showing cross-section through the large-scale pockmark P1, Gulf Of Thailand. 64 SBP profile line (Figure 4.15) Two crater-shaped features within double pockmark P2 Mounds consisting of very soft clay (Unit A) Seabed consisting of stiff silt-clay (Unit B) covered by thin [~0.5m] very soft clay layer (Unit A) Figure 4.14. MBES image showing double pockmark P2, central Gulf of Thailand. Figure 4.15. Single channel seismic (SBP) image showing a cross-section through twin pockmark P2. The red rectangle indicates the location of gravity core DC 66B. 65 4.3 Central-West Basin 4.3.1 Bathymetry and Seabed Morphology The water depth of this part of the study area varies from 53.5 to 62.5 m below MSL. The seabed is undulating as the topography of the seafloor is determined by the presence of moderate to high relief mud mounds with up to 7 m of relief. These elongate forms trend in a NNW-SSE direction, with an average orientation of their major axis measured as 147º. The areas where mounds are absent are characterised by a flat seabed, very gently sloping towards the south-east. 4.3.2 Lithology and Sub-bottom Features A geological model of the area, based on the seismic cross-section trending NNE-SSW, is presented in Figure 4.16. Three main units forming the core of shallow subbottom stratigraphy are recognizable within the top 30 m of the sub-seabed deposits. The lowest of these units, Unit C, is composed of low reflectivity sediments. The thickness of these deposits exceeds 15 m. The base of this unit was not defined due to attenuation of the SBP signal. Figure 4.16. Model of shallow geology of Gulf of Thailand central west basin, based on SBP data. 66 The overlying Unit B consists mainly of firm to stiff clays and silts. This substratum in the local stratigraphic nomenclature is defined as the Stiff Clay Member. Unit B is up to 20 m thick. Unit B on the seismic records displays a few high reflectivity horizons with minor channels and several weaker internal reflectors. The uppermost sediments are soft clayey silt deposits of Unit A. They are interpreted as the Holocene Bangkok Clay Member. These sediments occur mostly as soft muddy mounds up to 7 m thick. In areas lacking mud mounds, Unit A appears as a thin veneer of highly water-saturated mud, 0–0.5 m thick. The base of Unit A is a flat unconformity, R1, which is considered the boundary between Upper Pleistocene and Holocene sediments. The detailed description of gravity cores from the central west basin, including location of the X-radiographed slabs and samples collected for lab analyses, are shown on Figure 4.17. The cores GC 3 and GC 4 show the structure and type of sediments from the uppermost part of the muddy mounds (Unit A). The substratum beneath these mounds (Unit B) covered by the superficial sediment (Unit A) is represented by cores GC 1, GC 2, and GC 5. The uppermost part of the GC 5 core represents gravelly sediments penetrated in a pockmark cluster. The X-radiographs of cores GC 3 (Figure 4.18) and GC 4 (Figure 4.19), illustrate the highly bioturbated and deformed internal structure of the mounds. Deformed laminae (Figure 4.18: RTG1, RTG5, and Figure 4.19: RTG4) are interpreted to be related to fluid escape. The homogenised mud (Figure 4.18: RTG2) is considered to reflect extensive bioturbation. There are other bioturbation structures within the X-rayed sections (Figure 4.18: RTG1, RTG5, and Figure 4.19: RTG3). A nearly horizontal burrow with spreite (Teichichnus?) was noted in one section (Figure 4.18: RTG5). All X-rayed sections contain small fragments (up to a few mm in size) of bivalve shells, whereas section RTG1 also contains a gastropod shell. The shelly matter is distributed irregularly within the sediment, mostly as isolated clasts, and sporadically as shelly patches. The X-radiograph of core GC 1 (Figure 4.20), shows the contact between sediments of Unit A and Unit B. The sediments of Unit B, visible in the middle and lower part of the core, are rich in carbonate-ferruginous concretions. The largest one, 3 cm in diameter, is not visible on the X-radiograph, as it has been removed in order to allow undisturbed cutting of the core. Smaller concretions, up to granule size (Figure 4.20: RTG7) show angular shapes and occur as either chaotically distributed isolated pieces, or linear streaks, up to 2 cm long and several millimetres wide. The angular shape and streaked occurrence of the concretions appear to indicate their in-situ growth within the burrows, fissures, or empty spaces of the sediment. The sediment is mottled, (Figure 4.20: RTG7), and shows patches of yellowish and greenish silty mud interspersed with brownish plant matter inclusions. The mottled structure and presence of ferruginous minerals reflects 67 the formation of a lateritic palaeosoil. The occurrence of the larger carbonate concretions, of angular to sub-angular shape, may indicate subsequent influence of a marine environment as a form of reworking, bioturbation and calcite crystallisation during and after marine transgression. Figure 4.17. Gravity core logs oriented in reference to sea level, showing detailed location of XRD sections and lab analysed samples. Cores GC 3 and GC 4 represent upper parts of the silty mounds, while cores GC 1 and GC 2 represents areas between the mounds, and core GC 5 represents the pockmark cluster area. 68 Figure 4.18. X-radiograms and photographs of selected GC 3 core slabs (Unit A): RTG 5 (left), RTG 2 (middle), RTG 1 (right). 69 Figure 4.19. X-radiograms and photographs of selected GC 4 core slabs (Unit A): RTG 3 (left), RTG 4 (middle), RTG 6 (right). 70 Figure 4.20. X-radiogram and photograph of GC 1 core slab (Unit A/Unit B): RTG 7. The granulometric analyses of core samples GC 3 (Figure 4.21) and GC 4 (Figure 4.22) show vertical variation of grain size within the muddy mounds. The tests of remaining core samples (Figure 4.23 and Figure 4.24) demonstrate the difference in sediment type at the boundary between Units A and B and the characteristics of the muddy matrix in the pockmark cluster. The samples collected from the mounds (Unit A) are relatively uniform in composition. Their main fraction is silt, with clay as a secondary constituent. No significant grain size variations were noted downcore. The only exception to this pattern was sample G1, where fine calcareous sand was a secondary fraction. A similar composition was also noted in the surficial samples of Unit A (G10a, G10b, G14), located near the R1 horizon, and sample G15 representing Unit B. Two other samples assigned to Unit B (G12, G19) are composed of mainly of coarse silt and fine sand, and minor amounts of clay and fine silt. The grain-size distribution curve of sample G16, the 71 sediment matrix from a pockmark cluster, shows three peaks, i.e., 100 µm corresponding to coarse silt and fine sand, 5 µm being equivalent of fine silt and clay, and 600 µm matching medium to coarse sand fractions. The granulometry tests are summarised on the triangle diagram (Figure 4.25), which shows grain size distribution of the whole population of samples in reference to the various sediment units and seabed forms. Table 4.4 contains the results of statistical analysis. The whole population of samples is typified by poor or very poor sorting, based on the values of inclusive graphic standard deviation (Folk, 1974). Poor sediment sorting is displayed by sample G15 (Unit B), and four samples from mud mounds. All other samples reveal very poor sorting. Sample G16 shows the poorest sorting among all tested deposits. Grain-size distributions tend to be nearly symmetrical, except for two samples from Unit B that are strongly coarse-skewed. The measurements of kurtosis presented variation of samples from platycurtic to leptokurtic, but no major trend in reference to the features or units was noted. G3: core GC 3, 1.0 m depth, Unit A Volume (% ) Particle Size Distribution 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 0.01 Fraction Colloid Clay Silt Sand BS 5930: Shepard's: Folk's: 0.1 1 10 100 1000 ISO-14688-1 [%] 4.4 10.8 76.0 8.7 Wentworth [%] 4.4 28.6 58.3 8.7 SILT Clayey SILT SILT 3000 Particle Size (µm) G5: core GC 3, 2.5 m depth, Unit A Volume (% ) Particle Size Distribution 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 0.01 Fraction Colloid Clay Silt Sand BS 5930: Shepard's: Folk's: 0.1 1 10 100 1000 ISO-14688-1 [%] 4.9 12.3 75.9 6.9 Wentworth [%] 4.9 31.6 56.5 6.9 SILT Clayey SILT MUD 3000 Particle Size (µm) Figure 4.21. Grain size distribution curve (left) and table (right) of samples G3 and G5 from core GC 3, showing vertical variation of sediments within the upper part of a mud mound. Sediment type defined in the table is based on BS 5930:1999, Folk (1974), and Shepard (1954) classifications, respectively. 72 G7: core GC 4, 0.5 m depth, Unit A Volume (% ) Particle Size Distribution 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 0.01 Fraction Colloid Clay Silt Sand BS 5930: Shepard's: Folk's: 0.1 1 10 100 1000 ISO-14688-1 [%] 4.1 11.3 77.9 6.7 Wentworth [%] 4.1 30.0 59.2 6.7 SILT Clayey SILT MUD 3000 Particle Size (µm) G8: core GC 4, 1.5 m depth, Unit A Particle Size Distribution 5 Fraction Volume (% ) 4 Colloid Clay Silt Sand 3 2 BS 5930: Shepard's: Folk's: 1 0 0.01 0.1 1 10 100 1000 ISO-14688-1 [%] 4.6 12.4 79.4 3.6 Wentworth [%] 4.6 32.2 59.5 3.6 SILT Clayey SILT MUD 3000 Particle Size (µm) G9: core GC 4, 2.7 m depth, Unit A Particle Size Distribution Fraction Volume (% ) 5 Colloid Clay Silt Sand 4 3 2 BS 5930: Shepard's: Folk's: 1 0 0.01 0.1 1 10 100 1000 ISO-14688-1 [%] 2.0 6.5 88.2 3.3 Wentworth [%] 2.0 20.0 74.7 3.3 SILT Clayey SILT SILT 3000 Particle Size (µm) Figure 4.22. Grain size distribution curve (left) and table (right) of samples G7, G8, and G9 from core GC 4, showing vertical variation of sediments within the upper part of a mud mound. Sediment type defined in the table is based on BS 5930:1999, Folk (1974), and Shepard (1954) classifications respectively. 73 G10a: core GC 1, depth 0.1 m, Unit A Particle Size Distribution Volume (% ) 4 Fraction 3.5 3 Colloid Clay Silt Sand 2.5 2 1.5 BS 5930: Shepard's: Folk's: 1 0.5 0 0.01 0.1 1 10 100 1000 ISO-14688-1 [%] 3.1 9.2 70.9 16.9 Wentworth [%] 3.1 25.4 54.7 16.9 SILT Clayey SILT Sandy SILT 3000 Particle Size (µm) G10b: core GC 1, depth 0.15 m, Unit A/B (R1) Volume (% ) Particle Size Distribution 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 0.01 Fraction Colloid Clay Silt Sand BS 5930: Shepard's: Folk's: 0.1 1 10 100 1000 ISO-14688-1 [%] 2.1 7.1 76.0 14.7 Wentworth [%] 2.1 21.3 61.8 14.7 SILT Clayey SILT Sandy SILT 3000 Particle Size (µm) G14: core GC 1, depth 0.3 m, Unit A/B (R1) Particle Size Distribution 5 Fraction Volume (% ) 4 Colloid Clay Silt Sand 3 2 BS 5930: Shepard's: Folk's: 1 0 0.01 0.1 1 10 100 1000 ISO-14688-1 [%] 4.4 10.8 76.0 8.7 Wentworth [%] 4.4 28.6 58.3 8.7 SILT Clayey SILT SILT 3000 Particle Size (µm) G12: core GC 1, depth 0.6 m, Unit B Particle Size Distribution Fraction Volume (%) 5 Colloid Clay Silt Sand 4 3 2 BS 5930: Shepard's: Folk's: 1 0 0.01 0.1 1 10 100 1000 ISO-14688-1 [%] 5.1 7.9 51.8 35.2 Wentworth [%] 5.1 17.8 41.9 35.2 Sandy SILT Sandy SILT Sandy MUD 3000 Particle Size (µm) Figure 4.23. Grain size distribution curve (left) and table (right) of samples G10a, G10b, G14 and G12 from core GC 1, showing vertical variation of sediments on the boundary between Units A and B. Sediment type defined in the table is based on BS 5930:1999, Folk (1974), and Shepard (1954) classifications, respectively. 74 G15: core GC 2, depth 0.7 m, Unit B Volume (% ) Particle Size Distribution 7 Fraction 6 Colloid Clay Silt Sand 5 4 3 BS 5930: Shepard's: Folk's: 2 1 0 0.01 0.1 1 10 100 1000 ISO-14688-1 [%] 3.1 12.6 82.0 2.3 Wentworth [%] 3.1 38.9 55.7 2.3 SILT Clayey SILT MUD 3000 Particle Size (µm) G16: core GC 5, depth 0.1 m, pockmark cluster (R1) Volume (% ) Particle Size Distribution 4 3.5 Fraction 3 2.5 Colloid Clay Silt Sand 2 1.5 BS 5930: Shepard's: Folk's: 1 0.5 0 0.01 0.1 1 10 100 1000 ISO-14688-1 [%] 0.9 4.0 51.1 44.1 Wentworth [%] 0.9 14.4 40.6 44.1 Sandy SILT Silty SAND Sandy SILT 3000 Particle Size (µm) G19: core GC 5, depth 0.2 m, Unit B Particle Size Distribution 7 Fraction Volume (% ) 6 Colloid Clay Silt Sand 5 4 3 BS 5930: Shepard's: Folk's: 2 1 0 0.01 0.1 1 10 100 1000 ISO-14688-1 [%] 2.2 4.7 51.3 41.8 Wentworth [%] 2.2 12.1 43.9 41.8 Sandy SILT Sandy SILT Sandy SILT 3000 Particle Size (µm) Figure 4.24. Grain size distribution curve (left) and table (right) of sample G15 (core GC 2), and samples G16 and G19 (GC 5 core), showing sediment types of Unit B (G15, G19) and muddy matrix from pockmark cluster (R1). Sediment type defined in the table is based on BS 5930:1999, Folk (1974), and Shepard (1954) classifications, respectively. SAND CLAY SILT Figure 4.25. Grain size distribution triangle of all analysed samples assigned to corresponding units/seabed forms (based on Udden-Wentworth grain scale). 75 Table 4.4. Statistical parameters of the grain size of analysed sediment samples. Methodology of calculations after Folk (1974). Sample/unit Sediment type (Folk) Median Mode Graphic mean (Folk) [Φ] [Φ] [Φ] [Φ] Sorting Inclusive graphic standard deviation (Folk) Inclusive graphic skewness (Folk) Kurtosis (Folk) G3 Mound (Unit A) SILT 7.6 8.25 7.6 1.82 Poorly sorted 0.04 Near symmetrical 0.896 Platykurtic G5 Mound (Unit A) MUD 7.7 8.5 7.6 2.03 Very poorly sorted 0.11 Fine skewed 0.933 Mesokurtic G7 Mound (Unit A) MUD 7.6 8.5 7.5 1.99 Poorly sorted 0.10 Near symmetrical 0.937 Mesokurtic G8 Mound (Unit A) MUD 7.8 8.5 7.7 1.83 Poorly sorted 0.08 Near symmetrical 1.398 Leptokurtic G9 Mound (Unit A) SILT 6.9 5.75 7.0 1.75 Poorly sorted -0.13 Coarse skewed 0.902 Mesokurtic G10a Unit A Sandy SILT 7.1 8.25 6.9 2.33 Very poorly sorted 0.11 Fine skewed 0.867 Platykurtic G10b R1 Sandy SILT 6.8 6.5 6.8 2.19 Very poorly sorted 0.01 Near symmetrical 1.018 Mesokurtic G12 Unit B SILT 5.2 3.65 6.0 2.55 Very poorly sorted -0.41 Strongly coarse-skewed 0.742 Platykurtic G14 R1 Sandy MUD 7.7 7 7.6 2.01 Very poorly sorted 0.15 Fine skewed 1.103 Leptokurtic G15 Unit B MUD 8.2 7.75 8.2 1.30 Poorly sorted 0.01 Near symmetrical 1.037 Mesokurtic G16 Pockmark cluster (R1) Sandy SILT 5.3 3.3 5.2 2.93 Very poorly sorted 0.04 Near symmetrical 0.784 Platykurtic G19 Unit B Sandy SILT 5 3.55 5.7 2.25 Very poorly sorted -0.40 Strongly coarse-skewed 0.832 Platykurtic 76 Five samples of carbonate-ferruginous concretions have been subjected to XRD-analysis for mineral content. Three samples have been extracted from the R1 surficial deposits (gravity core GC 1, samples G20, G21a, G21b) representing the transition between Units A and B. Two samples have been extracted from the muddy gravel lag of the pockmark cluster area (GC 5 core, samples G18a, G18b), and assigned to horizon R1. The detailed location of all samples is shown on Figure 4.17. All the analysed concretions were brown with variety of tones, angular to sub-angular in shape. All of the samples reacted with a diluted hydrochloric acid. X-Ray diffractograms (Table 4.5) show similar composition for all samples, with quartz and calcite being the dominant constituents. The secondary minerals for sample G18a are illite and microcline, and for other samples illite, kaolinite and plagioclase. Goethite was present in all samples except G18a and G18b, where its occurrence was limited to traces. The presence of siderite was detected in sample G21b. Results of these analyses indicate that ferruginous minerals occur in these concretions as a minor component. They are responsible of the colour of the concretions, but quartz and calcite dominate by volume. The concretions from the pockmark cluster (G18a, G18b) are characterised by more angular shapes and a lesser amount of ferrous minerals. Table 4.5. Interpretation of XRD analyses of carbonate-ferruginous concretions. Detailed locations of the samples are presented on Figure 4.17. Detected minerals Quartz Plagioclase Microcline Kaolinite Calcite Illite Siderite Goethite Sample Core number Sample type number and depth +++ - ++ - +++ ++ - + +++ + - ++ +++ ++ - + Unit A/B (R1) +++ + - + +++ ++ - ++ GC 1, 0.5 m Unit B +++ ++ - ++ +++ ++ - ++ GC 1, 0.5 m Unit B +++ ++ - ++ +++ +++ ++ ++ G18a Concretion GC 5, 0.1 m G18b Concretion GC 5, 0.1 m G20 Concretion GC 1, 0.1 m G21a Concretion G21b Concretion Unit Pockmark cluster (R1) Pockmark cluster (R1) Note: +++ dominant, ++ present, + traces 77 Figure 4.26. Microscope images showing internal structure of carbonate concretions: A & D sample G21b (Unit B); B – sample G18a (pockmark cluster); C – sample G20 (R1 surface). A, B – parallel nicols; C, D – crossed nicols. A. Two generations of calcite cement: (1) brown calcite matrix coloured by goethite/siderite pigment, (2) recrystallized white calcite without pigment; and pores filled by quartz mineralization; B. Quartz silt intraclasts enclosed by calcite cement, partially pigmented by goethite; C. Marine microfossil (Foraminifera) preserved within concretion; D. Two types of calcite cement can be seen, (1) brown coloured calcite micrite with goethite/siderite pigment, (2) white recrystallized calcite (without pigment); and quartz grain. The cement of the all concretions is composed of calcite in the form of a micrite. Two generations of calcite were noted. The first generation is a micrite with brownish colouration originating from the presence of goethite/siderite (Figure 4.26D). The second generation is formed by an early diagenetic calcite, which occurs mainly as veins of a pale-grey/white colour, originating from recrystallization of the primary micrite. The recrystallization of the micritic calcite led to removal of the original ferruginous pigment (Figure 4.26A). The empty pores within these veins are locally filled by quartz mineralization. The intensive calcite recrystallization areas (samples G20, G21a, G21b) are often associated with ferruginous aggregates, being the product of the iron ion’s remobilisation. Quartz, the second (after calcite) most abundant mineral, occurs mainly in the form of intraclasts. The shape of quartz silt grains varies from angular to rounded (Figure 4.26B) to grains (samples G18a, G18b). The angular shaped particles indicate 78 limited or no transport. The rounded and sub-rounded grains, along with marine microfossils, signify a marine environment. The other minor compounds within concretions like feldspar grains, clay aggregates or fossil remnants, occur as isolated intraclasts. Analysis of the sand fraction extracted from the mud samples of Unit A has shown the dominance of quartz grains and calcareous bioclasts. The substantial amount of dark minerals was also noted. The sand fraction of Unit B is composed mainly of quartz and ferruginous grains and a minor amount of calcareous and dark mineral grains. The gravel fraction of Unit B samples consists of carbonate-clayey-ferruginous-silica concretions and shell fragments. A similar composition occurs within the samples from gravel lags of the R1 surface. The sand fraction of sediments from the R1 surface is a mixture of grains associated with both upper (A) and lower (B) units. The fossils collected from the aforementioned samples are shells of foraminifers (classes Rotalidia and Miliolidia), gastropods, bivalves, echinoid plates and needles, and fish teeth. The detailed list of the identified families, genus and species is presented on Table 4.6. The identification numbers included in the table refers to Figure 4.27, Figure 4.28, Figure 4.29, and Figure 4.30, where photographs of the fossils are shown. The majority of the samples share a similar population of fossils. The most abundant microfossils are foraminifers with calcareous, porcelanous, or agglutinated shells, and micro-molluscs. The latter are represented mainly by bivalves belonging to the families Veneridae, Arcidae and Corbulidae, and gastropods of the Costellariidae family. Larger fossils are primarily composed of bivalves and gastropods. The identified population represents mostly benthonic marine fauna, characteristic for a shelf environment. Few families, such as Arcidae and Hauerinidae show also estuarine and freshwater affinities. No significant differences between the fossil populations of Unit A and those associated with the R1 boundary were noted in terms of composition. The main difference is the shell size and abundance, which are much higher within the gravel lags spread over the R1 surface. The uppermost part of Unit B contains an analogous population of microfauna, but the number of fossils is much smaller and decreases rapidly downwards. In sample G15, taken from Unit B located 0.7 m below seabed and around 0.4 m below the R1 surface, foraminifers are almost absent, which may indicate that microfauna of the uppermost part of the Unit B derives from an overlying horizon and was embedded into the sediment, possibly due to bioturbation. 79 Table 4.6. Environment Marine Marine List of fossils identified within the sediment samples. Habitat Seamounts and knolls Depth range: 0 – 529 m Id Class Family Genus Species Sample Id Core Depth Unit G1 DC3 0.1m A G3 DC3 1m A G4 DC3 2.1m A G5 DC3 2.5m A G7 DC4 0.5m A G8 DC4 1.5m A G9 DC4 2.7m A * Present, ** Abundant G10a G11 G10b G14 DC1 DC1 DC1 DC1 0.1m 0.1m 0.2m 0.3m A A R1 R1 schroeteriana ** ** ** ** ** ** ** ** F2 Elphidiidae Elphidium crispum? ** * ** ** ** ** ** * * * ** ** ** ** ** ** ** ** * ** ** * Marine Benthic F3 Calcarinidae Calcarina spengleri? Marine, lagoons, reefs Benthic F4 Elphidiidae Elphidella? ? * * ** ** ** F5 Almaenidae Anomalinella rostrata ** ** * * ** Hauerinidae Quinqueloculina ? * * ** * ** * Marine Miliolata Benthic F6 Marine, estuarine Benthic F7 Hauerinidae Quinqueloculina ? ** * ** ** ** ** * ** Marine, estuarine Benthic F8 Hauerinidae Quinqueloculina ? * ** ** ** ** ** ** ** Marine Benthic F9 Spiroloculinae Spiroloculina Sp ** ** ** ** Marine Benthic F10 Rotalitata Textulariida (Order) ? * ** * ** Marine Benthic F11 Rotalitata Nummulitidae Operculina ammonoides Marine Benthic F12 Miliolata Cribrolinoididae Adelosina ferrusaci? * ** * * * * * Marine F13 Rotalitata Rosalinidae? Glabratellidae? Marine F14 Miliolata Hauerinidae F15 ? Marine S1 Gastropoda Marine * Lachlanella G19 DC5 0.2m B ** ** ** * * * ** * * ** ** G17b DC5 0.1m A(pc) ** ** * * * * * ** ** * ** * * corrugata * ** * * Turridae Gemmula S2 Mangeliidae ? S3 Borsoniidae Microdrillia ? Marine S4 Pseudomelatomidae Inquisitor ? Marine S5 Pseudomelatomidae Inquisitor ? Marine S6 Pyramidellidae Turbonilla ? Marine * G17a DC5 0.1m A(pc) ** Marine, estuarine Depth range: 50 – 500 m G16a DC5 0.1m A(pc) Pseudorotalia Rotalitata ** G15 DC2 0.7m B Rotalidiae F1 ** G12 DC1 0.6m B * rarimaculata * * * * * * 80 S7 Pyramidellidae Eulimella toshikazui * * S8 Nassariidae Nassarius (Zeuxis) eximius * * S9 Nassariidae Nassarius (Zeuxis) castus Marine S10 Conidae Conus acutangulus Marine S11 Skeneinae ? Marine S12 Skeneinae Cyclostrema novemcarinatum * S13 Cylichnidae Cylichna sibogae * Marine Marine Marine Marine Benthic (mud snail) Benthic (mud snail) Benthic * * * Marine S14 Costellariidae Marine S15 Rissoidae Zebina reclina Marine S16 Pyramidellidae Odostomia goniostoma? Marine S17 Raphitomidae? Hemilienardia? Marine S18 Ringiculidae Ringicula Marine S19 Cylichnidae Cylichna kawamurai Marine S20 Mangeliidae Ithycythara infulata Marine B1 Glycymerididae Glycymeris reevei B2 Yoldiidae Orthoyoldia lepidula B3 Veneridae Pitar chordatum B4 Veneridae Circe scripta Marine B5 Veneridae Lioconcha ? Marine B6 Veneridae Lioconcha ? Marine B7 Veneridae Veremolpa mindanensis B8 Arcidae ? B9 Noetiidae Arcopsis ? B10 Corbulidae Corbula ? * B11 Corbulidae Corbula ? * Marine Marine Marine Marine, fresh water Marine Marine Marine Benthic Demersal zone, 10 - 120 m Depth range: 10 – 60 m Shallow water Depth range: 0.3 – 150 m Benthic, 0 - 200 m Benthic, 0 - 200 m * * Vexillum (Costellaria) Bivalia * * * * ** ** * * * * * * * * * * * * * * * * * * ** * * ** ** * * * * * * ** * * * ** ** ** * ** ** * ** * * ** * * ** ** * ** * * * * ** 81 Marine B12 Corbulidae Corbula? B13 Ostreidae ? Marine B14 Veneridae Gouldia? Marine B15 Arcidae Striarca Marine B16 Lucinidae Epicodakia Marine B17 Nuculidae Nucula Marine B18 ? B19 ? Marine Intertidal Marine Marine Marine Marine, fresh water Intertital to deep water Intertital to deep water E1 Echinoidea ? E2 Echinoidea ? P1 Pisces ? * rotalis? ** ** * * * * ** ** zebuensis ** ** * ** * * * ** * * ** ** * ** * * paulula * * * * * 82 Figure 4.27. Plate showing foraminifers found within sediment samples. Identification of these fossils is presented in Table 4.6. 83 Figure 4.28. Plate showing gastropods found within sediment samples. Identification of these fossils is presented in Table 4.6. 84 Figure 4.29. Plate showing bivalves found within sediment samples. Identification of these fossils is presented in Table 4.6, 85 Figure 4.30. Plate showing micro-molluscs found within sediment samples. Identification of these fossils is presented in Table 4.6. 86 The radiocarbon dates (Table 4.7, Figure 4.31) derived from samples G4 to G8, representing upper part of the mud mound/ridge, point to a Holocene age. The dates vary from 755±28 cal yr BP for the shallowest horizon (0.5 m below seabed) to 5 539±54 cal yr BP for a shell located 2.1 m below seabed. Microfauna derived from 2.7 m below seabed were dated to 4 843±9 cal yr BP, which indicates intensive sediment mixing and bioturbation. Samples G10b and G11, from an area between mounds where the Unit A1 sediments are mixed with gravels at the R1 surface, yielded ages of 6 365±46 cal yr BP and 1 092±81 cal yr BP, respectively. These dates show that both erosive processes preventing fine sediment deposition and intensive sediment mixing takes place across much of the seabed under present-day conditions. Table 4.7. Sample Radiocarbon ages of shells and microfauna of the central-west basin. Water depth [m] Depth in core [m] Unit Material dated Calibrated 14C age ± error 1Σ [years BP] 5587±7 (20%) 5523±2 (2.9%) 5500±15 (45.4%) G4 56.1m 2.1m Unit A1 Gastropod shell G5 56.1m 2.7 m Unit A1 Mollusc shells 4843±9 G7 56.0m 0.5 m Unit A1 Bivalve shells 755±28 G8 56.0m 1.4 m Unit A1 Foraminifers and mollusc shell detritus 3140±36 (52.3%) 3087±10 (15.9%) G10b 62.1m 0.1 m Unit A1/R1 Mollusc shells 6399±11 (19.8%) 6347±28 (48.4%) G11 62.1m 0.1 m Unit A1/R1 Gastropod shell 1166±7 (11%) 1127±19 (23.3%) 1073±18 (27.5%) 1016±5 (6.4%) G12 62.1m 0.6 m R1/Unit B Foraminifers and mollusc shell detritus 5428±18 (49.5%) 5318±7 (18.7%) G16a 62.4m 0.1 m Unit A1/R1 (pockmark cluster) Various shell detritus 24497±161 G17a 62.4m 0.1 m Unit A1/R1 (pockmark cluster) Venus 442±15 (23.5%) 370±21 (31.2%) 328±8 (13.5%) G17b 62.4m 0.1 m Unit A/R1 (pockmark cluster) Ostrea 1340±25 G19 62.4m 0.2 m Unit A/R1/Unit B (pockmark cluster) Various shell detritus 1297±12 Similar results were also obtained for samples G16a to G19 from a pockmark cluster and the exposed R1 surface. The carbonate fossils from this location yielded late Holocene (389±69 cal yr BP, and 1 340±25 cal yr BP) and late Pleistocene (24 497±161 cal yr BP) ages. The younger dates associate with contemporary inhabitants of the seabed, the older date seems to signify fossils washed in during the late Pleistocene 87 (Marine Isotope Stage 2), when sea level on the Sunda Shelf was relatively low and fluctuated between 50 and 90 m below modern sea level (Hanebuth, 2000; Hanebuth et al., 2002). These fossils could have been redeposited during the last marine transgression. Figure 4.31. Gravity core logs oriented with reference to sea level, showing locations and ages of radiocarbon dated samples. Sample G12 provides the age for microfossils from the topmost part of Unit B, just below the R1 surface. The age of 5 378±67 cal yr BP points to the Holocene period. Analysis of the microfauna from the uppermost part of Unit B showed a similar population 88 of microfossils to that of the present-day marine muds. The number of these fossils rapidly decreases downwards in Unit B, and therefore they are not considered as reliable indicators of the age of Unit B. 4.3.3 Seabed Features The major seabed features are sediment mounds composed of very soft clayey silt containing marine shell fragments. The mounds overlie a flat stiff clay/silt substratum. One significant pockmark cluster, approximately of 300 m in diameter, was found in the area (Figure 4.32). Coarse sediments Pockmark cluster 60 m Figure 4.32. SSS record showing pockmark cluster within central west basin. The high reflectivity sonar patches observed within the cluster are interpreted to represent coarser sediments. The gravity core sample recovered from this patch (GC 1) confirmed that these sediments consist of silty shelly gravel with well-cemented carbonate-ferruginous crusts. The XRD analysis showed quartz and calcite as the dominant minerals of these crusts. The minor constituents were clay minerals, feldspars and traces of goethite acting as ferruginous pigment. The granulometric analysis of the fine fraction (G16) reveals very poorly sorted sandy clayey silt (Table 4.4, Figure 4.24). 89 Minor seabed features in the area include isolated pockmarks, depressions, unknown sonar contacts and numerous trawl scars. 4.4 South-West Margin (Offshore Songkhla) 4.4.1 Bathymetry and Seabed Morphology The water depth within this part of the study area ranges from 22.0 m to 24.3 m below MSL. The seabed is generally flat with no significant bathymetric anomalies or trends. The seafloor, comprising part of the basin slope, dips eastwards with an angle of less than 0.1°. 4.4.2 Seabed Features The seabed is generally smooth and featureless. A thin layer of soft clay blankets the entire area, levelling all basal irregularities. Minor features within the area are isolated pockmarks, coarser sediment patches or features related to human activity, like trawl and anchor scars, debris or jack-up rig footprints. 4.4.3 Lithology Two major horizons were identified within the site: Unit A, composed of soft marine muds/clays, and Units B-C composed of stiff to hard clays. Based on the lithology and stratigraphic position Unit A corresponds to the Bangkok Clay Formation (Table 4.1), and Units B and C correlate with the Stiff Clay Member (Table 4.1). The borehole log shown in Figure 4.33 indicates that, despite vertical variations on seismic profiles, sediments below Unit A are basically similar and can be treated as one package. The latter is composed of stiff clays grading downwards into hard clay interbedded with a single layer of sand. The surficial layer of soft clay is typically 1-2 m thick, except in the areas where soft clay sediments fill the palaeochannels incised into the substratum. The thickness of Unit A within the channels may reach 9 m. The stiff and very stiff clays occur below the R1 unconformity, interpreted as the boundary between Holocene and upper Pleistocene sediments. The base of the very stiff clay was found 35 m below seabed. Beneath this horizon a series of hard clays occur. The 5 m thick sand layer was found within the very stiff clay sequence at depths between 18.6 m and 23.7 m. These sand deposits may represent either coastal sediments, or a channel fill within Unit B and C. The conceptual model of shallow geology across the site, based on single channel seismic profile and borehole log, is presented on Figure 4.33. 90 Figure 4.33. Model of shallow geology of south-west margin of Gulf of Thailand, based on SBP data and drilling log. 91 4.4.4 Sub-bottom Features Numerous palaeochannels occur in the area. The youngest channel is completely filled by soft clay (Unit A). It was mapped across the whole area, being approximately 200 m wide, 9 m deep, and oriented in a N-S direction. Other channels were found within the stiff clay series (Units B-C). The system of channels is well developed, many of them are connected to each other, and some are incised from above by a subsequent generation of channels. However, a detailed reconstruction of the system is difficult due to the scarcity of available data. The orientation of these channels varies; however, the general observed trend is from west to east. The dimensions of the channels normally do not exceed 500 m in width and 15 m in depth. 4.5 North-South Profile 4.5.1 Bathymetry and Seabed Morphology The water depth along the surveyed route ranges from 15.0 m to approximately 80.0 m below MSL. The costal sections and basin margins are characterised by relatively flat sea floor dipping southwards. The seafloor below 50 m is undulating due to occurrence of the mound and trough topography. The highest mounds, up to 7 m high, occur at water depths between 50 m and 60 m below MSL. The mounds located below 60 m water depth do not stand more than 5 m above the surrounding seabed. The lowest relief mounds were observed within the deepest part of the basin. The mounds along the route are regularly oriented in a NW-SE direction. The main bathymetric anomalies are caused by the mentioned mounds; however, sporadic minor water depth variations are related to pockmarks. 4.5.2 Lithology The bottom of the described sequence is represented by Unit C, which is composed of generally flat, interbedded low and high reflectivity horizons. The lithology of Unit C was not ground-truthed due to limited vibro-corer penetration. Geoacoustic profiles suggest that Unit C is composed mostly of marine clayey sediments interbedded with silty and sandy deposits of deltaic and alluvial origin. The top of Unit C is marked by reflector R2. The deposits overlying unconformity R2 are Unit B sediments, consisting of stiff clays and silts, ascribed by the author to the upper part of Stiff Clay Member (Table 4.1), based on lithological characteristics and stratigraphic position. The uppermost part of Unit B forms a stiff, relatively flat and widespread surface. Numerous channels are common at this bounding horizon. These channels are related to palaeorivers and 92 palaeodepressions. They incise the substratum down to various depths and they are typically filled by flat-bedded or inclined mud packets. In most cases, bedding geometry concordantly follows the shape of the channel bottom (Figure 4.34). The top of Unit B is a flat unconformity, R1, representing the boundary between Upper Pleistocene and Holocene sediments. SW NE Unit A1 R1A Unit A2 R1 Unit B - channel R2 10 ms ~ 8 m 100 m Figure 4.34. Shallow seismic cross-section 170 km from northern coastline. Central basin of Gulf of Thailand. The uppermost sediments ascribed to Unit A, are composed of soft marine clays and silts. They are interpreted as Holocene, marine sediments corresponding to the Bangkok Clay Formation (Table 4.1). In the central part of the basin, at water depths beyond 50 m, these sediments occur mostly as soft mud mounds, up to 7 m in height. In the areas where mounds are absent, Unit A consists of a thin veneer of soft clay, up to 0.5 m thick. Water depths in the coastal and basin margin sections are less than 50 m, in these regions the seabed is characterised by the flat top of Unit A. The soft mud that occurs here is mostly the fills of seabed depressions and channels. It also forms smooth, uniform mud sheet up to several metres thick. In the areas where soft mud is absent, the stiff substratum of Unit B crops out. The thickness of soft clay along the route ranges from null to over 20 m. Unit A can be divided into two subunits (Figure 4.34). Subunit A2 consists of soft to firm flat-bedded clay sediments. These sediments were formed in a shallow-marine environment located above the storm wave base, or alternatively, they represent the 93 remains of early Holocene prodeltaic clays. Subunit A1 is composed of very soft to soft mud mounds and occurs only in deep water sections. 4.5.3 Seabed Features Shallow Water Section (15-50 m below MSL) Two types of seabed characterise this section. The first type is a flat seafloor composed of stiff silts and clays, locally enriched in sandy fractions. The second type consists of unconsolidated soft marine clays that occur in various forms, such as the fills of small seabed depressions, low relief mounds, and as a thin uniform layer mantling the stiff substratum. The seafloor in this zone is generally smooth. Minor seabed features are related to fishing activity, e.g. fish traps or trawl scars. Natural features, like isolated pockmarks and low relief mud mounds, are rare and usually related to soft sediments. The coarser sediment patches are occasional and generally related to stiff clay or channel infill sediments. Deep Water Section (50-80 m below MSL) The most abundant seabed features in this part of basin are mounds and ridges composed of very soft clay containing marine fossils. These features overlie a flat stiff clay/silt surface. The highest mud mounds occur at depth between 55 m to 60 m below MSL. The relief of these mounds slightly decreases with increasing water depth. A graph illustrating distribution of mud mounds and ridges along the surveyed route is presented on Figure 4.35. The other observed features in this section are pockmarks, coarse sediment patches and pockmark clusters. The diameter of a typical pockmark cluster ranges between 50 m and 200 m. These clusters are often associated with the high sonar backscatter patches indicating concentrations of coarser material. Small isolated pockmarks are numerous and mostly associated with soft clays. Trawl scars, anchor scars and undefined items of debris are also present. 4.5.4 Sub-bottom Features The most prominent sub-bottom geological features observed in the area are palaeochannels, gas chimneys, acoustic masking zones and shallow faults. The channels occur mostly within Unit B. The internal structure of the channel fills varies from flat bedded through inclined bedded to non-structured, massive or high reflectivity deposits. 94 0 -10 -20 Water depth [m] -30 -40 -50 -60 -70 -80 -90 0 50 100 150 200 250 300 350 400 450 500 550 600 650 Distance from northern coastline [km] Lo w rel ief mud mo unds Medi um r elief mud mound s High re lief mu d mo und s Figure 4.35. Presence of mud mounds and ridges along North-South profile across Gulf of Thailand. Several channels are associated with acoustic blanking, which typically originates from the presence of either biogas, or high reflectivity coarse detritus. The upper parts of the channels are often filled by flat-bedded clays of subunit A2 (Figure 4.34). The bottom sediments, showing various type of bedding, may be interpreted as pre-Holocene alluvial and deltaic deposits. The upper flat-bedded clays are probably shallow marine sediments which accumulated within hollows in the terrain during the Holocene marine transgression. The channels may be ascribed to the palaeodrainage basin of the Chao Phraya River. Several areas of acoustic masking were noted along the survey route. The masking mostly occurs below the channels. It may be caused by the high reflectivity/scattering properties of the channel fill sediments or be related also to shallow gas presence. A few columnar structures about one hundred metres in diameter, interpreted as gas chimneys, were also noted (Figure 4.37). Several shallow normal faults were found along the route. These faults are located mostly in central part of the basin. Estimated fault throws are less than 5 m. The faults occur within the deposits of Unit C (Figure 4.36). The uppermost sediments are generally not faulted, as these sediments are unconsolidated and not prone to brittle deformation. The displacement on the mentioned faults gradually increases with depth (Figure 4.36), which points to their synsedimentary activity. 95 NW SW Unit A1 R1 Unit B R2 Unit C 10 ms ~ 8 m 100 m FAULT Figure 4.36. Shallow seismic cross-section 450 km from northern coastline showing a growth fault within Unit C. Central basin of Gulf of Thailand. NW SW R1 R1A Unit A1 R2 Unit B Unit A2 channel Unit C 10 ms ~ 8 m Gas chimney 100 m Figure 4.37. Shallow seismic cross-section 600 km from northern coastline showing a gas chimney below a palaeochannel displaying symmetrical fill. Central south basin of Gulf of Thailand. 96 4.6 Mouth of the Gulf (Offshore Vietnam) 4.6.1 Bathymetry and Seabed Morphology All sites within this region show that seabed is very flat and regular with no significant bathymetric anomalies. The water depth ranges between 51 m and 54 m below MSL. Negligible local seabed undulations are less than 0.5 m high. 4.6.2 Seabed Features The seabed is generally smooth and featureless. The only observed features are trawl scars and isolated pockmarks occasionally surrounded by a thin patch of coarser sediments. The surficial sediments consist of a thin layer of soft clay, covering stiff grey clay containing some plant matter partings. 4.6.3 Lithology and Sub-bottom Features The sub-bottom sediments can be divided into two units. The lower part of the sequence can be ascribed to one stratum, which encompasses Units B and C. Channels, or depressions observed within the sub-seabed sequence, are filled by flat-bedded sediments interpreted as mud. No other features were noted. The uppermost part of Unit A consists of layer of soft grey clay, 0.5 to 2 m thick, which is underlain by a sequence of stiff to hard clay deposits with some minor partings of silt and occasional organic (plant matter) or sandy lenses. 4.7 Central Basin (Offshore Cambodia) 4.7.1 Bathymetry and Seabed Morphology The water depth within this part of the study area ranges between 68 m and 74 m below MSL. The seabed is generally flat with minor undulations related to low relief mud mounds and pockmarks. The mud mounds occur occasionally and are not higher than 3 m. 4.7.2 Seabed Features The most common seabed type is cohesive mud (stiff clay/silt) covered by a very thin veneer of soft marine clay. The morphology of the seabed is characterised by numerous isolated eyed pockmarks, typically associated with highly reflectivity patches. Pitted 97 seabed areas (consisting of clusters of very tiny depressions) are also common. These pitted seabed fields are usually oval and around 100 m in diameter. The other features, occurring rather scarcely, comprise well-developed, irregular pockmark clusters up to 500 m in diameter. These are usually composed of several high reflectivity sonar patches, indicating coarser sediment accumulations. Isolated high reflectivity patches are also widespread in the area. The second type of morphology comprises low relief mud mounds. The mounds appear where the thickness of surficial soft marine mud exceeds 2 m. They are rather irregularly distributed with elongation in a predominantly NW-SE direction. The high reflectivity patches related to pockmarks are less abundant within the areas of soft muddy seafloor. 4.7.3 Lithology and Sub-bottom Features The sub-bottom sediments can be divided into three main units. The lowermost deposits have been distinguished as Unit D. No lithological information regarding Unit D was available. The top of Unit D is defined by reflector R7, which represents a significant unconformity. Unit D is overlain by a sequence of stiff to hard clay deposits locally interbedded with thin silt and sand layers. This sequence can be ascribed as one stratum corresponding to Unit C in other parts of the basin. Unit C has been subdivided into five subunits numbered from C5 at the bottom to C1 at the top and separated from each other by unconformities R6 to R3. Unit C contains numerous incised palaeochannels, numbered from P5 to P2. Unit B was not identified in this area. The uppermost section, ascribed to Unit A, consists mainly of medium grey soft clay and represents Holocene marine sediments, 0.5 to 2 m thick. Pleistocene Clastic Sediments (Unit C) The studies of geoacoustic data (Figure 4.40 and Figure 4.41) in conjunction with borehole logs (Figure 4.38 and Figure 4.39) show that Unit C is composed of mostly marine/prodeltaic clayey sediments inter-bedded with silty and sandy horizons representing shallower facies. The detailed description and interpretation of the subunits is presented below. Unit C5 is located between reflectors R7 (base) and R6 (top). It consists of a 15-20 m thick layer of flat-bedded clays, dipping gently towards the west. In the eastern side of the study area, the upper part of Unit C5 is composed of ca. 5 m thick, upward coarsening layers of silt and sand with mud intraclasts within the topmost section. The sand horizon is bounded at the top by erosive unconformity R6. Clays of Unit C5 are 98 interpreted here as prodeltaic/open bay deposits showing characteristics of a highstand system tract (HST), while the silty and sandy succession in the east may indicate a gradationally based shoreface succession formed during the following lowstand regression (LST). Palaeochannel P5, incised in C5 clays correspond, according to the author, to the tidally influenced channels of a delta plain (LST). The top of units P5 and C5 is an erosive unconformity, R6, cut by ravinement processes during the ensued transgression period. The most common E-W orientation of the P5 channels, westward dipping of clay beds and location of coastal facies deposits within eastern part of the area indicates that the direction of sediment transport during the formation of Units P5 and C5 was generally from the east. Figure 4.38. Borehole logs of the central basin, Gulf of Thailand – part I. 99 Figure 4.39. Borehole logs of the central basin, Gulf of Thailand – part II. Unit C4 overlies reflector R6 and is bounded at the top by reflector R5. It is composed of deltaic/prodeltaic clays and silts inter-bedded with sandy horizons, probably representing storm beds. The lithology of Unit C4 varies spatially. Unit C4 within the north-east part of the survey area is represented by a 40 m thick packet of clays and silts with large scale sigmoidal bedding characteristic of a delta front environment (Figure 4.40). The dip direction of this sigmoidal bedding suggests that sediment transport was, in general, from the east. The eastern part of survey area is characterised by a succession of clays interbedded with (and passing upwards into) sands and silts, interpreted as a former coastal facies. This package is up to 20 m thick. 100 Figure 4.40. Shallow geological model for the central basin, Gulf of Thailand, based on SBP data (line SL-06) and drilling logs. The inclined stratification of Unit C4 represents deltaic clinoforms. 101 Figure 4.41. Shallow geological model for the central basin, Gulf of Thailand, based on SBP data (line SL-09) and drilling log. 102 Unit C4 in the western part of the area consists of ca. 20 m thick layer of clay, which thins towards the west. These deposits are generally flat bedded or gently westward dipping and are interpreted as laid down in a prodeltaic setting. The spatial distribution of facies, geometry of sigmoidal bedding and the westward decrease of deposits thicknesses all indicate that these sediments were sourced from the east. A palaeochannel P4 which is about 30 m deep and cuts into Units C4 and C5, occurs within the south-eastern part of the study area. The channel axis trends mainly N-S and is nearly perpendicular to the dip direction of the sigmoidally-bedded sediments. This suggests a major change in sediment transport direction took place within the basin after sedimentation of Unit C4. The top of Units C4 and P4 is surface R5. Reflector R5 is related to a sharp lithological boundary between the sands of Unit C4 and the overlying clays of Unit C3 within the south-east part. In other areas, R5 manifests as a minor boundary between various types of clay. These spatial variations of Unit C3 indicates that regressive succession within the eastern part of the area reflects local conditions, such as high sediment supply through palaeo-Mekong delta, rather than any eustatic sea level changes. This may also explain the change of the transport direction after deposition of Unit C4, as the emplacement of a 40 m thick packet of delta front sediment had to produce a significant change in basin morphology. Unit C3 overlies reflector R5 and consists mostly of clay deposits. Unit C3 in the south-eastern part of study area is a complex stack of numerous, irregular channels filled by clay sediments. These channels are interpreted as fluvial-estuarine in origin. The largest channel, named P3, is 30 m deep and incises Units C4 and C5. Its infill comprises clays containing plant material at the bottom, sand interbeds in the middle part, and clays with sand pockets in the upper part. This succession is interpreted as belonging to a transgressive systems tract (TST). In facies terms, it begins with salt marshes/tidal flats and ends with estuarine to openbay deposits. The mean orientation of the channel is NE-SW. Unit C3 in the northern part is composed of a 5-20 m thick layer of clay, which is truncated at the top by reflector R4. Clays of Unit C3 in the western part of the area are 10-30 m thick and form one amalgamated division C2-C3; reflector R4 is not present here. These sediments are interpreted as a bay fill. Unit C2 consists of flat-bedded clays and silts that are limited at the base by reflector R4 and at the top by reflector R3. The thickness of these sediments varies from null to 10 m. Unit C2 within the western part of study area forms one division together with the clays of Unit C3. Unit C2 within the eastern part fills local palaeochannels (P2) oriented in a N-S or NE-SW direction. Unit C2 has been interpreted by the author as one regressive-transgressive sequence, where marine clays represent deposits of the HST 103 period, channel sediments the LST period, and reflector R3 represents an erosional transgressive surface. Unit C1 is defined to extend from reflector R1 down to a major channelized surface referred as reflector R3. It is composed of firm to stiff clays and silts and has an average thickness of 5 m. The maximum thickness of 30 m is observed within the deepest palaeochannels. The N-S orientation of numerous palaeochannels indicates transport of sediments from the north, which can be traced to the palaeo-Chao Phraya River. The bottom of Unit C1, represented by channelized surface R3, points to fluvial incision into the emergent shelf. The signs of continuous sedimentation of clays and silts from the channel base up to the R1 surface suggest a gradual change from fluvial (LST) to estuarine-marine facies (TST) during deposition of Unit C1. The erosional unconformity R1, marking the top of Unit C1, displays a predominantly flat, horizontal character, and may be correlated with the last Pleistocene lowstand period followed by the Holocene marine transgression. Pleistocene/Holocene unconformity R1 can be correlated across other areas of the Gulf of Thailand; thus, it is a key surface in the Quaternary stratigraphy of the whole basin (Table 4.1). Holocene Marine Clays (Unit A) Unit A consists of those surficial sediments overlying surface R1. It is clearly visible on the sub-bottom profiler as the topmost transparent layer composed of very soft to soft silty clay with marine fossils. The thickness of Unit A within the study area varies from null to 2 m. The thickest parts are related to areas of elongated low relief mud mounds, in those areas where Unit A is absent surface R1 is exposed at the seabed and covered by a veneer of a semi-fluid mud. The sediments of Unit A are interpreted as very soft to soft marine clays deposited during the Holocene (HST). 104 5 INTEPRETATION 5.1 Stratigraphy of Upper Pleistocene and Holocene of the Gulf of Thailand The morphology of the surficial and sub-seabed features within the Gulf of Thailand is strongly related to lithology of the sediments and the latest geological history of the basin. Hence, the shallowest stratigraphy of the Gulf of Thailand has been thoroughly analysed in this paper. As mentioned earlier, this tripartite stratigraphy comprises (Puchała et al., 2011): Unit C – Upper Pleistocene sedimentary sequence, Unit B – high reflectivity horizon of palaeosoils and stiff silts/clays formed during initial stage of last marine transgression, and Unit A – Holocene marine muds. These three divisions occur in the majority of the Gulf of Thailand basin. 5.1.1 Unit C - Upper Pleistocene sedimentary sequence Unit C is a sequence of marine to prodeltaic clayey sediments interbedded with silty and sandy horizons representing possibly storm deposits. In most parts of the basin, this sequence is typified by horizontal bedding and reveals numerous disconformities, often associated with minor palaeochannels. The deposits of Unit C stratigraphically correlate with Units 4 to 9 of the central Sunda Shelf area (Hanebuth et al 2002) and the Chao Phraya Formation of the Lower Central Plain, Thailand (Dheeradilok, 1995; Sinsakul, 2000; Biswas, 1973). A complex deltaic sequence within Unit C occurs in the central part of the basin (offshore Cambodia), where it was divided into subunits C5 to C1, and respective palaeochannel systems P5 to P2. The lower part of the sequence (Units C4-C5, P5) starts with prodeltaic and open bay muds and clays, followed by gradationally coarsening upwards sediments indicating a shoreface succession, and numerous incised palaeochannels filled by delta plain muds, truncated by erosional ravinement. The whole sequence (C5, P5) together with overlying thick packet of sigmoidally bedded delta front muds (Unit C4), provides evidence of westward progradation. The lack of major rivers on the north-eastern coast of the current Gulf of Thailand suggests that these deposits record an activity of the palaeo Mekong river. The present-day Mekong River delta, which enters the South China Sea, is of Holocene age (Ta et al., 2002). It includes minor distributary channels of early-middle Holocene age, which were discharging sediments towards the Gulf of Thailand (Figure 4 in Ta et al., 2002). The presence of such a large delta front deposit in the central basin of the gulf indicates that the Mekong discharged indeed a significant part of its sediment load towards the Gulf of Thailand during the late Pleistocene. The upper part of Unit C (C3-C1, P4-P2) represents at least two 105 transgressive-regressive cycles and was fed from a northern source, which can reasonably be linked to the palaeo Chao-Phraya river. The Unit C sequence in the Gulf of Thailand is bounded at the top by either unconformity R2 (central south basin, central north basin), or ravinement surface R1 related to the early Holocene marine transgression (central basin), or is marked by a gradual transition to Unit B (mouth of the Gulf of Thailand, margins of the basin). 5.1.2 Unit B – Uppermost Pleistocene lateritic palaeosoil Unit B is a key marker in the shallow stratigraphy of the Gulf of Thailand because it spreads over almost the entire area of the Gulf as the first very strong seismic reflector. The unit is typically several metres thick and is interpreted to contain, from the top, Late Pleistocene lateritic soil, influenced by marine reworking. The soil-like horizon consists of mottled orange and grey stiff silt/clay with numerous ferruginous or carbonate authigenic concretions. The plant matter pocket extracted from Unit B sediment from the central south basin dates soil formation at 20 cal kyr BP, which correlates well with analogous deposits from the Sunda Shelf, dated to 21-23 kyr (Hanebuth et al., 2002). This palaeosoil can also be correlated with top of the Stiff Clay Member from the Lower Central Plain area (Sinsakul, 2000), which provides evidence of lateritic weathering. The microfauna shells from Unit B yielded ages of 8.2 and 6.5 cal kyr BP for the central south basin, and 5.4 cal kyr BP for the central west basin. These dates are inconsistent with the assumed age of soil formation. However, the shelly matter for dating was available only from the topmost part of Unit B, which was affected across the R1 surface by Holocene transgression and accordingly prone for contamination. Origin of ferruginous concretions The ferruginous-clayey concretions of the central-south basin, occurring within the orange-flamed soil of Unit B, are interpreted to be a result of laterization of the sediments exposed above sea level. Such a process typically occurs in hot and wet subtropical monsoon areas (Hill et al., 2010) and is related to repeated changes of reducing and oxidising conditions (Brown et al., 2003). Hanebuth et al. (2003) suggested that this processes affected a marshy palaeo-environment in the Sunda Shelf during the last sealevel fall. The mechanism of lateritic weathering is related to leaching of more soluble ions from parent rocks (Whittington and Muir, 2000), by percolating rain water during the wet season, followed by transport of the resulting solution upwards by capillary motion and precipitation of leached compounds near the soil surface during the dry season. The 106 precipitated soluble salts are washed away during next wet season. This mechanism causes depletion of the easily leached ions of Na, K, Ca, Mn and Mg from the reaction zone subjected to water table fluctuations enriching the mother rock in residual Al and Fe compounds (Friedman and Sanders, 1978). The rounded and sub-rounded quartz grains, pellets, ooids, fossil remnants and clay minerals occurring in these concretions are the relics of the parent sediments, while the goethite and lepidocrocite cement is the product of laterization. The sub-rounded shape of ferruginous concretions found on the surface of Unit B records the reworking of the laterized soil in a marine environment. Origin of carbonate-ferruginous concretions The carbonate-clayey-ferruginous concretions of the central-west basin, despite a different chemical composition, occur in the analogous mottled orange-flamed soil as the ferruginous concretions. The characteristics of the soil and the stratigraphic position suggest a similar origin for both these forms, i.e., lateritization. The differences in the chemical composition may be explained by various environmental conditions. While the ferruginous concretions form at the sites where the soluble ions are leached from, the calcite concretions may form in places where these compounds precipitate. Moreover, these concretions are composed of two generations of calcite cement. The first one, pigmented brown by iron oxides is related to lateritization within the soil. It cements the silica, calcareous grains and clayey minerals of the mother sediment. The younger generation of calcite, without the brown pigment, fills the cracks and pores. This second generation of calcite, which crystallised in different conditions, is interpreted to have been formed in a marine environment, or an intertidal zone, after flooding of palaeosoils. 5.1.3 Reflector R1 – Lower Holocene transgressive surface The R1 surface is a disconformity between stiff sediments of Unit B and soft Holocene marine muds of Unit A, occurring over the whole area of the gulf. The contact is sharp, locally with thin shelly gravel layer lying on the top of stiff silts/clays. Holocene marine muds (Bangkok Clay Member), together with the underlying more consolidated Upper Pleistocene to Lower Holocene muds (Stiff Clay Member) are widespread in the Lower Central Plain, Thailand (Biswas, 1973; Sinsakul, 2000), and are expected to continue seaward across the entire area of the central basin of the Gulf of Thailand. The R1 surface separating these two units is the youngest and one of the most important stratigraphic horizons on the Sunda Shelf (Hanebuth and Stattegger, 2004). It is easily 107 recognizable on high-resolution seismic records (SBP), as it separates low reflectivity transparent clays from highly reflective pre-Holocene stiff silts/clays. The strong expression of this boundary on geoacoustic profiles is related to a distinct change in sediment properties. The marine muds overlying reflector R1 are relatively uniform, unconsolidated and highly saturated, hence acoustic energy can penetrate them easily and only a minor amount of the signal is reflected. The sediments underlying R1 are more consolidated, cohesive and abound in silt and sand layers. Ferruginous concretions and ferruginous/carbonate cements within these sediments are also common. Moreover, coarse-grained lags related to erosive ravinement are concentrated at the R1 surface. The consolidated nature of clays/silts below R1, as well as the abundance of ferruginous concretions within them, causes the uppermost part of the stiff clay to be much more reflective than the overlying deposits. The R1 surface is also well defined in the boring data, as it corresponds to a rapid increase in sediment consolidation, which can be clearly identified by the sharp increase in shear strength. Stratigraphically, the R1 surface is related to a period following the Last Glacial Maximum (LGM), when eustatic sea level was >100 m lower than the present sea level, so that the whole Gulf of Thailand area was exposed (Voris, 2000). This led to intensification of erosive processes on the newly exposed shelf (Stiff Clay Member). Fluvial channels were formed across the emergent shelf, while weathering and soil formation mostly occurred on the interfluves. The latter process resulted in numerous ferruginous concretions and plant matter accumulated in pockets at the top of the Stiff Clay Member. At the end of the Last Glacial Period (LGP) the area experienced a rapid eustatic rise manifested as a marine transgression across the Gulf of Thailand. The R1 surface is in many places overlain by coarse-grained lag composed of sub-angular goethite concretions and marine shells fragments, which reflects abrasive processes occurring during transgression. Consequently, the R1 marker is interpreted as a ravinement surface generated the early Holocene marine transgression. The transgressive ravinement began in the outer and central Sunda Shelf at 14-19 kyr BP and 11-11.5 kyr BP, respectively (Hanebuth et al., 2002, 2003) and reached the central part of the Gulf at 10.2-10.8 cal kyr BP (Puchala et al., 2011). The end of transgression in the head of the Gulf of Thailand, defined by the pick of the sea level in Holocene, is dated to 7-8 kyr BP (Tanabe et al., 2003). The Holocene marine transgression following the Last Glacial Maximum resulted in flooding of tropical forests covering the area of the Gulf of Thailand. These forests, as well as mangroves and salt marshes located near to river mouths, delivered a significant amount of plant matter, which was washed away and mixed with marine clays. Locally preserved peat deposits in the Lower Central Plain, occurring between the sediments of 108 the Upper Pleistocene (Stiff Clay Member) and Holocene (Bangkok Clay), are known as the ’’Basal Peat” (Sinsakul, 2000). The analogous peat deposits from the central south basin Gulf of Thailand and located at the base of the Holocene marine clays above the R1 surface, dated to 14.4 cal kyr BP, are considered to be redeposited sapropelic peats (Puchala et al., 2011). 5.1.4 Unit A – Transgressive systems tract and Holocene marine muds Unit A represents the Holocene marine sediments. It has been divided into two subunits, A1 and A2, separated by ravinement surface R1A. Unit A2 in the central south area represents prodeltaic deposits of the palaeo Kelantan River, whereas within the isolated areas of the northern basin it occurs as estuarine muds filling pre-Holocene palaeochannels. Because these sediments show downlapping termination against the R1 ravinement surface they are interpreted to record a short highstand interlude allowing for coastal progradation within the overall Holocene transgressive systems tract (Puchała et al., 2012). No radiocarbon dates of these sediments were available; hence, considering the stratigraphic position only, they may correlate with either Units 2 and 3 from the Sunda Shelf (Hanebuth et al 2002), dated to 14-19 kyr BP, or the lowest part of Unit 1 dated to 11-13 kyr BP. Unit A1 represents the present-day marine muds that mantle the entire gulf. The age of these sediments was dated to 2.9-10.3 cal kyr BP for central south area, and 0.7-6.4 cal kyr BP for central west area. All these deposits are the uppermost sediments; thus, the range of the dates can be extended to the present time. The analogous deposits from the Sunda Shelf (Unit 1, Hanebuth et al 2002) are dated to 0-11 kyr BP (Hanebuth et al., 2003). The Unit A1 sediments correlate with Bangkok Clay Formation in the Lower Central Plain (Dheeradilok, 1995). 109 5.2 The origin, shape and distribution of pockmarks of the Gulf of Thailand 5.2.1 Pockmarks morphology and fluid seeps Small unit pockmarks The analysis of small unit pockmarks from the Gulf of Thailand area has shown that these features are dispersed with similar density in areas of analogous seabed characteristics (Figure 5.1). The sub-bottom profiles beneath these small isolated pockmarks show no significant connection to any major geoacoustic turbulence and/or blanking. Neither anisotropy, nor any specific pattern of pockmark distribution that could suggest a relationship between these depressions and local geology has been observed. The areas of unit pockmarks, developed on large surfaces of soft mud, are characterised by the coexistence of small circular pits within fields of larger elongated pockmarks. This observation suggests that the origin of these circular features can be related to smallscale dehydration processes within unconsolidated seabed and near seabed sediments. The elongation of pockmarks along a SE-NW axis may reflect a bottom current-influence, as originally these pockmarks were circular in shape and then become larger and elongated. These processes are discussed in Section 6.1 The comparison of a density and distribution pattern of the unit pockmarks within soft mud (Figure 5.1) and the analogous eyed pockmarks within stiff sediment (Figure 5.2) points to the higher density and spatial dispersion of the unit pockmarks. This can be explained by a higher volume of escaping water from soft unconsolidated sediments than from partially compacted stiff silts/clays. The more intense the dewatering of the sediment, the higher number of discharge events, and hence more pockmarks are formed. The lesser compaction of the soft mud allows easier escape of the pore water from any place, which results in a more dispersed distribution pattern. Moreover, the number of the observed unit pockmarks is probably lower than the number of total expulsion events as the erosive regime causes elongation and merging of the adjacent pockmarks, reducing the preservation potential of any individual feature. 110 Analysed area Number of observations (unit pockmarks) Average density Observed mean distance 7 km2 843 120 per km2 62.9 m Figure 5.1. Distribution pattern based on Average Nearest Neighbour method (top) and Standard Directional Ellipse (bottom) of unit pockmarks, calculated in ArcGIS software. 111 Analysed area Number of observations (eyed pockmarks) Average density Observed mean distance 9 km2 578 64.2 per km2 80.0 m Figure 5.2. Distribution pattern based on Average Nearest Neighbour method (top) and Standard Directional Ellipse (bottom) of eyed pockmarks, calculated in ArcGIS software. 112 Large isolated pockmarks While unit pockmarks might be created by very small amounts of water escaping from unconsolidated sediments, the formation of large pockmarks needs much more fluid. Referring to sub-bottom profiles (Figure 4.13 and Figure 4.15), one can observe acoustic signal refraction/disturbance right beneath large-scale pockmarks. Acoustic refraction displays significant changes in acoustic impedance and it can indicate either an internally structured horizon (Lekkerkerk et al., 2006), or the presence of a solid-fluid interface within the seabed. A shallow gas (or liquid) body underneath pockmark can be a potential source of fluid, which seeps into seawater. During emanation of a significant amount of fluid, the resultant pressure could cause unconsolidated particles of clay to enter suspension or colloid form and migrate with water away from the pockmark (Figure 5.3). Small crater-shaped features detected within both pockmarks (Figure 4.14), interpreted as vents, appear to confirm this assumption. Pockmarks clusters The origin of pockmark clusters (pockmark fields) is analogous that of single pockmarks. A conceptual model of pockmark cluster form central basin, Gulf of Thailand (coring location 37A) is presented in Figure 5.3. The assumed processes occurring within this feature are related to fluid migration and seepage from the sediment. The rapid escape of fluids from the seafloor to the overlying water column causes erosion within the immediate vicinity of a seep. The result of this process is the formation of a depression within the seabed. The lightest fractions of sediment may be washed out and, as particles become suspended in the water, migrate away from the seep location. If the process is prolonged and intensive, clay and silt fractions could be removed and field of numerous depressions formed. Finally, a cluster of pockmarks may be formed, partially filled by remaining heavy fractions of original sediment, i.e., those particles that cannot enter suspension. The example of the pockmark field at the 37A coring location shows that such heavy fractions are mixtures of shells and goethite concretions, which can be visible on SSS records as high backscatter patches. The comparative analysis of goethite concretions from pockmark cluster 37A and sample DC40 (representing underlying stiff silt of Unit B), shows that concretions from the pockmark field are larger and some of them are more rounded. It is likely that concretions from pockmark cluster 37A might have been originally larger being a part of the gravel lag on the Holocene ravinement surface. Due to the erosive influence of 113 escaping fluids, the soft clay cover is currently removed and hence the underlying coarser deposits are exposed at the seabed. Figure 5.3. Model of pockmark cluster 37A. Central Basin, Gulf of Thailand. Section A shows possible mechanism of erosion caused by fluid seeps and explains also coexistence of various fossils, concretions and clay within pockmark field. Section B shows final effect of process shown in Section A. 114 5.2.2 Pockmarks morphology and marine currents Pockmarks within soft clays of the southern central basin, Gulf of Thailand show a regular trend of elongation along a SE-NW axis (Table 5.8, Figure 5.4). The orientation of pockmarks across the whole research area is generally constant with only minor deviations from the main direction. Such a trend is commonly taken as an evidence for the activity of bottom currents oriented parallel to the direction of elongation (Hovland et al., 2002). The current erosion is perhaps the single significant, and certainly the most important, factor determining pockmark shape after its initial formation. According to Boe and others (1998) in areas subjected to strong bottom currents, whole pockmark fields can be aligned parallel to the dominant current direction. The elongated pockmarks, due to long-term regular erosion, may become very elongated depressions with high ellipticity. The observed geometry of elongated pockmarks is in general symmetric, or slightly asymmetric, with the steeper slope dipping towards the SE. The asymmetrical shape of pockmarks is related to stronger erosion on the downstream side of a pockmark by unidirectional bottom currents (Josenhans et al., 1978). Symmetrical pockmarks may be created due to erosion on both downstream and upstream sides by the eddy current related to water flow turbulences within the pockmark (Josenhans et al., 1978; Andersen et al., 2008). The predominant bottom currents propagate towards one major direction, or alternatively, belong to the category of reversing flows. The elongated pockmarks occur in the studied area within the soft clay sediments only, while pockmarks within the stiff clay/silt sediments are in general circular in shape. This indicates that the unidirectional bottom currents are able to shape the morphology of features within the soft unconsolidated clayey seabed, but they are not strong enough to cause significant erosion within more consolidated stiff clays and silts; hence depressions within these deposits remain in their original rounded shape. The bottom current velocity, essential to erode soft unconsolidated mud, can be estimated from Shield’s modified diagram (Figure 5.5). The minimum current velocity, which is required to suspend/fluidize mud, varies from 0.05 cm/s for uncompacted mud or compacted coarse silt to 2.5 cm/s for consolidated clay. This shows that the maximum speed of the bottom current within the study area does not exceed 2.5 cm/s, as depressions within consolidated clay-silt are not elongated. The composition of this stiff clay-silt horizon (Figure 4.5, Figure 4.23 and Figure 4.24) is not uniform, including typically approximately 20% clay and 40% silt, hence it is impossible to determine exact suspension/fluidization limits based on the presented diagram only. 115 Figure 5.4. Orientation of longest axis of unit pockmarks (violet roses), seabed depressions (red roses) and mud ridges (green roses) indicating dominant direction of bottom current within the Gulf of Thailand, against predominant surface wind direction (The wind directions based on Koompans 1972). 116 Table 5.8. Orientation of elongated pockmarks and seabed depressions indicating dominant direction of bottom current, southern central basin, Gulf of Thailand. Location Water depth [m below MSL] Number of measurements Average orientation Standard deviation of orientation Area 1 Area 2 Area 3 60-62 62-72 74-80 43 43 43 129,5° – 309,5° 128,3° – 308,3° 125,0° – 305,0° 4.5 4.1 7.2 * Detailed locations are presented on Figure 5.4 (violet “direction roses”) Figure 5.5. Diagram showing how fluidization velocity depends on grain size and consolidation of sediment (from Hovland and Judd 1988, redrawn from Lowe 1975). 5.2.3 Processes driving to further changes in pockmarks morphology The other processes that affect pockmark morphology are the merging of individual pockmarks and inward gravity sliding on unstable pockmark flanks (Hovland and Judd, 1988; Andresen et al., 2008). The merging of pockmarks along the axis of a mean bottom current direction is a significant factor in the Gulf of Thailand, as it leads to the formation of channels and runnels. The inward gravity sliding on pockmark flanks plays only a marginal role, because they are relatively stable due to the cohesion of the clayey sediments. Once seabed morphology starts developing into the mounds and runnels, the role of fluid seeps considerably decreases, as bottom current erosion dominates, being the major 117 factor in the further development. Also the direction of erosion gradually changes from vertical to lateral. The sideways erosion may be driven by presence of a secondary flow of helical pattern (Hickin, 2003), aiding the processes of hydraulic action and corrasion. 5.3 Holocene marine mud mounds and ridges 5.3.1 Soft mud mounds and ridges – morphology and spatial distribution Soft mud mounds and ridges are very common seabed features in the central depression, Gulf of Thailand. Typically, they occur within the areas deeper than 50 m below mean sea level. The spatial distribution of these features is shown on Figure 5.6. The mounds and ridges have been divided here into three classes: low-relief characterised by a height less than 2 m, medium-relief between 2 m and 5 m height, and high-relief, i.e., higher than 5 m. The author’s observations show that the maximum height of the soft mud mounds and ridges does not exceed 7 m above the ambient seabed. The length and width of the features vary from tens of metres to kilometres. The size of the mounds and ridges may vary from metres to kilometres in length and width. The shape of these features is generally elongated in a NW-SE direction. Large mounds, hundreds of metres to kilometres in extent, are characterised by a generally flat top, commonly with a pockmarked surface, and relatively short, steep slopes. Small mounds, tens to hundreds of metres in size, show an elliptical shape, similar to elongated tidal bars. The most elongated bedforms are referred here to as mud ridges. The mounds and ridges are composed of very soft to firm, greenish grey mud admixed with a small amount of marine fossils. The consolidation of the mud gradually increases downwards, along with a decrease in water content of the sediment. The base of the features is the top of the pre-Holocene stiff clay/silt, which is in places covered by a shelly gravel lag. The morphology of the mounds/ridges varies depending on the water depth, the thickness of the Holocene mud layer, and the strength of the bottom current. They generally occur in water depths below 50 m, however occasionally low-relief mounds may be observed in shallower waters. The graph presenting the relationship between water depth and the relief of mounds/ridges is shown on Figure 4.35. The most favourable conditions for the high-relief features exist in water depths between 60 and 70 m. At water depths beyond 70 m below mean sea level, the number and relief of the mounds decreases distinctly. A discussion regarding phenomenon associated with the >50 m water depth zone within the Gulf of Thailand has been presented in Section 6.1.5. It seems obvious that the limitation in the presence of most mounds and ridges to water depths below the 118 50 m isobath is closely associated to the occurrence of elongated pockmarks within the same zone, and can be related to the erosive action of bottom currents below the thermohalocline. The decrease in the relief of mud mounds beyond 70 m water depth seems to correlate with the limited thickness of the Holocene marine mud layer; thus, only medium- and low relief mounds and ridges can develop. Moreover, the deepest areas are located nearer to the South China Sea so the strongest bottom currents may be expected there. The coexistence of strong currents and the thin veneer of soft clay results in faster erosion of the mud mounds, hence these features are smaller and less numerous. An important factor determining the distribution of the mud mounds is the thickness of the Holocene sediments. These features were not observed within the areas where thickness of soft mud was greater than 7 m. It is likely that the main reason of this is the insufficient amount of coarse particles and a greater cohesion within the lower part of the thick mud/clay layers. The coarse particles that are able to sustain effective erosion of these fine sediment sheets play an important role in formation of the mud mounds and ridges. The relatively thin layers of mud may contain a higher percentage of calcareous fragments and be less cohesive, because they were deposited in settings typified by relatively low sedimentation rate of suspended fines. Moreover, exposing of a coarser material from the underlying horizons with advancing fluid seepage is more probable when a layer of mud is thin. 5.3.2 Orientation of mud mounds and ridges The orientation of the elongated axis of the mounds and ridges within the whole Gulf of Thailand is generally NW-SE. This direction slightly varies from NNW-SSE in the central part to NW-SE in the southern part of the Gulf (Figure 5.4; Table 5.9). The slight difference in mound orientations in the central and southern gulf’s parts reflects variations in the seabed morphology. 119 Figure 5.6. Spatial distribution of mud mounds and ridges in the Gulf of Thailand. The magenta hatch indicates areas outside survey corridors where occurrence of mud mounds and ridges is expected. 120 Table 5.9. Summarised results of mud mounds/ridges and depressions/subaqueous channels orientation for three different areas of the Gulf of Thailand. Location Central South Basin Mounds/ Ridges Central South Basin Depressions/ Channels West Basin Depressions/ Channels West Basin Mounds/ Ridges Central Basin Mounds/ Ridges 60-80 60-80 55-60 55-60 70-80 38 36 103 148 73 140.5 140.1 154.8 156.6 157.6 6.5 19 15 14.2 5.1 Water depth [m b MSL] Number of measurements Mean orientation [°] Standard deviation [°] The orientation measurements of elongated unit pockmarks in the central south basin show a NW-SE trend with the mean azimuth of 127°-307° (Table 5.8). The elongation of mounds and depressions within the central south (Table 5.9) shows a generally similar SENW trend with the mean orientation of 140 o-320o, which is slightly deviated toward a N-S direction compared to the unit pockmarks of the west basin. The comparison between the orientation of mounds and depressions within the western part of the basin shows a very good correlation, with the mean azimuths of 155 o -335o and 157o-337o for the depressions and mounds, respectively (Figure 5.4). The orientation of elongated unit pockmarks, depressions and mud mounds shows generally the same trend for all of these features, which is NW-SE for the southern part of the basin and NNW-SSE for the central part of the basin. The correlation of elongation and coexistence of these features within the same area indicates that the origin of these features is rooted in the same set of environmental conditions. According to Boe and others (1998) and based on the arguments presented in sections 5.2.2 and 6.1, the main environmental control was exerted by a regular, strong bottom current directed parallel to the bedforms’ elongation. In this case the current is also aligned with the main axis of the basin. Comparable, but of a much smaller scale, are more elongated systems of subparallel pairs of ridges and runnels in the soft and consolidated cohesive sediments. These are common in subtidal and coastal waters, whereas large-scale systems, 1-2 m high, tend to occur in the deep ocean (Carting et al., 2009). The ridge-runnel system is thought to be an erosive-depositional topography that originates from a periodic flow oriented parallel to the linear bedforms. The fluid evorsion, aided by high density suspension of fine sediment are the main factors responsible for initiating and maintaining the runnels (Williams et al., 2008). This process is accompanied by sediment accretion, which is the main cause of ridge formation (Carting et al., 2009). 121 5.3.3 Age of the mud mounds and ridges The age of the mounds and ridges was determined by radiocarbon dating. The ages in the upper part of the mound in the central-west basin ranged from 755±28 cal yr BP (0.5 m below the top) to 5539±54 cal yr BP (2.1 m below the top). The lower part of the mound was not dated; however the oldest Holocene fossils from the horizon underlying the mud sheet yielded age 6365±46 cal yr BP, which may point to an initial stage of sedimentation of neritic mud. Radiocarbon dating in the central-south basin yielded a date of 2945±55 cal yr BP for the upper part of the mound and 6585±50 to 10593±164 cal yr BP for the horizon underlying the mud sheet. 5.4 Evolution of seabed morphology 5.4.1 Pockmark cluster evolution The average pockmark clusters in the Gulf of Thailand are of several hundred metres in diameter. In contrast to the unit pockmarks and eyed pockmarks, the pockmark fields are related to larger scale seepage processes. They can be formed when fluid releases occur repeatedly within the same area. Pockmark fields can be formed in areas where shallow sediments are gas-charged, or where shallow geology allows the migration of fluids from deeper sources. In addition, the thickness of superficial soft clay should be relatively limited, so each depression within the cluster can retain its shape. The evolution of these features is analogous to eyed pockmarks, where fine particles from the dynamic layer are washed away by seepage processes and bottom currents, and coarser material from uppermost sediment as well as gravel lags from stiff silt/soft clay interface can remain in situ, forming high reflectivity patches. As the concentration of pockmarks in these fields is very high and the fluid seeps occur frequently, the overall erosive power is higher, so that a higher proportion of coarse material remains at the seabed. Moreover, periodic delivery of fluids, usually methane, often results in the inorganic and/or biologically mediated calcium carbonate precipitation (Hovland et al., 2012). These methane derived authigenic carbonates can cause the cementation of the sea floor. The shape of the pockmarks within the cluster remains typically circular, as they occur within relatively cohesive stiff clay/silt; hence the erosive force of bottom currents is not great enough to substantially modify their geometry. 122 5.4.2 Evolution of large pockmarks Large pockmarks observed by the author in the Gulf of Thailand are isolated depressions with diameters up to 300 m and depths reaching 12 m. These features, like the pockmark clusters, are interpreted to be related to shallow gas or deeper source fluid escape. However, instead of numerous depressions scattered within the area of a cluster, the large isolated pockmarks are characterised by a single vent in the centre of a depression. The large pockmarks may be formed within both major seafloor types of the Gulf of Thailand. However, a pockmark’s shape and its evolution reflects the nature of the host sediment. The pockmarks within the stiff silt are relatively shallow and are characterised by circular shape, related to the original fluid expulsions. The evolution of the depression is related mostly to the intensity of seepage processes. If seeps occur often and the amount of released fluid is significant, the radius and the depth of pockmark tend to increase. No significant influence of bottom currents on the shape of pockmarks within consolidated sediments was noted. Pockmarks within the soft mud are larger and deeper, as unconsolidated fine particles enter suspension more readily and are freely washed away during fluid release. The evolution of these features is related to two main factors: (1) the frequency and intensity of fluid seepage, which causes deepening and enlargement of the depression, and (2) the speed and direction of the local bottom current, which causes the elongation of a depression. When both of these processes co-occur, the lower part of the pockmark retains its original circular shape, but the upper part of the depression becomes elongated along the dominant current direction. 5.4.3 Unit pockmark evolution and further development of the seabed morphology The unit pockmarks, also known as pits (Harrington, 1985), originate as circular features formed during a single event of fluid release from the soft unconsolidated sediment (Hovland and Judd, 1988). The release of fluids occurs mostly as a result of dehydration processes during the sediment consolidation (early compaction). During consolidation of large sheets of homogeneous marine mud/clay, the characteristic pattern of randomly distributed pits forms on the mud surface. Such pitted patterns are widely observed on the clayey seafloor. However, within the central and southern parts of the Gulf of Thailand basin most of these pockmarks display an elongated shape, except of the smallest ones (youngest), which are rounded. 123 The elongation of the majority of depressions indicates that these features exist under an erosive current regime, which has modified their original circular shape. Figure 5.7A shows a pattern of circular and elongated seabed pits within the surface of the soft mud layer. The circular depressions are interpreted as relatively young features, still preserving their original shape, while the elongated pockmarks are interpreted as features old enough to have been reshaped by bottom currents. The next stage of elongated pockmark evolution is the formation of large elongated depressions. The Figure 5.7B shows the co-occurrence of circular pockmarks, pockmarks in various stages of elongation, merged elongated pockmarks, and large elongated depressions with flat bottoms. All these features seem to develop one from another under the same environmental conditions. The relatively young, immature features grow mainly along the mean current direction and tend to form linear patterns, which become minor channels. More mature features also develop downwards until they reach the stiff bottom, and to some extend sideways causing the broadening of the channels. The abundance of broad depressions with flat bottom and steep banks as well as a large quantity of the young minor forms coupled very limited number of narrow runnels, indicate that the intermediate step of developing depressions takes a relatively short time. This suggests that the level of scouring may increase after reaching the stiff bottom (the base of the soft mud), where the increased concentration of coarser particles enhances erosive processes. Moreover, the presence of newly formed long runnels and channels focusing bottom currents, allows for the development of secondary flows (e.g. helicoidal flow), which may contribute to the sideways erosion (Hickin, 2003). This is why large, long depressions are likely to expand faster than smaller immature features. Figure 5.7C shows a stage of seabed evolution, in which the eroded area of depressions covers as much of the seafloor as the remaining soft mud residues. These residues, very common features within the Gulf of Thailand, are called mud mounds and ridges. The following step of seabed morphological development leads to a further flattening of the seabed. The progress of erosion causes the denudation of the mud mounds until there is localised exposure of the underlying stiff materials at seabed. The final result of these processes is a mature seabed morphology shown on Figure 5.7D. This morphology is characterised by flat seabed composed of consolidated stiff clay/silt with numerous eyed pockmarks. The “eyes” of these features within the Gulf of Thailand are related to coarse sediment patches inside and around depressions. These pockmarks originate from fluid seeps and remain circular in shape as they are formed within cohesive sediment, which is consolidated enough to prevent significant bottom current erosion. 124 The stages of this evolution (Figure 5.8) are discussed below. These stages have been generally split up in time; however this variation also takes place in space. A spectrum of morphologies exists between the end states of perfectly intact soft clay and totally removed soft clay. The present-day location of the seabed on this spectrum depends on both its temporal and spatial position. Pre-erosive stage (early Holocene) This stage corresponds to a time when sedimentation of marine muds and clays was the dominant process within the central basin. This resulted in the formation of broad and thick sheets of soft clays that covered the basal Holocene ravinement surface. This period may be linked to an early Holocene initial stage of transgression, when the sea level was still relatively low, so that sediment delivery from the shore to the central basin was more intense than it is at present. This stage also defines the beginning of mud consolidation associated with initial dewatering and gas escape from the soft sediments. As long as the rate of sedimentation was high, the occasional pockmarks were covered by the freshly-settled clay rather than continuing their development at the seabed surface. The erosion forces during this stage were different from those acting today because the shallow water conditions dominated in the basin. Moreover, the shallow water depth did not allow for the development of a full stratification in water column and prevented any significant inflow from the South China Sea. Formation of elongated pockmarks (mid-late Holocene) This is the first stage of bottom regime when the rate of erosion was higher than that of deposition. This allowed small pockmarks to remain uncovered on the seabed and be shaped by bottom currents (Hovland et al., 2002). The unidirectional bottom-current regime resulted in the elongation of depressions parallel to the direction of the main current (Figure 5.7A). This type of morphology can currently be observed in some areas of the Gulf of Thailand, where the sheet of soft mud is relatively thick and broad. The low quantity of coarser particles at the sea bottom resulted in relatively limited erosion, as the main source of erosion is hydraulic action, without significant participation of corrasion (abrasion) which is a more effective method of erosion (Clowes and Comfort, 1982). Such a weak bottom current is only able to wash out the topmost, less cohesive, semi-fluid part of the clay (Lowe, 1975), thus erosion of the seafloor progresses very slowly and pockmarks remain relatively small. 125 B A 200 m 200 m C 150 m D 150 m Figure 5.7. Evolutionary stages of mud mounds and ridges formation in the Gulf of Thailand, central basin. A - elongated pockmarks within soft clay - stage 2; B - elongated depression within soft clay - stage 3; C - mud mounds and ridges - stage 4; D – flay stiff seabed with eyed pockmarks– stage 5). 126 Figure 5.8. Model of seabed morphology development in the Gulf of Thailand, central basin. 127 Enlargement of depressions and channel development (mid-late Holocene) The size of pockmarks influenced by scouring increases along with current strength (Boe et al., 1998), mainly in the direction of main bottom current. This type of morphology can be found in areas where the thickness of unconsolidated mud is relatively limited, so that the bottom current is able to wash out coarser sediment (fine sand) accumulated near the base of the clay layer. These fine sandy particles, transported near the seabed enhance the abrasive force of the bottom currents and this force grows further when the sand content in the boundary layer increases (Clowes and Comfort 1982). Such type of erosion is driven by a combination of hydraulic action and abrasive action of the sand particles dragged along the seafloor. In response to rising Holocene sea levels, the deepest and most elongated depressions evolved locally into subaqueous channels providing confinement to the bottom current flow (Figure 5.7B).These features are also called "current moats" (Cartwright, 1995). In the Gulf of Thailand such moats occur rather occasionally and are characterized by a relatively flat bottom, where the stiff clay/silt substratum is either exposed, or covered by a thin veneer of loose, mobile fine sand. The banks of these channels form edges to soft mud sheets or mounds. The depth of the channels is usually equal to the thickness of the adjacent soft mud layers. This kind of seabed morphology is currently very common in central Gulf of Thailand. Development of mud mounds and ridges (mid-late Holocene) This stage is typified by the development of elongated mud mounds and ridges separated by runnels (Figure 5.7C). The continued erosion of the seabed causes further widening and broadening of depressions and subaqueous channels. Adjacent depressions tend to merge together and the remnants of uneroded soft mud stand up as ridges or mounds. The mounds lie on the stiff silt substratum and are elongated in the same direction as the depressions. This is the most typical morphology of the Gulf of Thailand seabed. Flattening of the seabed (mid-late Holocene) The final stage of seabed development involves the flattening of the bottom that is a natural consequence of long-term erosion of the mud mounds. This ultimate state happens when all of the unconsolidated muds are washed away. Such mature morphology (Figure 5.7D) may have appeared also in an earlier stage, if the layer of unconsolidated soft sediments was originally very thin, or the bottom current was 128 strong. It is also the primary seabed morphology in areas where soft mud was never deposited. This morphology is characterised by a relatively smooth, flattened surface with scattered, isolated eyed pockmarks. These pockmarks were formed within the consolidated cohesive sediment, which is more resistant to current erosion (Lowe, 1975); hence, these pockmarks keep their original rounded shape. This morphology is most common within the deepest part of the basin where sediment supply was lowest and soft clay veneer thinnest. 129 6 DISCUSSION OF RESULTS 6.1 Oceanographic conditions and formation of elongated pockmarks The presence of broadly distributed and regularly oriented, elongated, symmetrical or slightly asymmetrical (downstream side at SE) pockmarks is an indicator of the existence of regular bottom currents flowing SE to NW. The analysis of oceanographic data from available publications show that the marine current pattern within the Gulf of Thailand is complex, and there is no single current that could be responsible for the orientation of these pockmarks. Moreover, the existing models presenting marine currents of the Gulf of Thailand are often at variance with each other. These issues are discussed below. 6.1.1 Bottom currents and wind direction The direction and characteristics of wind-driven currents within the Gulf of Thailand are related mainly to monsoon winds. Winds from the south-west dominate between May and September, while the north-east monsoon wind dominates from November to February. These two seasons are separated by a relatively quiet period of transition between the opposing monsoons. Figure 2.11 shows a general correlation between wind and surface current direction in the South China Sea. Although this correlation is observed mostly within the main basin of South China Sea, the Gulf of Thailand area shows a different pattern, due to its limited size and connection to the South China Sea. It seems that surface currents within the Gulf of Thailand are more affected by in- and outflows from the South China Sea, than those generated by the monsoons. The analyses of data from current meter stations located in northern, central, and southern parts of the Gulf of Thailand, near the axis of the basin, show that the direction of the observed currents was not primarily dependent upon the simultaneous local winds (Robinson, 1974). A comparison between main wind direction and bottom current direction stemming from the elongation of pockmarks (Figure 5.4), shows no direct correlation between these two phenomena. The general direction of the monsoons is mainly NE-SW, while observed direction of bottom currents based on the seabed features is perpendicular to that. However, this 90° deviation could be theoretically explained by influence of factors like: (1) Coriolis force which deviates wind-driven surface currents by 45° clockwise in the northern hemisphere from the dominant wind direction and (2) Ekman transport, which deviates bottom currents by a certain angle in relation to surface current. 130 However, the oceanographic data as well as the orientation of seabed features at various depths did not prove any major impact of these two processes onto the direction of bottom currents within the Gulf of Thailand. Firstly, both Coriolis force and Ekman transport tend to create curved rather than linear shapes, which is contrary to the observed patterns of pockmark elongation. Secondly, Ekman transport causes the decrease of current speed downwards, while the strongest bottom currents, able to shape the pockmarks, are active in water depths beyond 50 m. The third strand of evidence against correlation between the winds and bottom currents of the Gulf of Thailand is the contrary orientation of pockmarks in the Gulf of Thailand and the South China Sea. Both of these areas are under the influence of the same monsoon winds, but the trends of pockmarks from these two areas are different. The Gulf of Thailand pockmarks show NW-SE elongation, while pockmarks located south of the Gulf of Thailand are generally circular in shape without any significant elongation. 6.1.2 Tidal currents – theory and definitions Another and one of the most probable options for the origin of the elongated pockmarks is tidal activity. A summary of tidal mechanics given below is based on Bowditch (1995). The tidal current is a periodic, horizontal movement of the water related to tides. The water flow during rising of the tide is called the flood current and the movement during falling of the tide is referred to as the ebb current. Tidal forces are mainly generated by the gravitational pull of the Moon and Sun; however, the Moon exerts a stronger influence than the Sun. Three major types of tides are recognizable. (1) The semidiurnal tide is characterised by two high and two low waters each tidal day, with minor differences between high and low waters. (2) The diurnal tide is typified by single high and low water each tidal day. (3) The mixed tide is a combination of diurnal and semidiurnal oscillations. The characteristics of tides vary across the year depending on Moon and Sun position and declination. When the Earth, Moon and Sun are aligned, the tractive forces of the Moon and Sun combine causing spring tides, which are of a greater range than average. A perigean tide is another type of tide that is greater range than average. It occurs when Moon is at the point of its orbit that is nearest to the Earth. When the Moon is in first and third quarters the neap tides occur, which are less than average in range. These variations are related to the semidiurnal portion of the tide. The diurnal portion variations are related to Sun’s and Moon’s declination changes. The direction of tidal flow is generally related to the morphology of the basin. In open seas lacking major barriers, the tidal current is rotary. It means the water flows continuously in a direction determined by Earth’s rotation. This rotary current is naturally 131 clockwise in the Northern Hemisphere and counter-clockwise in the Southern Hemisphere, although local conditions may modify these patterns. The speed of rotary currents typically varies across the tidal cycle, with the two maximums in relatively opposite directions, and two minimums about halfway in between them. In straits and complex shorelines, where the flow is more or less confined, the tidal current is reversing. It means that it flows alternately in two opposite directions, with a short period of slack in between when there is little or no water movement. The velocity of reversal currents varies from zero at slack water to a maximum about midway between the slacks. The maximum of the reversal current oriented towards shore is called “strength of flood” and the other maximum oriented towards open sea is called “strength of ebb”. The strengths increase and decrease during periods of two weeks, month and a year, achieving the maximum speeds during spring and perigean tides. In inland tidal estuaries, the speed of tidal currents varies across the channel, being usually higher in midstream than near shore. Vertically, the speed of the ebb strength decreases towards the bottom, contrary to the flood strength, which is stronger at subsurface depths and decreases towards the surface. 6.1.3 Tidal currents and seabed morphology Numerous oceanographic publications describe various tidal patterns within the Gulf of Thailand. The study by Wattayakorn and others (1998) illustrates domination by strong, predictable, generally shore-parallel tidal currents within the Gulf of Thailand. The rotary movement of the tidal current is not as simple as semi-diurnal phase propagates clockwise in the central part of Gulf of Thailand, opposite to the phase of diurnal tides, which are counter-clockwise (Yanagi and Takao, 1998). Any correlation between the shape of seabed features and tidal currents is not obvious. Firstly, the pockmarks are elongated. This suggests the presence of erosive bottom currents from two opposite directions, which is generally consistent with the nature of tidal currents within estuaries, straits and narrow basins connected with oceans. On the other hand, the circular tidal motion within the Gulf of Thailand, which is typical to open seas, does not agree with the linear trend of seabed feature elongation. However, there are some possibilities, which could explain coexistence of these two phenomena. Fang and others (1999) showed that the tidal motion in the Gulf of Thailand is predominantly maintained by the energy flux from the Pacific Ocean through the South China Sea. This results in relatively strong diurnal tidal currents at the entrance to the Gulf. This means that the main tidal flow within the Gulf of Thailand is directed towards and away from the South China Sea. This is the SE-NW direction common to the axis of seabed feature elongation. 132 The two main processes determining the direction of tidal patterns in the Gulf of Thailand are: (1) reverse motion in and out of the South China Sea and (2) deviation of this motion by earth’s rotation, resulting in rotation flow of the tidal currents. When the orientation of these forces is considered a simple conclusion can be made that the first process is in agreement with the pockmark elongation trend, while the second process is in opposition. The Gulf of Thailand is a sort of intermediate form between open sea and narrow estuary and is characterised by both of these processes influencing the tidal pattern. Hence, the tidal current rose can be presented as an ellipse with the maxima oriented in SE-NW directions. This simplified theoretical model, which is not in opposition with oceanographic data published by Robinson (1974), appears to explain the coexistence of linear features originating from tidal currents with a circular tidal pattern presented by oceanographic charts. In particular: The mentioned current ellipse indicates that tidal currents reach the highest speed when the main current direction is oriented NW-SE. These currents, being stronger than an average, may have an ability to shape the seabed, so that the elongated depressions could be formed during these periods; During periods where the strength of tidal currents decreases, the direction of these currents changes according to rotary pattern. These currents may be too weak to cause efficient erosion of the seabed, so they would not significantly affect pockmark morphology. The conformation of these hypotheses by oceanographic data could indicate tidal currents as a potential factor shaping morphology of seabed in water depths beyond 50 m within the Gulf of Thailand. However, analyses provided by Robinson (1974) from current meter stations in northern, central, and southern parts of the Gulf of Thailand near the axis of the basin have shown that the wind and the water density factors are more influential than tidal effects. The geometry of the pockmarks, which are either symmetrical or slightly asymmetrical, with a more eroded NW side, suggests that flow from the south-east is more influential than the opposing direction. This could indicate that flood strength which is typically stronger at the seabed (Bowditch, 1995), is the dominant erosive tidal factor, with the ebb current being secondary. The water flooding the Gulf of Thailand from the South China Sea is colder and more saline. It will therefore tend to flow near the bottom, while the water returning with the ebb current will generally migrate near the surface. 133 6.1.4 Density-driven currents and seabed morphology A density–driven current is the water flow related to variation in water density, which is mainly controlled by temperature and salinity. This type of current occurs in the Gulf of Thailand and varies seasonally (Yanagi et al., 2001). The most intensive currents develop during April and May. During this period, the temperature of the water within the upper layer increases due to intense surface heating and relatively light winds. At the same time, the water within deeper parts remains relatively cold and heavy, so that vertical stratification builds up. The concurrence of relatively light and warm water within the surface zone of the Gulf of Thailand with much denser water of the South China Sea results in the development of estuarine circulation and significant density-driven currents. The surface water flows towards the South China Sea, while the bottom water enters the Gulf of Thailand. These currents, induced by the horizontal density gradient between the head of the Gulf of Thailand and the South China Sea, occur throughout the entire stratified season from March to October (Yanagi et al., 2001). The strength of these currents has a main peak in April and a secondary peak in October, when the delivery of fresh water into the Gulf of Thailand is highest, due to the maximum river discharge (Snidvongs, 1998). Between November and February, the stratification vanishes due to significant vertical mixing related to the presence of strong monsoons. The direction of these density-driven currents is in general consistent with the orientation axis of elongated pockmarks. This observation suggests that density-driven currents are one of the potential forces shaping seabed morphology in the deep zone of the Gulf of Thailand. The estuarine circulation is characterised by a bottom current flowing in a north-westerly direction and surface current flowing to the south-east. The slight asymmetry of several pockmarks as well as their presence in the deeper section of the basin, suggest that a north-westerly bottom current may be responsible for shaping these seabed features. However, the density-driven currents display generally low velocities; hence, the erosive potential of this type of flow appears to be limited. Although this seems to deny density-driven currents as the main cause of pockmark elongation, it is not unlikely that they may enhance the erosive action of other types of current, such as tidal flood flows, contributing to the formation of the observed seabed morphology. 134 6.1.5 Seabed morphology and thermohalocline The stratification of the sea water in Gulf of Thailand mentioned in the previous section was reported by many oceanographers (Pukasab and Pochanasomburana 1957; Robinson, 1974, Yanagi et al., 2001) and confirmed by sound velocity profiles analysed by the author. In general, two different water layers can be observed. The top layer is characterised by a higher temperature and lower salinity, which reflect fresh water delivery due to heavy precipitation and river discharge, as well as the influence of the sun warming up surficial waters. The bottom layer is cooler and characterised by higher salinity as it flows in from the open South China Sea. This bottom layer fills mostly deep parts of central basin in water depths beyond 50 m (Robinson, 1974). The boundary between two layers is defined by a thermocline and halocline, which typically co-occur at a water depth of 50 m and varies seasonally. The typical temperature and salinity difference between these both layers is around 2°C and 2‰ respectively. The presence of this thermohalocline at 50 m below sea level correlates with the occurrence of elongated pockmarks, which are widespread beyond this water depth. This fact may indicate that currents causing elongation of seabed features relate to inflow of cooler and heavier waters from the South China Sea below the thermohalocline. 6.1.6 Seabed morphology and internal waves Internal waves on the interface between two layers of water of different densities have been described from the Gulf of Thailand by Pukasab and Pochanasomburana (1957, fide Robertson, 1974). These seiches are characterised by a period of approximately 40 minutes and their very long wavelengths seem preclude any significant impact on the formation of elongated pockmarks. 6.1.7 Bottom currents and the nature of oceanographic data It has to be emphasized however, that the available oceanographic data do not permit clear definition of the characteristics of bottom currents within the Gulf of Thailand. This fact is related to the methodology behind standard oceanographic measurements, which are focused usually on the surficial part of the basin rather than water masses close to seabed. Most publications present a generalised current scheme for the whole basin rather than separate solutions for different bathymetric ranges. The analysis of seabed morphology within central depression of the Gulf of Thailand shows no correlation between these generalised current patterns and the elongation of seabed features. This shows a need for oceanographic studies of sea currents beyond the 50 m 135 water depth thermohalocline, which is required for a better understanding of specific environmental conditions within this zone of the Gulf of Thailand. 6.1.8 Summary According to Robinsons (1974), the data from current meter stations located within the Gulf of Thailand near the axis of the basin indicate that the prevailing currents associated with wind forcing and density contrasts appear to overprint tidal effects. The review of various types of currents occurring in the Gulf of Thailand, given above, shows that there is no single current type that can readily explain the formation of elongated seabed features. The arguments presented here show that pockmark elongation is probably caused by a combination of tidal currents (predominantly flood current) and density-driven currents, both related to water inflow and outflow between the South China Sea and Gulf of Thailand. The erosion may occur normally during tidal maxima and during the March-October season when the strongest density-driven currents occur. During the rest of the year, currents seem to be relatively weak and unlikely to have significant erosive capacity. 136 7 CONCLUSIONS (1) Three main stratigraphic units were identified in the shallowest deposits of the central part of the Gulf of Thailand, based on seismic data, gravity cores, geotechnical logs and radiocarbon dating. Unit C is an Upper Pleistocene assemblage of transgressiveregressive sequences. Unit B represents marine sediments, which were subjected to lateritization during the last glacial maximum and were subsequently reworked during the Holocene transgression. Unit A comprises Holocene marine muds and clays. This tripartite stratigraphy is consistent across the whole basin and shares similarity to the Quaternary sequence known from the Lower Central Plain of Thailand, where marine sediments of Unit A correlate with the Bangkok Clay Formation and deposits of Unit B with the Stiff Clay Member of the Chao Phraya Formation, respectively. (2) Unit C in the central gulf contains Upper Pleistocene prodeltaic and delta front sediments up 40 m thick, which were derived from an easterly source and are interpreted as the deposits of the palaeo-Mekong River. The other sediments of Unit C are either of marine origin or delivered from a northerly source, interpreted as deposits of the palaeo-Chao Phraya River. (3) Units B and A are separated by erosional unconformity R1, which reflects a Holocene transgressive ravinement and forms one of the most important and easily recognizable stratigraphic horizons across the Gulf of Thailand. This surface is expressed as a strong seismic reflector at the soily top of Unit B, mantled locally by transgressive shell-gravel pavement below the seismically transparent marine muds of Unit A. The radiocarbon ages of shell detritus derived from the transgressive lag date the ravinement process at 10.4-10.6 cal kyr BP for the central-southern part of the basin and >6.5 cal kyr BP for the western area. These ages are consistent with the rapid, though diachronous, northward spread of the Holocene transgression, which began in the central Sunda Shelf at 11–11.5 cal kyr BP and reached the head of the Thailand Bay at 8–7 cal kyr BP. (4) The modern seabed morphology within the deepest part (below 50 m water depth) of the Gulf of Thailand basin can be divided into three main types: (i) a thick uniform layer of the Holocene soft marine mud, (ii) mounds and ridges of the Holocene soft marine mud, and (iii) flat pre-Holocene stiff silt surface veneered with thin semi-fluid mud. These morphological types are heavily pockmarked and linked by a variety of intermediate forms. This complex seabed topography is interpreted to reflect an interaction between sediment dewatering and the erosional activity of the present-day bottom currents. The sediment dewatering and fluid seepage result in the formation of numerous, randomly or distributed small pits and pockmarks. The sporadically occurring 137 larger pockmarks and fields of pockmarks reflect the emanation of gas or other fluids, which are sourced in deeper parts of the sediment column and are released in rapid bursts, or in a long-term, gradual way. (5) The erosional activity of bottom currents results in an increased elongation of these features, which originated initially as circular forms within the unconsolidated muds. The long-term erosion imposed by currents of stable orientation modifies small, elongated pockmarks into long runnels and depressions, and ultimately, leads to formation of the most common present seabed morphology of the Gulf of Thailand that is represented by large fields of elongated mounds and ridges, as well as the residual outliers of un-eroded mud and clay sheets. (6) The development the seabed morphology proceeded in five evolutionary steps. These comprise (i) pre-erosive phase (early Holocene) when broad and thick sheets of soft clays covered the basal Holocene ravinement surface, (ii) formation of elongated pockmarks (mid-late Holocene); (iii) enlargement of depressions and channel development (mid-late Holocene), (iv) development of mud mounds and ridges (mid-late Holocene), and finally, (v) flattening of the seabed (mid-late Holocene) representing the ultimate output of prolonged erosion of mud mounds. (7) The mud ridges and runnels located in-between show a stable NW-SE alignment. This direction follows the basin axis trend and is generally persistent throughout the whole basin. Such elongation of pockmarks, mud mounds and ridges is believed to be generated by the combination of tidal currents, predominantly flood current, and density-driven currents, both related to water in- and outflow between the South China Sea and Gulf of Thailand. The erosion intensifies most likely during tidal maximums and during the March-October season when the strongest thermohalocline develops at ca 50 m water depth. The thermohalocline separates a lower water mass, dominated by unidirectional bottom currents, from upper one, where wind-driven and ebb-flow tidal currents subjected to Coriolis force result in a multidirectional water circulation. (8) The available current meter data from the Gulf Thailand are biased towards the surface water circulation and cannot simply be extrapolated to define the behaviour of a water mass extending below the 50 m isobath. The results of the present contribution emphasize the importance of bottom currents in creating the deeper-water bottom morphology of the gulf and highlight the significance of water stratification in this process. This also indicates a need for erecting a separate circulation model for waters located below the thermohalocline. (9) The present dissertation brings the first detailed account on the morphology and evolution of the pockmarked mud mounds and ridges in Gulf of Thailand. Similar sea 138 floor topography has been reported from many shelf basins, though in none of these is the scale of mud ridges so large and their orientation as stable as in this area. The uniqueness of the Gulf of Thailand in this respect appears to reflect a combination of the presence of an unconsolidated mud veneer upon compacted mud at seabed with the activity of persistent unidirectional bottom currents, which were confined along the basin axis and decoupled by the thermohalocline from any major influence of the wind-driven surficial circulation. 139 REFERENCES Aksornkoae, S., 1993. Ecology and Management of Mangroves. IUCN, Bangkok, Thailand, 176 pp. Andresen, K.J., Huuse, M., and Clausen, O.R., 2008. Morphology and distribution of Oligocene Miocene pockmarks in the Danish North Sea - implications for bottom current activity migration. Basin Research 20, 445–466. Anikiyev, V.V., Zaytsev, O.V., Trinh, The Hieu, Savil’Yeva, I.I., Starodubtsev, Y., and Shumilin, Y.N.. 1986. Variation in the time-space distribution of suspended matter in the coastal zone of the Mekong River. Oceanology 26, 725–729. Biswas, B., 1973. Quaternary change in sea- level in the South China Sea. Geological Society Malaysia Bulletin 6, 229-256. Boe, R., Rise, L., and Ottesen, D., 1998. Elongate depressions on the southern slope of the Norwegian Trench (Skagerrak): morphology and evolution. Marine Geology 146, 191-203. Bowditch, N., 1995. The American Practical Navigator. An Epitome of Navigation. National Imagery and Mapping Agency, Bethesda, Maryland. Chapter 9: Tides and Tidal Currents, 143-164. Brown, D.J, Helmke, P.A., and Clayton, M.K., 2003. Robust geochemical indices for redox and weathering on a granitic laterite landscape in Central Uganda. Geochimica et Cosmochimica Acta 67, 2711-2723. BS 5930:1999, 1999. Code of practice for site investigations. British Standards Institution. ISBN 9780580616228. Carling, P.A., Williams, J.J., Croudace, I.W., and Amos, C.L., 2009. Formation of mud ridge and runnels in the intertidal zone of the Severn Estuary, UK Continental Shelf Research 29, 1913-1926. Cartwright, J., 1995. Seismic-stratigraphical analysis of large-scale ridge-trough sedimentary structures in Late Moicene to Early Pliocene of the central North Sea. IAS Special Publication 22, 285-303. Chappell, J., Omura, A., Esat, T., McCulloch, M., Pandolfi, J., Ota Y., and Pillans, B., 1996. Reconciliation of late Quaternary sea levels derived from coral terraces at Huon Peninsula with deep-sea oxygen isotope records. Earth and Planetary Science Letters 141, 227-236. Cheevaporn, V., and Menasveta, P., 2003. Water pollution and habitat degradation in the Gulf of Thailand. Marine Pollution Bulletin 47, 43–51. Clowes, A., and Comfort, P., 1982. Process and landform: An outline of contemporary geomorphology. Edinburgh: Oliver and Boyd, 289 pp. 140 Cox, J. B., 1968. A review of the engineering characteristics of the recent marine clays in southeast Asia. Asian Institute of Technology Research Report 6, Bangkok. Dheeradilok, P., 1995. Quaternary coastal morphology and deposition in Thailand. Quaternary International 26, 49-54. Emery, E.O., and Niino, I., 1963. Sediments of the Gulf of Thailand and Adjacent Continental Shelf. Geological Society of America Bulletin 74, 541–554. Encarta, 2006a. “World Atlas.” Microsoft® Encarta® 2006 [DVD]. Microsoft Corporation, 2005. Encarta, 2006b. “Thailand.” Microsoft® Encarta® 2006 [DVD]. Microsoft Corporation, 2005. Encarta, 2006c. “Cambodia.” Microsoft® Encarta® 2006 [DVD]. Microsoft Corporation, 2005. Encarta, 2006d. "Monsoon." Microsoft® Encarta® 2006 [DVD]. Microsoft Corporation, 2005. Encarta, 2006e. Department of Botany, Natural History Museum. "Mangrove." Microsoft® Encarta® 2006 [DVD]. Microsoft Corporation, 2005. EN ISO-14688-1:2002, 2002. Geotechnical investigation and testing - Identification and classification of soil - Part 1: Identification and description. International Organization for Standardization, 12 pp. Fang, G., Kwok, Y-K., Yu, K., and Zhu, Y., 1999. Numerical simulation of principal tidal constituents in the South China Sea, Gulf of Tonkin and Gulf of Thailand. Continental Shelf Research 19, 845-869. Folk, R.L., and Ward W. C., 1957. Brazos River bar: a study in the significance of grain size parameters. Journal of Sedimentary Petrology 27, 1, 3-26. Folk R.L., 1974. Petrology of Sedimentary Rocks, Second Edition. Hemphill's, Austin, 170 pp. Friedman, G.M., and Sanders, J.E., 1978. Principles of Sedimentology. New York. Wiley & Sons, Inc., 792 pp. Fugro NV, 2001. Geophysical & geotechnical techniques for the investigation of nearseabed soils and rock, Fugro NV, 2001, www.fugro.com. Fugro, 2006. Fundamentals of High Resolution Sub-bottom Profiling. Internal technical note. Fugro Survey Pte Ltd, Singapore, 69 pp. Fugro/Starfix.HP, 2003. Starfix-HP TM SPM2000A brochure. Internal technical note. Fugro Survey Pte Ltd, Singapore, 2 pp. Gorshkov, S.G, Faleev, V.I., and Demin, L.A (Eds.). 1974. Atlas Okeanov (Atlas of Oceans). Vol 1: Tikhii Okean (The Pacific Ocean). Glavnoe upravlenie navigacii i okeanografii Ministerstva oborony Vaenno-morskoj flot, Moscow, 306 pp [in Russian]. 141 Hanebuth, T.J.J., 2000. Sea-level changes on the Sunda Shelf during the last 50,000 years. Ph.D.Thesis. Christian-Albrecht-Universitat zu Kiel, Kiel, 104 pp. Hanebuth, T.J.J., and Stattegger, K., 2004. Depositional sequences on a late Pleistocene–Holocene tropical siliciclastic shelf (Sunda Shelf, southeast Asia). Journal of Asian Earth Sciences 23, 113–126. Hanebuth, T.J.J., Stattegger, K., and Saito, Y., 2002. The architecture of the central Sunda Shelf (SE Asia) recorded by shallow-seismic surveying. Geo-Marine Letters 22, 86-94. Hanebuth, T.J.J., Stattegger, K., Schimanski, A., Ludmann, T., and Wong, H. K., 2003. Late Pleistocene forced-regressive deposits on the Sunda Shelf (Southeast Asia). Marine Geology 199, 139-157. Harrington, P.K., 1985. Formation of pockmarks by pore-water escape. Geo-Marine Letters 5, 193-197. Heward, A.P., Chuenbunchom, S., Makel, G., Marsland D., and Spring, L., 2000. Nang Nuan oil field, B6/27, Gulf of Thailand: karst reservoirs of meteroic or deep burial origin? Petroleum Geoscience 6, 15–27. Hill, I.G., Worden, R.H., and Meighan, I.G., 2000. Geochemical evolution of a palaeolaterite: the Interbasaltic Formation, Northern Ireland. Chemical Geology 166, 65–84. Hickin, E.J., 2003. Meandering Channels, in, Middleton, Gerard V., Encyclopedia of Sediments and Sedimentary Rocks, New York: Springer, pp. 430-433. Hovland, M., 2003. Geomorphological, geophysical, and geochemical evidence of fluid flow through the seabed. Journal of Geochemical Exploration 78-79, 287-291. Hovland, M., Gardner, J.V., and Judd, A.G., 2002. The significance of pockmarks to understanding fluid flow processes and geohazards. Geofluids 2, 127-136. Hovland, M., Jensen, S., and Fichler, C., 2012. Methane and minor oil macro-seep systems - their complexity and environmental significance. Marine Geology 332– 334, 163–173. Hovland, M., and Judd, A.G., 1988. Seabed Pockmarks and Seepages, Impact on Geology, Biology and the Marine Environment. Graham and Trotman, London, 293 pp. Hughen, K. A., Lehman, S., Southon, J., Overpeck, J., Marchal, O., Herring, C., and Turnbull, J., 2004. C-14 activity and global carbon cycle changes over the past 50,000 years. Science 303, 202-207. Hutchison, C. S., 1983. Multiple Mesozoic Sn-W-Sb granitoids of Southeast Asia. In, J. A. Roddick, Circum-Pacific plutonic terrains, Geological Society of America Memoirs 159, 35–60. 142 Jardine, E., 1997. Dual petroleum systems governing the prolific Pattani Basin, offshore Thailand. Petroleum systems of S.E. Asia and Australasia Conference, Jakarta, May 21–23, 1997, 351–363. Josenhans, H.W., King, L.H., and Fader, G.B., 1978. A side scan sonar mosaic of pockmarks on the Scotian Shelf. Canadian Journal of Earth Sciences 15, 831-840. Kealhofer, L., 2002. Changing perceptions of risk: The development of agro-ecosystems in Southeast Asia. American Anthropologist 104, 178–194. Kongsangchai, J., 1995. Problems of mangrove degradation in Thailand. In: Ecology and Management of Mangrove Restoration and Regeneration in East and Southeast Asia. Proceeding of ECOTONE IV 18-22 January 1995. Wang Tai Hotel, Surat Thani, Thailand, 119-128. Koompans, B.N., 1972. Sedimentation in the Kelantan Delta (Malaysia). Sedimentary Geology 7, 65-84. Latypov, Yu.Ya., 1995. Korally skleraktinii V’etnama. Ch. III. Faviidy, Fungiidy (Scleractinian Corals of Vietnam. P. III. Faviidae, Fungiidae), Moscow: Nauka. Latypov, Yu.Ya., 2003. Reef-Building Corals and Reefs of Vietnam: 1. The Gulf of Thailand. Russian Journal of Marine Biology, 29, Suppl. 1, S22–S33. Lekkerkerk, H-J., Van der Velden, R., Roders, J., Haycock, T., De Vries, R., Jansen, P., and Beemster, C., 2006. Handbook of offshore surveying. Acquisition & Processing. Vol. 2, Clarkson Research Services Ltd, London, 313 pp. Leloup, P.H., Lasassin, R., Tapponier, P., Scharer, U., Dalai, Z., Xiaohan, L., Liangshang, Z., Shaocheng, J., and Trinh, P.T., 1995. The Ailao Shan–Red River shear zone (Yunnan, China), Tertiary transform boundary of Indochina. Tectonophysics 251, 3– 84. Lian, H.M., and Bradley, K., 1986. Exploration and development of natural gas, Pattani Basin, Gulf of Thailand. Transactions of the Fourth Circum-Pacific Energy and Mineral Resources Conference, Singapore, pp. 171–181. Loi, Tran Ngoc, 1965. "Les Conditions éologiques de la Mer de Chine Méidionale", Contributions Institute of Oceanography Nhatrang 83, 249–267. Lowe, D.R., 1975. Formation of water escape structures in coarse-grained sediments. Sedimentology 22, 157-204. Moh, Z.C., Nelson, J.D., and Brand E.W., 1969. Strength and deformation behaviour of Bangkok Clay. In: Proceedings of the Seventh International Conference on Soil Mechanics, Vol. 1, Mexico City, 287 pp. Oudom-Ugsorn, P., Bunopas, S., Achalabhuti, C., and Polachan, S., 1986. Petroleum potential in pre-Tertiary basins of Thailand. ASCOPE, Indonesia, October 12, p. 1997. 143 Packham, G., ,1996,. In, R. Hall and D. J. Blundell, Cenozoic SE Asia: Reconstructing its aggregation and reorganization, Tectonic evolution of Southeast Asia. London: Geological Society of London Special Publication 106, 123-152. Pham, Hoang Ho, 1985. Vegetation of Phu Quoc Island (Thuc Vat Dao Phu Quoc). Ho Chi Minh Publishing Co., pp. 128–156. Pigott, J.D., and Sattayarak, N., 1993. Aspects of sedimentary basin evolution assessed through tectonic subsidence analysis. Example: northern Gulf of Thailand. In B. K. Tan, T. T. Khoo, A. J. Barber, Y. Cheng, J. F. Lovering and A. A. Meyerhoff, Geosea VII Proceedings, pp. 407–420. Polchan, S., and Sattayarak N., 1989. Strike-slip tectonics and the development of Tertiary basins in Thailand. International Symposium on Intermontane Basins: Geology and Resources, Chiang Mai, Thailand, pp. 243–253. Puchała, R., Porębski, S.J., Śliwiński, W., and August, C., 2011. Pleistocene to Holocene transition in the central basin of the Gulf of Thailand, based on geoacoustic survey and radiocarbon ages. Marine Geology 288, 103-111. Pukasab, P., and Pochanasomburana, P., 1957. The types of tides and mean sea level in the Gulf of Thailand. Hydrographic Dept., Royal Thai Navy, 7 pp. Rao, A.N., 1986. Mangrove ecosystems of Asia and the Pacific. In: Mangrove of Asia and the Pacific: Status and Management. Tech. Rep. UNDP/UNESCO, 48 pp. Rau, J.L., and Nutalaya, P., 1983. Geology of the Bangkok clay. Geological Society of Malaysia Bulletin 16, 99–116. Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Bertrand, C.J.H., Blackwell, P.G., Buck, C.E., Burr, G.S., Cutler, K.B., Damon, P.E., Edwards, R.L., Fairbanks, R.G., Friedrich, M., Guilderson, T.P., Hogg, A.G., Hughen, K.A., Kromer, B., McCormac, G., Manning, S., Bronk, Ramsey, C., Reimer, R.W., Remmele, S., Southon, J.R., Stuiver, M., Talamo, S., Taylor, F.W., Van der Plicht J., and Weyhenmeyer, C.E., 2004. IntCal04 Terrestrial Radiocarbon Age Calibration, 0–26 cal kyr BP". Radiocarbon 46, 1029-1059. Reimer, P.J., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk Ramsey, C., Grootes, P.M., Guilderson, T.P., Haflidason, H., Hajdas, I., Hatte, C., Heaton, T.J., Hoffmann, D.L., Hogg, A.G., Hughen, K.A., Kaiser, K.F., Kromer, B., Manning, S.W., Niu, M., Reimer, R.W., Richards, D.A., Scott, E.M., Southon, J.R., Staff, R.A., Turney, C.S.M., and Van der Plicht, J., 2013. IntCal13 and Marine13 Radiocarbon Age Calibration Curves 0-50,000 Years cal BP. Radiocarbon 55, 1869-1887. Reson, 2002. SeaBat 8101 Multibeam Echosounder System. Operator’s manual. Version 3.01. Reson Inc., 148 pp. 144 Robinson, M.K., 1974. The physical oceanography of the Gulf of Thailand, Naga Expedition; Naga Expedition, Cruise S11. UC San Diego: Scripps Institution of Oceanography. Retrieved from: http://escholarship.org/uc/item/4mf3d0b7 Royal Thai Navy, 1995. Hydrographic sheet of the Gulf of Thailand: east coast. Hydraulic Department, Royal Thai Navy. Royal Thai Navy, 1996. Hydrographic sheet of the Gulf of Thailand: west coast. Hydraulic Department, Royal Thai Navy. Shackleton, N.J., 1987. Oxygen isotopes, ice volume and sea level. Quaternary Science Reviews 6, 183-190. Shepard, F.P., 1954. Nomenclature based on sand-silt-clay ratios. Journal of Sedimentary Petrology 24, 151-158. Sinsakul, S., 2000. Late Quaternary geology of the Lower Central Plain, Thailand. Journal of Asian Earth Sciences 18, 415-426. Snidvongs, A., 1998. The oceanography of the Gulf of Thailand: Research and management policy.. In, SEAPOL Integrated Studies of the Gulf of Thailand, Vol. 1, ed. by D. M. Johnston, Southeast Asian Programme in Ocean Law, Policy and Management, 1–68. Somboon, J.R.P., 1988. Paleontological study of the recent marine sediments in the lower central plain, Thailand. Journal of Southeast Asian Earth Sciences 2, 201–210. Somboon, J.R.P., and Thiramongkol, N., 1992. Holocene highstand shoreline of the Chao Phraya Delta, Thailand. Journal of Southeast Asian Earth Sciences 7, 53–60. Songtham, W., Watanasak, M., Insai, P., 2000. Holocene marine crabs and further evidence of a sea-level peak at 6,000 years BP in Thailand. In: The Comprehensive Assessments on Impacts of Sea-Level Rise. Proceedings of the Thai-Japanese Geological Meeting, Department of Mineral and Resources, Thailand, pp. 89–97. Srisuksawad, K., Porntepkasemsan, B., Nouchpramool, S., Yamkate, P., Carpenter, R., Peterson, M.L., and Hamilton, T., 1997. Radionuclide activities, geochemistry, and accumulation rates of sediments in the Gulf of Thailand. Continental Shelf Research 17, 925-965. Stansfield, K., and Garrett, C., 1997. Implications of the salt and heat budgets of the Gulf of Thailand. Journal of Marine Research 55, 935-963). Suensilpong, S., Nakornsri, N., Dheeradilok, P., Lumjuan, A., Tansathien, W., and Vimuktanandana, S., 1982. Geological Map of Thailand. Geological Survey Division. Department of Mineral Resources, Thailand, Scale 1: 1,000,000, 2 sheets. Ta, T.K.O., Nguyen, V.L., Tateishi, M., Kobayashi, I., Tanabe, S., and Saito, Y., 2002a. Holocene delta evolution and sediment discharge of the Mekong River, southern Vietnam. Quaternary Science Reviews 21, 1807-1819. 145 Ta, T.K.O., Nguyen, V.L., Tateishi, M., Kobayashi, I., Saito, Y., and Nakamura, T., 2002b. Sediment facies and late Holocene progradation of the Mekong River Delta in Bentre Province, southern Vietnam: an example of evolution from a tide dominated to a tide- and wave-dominated delta. Sedimentary Geology 152, 313-325. Tanabe, S., Saito, Y., Saito, Y., Suzuki, Y., Sinsakul, S., Tiyapairach, S., and Chaimanee, N., 2003. Stratigraphy and Holocene evolution of the mud-dominated Chao Phraya delta, Thailand. Quaternary Science Reviews 22, 789–807. Tan, B.K., 1996). ‘Suture Zones’ in peninsular Malaysia and Thailand: implications for palaeotectonic reconstruction of southeast Asia. Journal of Southeast Asian Earth Sciences 13, 243-249. Thampanya, U., Vermaat, J.E., Sinsakul, S., and Panapitukkul, N., 2006. Coastal erosion and mangrove progradation of Southern Thailand. Estuarine, Coastal and Shelf Science 68, 75-85. Voris, H. K., 2000. Maps of Pleistocene sea levels in Southeast Asia: shorelines, river systems and time durations. Journal of Biogeography 27, 1153-1168. Watcharanantakul, R., and Morley, C.K., 2000. Syn-rift and post-rift modelling of the Pattani Basin, Thailand: evidence for a ramp-flat detachment. Marine and Petroleum Geology 17, 937–958. Wattayakorn, G., King, B., Wolanski, E., and Suthanaruk, P., 1998. Seasonal dispersion of petroleum contaminant in the Gulf of Thailand. Continental Shelf Research 18, 641-659. Wentworth, C.K., 1922. A scale of grade and class terms for clastic sediments. Journal of Geology 30, 377-392. White, J.C., Penny, D., Kealhofer, L., and Maloney, B., 2004. Vegetation changes from the late Pleistocene through the Holocene from three areas of archaeological significance in Thailand. Quaternary International 113, 111–132. Whittington, B.I., and Muir, D., 2000. Pressure acid leaching of nickel laterites: a review. Mineral Processing and Extractive Metallurgy Review 21, 527-600. Williams, J.J., Carling, P.A., Amos, C.L., and Thompson, C., 2008. Field investigation of ridge–runnel dynamics on an intertidal mudflat. Estuarine, Coastal and Shelf Science 79, 213–229. Wolanski, E., Nguyen, N H., Le Trong, D., Nguyen, H N., and Nguyen, N. T., 1996. Finesediment dynamics in the Mekong River Estuary, Vietnam. Estuarine, Coastal and Shelf Science 43, 565-582. Woodroffe, C.D., 2000. Deltaic and estuarine environments and their Late Quaternary dynamics on the Sunda and Sahul shelves. Journal of Asian Earth Sciences 18, 393– 413. 146 Yanagi, T., and Takao, T., 1998. Clockwise phase propagation of semi-diurnal tides in the Gulf of Thailand. Journal of Oceanography 54, 143-150. Yanagi, T., Sachoemar, S.I., Takao, T., and Fujiware, S., 2001. Seasonal variation of stratification in the Gulf of Thailand. Journal of Oceanography 57, 461-470. 147