Reef Growth and Maintenance WAMSI Kimberley Marine Research

Home , Discordant coastline, Turtle Reef

Reef Growth and Maintenance Lindsay B Collins1,3, Alexandra Stevens1,3, Mick O’Leary 2,3, Giada Bufarale 1,3, Moataz Kordi 1,3, Tubagus Solihuddin 1,3 1Department of Applied Geology, Western Australian School of Mines, Curtin University, Perth, Western, Australia 2Department of Environment and Agriculture, Curtin University, Perth, Western, Australia 3Western Australian Marine Science Institution (WAMSI), Perth, Western Australia, Australia

WAMSI Kimberley Marine Research Program Report Project 1.3.1

June 2016

Reef Growth and Maintenance

WAMSI Kimberley Marine Research Program Initiated with the support of the State Government as part of the Kimberley Science and Conservation Strategy, the Kimberley Marine Research Program is co-invested by the WAMSI partners to provide regional understanding and baseline knowledge about the Kimberley marine environment. The program has been created in response to the extraordinary, unspoilt wilderness value of the Kimberley and increasing pressure for development in this region. The purpose is to provide science based information to support decision making in relation to the Kimberley marine park network, other conservation activities and future development proposals.

Ownership of Intellectual property rights Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this publication is owned by the Western Australian Marine Science Institution and Curtin University.

Copyright © Western Australian Marine Science Institution All rights reserved. Unless otherwise noted, all material in this publication is provided under a Creative Commons Attribution 3.0 Australia Licence. (http://creativecommons.org/licenses/by/3.0/au/deed.en)

Legal Notice The Western Australian Marine Science Institution advises that the information contained in this publication comprises general statements based on scientific research. The reader is advised and needs to be aware that such information may be incomplete or unable to be used in any specific situation. This information should therefore not solely be relied on when making commercial or other decision. WAMSI and its partner organisations take no responsibility for the outcome of decisions based on information contained in this, or related, publications.

Front cover images (L-R)

Image 1: Satellite image of the Kimberley coastline (Landgate) Image 2: Getting ready to lift core PN5, with assistance from Bardi Jawi Rangers. (Image: Tubagus Solihuddin) Image 3: Humpback whale breaching (Image: Pam Osborn) Image 4: Getting ready for coring with help from Bardi Jawi Rangers and Erin McGinty from KMRS. (Image: Tubagus Solihuddin) Reef Growth and Maintenance

Year of publication: 2016

Metadata: http://catalogue.aodn.org.au/geonetwork/srv/eng/metadata.show?uuid=7ab491d2-9507-428c-aed1- 091d2aaed521

Citation: Collins et al (2015). Final Report of Project 1.3.1 of the Kimberley Marine Research Program Node of the Western Australian Marine Science Institution, WAMSI, Perth, Western Australia, 246pp.

Author Contributions: LC was project leader, fieldwork co-leader for the 2013 seismic fieldwork and provided technical advice, AS compiled milestone reports, managed project data, provided logistic support, technical advice and assistance and participated in the 2013 seismic fieldwork. MO was fieldwork co-leader for the 2013 fieldwork and fieldwork leader for the 2014 coring fieldwork and provided technical advice. MK undertook spatial analysis of satellite imagery and orthophotography of the Kimberley Bioregion, created the geodatabase, ReefKIM and participated in the 2013 and 2014 fieldwork. GB undertook analysis of the seismic profiles collected during the 2013 field work, assisted with report compilation, provided logistic support and participated in the 2013 fieldwork. TS undertook analysis of the Cockatoo Island minepit data and the cores collected during the 2014 fieldwork and participated in the 2013 and 2014 fieldwork.

Corresponding author and Institution: M O'Leary ([email protected]), Department of Applied Geology, Western Australian School of Mines, Curtin University, Bentley, WA, Australia.

Funding Sources: This project was funded by the Western Australian Marine Science Institution Joint Venture Partners as part of the WAMSI Kimberley Marine Research Program, a $30M program with seed funding of $12M provided by State government as part of the Kimberley Science and Conservation Strategy.

Competing Interests: The commercial investors and data providers had no role in the data analysis, data interpretation, the decision to publish or in the preparation of the manuscript. The authors have declared that no competing interests exists.

Kimberley Traditional Owner agreement: This research was enabled by the Traditional Owners through their advice, participation and consent to access their traditional lands.

Acknowledgements: This study is part of the Kimberley Reef Geomorphology Project 1.3.1 which has been funded by the Western Australian State Government through the Western Australian Marine Science Institution (WAMSI). We are grateful to Traditional Owners of the Kimberley land (the Bardi Jawi, Mayala and Dambimangari people) for their assistance, advice and consent to access their traditional lands. We also wish to thank the following: The Kimberley Marine Research Station, in particular Erin McGinty and James Brown, for providing vessels and logistic and in kind support for marine operations and access to research facilities; WA Museum, in particular Clay Bryce and Zoe Richards for providing advice and ground truth data through the WA Museum/Woodside Collection Project (Kimberley) 2008 – 2011; Kimberley Media for providing quality site pictures; The Department of Parks and Wildlife (DPaW) for providing VHR orthophotographs and HR satellite images; Mark Hardman (Fugro Satellite Positioning Pty Ltd) for supplying the DGPS; Neil MacDonald (AAEngineering Ltd) and Western Advance for the equipment support; Giovanni De Vita for his technical advice; Pluton Resources (particularly Jeremy Bower and Anson Griffith) are thanked for providing access to parts of their Cockatoo Island Mining Tenement and for logistic support during the study; David Blake from MScience for fieldwork assistance; MScience is thanked for providing access to marine video of the reef; Geoscience Australia (GA) for providing DEMs and GIS data; The United States Geological Survey (USGS) for providing Landsat images; The Geological Survey of Western Australia (GSWA) for providing digital geological maps. Finally, it must be noted that this research was completed in an area where the Traditional Owners have a rich cultural history of climate, land and environment based on thousands of years of habitation. It is important to consider that broad understanding alongside the modern science presented here.

Collection permits/ethics approval: No collection occurred in the production of this report.

Reef Growth and Maintenance

Contents

TABLE OF FIGURES TABLE OF TABLES EXECUTIVE SUMMARY ...... I IMPLICATIONS FOR MANAGEMENT ...... I 1 INTRODUCTION ...... 1

1.1 STUDY AREA ...... 1 1.2 STUDY SCOPE AND OBJECTIVES ...... 2 1.3 BACKGROUND ...... 3 Project Objectives and Activities ...... 3 Timeline of events ...... 3 2 PROJECT ESTABLISHMENT...... 4

2.1 PROJECT OBJECTIVES AND ACTIVITIES COMMENCED ...... 4 2.2 PROJECT ACTIVITIES ...... 4 Establishment of a Coral Reef geodatabase ...... 4 Map of Islands and Reef Distribution ...... 5 Seabed & bathymetric maps in GIS compatible format ...... 5 Preliminary morphologic classification of the reefs of the Kimberley Region ...... 6 Sea-level curve and drowning history for the Kimberley Region ...... 6 Drowning Model ...... 8 2.3 CONCLUSIONS ...... 9 2.4 REFERENCES ...... 10 Data sources ...... 10 3 MATERIALS AND METHODS – DATA MANAGEMENT ...... 11

3.1 DATA ACQUISITION ...... 12 3.2 DATA SOURCES ...... 13 Remotely sensed images ...... 13 Landsat Imagery ...... 13 High Resolution Satellite Images ...... 15 Aerial photography (Orthophotos) ...... 15 Maps and Charts ...... 15 Ground Truth ...... 15 Sub-Bottom Profiling ...... 15 Sediment Samples ...... 15 Isotopic Data ...... 16 Cores ...... 16 Data Integration ...... 16 3.3 DATA PROCESSING ...... 17 Pre-processing ...... 17 Metadata ...... 17 3.4 FILE FORMATS AND MANAGEMENT ...... 17 File Formats ...... 17

Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

File Naming ...... 17 Directories ...... 18 Version Control ...... 18 3.5 DATA ANALYSIS ...... 18 3.6 DATA STORAGE ...... 19 3.7 DATA ACCESSIBILITY ...... 19 3.8 FUTURE DATA ACCESSIBILITY ...... 19 3.9 COPYRIGHT AND INTELLECTUAL PROPERTY ...... 19 Copyright ...... 19 Intellectual Property ...... 20 3.10 CONCLUSION ...... 20 3.11 REFERENCES ...... 21 3.12 DATA SOURCES...... 21 4 MAPPING AND GEOMORPHIC CLASSIFICATION OF THE KIMBERLEY REEFS, NORTH WEST AUSTRALIA 22

4.1 INTRODUCTION ...... 22 Reef Mapping and Classification ...... 22 North West Australian Reef Systems ...... 23 Study area ...... 24 4.2 METHODOLOGY ...... 24 Datasets ...... 24 Field surveys ...... 24 Mapping approach ...... 25 4.3 RESULTS ...... 27 Coastline, island and reef mapping ...... 27 Geomorphic classification of the Kimberley Reefs ...... 30 Oceanic Shoals Bioregion...... 31 Kimberley Bioregion ...... 32 Reef distribution and size ...... 35 Reef distribution by latitude ...... 37 4.4 DISCUSSION ...... 37 4.5 ACKNOWLEDGEMENTS ...... 39 4.6 REFERENCES ...... 39 5 MAPPING INTRA-REEF GEOMORPHOLOGY AND ASSOCIATED SUBSTRATES AND HABITATS OF THE KIMBERLEY BIOREGION, NORTH-WESTERN AUSTRALIA...... 41

5.1 INTRODUCTION ...... 41 Remote sensing ...... 42 5.2 METHODOLOGY ...... 43 Study area ...... 43 Remote sensing datasets ...... 43 Ground truth data ...... 45 Data processing ...... 46 Accuracy assessment ...... 48 5.3 RESULTS ...... 48 Fringing reefs ...... 48 Maret Islands...... 48

Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Cape Londonderry ...... 49 Molema Island ...... 50 Bathurst and Irvine Islands ...... 51 Tallon Island ...... 53 Planar reefs...... 53 Montgomery Reef ...... 54 Adele Reef ...... 54 Long Reef ...... 55 High intertidal reefs ...... 57 Similarities and differences between reefs in the study area ...... 57 5.4 DISCUSSION AND CONCLUSIONS ...... 59 5.5 ACKNOWLEDGEMENTS ...... 60 5.6 REFERENCES ...... 61 6 REEFKIM: AN INTEGRATED GEODATABASE FOR SUSTAINABLE MANAGEMENT OF THE KIMBERLEY REEFS, NW AUSTRALIA ...... 63

6.1 INTRODUCTION ...... 63 Reef geodatabases ...... 63 Crowdsourcing data ...... 64 6.2 METHODOLOGY ...... 64 Data sources ...... 65 Data processing ...... 66 Data integration ...... 66 Data accessibility ...... 66 6.3 RESULTS ...... 67 6.4 DISCUSSION ...... 73 6.5 CONCLUSIONS ...... 74 6.6 ACKNOWLEDGEMENTS ...... 74 6.7 REFERENCES ...... 74 7 QUATERNARY ONSET AND EVOLUTION OF KIMBERLEY CORAL REEFS REVEALED BY HIGH RESOLUTION SEISMIC IMAGING ...... 76

7.1 INTRODUCTION ...... 76 7.2 ENVIRONMENTAL SETTING AND GEOLOGY ...... 77 7.3 MATERIALS AND METHODS ...... 78 7.4 RESULTS AND DISCUSSION ...... 79 Seismic Facies Analysis ...... 79 Inner Shelf Reefs ...... 79 Mid Shelf Reefs...... 80 Seismic Calibration ...... 81 Reef Classification Scheme ...... 82 High Intertidal, Intertidal and Subtidal Reefs ...... 83 Fringing, Inter-island Reef...... 83 Fringing, Bay Head Reefs ...... 89 Planar, Coralgal Reefs ...... 90

Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Patch Reefs and Shoals ...... 96 Sea Level Changes and Subsidence ...... 97 7.5 CONCLUSIONS ...... 99 7.6 ACKNOWLEDGEMENTS ...... 100 7.7 REFERENCES ...... 100 8 COCKATOO ISLAND REEF STUDY ...... 103

8.1 INTRODUCTION ...... 103 Oceanography of the Kimberley Region ...... 104 8.2 LOCATION AND METHODOLOGY ...... 105 Geophysical Surveys ...... 105 Reef Mapping ...... 105 Ore Pit Mapping ...... 105 Geochronology ...... 105 8.3 RESULTS ...... 106 Living coral community zonation ...... 106 Stratigraphy and palaeoecology ...... 108 Reef Geochronology and Growth History ...... 110 Reef Architecture and Seismic Structure ...... 114 8.4 DISCUSSION ...... 115 Evidence of neotectonic subsidence along the Kimberley Coast and offshore shelf ...... 115 Holocene reef growth ...... 116 Comparisons with other Holocene Reef Systems ...... 117 8.5 CONCLUSIONS ...... 119 8.6 ACKNOWLEDGEMENTS ...... 120 8.7 REFERENCES ...... 120 9 HOLOCENE REEF ACCRETION STYLES IN THE BUCCANEER ARCHIPELAGO...... 122

9.1 INTRODUCTION ...... 122 9.2 FIELD SETTINGS ...... 122 Biogeography ...... 122 9.3 GEOLOGY ...... 122 9.4 OCEANOGRAPHY ...... 123 9.5 METHODS ...... 124 Reef coring ...... 124 Core logging and sampling ...... 124 14C coral dating ...... 124 9.6 RESULTS ...... 124 Reef Stratigraphy ...... 124 Bathurst-Irvine Island ...... 124 Sunday Island ...... 127 Tallon Island ...... 129 Reef geochronology and Growth History ...... 132 Bathurst-Irvine Island ...... 132 Sunday Island ...... 134 Tallon Island ...... 136

Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

9.7 DISCUSSION ...... 139 Reconstruction of Holocene sea-level and growth history ...... 139 Chronology of Holocene reef build-up ...... 140 Rate and mode of Holocene reef accretion ...... 141 9.8 CONCLUSION ...... 142 9.9 REFERENCES ...... 142 9.10 ADELE ISLAND ...... 144 Ecological Characteristics ...... 147 Core Summary ...... 147 10 MANAGEMENT IMPLICATIONS AND RECOMMENDATIONS...... 149 11 CONCLUSIONS ...... 149

11.1 REEF MAPPING AND GEOMORPHOLOGY ...... 149 11.2 KIMBERLEY REEF SEISMIC STRATIGRAPHY ...... 149 11.3 REEF GEOLOGY, STRATIGRAPHY AND EVOLUTION ...... 150 12 COMMUNICATION ...... 151

12.1 PUBLICATIONS ...... 151 12.2 MEDIA AND INTERNET ...... 152 13 APPENDICES ...... 153

APPENDIX I ...... 153 APPENDIX II ...... 157 APPENDIX III ...... 159 APPENDIX IV ...... 165 APPENDIX V ...... 214 APPENDIX VI ...... 218 APPENDIX VII ...... 222 APPENDIX VIII ...... 236

Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Table of Figures

Figure 1 – Study Area, Arrows identify key study sites ...... 1 Figure 2. Sample output of digitised features ...... 5 Figure 3 – Preliminary reef classification scheme based on adaptation of previous studies and geoscientific data from this study ...... 6 Figure 4 – Sea-level curve estimation for Kimberley Bioregion at the Last Glacial Maximum ...... 7 Figure 5 – Model sea level curve based on data presented in Figure 4...... 7 Figure 6 – Predicted SL at 21.5k years BP about 120 m below present SL...... 8 Figure 7 – Predicted SL at 12.8k years BP about -50 m below present SL...... 9 Figure 8. Data Management Flowchart ...... 11 Figure 9 – A map of the study area, showing the main source of datasets...... 12 Figure 10 – Data Integration ...... 16 Figure 11 – Example of linking map features to an attribute table ...... 18 Figure 12 – Map of the North West Shelf (NWS) showing the spatial distribution of reefs, bioregion boundaries and the continental shelves (ramp) subdivision...... 23 Figure 13 – Detail of features evident in high resolution images...... 25 Figure 14. Landsat images of reefs near Jones Island (north KIM) acquired over three different dates ...... 26 Figure 15. Methods employed to map the coastline and islands and to identify reef locations and dimensions...... 27 Figure 16 – A digitised map showing the complexity of coastline features and island distribution along the Kimberley Bioregion coast...... 28 Figure 17 – Number and size of islands across the shelf...... 30 Figure 18 – A geomorphological typology classification scheme of the North West Shelf...... 31 Figure 19 – Morphology of slope atolls of the outer ramp and OSS Bioregion...... 32 Figure 20 – Morphology of fringing reefs of the KIM Bioregion...... 33 Figure 21 – Low intertidal planar reefs ...... 34 Figure 22 – Reef distribution in the Kimberley Bioregion by type: ...... 35 Figure 23 – Number and size of reefs across the shelf...... 36 Figure 24 – Map of the Kimberley Bioregion showing the location of the reefs in this study: ...... 42 Figure 25 – Satellite images used to map intra-reef geomorphology and associated substrates for the targeted reefs of this study: ...... 44 Figure 26 – Schematic representation of typical geomorphological zones boundaries and associated habitats and substrate cover in the Kimberley Bioregion...... 46

Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 27 – (a) Habitats/substrate classification map of the Maret Islands Reef; (b) satellite image of Maret Islands with ground-truth points highlighted; (c) geomorphic zones map...... 49 Figure 28 – (a) Satellite image of Cape Londonderry—the classification map was verified using very high-resolution (<1 m spatial resolution) aerial photography; (b) geomorphic zones map; (c) habitats/substrate classification map of Cape Londonderry Reefs...... 50 Figure 29 – (a) Satellite image of Molema Island (Turtle Reef) with ground-truth points highlighted; (b) geomorphic zones map; (c) habitats/substrate classification map of Turtle Reef...... 51 Figure 30 – (a) Habitats/substrate classification map of the reefs between Bathurst and Irvine Islands; (b) satellite image of Bathurst and Irvine Islands with ground-truth points highlighted; (c) geomorphic zones map...... 52 Figure 31. (a) Satellite image of Tallon Island, with ground-truth points highlighted; (b) geomorphic zones map; (c) habitats/substrate classification map of Tallon Island reefs...... 53 Figure 32 – (a) Habitats/substrate classification map of Montgomery Reef; (b) satellite image of Montgomery Reef with ground-truth points highlighted; (c) geomorphic zones map...... 54 Figure 33 – (a) Habitats/substrate classification map of Adele Reef; (b) satellite image of Adele Reef with ground-truth points highlighted; (c) geomorphic zones map...... 55 Figure 34 – (a) Habitats/substrate classification map of Long Reef; (b) satellite image of Long Reef, with ground-truth points highlighted; (c) geomorphic zones map...... 56 Figure 35 – Common characteristics of high intertidal reefs: ...... 57 Figure 36 – Methodological scheme of data acquisition, processing, integration and storage for the Kimberley reef geodatabase (ReefKIM) ...... 65 Figure 37 – Spatial distribution map of the Kimberley reefs and islands compiled in ReefKIM...... 68 Figure 38 – Two maps of Adele reef:(a) intra-reef geomorphic zones and (b) distribution of habitats and substrates on the reef platform...... 70 Figure 39. Satellite image of Adele Reef showing locations of previous studies and work with links to their origins in attribute tables ...... 71 Figure 40 – Screenshot of an oblique view of the Kimberley Bioregion. Reefs are shown in vivid orange in Google Earth...... 72 Figure 41 – Screenshot showing photo locations (as yellow map pins). A pop-up window displays a reef photo with relevant information...... 72 Figure 42. Map of the southern Kimberley showing the geology of the region (after Tyler et al., 2012) and the marine bioregions (Integrated Marine and Coastal Regionalisation of Australia, ...... 77 Figure 43 – A) Cockatoo Island reef digital terrain model (DTM) showing a 3D representation of the seafloor bathymetry...... 81 Figure 44 – Reef classification scheme...... 82

Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 45. Bathurst – Irvine Reef...... 85 Figure 46. A) W – E section of the Bathurst – Irvine high intertidal fringing reef (width 7 km), showing two stages of platform growth, marginal sediment bodies, drowned reefs (insert, B) in the central elongate pool and basement topography...... 87 Figure 47. N – S section of Bathurst – Irvine Reef and adjacent embayment substrate...... 87 Figure 48. Molema Island and Turtle Reef.A) Landgate aerial photography provided by DPaW; SBP lines are marked in red. B) Rhodolith bank at reef crest, note the channel draining tidal water off the reef. C) Rhodolith bank at reef crest with drainage channel, looking towards inner platform. D) Platform surface showing scattered corals and mud veneer. E) Oblique view of Proterozoic islands, northern Turtle Reef crest and muddy channel bank complex in foreground. F) Oblique view of southern Turtle Reef, with sand bank in foreground...... 88 Figure 49 – SBP images of the NW channel bank complex associated with North Turtle Reef...... 89 Figure 50. Sunday Islands. A) Landgate aerial photography provided by DPaW; track plot of seismic profiles are marked with thin red lines. B) Seismic profile...... 90 Figure 51. A) Montgomery Reef, showing morphology (central Proterozoic island and surrounding sand sheet, upper and lower lagoon with rhodolith banks, and reef front ramp with crustose coralline algae). Note sand bodies and channels marginal to the platform and the Breakwater, which has shallow sandy substrates to the east and deeper water to the west. Landsat satellite image sourced from USGS-EROS; SBP lines are marked in red. B – D) View of the reef crest with pools of corals (Acropora) and sand patches. C – E) Aerial view of rhodolith banks (dark), showing complex channelling. F) Oblique view of Proterozoic central island, with surrounding sand sheets and two level terrace lagoon (upper and lower)...... 91 Figure 52. History of reef growth for the Breakwater from seismic profiles...... 92 Figure 53. Seismic sections showing structure of the SW margin of Montgomery Reef...... 93 Figure 54. A) Adele Reef. Landsat satellite image sourced from USGS-EROS; SBP lines are marked in red. B) Low tide view of reef flat showing rubble surface. C – D) Rhodoliths on reef flat, with an average diameter of 10 cm. E) View of dense community of corals (including Porites) covering the southern reef platform...... 94 Figure 55. Cross-sections showing multiple stages of reef buildup ...... 96 Figure 56. Seismic sections showing the structure of the SW margin of Montgomery Reef, and continuity of seismic horizons between an adjacent patch reef...... 97 Figure 57. Bathymetric and seismic profile across Adele group of platforms ...... 98 Figure 58. Map showing Bioregions modified from the Integrated Marine and Coastal Regionalisation of Australia (IMCRA) and geology of the Kimberley Region (After Griffin and Grey, 1990)...... 104

Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 59. Map showing Cockatoo Island geomorphic and substrate classification, based on aerial photography interpretation ...... 106 Figure 60. Living coral communities on the SW Cockatoo Island fringing reefs...... 107 Figure 61. Lithostratigraphic and chronostratigraphic summary of measured sections of Cockatoo mine-pit...... 108 Figure 62. Idealised stratigraphic column of Cockatoo mine-pit section (S2_P1)...... 110 Figure 63. Holocene vertical reef accretion and growth history curve for Cockatoo Island sections . 112 Figure 64. Composite growth history records for Kimberley region and GBR derived from coral sections ...... 113 Figure 65. Southwest Cockatoo Island SBP cross section showing two stages of reef development; Holocene and Last Interglacial, with a clear correlation to the measured island sections...... 114 Figure 66. SBP line with superimposed core log from adjacent mine pit, showing correlation between SBP units and lithological units in the subsurface...... 115 Figure 67. Map of research location and coring sites distribution ...... 123 Figure 68 – Bathurst-Irvine Island core logs. Inset map shows Bathurst-Irvine core locations...... 125 Figure 69 – Sunday Island core logs. Inset map shows Sunday Island core locations...... 128 Figure 70 – Tallon Island core logs. Inset map shows Tallon Island core locations...... 130 Figure 71. Holocene vertical reef accretion curves for Bathurst-Irvine Island...... 132 Figure 72 – Full cross section with isochrones and summary of Holocene reef facies for the inter-island fringing reef of Bathurst-Irvine Island ...... 133 Figure 73. Holocene vertical reef accretion curve and polynomial curve of best-fit for Sunday Island...... 134 Figure 74 – Full cross section with isochrones and summary of Holocene reef facies for (A) Northern and (B) Southern inter-island reef platform of Sunday Island ...... 135 Figure 75. Holocene vertical reef accretion curve and polynomial curve of best-fit for the Tallon Island reefs...... 136 Figure 76 – Full cross section with isochrones and summary of Holocene reef facies for the Tallon Island reefs...... 137 Figure 77. Composite sea level records for Kimberley Shelf from the last 10,000 years BP derived from coral in core from Tallon reef, Sunday reef, Bathurst-Irvine reef (this study), coral exposure from Cockatoo Island mine-pit (Solihuddin et al., 2015), coral in core from Scott Reef (Collins et al., 2011), cemented coral shingle pavement from Abrolhos reefs (Collins et al., 2006), and corals in core from Morley Island Abrolhos (Eisenhauer et al., 1993)...... 140 Figure 78. Adele Island showing planned coring locations. SBP survey in red...... 145 Figure 79. Adele Island showing actual coring locations...... 146

Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 80 – Adele Island Core Logs ...... 148

Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Table of Tables

Table 1 – Significant data and sources for the geodatabase ...... 4 Table 2 – Data Sources ...... 13 Table 3 – Data that have been used to produce Geomorphic, substrates and facies maps ...... 14 Table 4 – Ground Truth Sources ...... 15 Table 5 – Datasets and sources used in this study...... 24 Table 6 - Number and percentage of islands by latitude across the Kimberley Bioregion...... 28 Table 7 – Frequency and percentage of island by area category in the Kimberley Bioregion...... 29 Table 8 – Lengths and areas of reefs by type...... 36 Table 9 – Frequency of reef types by latitude in the Kimberley Bioregion...... 37 Table 10 – Landsat satellites and sensors and their specifications...... 45 Table 11 – Methodology for defining the geomorphological zones and associated substrate and habitats on Montgomery Reef...... 47 Table 12 – The overall accuracy of each reef classification by classes...... 48 Table 13 – Coverage area of substrates and habitats of selected reefs...... 58 Table 14 – Datasets used in this study and their sources ...... 65 Table 15 – Resultant feature classes included in ReefKIM ...... 67 Table 16 – Data sources used to produce geomorphic, substrate and facies maps for targeted reefs69 Table 17 – Characteristics and acoustic features of seismic units identified in the profiles...... 80 Table 18 – Classification of Kimberley Fringing Reefs...... 83 Table 19 – Radiocarbon dates from selected samples across Cockatoo mine-pit transects ...... 111 Table 20 – Summary of characteristics of turbid reefs from the GBR (after Browne et al., 2012) and the Cockatoo fringing reef ...... 118 Table 21 – Reef Facies of Bathurst – Irvine Islands...... 126 Table 22 – Reef Facies of Sunday Island...... 127 Table 23 – Reef Facies of Sunday Island...... 129 Table 24 – Average carbonate content percentage in sediment samples...... 131 Table 25 – Radiocarbon dates from selected samples across Tallon, Sunday, and Bathurst-Irvine reef...... 137 Table 26 – Table of Adele Island Coring locations...... 146

Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Executive Summary

Research activities initially concentrated on the development of a Kimberley coral reef geodatabase, using remotely sensed images to generate geomorphic, habitat and substrate maps of reef platforms. More than 860 nearshore reefs were mapped from Cape Londonderry to Cape Leveque with a total of 30 reefs studied in detail and their substrates documented across the Kimberley Bioregion. The prevalence of rhodolith dominated substrates on high intertidal reefs (both planar and fringing reefs) contrasts with the more coral dominated fringing reefs, with inner to mid-shelf planar reefs having some shared attributes with these contrasting categories. A sub-bottom profiling (SBP) field study of 9 reefs in the western Kimberley, the first detailed seismic study in the region, was undertaken in October of 2013. Over 294 km of SBP records were collected from a representative suite of reefs to determine reef growth history. Two seismic horizons were identified in the inshore reefs (Sundays Islands, Buccaneer Archipelago, where seismic calibration was available, Molema Island and Montgomery Reef), marking the boundaries between Holocene reef (Marine Isotope Stage 1, MIS 1, last 12 ky) commonly 10 – 15 m thick, and MIS 5 (Last Interglacial, 120 ky; LIG) and Proterozoic rock foundation over which Quaternary reef growth occurred. In offshore reefs (Adele Complex), two additional deeper acoustic reflectors were identified (possibly MIS 7 and MIS 9 or 11). The Kimberley reefs are not thin growths over bedrock as postulated recently. Common reef thickness is 10-15 metres with reef growth initiation occurring very soon after the inner continental shelf was flooded (~10,000 years ago) following post glacial sea level rise. A detailed geochronology has been established for the Holocene reef building phase based on data from Cockatoo Island, where a full Holocene reef sequence, which sits uncomformably over a last interglacial reef sequence (MIS 5e), has been identified. When reef growth history is compared to the Holocene sea level curve, it is shown that the reef accreted in a sea level catchup mode. The transition of reef building organisms was investigated by shallow coring (up to 6 m depth) in late 2014. The core study provided the first subsurface sedimentary samples for the key types of reef found in the southern Kimberley and sheds considerable light on their growth history. It has shown that the reefs are muddy in character, similar to the Cockatoo island fringing reef. Adele Reef, offshore, is distinctly sandier in character. Radiocarbon dating showed that, like Cockatoo Island, reef growth initiated soon after the inner shelf was flooded (~10,000 years ago). Early reef buildups are likely to be muddy with branching, plate, and massive corals. This is replaced on many reefs (particularly exemplified by Montgomery and Turtle Reefs) by coralline red algae (rhodoliths), small robust corals and coral rubble as reefs become intertidal at their surface. Multibeam surveys of reef flats discovered a new reef morphotype, the “High Intertidal Reef” which are uniquely characterised by having reef flats that have a surface elevation that sits above the level mid-tide level. Typically reef flats sit at the level of mean low water spring tide.

Implications for Management

• A scheme of reef classification, with GIS database of habitats, prepared from RS data. This will give preliminary data on key habitats and relative significance in Kimberley reefs for 30 Platforms • Coral veneers vs. Long lived reefs? Resolved, along with seismic architecture of Cockatoo fringing reef, Sunday I reefs, eg Tallon Reef; high intertidal reefs, Irvine/Bathurst, Molema, Montgomery; Inner Shelf planar reefs, Adele Group. • 3 stages of reef platform buildup (MIS1, 5 and 7) were identified using SBP, commonly 2 stages for inshore reefs. • Holocene sea level rise curve from Cockatoo Island, low stand channels, and drowning history model documented. • Cores in 2014: assessed community changes as reef grew up into tidal zone, ~last 1-10,000 years.Introduction

Kimberley Marine Research Program | Project 1.3.1 i

Reef Growth and Maintenance

1 Introduction

1.1 Study area The Kimberley coastal region is a large scale ria coast with a well-developed indented rocky shoreline and many offshore islands with unique geology and geomorphology. Regional structure has determined much of the contemporary landforms and related ecosystem development, and controlled the architecture of the drowned landscapes over which coral reef systems have developed. Thus there are strong geomorphological controls on reef development. Coral reefs have developed over a broad shelf as shelf edge, inner shelf and coastal reefs inshore, and whilst the offshore reefs have been studied little is known about the inshore reefs for over ca 400 km of this complex coastal zone.

Figure 1. Study Area, Arrows identify key study sites

Whilst reef morphology is believed to contain features similar to those identified from the Great Barrier Reef (GBR), regional reconnaissance mapping is need in the first instance to assess reef morphology within the complex framework provided by the Kimberley coast. While the offshore reefs have grown vertically as successive sea level events drowned the shelf during the Quaternary, the inshore reefs have interacted with complex geomorphology during coastal drowning, as well as terrestrial inputs, developing around and over rocky islands and coastal platforms, but patterns of geomorphology and substrates are poorly understood and little documented regionally. It is not known whether most reefs are veneers over rock platforms, or are long-lived features, recording process and patterns of growth, community composition and structure, hiatuses, sea levels and climate change through time. Access to such records is needed to determine long term responses to oceanographic and coastal processes as well as the history of climate change.

Kimberley Marine Research Program | Project 1.3.1 1

Reef Growth and Maintenance

1.2 Study scope and objectives The goal of this project was to gain a regional understanding of the geomorphology of the Kimberley coral reefs, including their interaction with different substrates, morphological patterns, distribution and relative exposure to terrestrial and other impacts. The most cost effective way to determine reef growth architecture is through shallow seismic surveys which will provide information on foundations, previous reef growth events and Holocene buildup, also providing site survey data for collection of Holocene cores. The Holocene reef record will be obtained by coring selected sites, to gain an understanding of sea levels and growth history responses, timing of events during the Holocene, reef building communities, resilience and climate change responses. This involved addressing the following: 1. How does the reef structure and architecture of the Kimberley reefs vary regionally and comparatively (i.e., Kimberley vs GBR) in response to coastal substrate controls, terrestrial inputs, types of foundations? 2. What are the contrasts between coastal, mid shelf and offshore reefs, and how is this reflected in reef substrates and oceanographic controls? 3. How have environmental conditions in the inshore Kimberley region changed over the last several thousand years and what has been the associated response of coral communities and growth patterns? 4. How have the interactions between substrates, sea levels, extreme macrotidal conditions, high turbidity and subsidence influenced the style and architecture of the Kimberley coral reefs. 5. How are these factors reflected in communities, “turn of – turn off” history, ecology and substrates 6. How many reefs and islands are there in the Kimberley, are they geographically significant features? 7. Do Kimberley reefs represent significant geological structures or more simple coral communities? 8. How rapidly have reefs grown over the Holocene, have types of reef building corals changed over this period? In summary, our objectives were: 9. Use remote sensing with limited ground truth checking to establish the regional geomorphology, growth patterns and substrates of the inshore Kimberley reefs; 10. Determine the seismic architecture of selected Kimberley reefs as part of an assessment of Holocene reef growth and relation to antecedent foundations to assess reef growth; 11. Obtain a Holocene record of sea level change, reef building communities, chronology and growth patterns and climate history of selected inshore reefs for comparison with the short term record, where suitable coral material is obtained; 12. Where possible, quantify reef flat elevations, reef topography and morphology. Our approach to answering these questions was based on previous reef studies at the Abrolhos, Ningaloo, Scott Reef and the Rowley Shoals, from which much has been learnt about reef growth patterns, communities and geomorphology, including long term adjustments by reefs to changing climates and sea levels. The information gained from the Oceanic Shoals Bioregion will serve as a reference for comparison with data to be obtained from the Kimberley reefs. The Applied Geology Department at Curtin University labs are equipped with software packages for the remote sensing studies, and we have developed a methodology for grouping of habitats and water-bottom sediments into different spectral classes and mapping of modern carbonate systems and reef environments by applying unsupervised classification to Landsat ETM+ multispectral images. Ground truth data is available for the Oceanic Shoals for cross-checking. However targeted high resolution imagery and some ground truth will be required for the Kimberley reefs.

2 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

1.3 Background During November 2012 a research term was assembled for Project 1.3.1 consisting of: Project Leader: Lindsay Collins; Project Researchers: Michael O’Leary; Alexandra Stevens; PhD Students: Giada Bufarale; Moataz Kordi, Tubagus Solihuddin; This project is working towards achievement of the goals and outputs of Project 1.3.1 whilst establishing linkages with the various projects within the overall Kimberly Research Program. Related interactive research links are being developed with other WAMSI projects with a view to co-ordinating outputs, developing operational synergies, and sharing information. Project Objectives and Activities Prior to commencement of acquisition of field data in 2013 time was invested in examining technical, equipment and logistic issues presented for SBP surveys and collection of coral reef cores for the remote and “tidally challenged” operational activities that will be required by this project. This is important also for site selection, which to some extent was dependent on other Kimberley project activities and priorities, and on ongoing discussions within WAMSI and with Industry groups. Ore pit mapping fieldwork was undertaken on a Holocene reef exposed in a mining pit on Cockatoo Island in July 2013 and a paper published on the findings in Marine Geology in 2015 (Chapter 8). In October 2013 SBP fieldwork was successfully completed and analysis of the data is complete with a paper summarising the findings submitted to Continental Shelf Research and currently in review (Chapter 7). Collection of cores was completed in October 2014, where possible, in the same locations as the SBP, all cores have been logged and sampled and samples sent for dating, dating results for all reefs, except Adele Reef, have been received (Chapter 9). The ‘ReefKIM’ database has mapped more than 800 reefs and 30 analysed in detail (Chapters 4, 5 & 6). Timeline of events November 2012 – Project Startup, GIS development commenced and in progress October 2013 – Seismic Survey November 2013 – Complete Project planning, Complete Fieldwork (Seismic) and Annual Data analysis (GIS Stage 1) Complete: Milestone report June 2014 – Annual data management completed (GIS Stage 2): Milestone Report July to October 2014 – Coring Fieldwork December 2014 - Complete Fieldwork (Coring): Milestone Report March 2015 – Finalise Data Analysis: Milestone Report August 2015 – Final Report

Kimberley Marine Research Program | Project 1.3.1 3

Reef Growth and Maintenance

2 Project Establishment

During November 2012 a research term was assembled for Project 1.3.1 consisting of: Project Leader: Lindsay Collins Project Researchers (Part Time): Michael O’Leary; Alex Stevens; PhD Students: Giada Bufarale, Moataz Kordi, Tubagus Solihuddin This project worked towards achievement of the goals and outputs of Project 1.3.1 whilst establishing linkages with the various projects within the overall Kimberly Research Program. Related interactive research links were developed with other WAMSI projects with a view to co-ordinating outputs, developing operational synergies, and sharing information.

2.1 Project Objectives and Activities Commenced Prior to commencement of acquisition of field data in early 2013 time was invested in examining technical, equipment and logistic issues presented for seismic surveys and collection of coral reef cores for the remote and “tidally challenged” operational activities that will be required by this project. This was important also for site selection, which to some extent was dependent on other Kimberley project activities and priorities, and on ongoing discussions within WAMSI and with Industry groups. In parallel with these investigations the focus was on literature review and information organisation, establishment of a georeferenced coral reef database for the Kimberley reefs, and related activities.

2.2 Project Activities Establishment of a Coral Reef geodatabase The development of a reef geodatabase for the Kimberley was needed as a research and management tool, and was an important task which continued as more information came to hand. Initial input was from a wide range of sources such as reports, publications, atlases, books, maps etc. in different forms (hard and soft copies). The data are in the public domain, sourced through collaboration, or some has been purchased. The data has been compiled into a data library and information that is related to this project has been extracted and rectified and entered into a database. After digitizing and geo-referencing, selected information was represented as feature classes and raster-based datasets in a GIS as data layers using ESRI’s ArcGIS software, creating a geodatabase of the Kimberley coral reefs. This geodatabase contains the most significant elements and conditions that have direct or indirect influence on reef growth in the region, such as environmental, geological, geomorphological, biological, chemical, physical, oceanographic and climatic factors (Table 1). These datasets can be updated and modified as new data becomes available, and presented as a dynamic map, figures and graphs associated with spatial information to facilitate more detailed analysis.

Table 1 – Significant data and sources for the geodatabase Datasets Source of data Representation Kimberley coastline Satellite images (USGS), Aerial photography (WAMSI) Polyline Islands Satellite images (USGS), Bathymetric charts (Australian Polygon Hydrographical Service), Geological maps (GSWA) Coral Reef Satellite images (USGS), Bathymetric charts (Australian Polygon Hydrographical Service), Geological maps (GSWA) Seabed Geomorphology Geoscience Australia Polygon Sea Surface Temp NOAA Polyline Bathymetric map Geoscience Australia, AHO Polyline Weather Australian Bureau of Meteorology Vary

4 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Map of Islands and Reef Distribution The locations of islands, reefs and other related geomorphic features of the coast were extracted using heads- up digitizing of satellite imagery, aerial photography, geological maps and bathymetric charts along the Kimberley coast. Image processing operations have been conducted on the satellite imagery using ENVI 4.3 software to enable data extraction to be undertaken. ArcGIS 10 was used to manipulate various images with other data sources. All maps and charts have been geo-referenced and projected according to the Australian standard (GDA 94) to produce a basemap. A range of features including coastline, islands, coral reefs, rock reefs, and areas of shoaling were precisely digitized to avoid any overlapping or mismatching between these features. The resulting features were stored in vector format. Information can be presented in different colours and symbols to facilitate discrimination between them (Figure 2). Each feature is linked to its attributes such as the feature name, area, location, date of survey, source of data, map scale and any other related information which have been saved in an attribute table. The outputs from analysis of the data will facilitate understanding of linkages between geological substrate, reef geomorphology and reef classification and distribution.

Figure 2. Sample output of digitised features

Seabed & bathymetric maps in GIS compatible format

Bathymetric charts of the Kimberley Bioregion and some targeted areas (e.g. Champagny Islands, Montgomery reef, Cockatoo Island and Sunday Island) have been extracted from a bathymetric grid of Australia provided by Geoscience Australia (2009). It is a high-quality 9 arc second (0.0025° or ~250 m at the equator) bathymetric grid of Australian waters using Mean Sea Level (MSL) as a datum. In addition, for some targeted areas, bathymetric charts from Royal Australian Navy (RAN) Hydrographic Office were also applied to derive more detail bathymetric chart in larger scale. This bathymetric chart uses lowest astronomical tide (LAT) as a datum and was mapped in different scale.

Kimberley Marine Research Program | Project 1.3.1 5

Reef Growth and Maintenance

A bathymetric contour map of the Kimberley coast was derived by combining several datasets of varying quality. The seafloor was interpolated using depths derived from navigational charts. At this stage, a Digital Terrain Model (DTM) has been generated for Montgomery reef and adjacent areas using ArcGIS. The reef slope and flat was approximated using data derived from WAMSI (2010). Preliminary morphologic classification of the reefs of the Kimberley Region A reef evolutionary or genesis model sensu Hopley (2007) has not been incorporated in this study, due to the simple fact that unlike the Great Barrier Reef, Kimberley reefs have not had the same level of geological and geomorphological investigation. Instead a geomorphological typology classification scheme using a simple hierarchy (primary, secondary and tertiary) of geometric criteria (Figure 3) has been developed for the Kimberley reefs, using the same categories as in Hopley, 1982 and Hopley et al., 2007. See Chapter 4 for detailed information.

Figure 3. Preliminary reef classification scheme based on adaptation of previous studies and geoscientific data from this study Sea-level curve and drowning history for the Kimberley Region Rising sea levels since the Last Glacial Maximum (SL at -120 metres at 18,000 years ago) drowned the Kimberley Shelf, provided different substrates and foundation types for reef growth, and generated the complex coastal morphology seen today. From existing sea level records for the Kimberley region and globally a “model sea level curve” has been assembled for the Kimberley so that the shelf drowning history can be reconstructed. The pattern of drowning of the shelf will be modelled and will be utilised to assess likely timing of onset and duration of reef growth during postglacial sea level rise, and this is providing information for site selection and the targeting of seismic and coring operations for 2013 field activities.

The sea-level curve estimation for the Kimberley Bioregion at the Last Glacial Maximum (LGM) (Figure 4) up to Holocene time was derived from previous studies conducted in several localities including the Abrolhos (Eisenhauer, A. et al., 1993, Collins, et al., 2005) and Sunda Shelf (Hanebuth, et al., 2000) as shown below:

6 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 4. Sea-level curve estimation for Kimberley Bioregion at the Last Glacial Maximum Figure 4 shows the best fit sea-level curve to be applied to the Kimberley Bioregion considering the proximity of the site to existing sea level data, tectonic displacements, and glacio-hydrostatic processes. A model sea-level curve (Figure 5) was established based on this curve to build a shelf drowning model:

Figure 5. Model sea level curve based on data presented in Figure 4. Note that the 14C dates of Hanebuth have been adjusted to derive comparable estimates with the remainder of the data (U-series dates on corals), the latter ages correlating most closely with calendar years.

Kimberley Marine Research Program | Project 1.3.1 7

Reef Growth and Maintenance

Drowning Model A shelf drowning model based on bathymetric charts and sea level history has been established for the Kimberley bioregion focusing on some targeted areas e.g. Champagny Island, Montgomery reef, Cockatoo Island, and Sunday Island. Starting from the -120 m depth contour, the drowning model is created for 20 m intervals up to present sea-level where every depth contour corresponds to “model age”. For example, the present -120 m depth contour is predicted as a coastline at U/Th 21.5k years BP, while the current -100 m depth contour is attributed to U/Th 19k years BP, etc.

Based on this drowning model, reefs on Champagny Island, Montgomery reef, Cockatoo Island, and Sunday Island are predicted to have started growing in the Mid-Holocene after 7.5k years BP, with the exception of reefs around the” breakwater morphology”, in the west-end of Montgomery reef. The sea-level rose dramatically and was reached current sea level at the postglacial highstand, about 6.5k years BP. This highstand remained steady at up to 1.5 m above present sea-level at 6k years BP and subsequently started to fall until present time.

Figure 6 and Figure 7 are examples of drowning model outputs for sea levels at -50 m and -120 m below present SL. The shelf drowning model will assist with determining seismic and reef coring targets taking into consideration the likely age of initiation of reef growth.

Figure 6. Predicted SL at 21.5k years BP about 120 m below present SL.

8 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 7. Predicted SL at 12.8k years BP about -50 m below present SL.

2.3 Conclusions During the first quarter of 2013 the project establishment phase of Project 1.3.1 was satisfactorily achieved and the creation of the reef geodatabase commenced.

Kimberley Marine Research Program | Project 1.3.1 9

Reef Growth and Maintenance

2.4 References

Collins, L. B., J.-X. Zhao, Heather Freeman. (2005). A high- Hopley, D. (1982). The Geomorphology of the Great Barrier precision record of mid–late Holocene sea-level Reef - Quaternary Development of Coral Reefs. Wiley- events from emergent coral pavements in the Interscience Publication, John Wiley and Sons Ltd., Houtman Abrolhos Islands, southwest Australia. New York. "Quaternary International 145–146(0): 78-85. Hopley D., Smithers, S., & Parnell, K. (2007). The Eisenhauer, A., G. J. Wasserburg, J. H. Chen, G. Bonani, L. B. geomorphology of the Great Barrier Reef: Collins, Z. R. Zhu, and K. H. Wyrwoll (1993). Holocene Development, diversity and change. Cambridge, UK: sea-level determination relative to the Australian Cambridge University Press. continent: U/Th (TIMS) and 14C (AMS) dating of coral WAMSI. (2010). Under Kimberley Water. CRAWLEY WA: cores from the Abrolhos Islands. Earth and Planetary Geonewmedia Science Letters 114(4): 529-547

Hanebuth, T., K. Stattegger, Pieter M. Grootes. (2000). Rapid Flooding of the Sunda Shelf: A Late-Glacial Sea- Level Record. Science 288(5468): 1033-1035.

Data sources

Satellite images: USGS (2012). United State Geological Survey, Earth Explorer, accessed 7/02/2012, http://earthexplorer.usgs.gov/

Bathymetric charts: AHS (2012), Australian Hydrographical Service, QuickCharts Australia West.

Geological maps: GSWA (2012), Geological Survey of Western Australia, Data and Software Centre accessed 20/02/2012, http://geodownloads.dmp.wa.gov.au/datacentre/datacentreDb.asp

Aerial photography: WAMSI.

10 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

3 Materials and Methods – Data Management

Evidence (in this case data) that is created, compiled or collected during the research process should be managed to ensure that it is accessible over long periods of time (UWA, 2014). Data management is an integral part of any research project process. Furthermore, it plays a crucial role in the success of the research outcomes as it usually deals with organisation of data during the life and beyond of the research, starting from acquiring data through data processing and analysing and concluding with extracting results and data archiving (Whyte & Tedds, 2011). Good management of data and other related sources makes the research process more efficient in both the short and long term. Moreover, data can be used, reviewed and developed in the future for research beyond the scope of the original research project. A data management plan for this project was developed based on a thorough review of existing plans from similar research projects. The plan relates to data that have been generated in digital formats, collected during fieldwork or hard copies that have been digitised. This chapter will outline how data will be managed during this research project’s lifecycle. It also includes management of other related sources such as textual sources, images, recordings, samples and much more. See Appendix I for definitions of the terminology used in this chapter

Figure 8. Data Management Flowchart

Kimberley Marine Research Program | Project 1.3.1 11

Reef Growth and Maintenance

3.1 Data Acquisition The required data sources for this project were varied, such as remotely sensed images, aerial photography, bathymetric charts, geological maps and sub-bottom profiles, sediment samples as well as ground truth. Some of these data were available from the Department of Applied Geology, Curtin University. Other data were obtained from local and international government authorities, such as The Department of Parks and Wildlife (DPaW), Geoscience Australia (GA), the Western Australia Marine Science Institute (WAMSI), the Geological Survey of Western Australia (GSWA), Western Australian Museum (WAM) and the United States Geological Survey (USGS). Other related data were extracted from a wide range of secondary sources such as reports, publications, atlases, books, maps etc. and some other data were purchased. All of these data have to be in digital format, so that they can be used as GIS data layers; if not they were digitised. Figure 9 shows the coverage of the data sources over the study area; Table 2 lists the sources.

Figure 9. A map of the study area, showing the main source of datasets.

12 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Table 2 – Data Sources

Datasets Source of data Representation Kimberley coastline Satellite images (USGS), Aerial photography (WAMSI) Polyline Satellite images (USGS), Aerial Photography (DPaW), Islands Bathymetric charts (Australian Hydrographical Service), Polygon Geological maps (GSWA) Satellite images (USGS), Aerial Photography (DPaW), Bathymetric charts (Australian Hydrographical Service), Coral Reef Geological maps (GSWA) Polygon, points WA Museum Woodside Collection Project (Kimberley) 2008- 2011, Brooke, 1995, 1997 and Wilson 2009, 2010 Seabed Geomorphology Geoscience Australia Polygon Sea Surface Temp NOAA Polyline Bathymetric map Geoscience Australia, AHO Polyline Sub-bottom profiles Applied Geology Dept. CU VSP Weather Australian Bureau of Meteorology Vary

3.2 Data Sources Remotely sensed images Remotely sensed (RS) images are used as a feasible source of data to analyse coral reefs. Currently, remote sensing platforms utilise powerful technologies, which increases the ability to detect coral reef morphology and identify habitats and substrates. Thus it has been employed to map coral reefs in a number of coral reef studies around the world due to its accuracy and cost effectiveness (Table 3).

Landsat Imagery

Landsat images have been extensively used for mapping coral reef geomorphology and inter-reef facies distribution in a number of recent coral reef studies around the world (Carvalho, & de Kikuchi, 2013; Madden et al., 2013; Rowlands et al. 2012; Hedley et al. 2012; Leon et al. 2012). Despite its moderate spatial resolution (30 m), this sensor’s greatest advantages are that it incorporates 7 spectral bands and is available at no cost, which means many scenes can be acquired and examined to ensure they suits the study requirements of low cloud cover, low water turbidity and low tide. Hence in a macrotidal environment such as the Kimberley coast where the tidal range can exceed 11 m above sea level in some areas, Landsat images remain the best option to remotely study reefs. Landsat 5 Thematic Mapper (TM) and Landsat 7 Enhanced Thematic Mapper (ETM+) images were used. More than 72 images which cover the entire study area were acquired covering various dates.

Kimberley Marine Research Program | Project 1.3.1 13

Reef Growth and Maintenance

Table 3 – Data that have been used to produce Geomorphic, substrates and facies maps for targeted reefs in the study area. LS = Landsat TM and ETM+ images, OP = Orthophotographs (Aerial ph.); HR = Higher Resolution satellite images; GT = Ground truth; SBP = Sub-bottom Profiling; BC = Bathymetric Charts; and PS = Previous Studies.

Data Sources Reef / Island LS OP HR GT SBP BC PS Adele Is. x x x x x x Albert Rf. x x x x Bathurst & Irvine Is. x x x x x Beagle Rf. x x x Browse Is. x x x x x Brue Rf. x x x x x Cape Londonderry x x x Cassini Is. x x x x x Champagny Is. x x x x Churchill Rf. x x x x Cockatoo Is. x x x x x x Colbert Is. x x x x x Condillac Is. x x x x De Freycinet Is. x x x x Hedley Is. x x x x x King Is. x x x Long Rf. x x x x x Maret Is. x x x x x Mavis Rf. x x x Molema Is. (Turtle Rf.) x x x x x x Montalivet Is.E x x x x x x Montalivet Is.W x x x x x x Montgomery Rf. x x x x x x Robroy Rf. x x x x Scott Rf. x x x x x Sunday Is. x x x x Tallon Is. x x x x White Rf. x x x x x x Wildcat Rf. x x x x x Woninjaba Is. x x x x

14 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

High Resolution Satellite Images

There is no doubt that higher resolution images are superior to lower resolution images. Also, classification obtained using higher resolution sensors images has twice the accuracy for detailed habitat statistics than lower resolution images (Mumby et al., 1997). However, as mentioned previously the Kimberley coast is tidally controlled with mostly turbid water which makes it a unique case. Thus high resolution images may not always be suitable; nonetheless they have been used where possible. Furthermore these images also been used for cross-check. There are a number of high resolution images from various sensors such as World View2, QuickBird2, GeoEye and Pleiades-1A used in this project.

Aerial photography (Orthophotos)

Orthophotographs are geometrically corrected aerial photographs mainly in true colour RGB 321, rarely they cover the Near Infra-Red (NIR) band also. Although they usually do not have as many spectral bands as satellites imagers do, their spatial resolution is very high (< 1 m). They can represent small ground features very well which allows one to identify more details. Thus they can be integrated with satellite imagery for verification and measuring purposes. More than 15 orthophotographs used in this project.

Maps and Charts Bathymetric charts are drawn by the Australian Hydrographic Office (AHO). The majority of these charts are based on field surveys which had been collected over different time periods. The main purpose of these charts is to alert mariners to an object poses a hazard to navigation. However, these charts were used in this study to locate occurrences of reef platforms and shoal areas. There are 12 charts covering the entire study area.

Ground Truth Field surveys can be used to verify the results of remote sensing images in particular areas. Georeferenced ground truth data was sourced from a number of surveys conducted by the WA museum as part of the WA Museum Woodside Collection Project (Kimberley) 2009-2012, Brooke, 1995, 1997 and Wilson & Blake 2011 and Wilson et al 2011 and are as follows in Table 4:

Table 4 – Ground Truth Sources Source Survey Date Wilson & Blake 12 points & transects 2011 Wilson et al 2011 Brooke 14 transects 1995, 1997 Woodside & WAM 121 points 2009 - 2012

Sub-Bottom Profiling Sub-bottom profiling (SBP) fieldwork was undertaken in October 2013, 266 lines covering 294 km were collected. Seismic data were digitally recorded in .sgy (Seg-Y) format (Rev 1), using SonarWiz 5 as acquisition and post- processing software. Navigation data are in .nav format. Sediment Samples Fieldwork in Cockatoo Island was conducted from 24 to 29 July 2013. There were 4 reef transects measured, 142 samples collected comprising 32 dating samples, 56 matrix samples, and 44 taxa samples. Position fixing was by RTK Trimble DGPS tied to mine site datum. Elevations from each transect were plotted relative to the Australian Height Datum (AHD). All the coral samples are now being stored in WA Museum (Zoe Richards), whilst matrix samples are being stored in the Department of Applied Geology’s sedimentology lab.

Kimberley Marine Research Program | Project 1.3.1 15

Reef Growth and Maintenance

Surficial sediment samples were also collected during seismic surveys using a pipe dredge in some locations i.e. Adele Island, Brue Reef, Bathurst and Irvine Island. Their positions were fixed by the handheld Garmin GPS and the samples are stored in the Department of Applied Geology. Isotopic Data 17 coral samples were collected for accelerator mass spectrometry (AMS) radiocarbon dating to establish a geochronologic record of reef accretion. This data is stored in the Department of Applied Geology Cores Cores were logged, photographed and sampled on site. Samples are stored in the Department of Applied Geology. Data Integration Data integration or data fusion is a process that often combines multiple data sources of the same object or feature to extract information or add value in order to produce a consistent, accurate, and useful new data (Figure 10). The expectation is that fused data is more informative and synthetic than the original inputs. Most of the datasets in this project are available in digital format as both raster and vector, so they can be used as GIS data layers with minor pre-processing (see 3.3.1). After pre-processing all data have been compiled into a data library and information that is related to this project has been extracted, rectified and entered into a database. The selected information is represented as feature classes and raster-based datasets in a GIS as data layers that can be used ArcGIS. This geodatabase also contains the most significant elements and conditions that influence reef growth in the region. Together it can show, in detail, the spatial distribution of coral reefs in the Kimberley region making it the most comprehensive dataset of coral reefs in this region. It also can be updated and modified as new data becomes available.

Figure 10. Data Integration

16 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

3.3 Data Processing Pre-processing This process is important for accuracy and reliability assurance of the data. Image processing operations have been undertaken on the satellite imagery using ENVI 4.3 software to enable data extraction to be undertaken. ArcGIS 10 was used to manipulate various images with other data sources. Maps and charts have been georeferenced and projected according to the Australian standard (GDA 94) for consistency and homogeneity. All remotely sensed images have been calibrated and atmospherically corrected. Moreover, some images were pre- processed to offset problems with the band data and to recalculate DN values.

All images, maps and charts have been geo-referenced and projected according to the Australian standard (GDA 94) to ensure accuracy and consistency of the map. A range of features including coastline, islands, reefs, rock reefs, and areas of shoaling were precisely digitised to avoid any overlapping or mismatching between these features. The resulting features were stored in vector format. Information can be presented in different colours and symbols to facilitate discrimination between map features.

Each feature is linked to its attributes such as the feature name, type, area, location, date of survey, source of data, map scale and any other related information which have been saved in an attribute table. The outputs from analysis of the data will facilitate understanding of linkages between geological substrate, reef geomorphology and reef classification and distribution.

Metadata Metadata is information that describes data. It can comprise many types of information on data which in this project include:

• - Data identification such as title, name, geographic region, creating date, source of information, owner or organization, and other elements. • - Data quality such as attribute, accuracy, consistency, method of production. • - Data spatial features such as map projections, grid system, datum and coordinate system See Appendix II for WAMSI metadata form The metadata generated for the project will conform to the ISO19115 Marine Community Profile metadata standard. The metadata generated will be compatible with the AODN GeoNetwork catalogue and will be harvested automatically every week by the AODN. (WAMSI DIMP). The following is taken from Curtin University’s Research Data Management Guide (Curtin University, 2014)

3.4 File Formats and Management File Formats Shp, gdb, geotiff, ASCII, SEGY

File Naming The use of good naming conventions provides a useful cue to the content and status of a file, including its version. The following examples highlight basic principles of file naming which will be followed by the project. • File name is concise and meaningful. • Sentence case including a capital letter for names and proper nouns. • Full name with family name in UPPER CASE. • Date format: yyyy-mm-dd. • Terms separated with a dash. Avoid punctuation.

Kimberley Marine Research Program | Project 1.3.1 17

Reef Growth and Maintenance

Directories Developing a system to organise files requires consideration of good naming conventions, consistency of terms used and a development of a coherent and consistent folder structure. This will ensure it is easy to locate, organise and navigate all files and versions • Consistent file names. • File hierarchy refers to the number of levels or sub-folders in the directory, folders are numbered, e.g. 001, 002 throughout folder structure. Version Control Version control is necessary when data is constantly updated and/or is accessed by more than one person and can be implemented by agreeing on a standard for naming files or folders. For example: Date can be part of the file name e.g. 2012-02-27. Changes are documented through the use of a superseded folder.

Reference - http://libguides.library.curtin.edu.au/content.php?pid=314427&sid=2573513

3.5 Data Analysis A Geographic Information System is a computer-based system that can deal with almost any type of information about features that can be referenced by geographical location. It is capable of handling both locational data and attribute data about such features and not only displays the locations of features or maps produced, but it a1so can provide a capability for recording and analysing descriptive characteristics about these features. For example, in this project the database not only represents a map of locations of reefs and islands but it also contains descriptive statistics of reef features. These attributes include information such as reef area, reef type, reef substrates and facies and so on. Figure 11 shows an example of attributes that can be associated with a given point, line, or area features. The strength of ArcGIS is its capacity for database management which enables the storage and manipulation of attribute data.

Figure 11. Example of linking map features to an attribute table

18 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

3.6 Data Storage This project requires a large amount of data storage because it contains many remote sensing images and digital maps. This data are being stored in the Department of Applied Geology at Curtin University as well as at IVEC, which is one of the main facilities for supercomputing, large scale data storage and visualisation in Western Australia. Also, data was backed up on a regular basis. They are kept on an external hard drive and protected by a password in multiple secured places. Hard copies of data including resources and references and progress reports are being stored in a locked cabinet at the department as well as at an off-site backup to ensure that data will not be lost in case of emergency. Data in both hardcopy and softcopy formats, and all other related information are accessible to authorised people only.

During the period of the project’s execution all data was be stored in the Department of Applied Geology, Curtin University. Additionally, backups of digital material were periodically updated on external hard drives, with access restricted to the researchers. All WAMSI project data is archived accordingly to WAMSI regulations.

Storage of original material and records will meet the guidelines of the Australian Code for the Responsible Conduct of Research (National Health and Medical Research Council et al, 2007) and will be held for a period of at least 5 years after publication of thesis and scientific papers.

Researchers will upload all data to the IVEC facility. The long-term (decades) storage of data will be arranged by WAMSI (WAMSI, 2013).

3.7 Data Accessibility Research team members shared their working files through Dropbox to ensure that all files are synchronised and everyone has the latest version of the files. In addition, there are many essential characteristics that can improve utilisation of a database in GIS and make it more practical. These characteristics include but are not limited to 1) ease of use, to enable user control and data access due to its user-friendly interface; 2) reliability, to avoid accidental loss of data; 3) security, to restrict data access to authorised users only; and 4) flexibility, to facilitate a wide-ranging of database queries and operations. For example, users can search for information using a query language which is known as SQL or Structured Query Language.

3.8 Future Data Accessibility Open source file types where possible – geotiff, KML, ASCII, JPEG2000. Proprietary software file types – ECW, GDB, SHP, SEGY.

3.9 Copyright and Intellectual Property The following information is taken from the WAMSI Data Information Management Plan (WAMSI, 2013).

Copyright After an embargo period of 18 months (to permit publishing), data will be available publicly under Creative Commons Attribution Non Commercial Share Alike (BY-NC-SA), subject to constraints.

If private companies wish to use the data for commercial purposes they need to contact the individual researcher(s)/organisations.

Data/data products will be available to State Government Departments upon request during the life of the project, subject to the use of those data being consistent with the Intellectual Property Policies of Curtin University.

Kimberley Marine Research Program | Project 1.3.1 19

Reef Growth and Maintenance

Intellectual Property All data collected as part of Ph.D. student research falls under the IP arrangements of Curtin University. WAMSI and the Universities will negotiate appropriate release of IP arrangements to allow the data to be made public in ways that do not compromise the student's IP or the public-good objectives of WAMSI.

3.10 Conclusion The 1.3.1 project researchers used information and experience gained from participation in WAMSI 1 within the current WAMSI 2 project research. Our data and newly generated products reflect the three areas of activity in the study; remote sensing, sub-bottom profiling, and associated sedimentological work. Consequently the data base consists of: • Remotely sensed and ancillary data used in development of a reef geodatabase for the Kimberley reefs • A data base of interpreted sub-bottom profiles from the almost 300 km of data collected from representative reef platforms in the southern Kimberley • Sedimentological and associated analytical data (particularly geochronology) obtained from core studies of selected representative coral reef systems. • Interpretive outputs – WAMSI reports, research papers and PhD theses.

Most of the information generated by the project will be nearing finalisation by the middle of 2015. We have discussed the project’s data storage and access requirements with Luke Edwards, Marine Data Manager, at IVEC and will discuss this with him further once data analysis has been completed and final data storage and access constraints have been determined.

20 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

3.11 References

Brooke B.P. 1995 Geomorphology, pp. 67-80, Part 4, in Rowlands, G., Purkis, S., Riegl, B., Metsamaa, L., Bruckner, Wells FE, Hanley JR and Walker DI, Survey of the A., & Renaud, P. 2012. Satellite imaging coral reef marine biota of the southern Kimberley islands. resilience at regional scale. A case-study from Saudi Western Australian Museum, Perth, unpublished Arabia. Marine pollution bulletin, 64(6), 1222-1237. report No. UR286. UWA. 2014. “UWA Research Data Management Plan”. Brooke B.P. 1997. Geomorphology of the north Kimberley University of Western Australia Accessed May 2014. coast, in Walker (ed) Marine biological survey of the http://lgdata.s3-website-us-east- central Kimberley coast. Western Australia. University 1.amazonaws.com/docs/380/615894/UWA_Data_m of Western Australia, Perth, unpublished report WA anagement_plan_final_Nov12_revFeb14.pdf, Museum Library No. UR377, 13-39. Whyte, A., Tedds, J. 2011. ‘Making the Case for Research Carvalho, R. C., & de Kikuchi, R. K. P. 2013. ReefBahia, an Data Management’. DCC Briefing Papers. Edinburgh: integrated GIS approach for coral reef conservation in Digital Curation Centre. Bahia, Brazil. Journal of Coastal Conservation, 1-14. Western Australian Marine Science Institution, Data Curtin University. 2014. “Research Data Management”. Information Management Plan. 2013. Western Curtin University. Accessed May 2014. Australian Marine Science Institution. http://libguides.library.curtin.edu.au/content.php?pid=31 Western Australian Museum Woodside Collection Project 4427&sid=2573513, (Kimberley) 2009-2012. Hedley, J., Roelfsema, C., Koetz, B., & Phinn, S. 2012. Wilson, B. & Blake, S. 2011. Notes on the origins and Capability of the Sentinel 2 mission for tropical coral biogeomorphology of Montgomery Reef, Kimberley, reef mapping and coral bleaching detection. Remote Western Australia. Journal of the Royal Society of sensing of environment, 120, 145-155. Western Australia, 94:107-119 Leon, J., Phinn, S., Woodroffe, C., Hamylton, S., Roelfsema, Wilson, B., Blake, S., Ryan, D. & Hacker, J. 2011. C., & Saunders, M. 2012. Data Fusion for Mapping Reconnaissance of species-rich coral reefs in a muddy, Coral Reef Geomorphic Zones: Possibilities and macro-tidal, enclosed embayment, Talbot Bay, Limitations. Paper presented at the 4th GEOBIA, Kimberley, Western Australia, Western Australia. Brazil. Journal of the Royal Society of Western Australia, Madden, R. H., Wilson, M. E., & O'Shea, M. 2013. Modern 94:251-265 fringing reef carbonates from equatorial SE Asia: An Satellite images: USGS (2012). United State Geological integrated environmental, sediment and satellite Survey, Earth Explorer, accessed 7/02/2012, characterisation study. Marine geology, 344, 163-185. http://earthexplorer.usgs.gov/ Mumby, P.J., Green, E.P., Edwards, A.J. and Clark, C.D., Bathymetric charts: AHS (2012), Australian Hydrographical (1997) Coral reef habitat mapping: how much detail Service, QuickCharts Australia West. can remote sensing provide? Marine Biology, V130(2): 193-202. Geological maps: GSWA (2012), Geological Survey of Western Australia, Data and Software Centre National Health & Medical Council, Australian Research accessed 20/02/2012, Council and Universities Australia. 2007. Australian http://geodownloads.dmp.wa.gov.au/datacentre/dat Code for the Responsible Conduct of Research. acentreDb.asp Commonwealth Copyright Administration, Attorney General's Department, Robert Garran Offices, National Circuit, Canberra, ACT.

3.12 Data sources High Resolution Satellite Images: DPaW Aerial photography: WAMSI and DPaW

Kimberley Marine Research Program | Project 1.3.1 21

Reef Growth and Maintenance

4 Mapping and Geomorphic Classification of the Kimberley Reefs, North West Australia

Adapted from Kordi M.N., Collins, L.B., O’Leary, M.J., & Stevens A.M., in review. Mapping and Geomorphic Classification of the Kimberley Reefs, North West Australia. International Journal of the Digital Earth.

4.1 Introduction Reef Mapping and Classification Coral reefs are widely distributed through the world’s tropical oceans and often rise abruptly from very deep water creating a major navigational hazard, but can also provide a safe anchorage. There are many famous examples of ships running aground on coral reefs, the wrecks of the VOC ship Batavia in 1629 and the HMS Pandora in 1779 being just two notable Australian examples (Green, 1975; Edwards et al., 2003). The increase in international shipping during the 18th and 19th century saw maritime nations and international trading companies seek ways to reduce shipping losses through improved charting, but also sought a scientific understanding of where reefs are likely to be encountered how they form. In fact few questions in 19th-century science aroused more controversy than the origin of coral reefs. So when HMS Beagle departed in 1831 on its 5-year journey of discovery around the world, it not only carried a young Charles Darwin but also secret instructions from the Admiralty requiring a detailed geological investigation on how coral reefs formed. The result was that in 1842 Darwin published The Structure and Distribution of Coral Reefs, the first of three major monographs arising from observations and data he collected during his voyage on the Beagle. This monograph put forth the theory of atoll formation through island subsidence and included the first detailed map of the distribution of different kinds of coral reefs through the Indo-Pacific and Caribbean region. This map represents the first global census of coral reefs and presented the first geomorphic reef classification scheme which is still in use today. Joubin (1912) expanded Darwin’s original reef census by producing a series of five 1:10,000,000 scale reef maps covering the entire globe. His work combined existing survey charts as well as observations and voyage reports from a range of sources. Following Joubin’s Treaties there was no other attempt to systematically map coral reefs on a global scale until the launch of the Landsat series of satellites in the 1970’s, which for the first time allowed the detection of coral reefs using low-resolution satellite imagery (Jupp et al., 1985), and the opportunity to map reefs globally. Most recently, the United Nations Environment Program (UNEP) funded the Global Reef Monitoring Network (GRMN) and the International Coral Reef Action Network to build and maintain a global reef GIS database ReefBase, which provides a repository for available knowledge about coral reefs. The Millennium Coral Reef Mapping Project is using a suite of high-resolution spaceborne remotely sensed imagery to systematic map and classify coral reefs (IMaRS-USF and IR, 2005; ReefBase, 2015) worldwide. The development of a geomorphic classification of coral reefs has been closely linked to the improvement of reef mapping techniques. For instance, three reef classes fringing, barrier, and atoll were identified by Darwin (1842) and formed the basis for his global map of coral reefs. In the 1920s, the advent of aerial photography allowed coral reefs to be viewed in plain view, allowing for a more detailed analysis of the spatial characteristics of reefs and mapping in greater detail. The pioneering geoscientific work on Australian reef classification was conducted by Fairbridge (1950, 1967), who recognised the role of antecedent topography, eustasy, and physical processes in generating reef morphology, first working on the Great Barrier Reef (GBR) and later on the reefs of northern Australia, including the Kimberley coast (Finkl, 2011). Subsequently, as knowledge of reefal processes increased, Hopley (1982) was able to develop an evolutionary reef classification scheme for the GBR. Accordingly, reef classification and reef typology at the global, regional and reefal scales have been dramatically improved (Andréfouët et al., 2006; Hopley et al., 2007; Rowlands et al. 2014). While these global scale mapping efforts have advanced our understanding of reef formation and growth, as well as providing information for monitoring of coral reef health and to support informed decisions about coral

22 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

reef use and management (Spalding, 2001; Wilkinson, 2008), they still lack the resolution required to provide realistic reef census data at regional and local scales. While efforts have been made to fill this gap, e.g., Hopley et al. (1989), there are still many reef regions of the world where the true numbers and extent of coral reefs are underestimated or unknown. One such area is the Kimberley region of NW Australia where there has only been a limited number of studies of fringing and nearshore reefs (e.g., Wilson, 2011; Wilson et al., 2011; Solihuddin et al., 2015) and no significant spatial dataset detailing the size and distribution of coral reefs and their attributes. North West Australian Reef Systems The Reefs of North West Australia occur within two distinct bioregions, the shelf edge Oceanic Shoals Bioregion (OSS) and the inner shelf Kimberley Bioregion (KIM) (Figure 12). The OSS, which includes the Rowley Shoals and Scott Reef have seen significant scientific investigations, due in part to their proximity to the Browse Gas Fields. In particular, the morphology and growth history of these reefs have been examined using a number of methodologies such as coring, U-series dating and vertical seismic profiling (Collins, 2011; Collins et al., 2011).

Figure 12. Map of the North West Shelf (NWS) showing the spatial distribution of reefs, bioregion boundaries and the continental shelves (ramp) subdivision. Boundaries of the study area, the Kimberley Bioregion (KIM), are highlighted with red lines. The Oceanic Shoals Bioregion (OSS) is at the seaward margin of the NWS.

The inner shelf Kimberley reefs have seen limited scientific study due in part to the geographic remoteness of the region and limited infrastructure. Early investigations by Teichert and Fairbridge (1948) noted that fringing reefs exist around the margins of many islands in the region, despite the normally unsuitable environment for coral build-up, including high sediment input, macrotidal regimes, highly turbid water and raised sea surface temperatures. Wilson (1972) and O’Conner (1989) observed reefs in the intertidal zone between shallow rocky shoals, along muddy shores, and in some bays. Geographic and geomorphic estimation of reef location, thickness, and reef area was provided by Brooke (1997), who demonstrated that fringing reefs were widespread along the Kimberley coast and could be classified into three reef forms: fringing reefs adjacent to a cliffed shore; reefs developing on bedrock edges; and large reef complexes. Recent studies of Montgomery Reef and Talbot Bay by Wilson (2011) and Wilson et al. (2011) revealed the unique nature of these reefs and their habitats and

Kimberley Marine Research Program | Project 1.3.1 23

Reef Growth and Maintenance

substrates and documented the relationship between the unique physical processes and reef geomorphology. Despite these studies, there are still many unknowns in our understanding of the fringing reefs of the KIM. This study aims to compile the first dedicated reef census and mapping dataset of the KIM. This spatial approach will allow for a detailed statistical assessment on the morphological characteristics and distribution of Kimberley reefs and for the development of a geomorphic classification scheme. Study area The KIM covers a massive 60,000 km2, stretching from Cape Londonderry (13˚S) in the north to Cape Leveque in the south (16°S). The KIM coast is characterised by deep inlets, capes and archipelagos forming a very complex coastline (Figure 12). It has extensive fringing coral reefs that exceed the Ningaloo Reef in their diversity and it supports a huge range of marine habitats (DEC, 2011). In some parts of the Kimberley coast the spring tide reaches more than 11 m, making it the highest tidal range of any coral reef system in the world and the second largest tide after Fundy Bay in Canada (Purcell, 2002; Wolanski and Spagnol, 2003). Kimberley coral reefs are the main geomorphic feature along the continental shelf between the latitudes of 12°S and 18°S (Collins, 2011). Currently, activities such as oil and gas extraction, mining, and tourism are increasing in this region, necessitating timely management to protect the marine environment (Wood and Mills, 2008).

4.2 Methodology Datasets There are two groups of datasets that have been utilised in this study: input datasets and derived datasets. The input datasets, including satellite images, orthophotographs and bathymetric maps, were acquired from different sources, while the derived datasets were extracted mainly from these input datasets (Table 5). All input datasets were assembled, georeferenced and projected to the Geocentric Datum of Australia (GDA94) using ESRI’s ArcGIS 10. Table 5 – Datasets and sources used in this study.

Datasets (raster) Sources a) Satellite images - Landsat 5 (TM) The United States Geological Survey (USGS)

- Landsat 7 (ETM+) The United States Geological Survey (USGS) - Landsat 8 (OLI) The United States Geological Survey (USGS) b) Orthophotographs Landgate, Western Australia Geoscience Australia (GA), c) Bathymetric maps Australian Hydrographical Office (AHO) Input datasets Input Datasets (feature classes) Description Reefs Major reefs occurring along the Kimberley Bioregion including reefs size, shape and types.

Coastline Land-water boundary mapped using the Mean Low Water Neap (MLWN) level to ensure that mangroves and reef flats are not included on the land. Islands Islands, islets, exposed rocks and cays of the Kimberley Bioregion as defined by the Geoscience Australia. Derived datasets Field surveys Ground-truth data, which included field surveys, observations and georeferenced field photographs for more than 300 points, were collected from primary and secondary sources using handheld GPS devices (WA Museum Woodside Collection Kimberley Project, 2009–2012; Wilson and Blake, 2011; Wilson et al., 2011, WAMSI 1.3.1 Reef Geomorphology Project, 2012 - 2015; Solihuddin et al., 2015) These observations, were made on various dates, encompassed descriptive geomorphological information, habitats and substrates, and were used as reference data for validation purposes.

24 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Mapping approach The Kimberley coastline has been mapped using a variety of different geomorphic and geological classification schemes. However, most existing maps of the Kimberley coast are derived from low or moderate spatial resolution data sources. Furthermore, due to the macrotidal systems in this region, defining land-water boundaries marked by tides can be complicated and may not represent the shoreline in a geomorphic aspect; thus, mangrove zones and sections of the reef flats are commonly classified as land on these maps. For this study, the marine/terrestrial boundary was remapped using the Mean High Water Neap (MHWN) level to ensure that mangroves and reef flats are not included on the land (Figure 13). A set of 24 orthophotographs that cover the entire Kimberley bioregion was used to accurately delineate the shorelines of mainland and nearshore islands. These orthophotographs were mainly in true colour and geometrically corrected, and had been acquired during MLWN tides. Although they usually do not have as many spectral bands as satellite images do, their spatial resolution is very high (< 1 m) and this resolution can bring out detailed features on the ground (Figure 13).

Land Water Water Reef flat Mangroves

Figure 13. Detail of features evident in high resolution images. The red line shows the coastline of this study which has been accurately delineated using high resolution Orthophotos. The white line represents a coastline from another study.

Kimberley Marine Research Program | Project 1.3.1 25

Reef Growth and Maintenance

Reef mapping approach The visibility of a reef feature depends on water clarity and exposure, however the extraordinary conditions of the Kimberley marine environment, including extremely high tides and water turbidity has the potential to prevent effective detection or mask reef features in this region. Hence mapping of reefs using remote sensing images alone was not going to be sufficient to detect the true number of reefs (Figure 14).

(a) (b) (c)

Figure 14. Landsat images of reefs near Jones Island (north KIM) acquired over three different dates, (a) at extremely high tide; (b) at high tide; and (c) at low tide.

Accordingly, a mapping process was developed for this study in order to determine reef locations precisely (Figure 15). First, 18 bathymetric maps which cover the study area were used as a base map. Although the bathymetric maps do not present reefs and coastal features at a geomorphic standard, they are a good source of information on reef locations and water depth. Areas of interest, such as islands, reefs and shoals, were roughly delineated using ArcMap 10.1. The digitised map features were then stored in a vector format (Figure 15). Next, a dataset was developed using Landsat TM, ETM+ and OLI images at a consistent ∼30 m resolution and using the first five spectral bands. More than 60 scenes covering the entire study area were acquired between 1999 and 2014. Regardless of the cloud cover density, reefs could be detected on sections of the images without cloud cover because each scene covers a relatively large area (170 km north-south x 183 km east-west). This data was then validated using ground truth and very high resolution orthophotos. RS images and digitised map features were imported to ArcMap 10.1 as data layers (Figure 15). RS images were then used as a base map and reef polylines were overlaid on these images to allow visual identification of coral reefs. All islands and reefs detected on an RS image were accurately recorded and classified according to the reef classification that has been developed for this study. Moreover, a reef that has multiple features (i.e. comprising more than one polygon) is recorded in the GIS database as a single reef.

26 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 15. Methods employed to map the coastline and islands and to identify reef locations and dimensions.

4.3 Results Coastline, island and reef mapping The geology of the Kimberley coast is characterised by a heavily jointed, faulted, and folded palaeo Proterozoic metasediment and volcanic’s which has resulted in a complex, highly discordant ria type coastline. The straight- line distance between Cape Londonderry and Cape Leveque is approximately 500 km, while the actual length of the coastline is 10 times longer, exceeding 5,000 km in length. However, if the lengths of island coastlines are taken into account, the total length of the Kimberley coastline is approximately 10,000 km (Figure 16).

Kimberley Marine Research Program | Project 1.3.1 27

Reef Growth and Maintenance

Figure 16. A digitised map showing the complexity of coastline features and island distribution along the Kimberley Bioregion coast.

A total of 2413 islands were mapped within the KIM, the vast majority being high rocky islands with only 31 low or reef islands recorded. Islands show significant geomorphologic complexity and spatial variability, with island numbers varying both latitudinally (Table 6) and across the shelf (Figure 17).

Table 6 - Number and percentage of islands by latitude across the Kimberley Bioregion.

Latitude (S°) Number % 13° 30′ - 14° 00′ 85 3.5 14° 00′ - 14° 30′ 441 18.3 14° 30′ - 15° 00′ 513 21.3 15° 00′ - 15° 30′ 385 16.0 15° 30′ - 16° 00′ 129 5.3 16° 00′ - 16° 30′ 784 32.5 16° 30′ - 17° 00′ 76 3.1 Total 2413 100

Islands are primarily clustered into two main regions with about 40% of the islands situated between 14°S and 15°S near the mainland coastline and in the Bonaparte Archipelago, while the other group comprising about 36% of the islands is found between 16°S and 17°S in the Buccaneer Archipelago (Table 6).

28 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

The vast majority of Kimberley Islands are relatively small in size; over 35% of islands have an area of less than 25 hectares (< 0.25 km2) and are little more than exposed rocks or islets, more than 75% of islands have an area of less than 1 km2. The number of islands decreases exponentially with size; with less than 2% of islands having an island area larger than 5 km2 (Table 7).

Table 7 – Frequency and percentage of island by area category in the Kimberley Bioregion.

Area (km2) Number % < 0.25 853 35.4 0.25 - 0.5 466 19.3 0.5 - 1.0 518 21.5 1.0 - 5.0 534 22.1 5.0 - 10 19 0.8 10 - 20 11 0.5 20 - 30 6 0.2 30 -40 2 0.1 40 -50 2 0.1 > 50 2 0.1 Total 2413 100

In terms of spatial variability, the number of islands decreases dramatically moving away from the coastline (Figure 17). Over 63% of all islands are found within a short distance (< 5 km) of the mainland coast, a further 30% of islands are found between 5 km and 20 km offshore and, only 7% of islands are found at distances greater than 20 km from the Kimberley coastline. While there appears to be no relationship between island size and distribution, approximately 13% of total island area is located within < 5 km of the mainland coast, this increases significantly to more than half of the island area (over 55%) between 5 km and 20 km from the mainland coast. Between 20 and 30 km from the coast island area steeply declines to reach about 24% of the total island area. An additional drop in total island area continues until it reaches its lowest value of < 2% between 30 and 40 km from the coastline.

Kimberley Marine Research Program | Project 1.3.1 29

Reef Growth and Maintenance

Figure 17. Number and size of islands across the shelf. The horizontal axis represents distances from the Kimberley mainland coastline. The left vertical axis represents the number of islands and the correlation between the number of islands and distance from the coastline is shown in the black solid line. The right vertical axis represents island areas in (km2) and the correlation between the island areas and distance from the coastline is shown in the grey dashed line. Geomorphic classification of the Kimberley Reefs A reef evolutionary or genesis model sensu Hopley (2007) has not been incorporated in this study, due to the simple fact that unlike the Great Barrier Reef, Kimberley reefs have not had the same level of geological and geomorphological investigation. Instead a geomorphological typology classification scheme using a simple hierarchy (primary, secondary and tertiary) of geometric criteria (Figure 18) has been developed for the Kimberley reefs, using the same categories as in Hopley, 1982 and Hopley et al., 2007.

30 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 18. A geomorphological typology classification scheme of the North West Shelf. The left section represents reefs from mainly the Oceanic Shoal Bioregions (OSS) and partially from the North West Shelf Bioregion (NWS). The right section represents reefs of the Kimberley Bioregion (KIM), (modified from Collins et al., 2015).

Oceanic Shoals Bioregion

The continental margin of the Kimberley has a ramp-like profile that can be subdivided bathymetrically into an outer, mid and inner ramp (James et al., 2004). The outer ramp (>120 m water depth) has a relatively steep seaward slope and corresponds to the OSS Bioregion. Reefs here are tower-like, rising from an old Miocene shoreline, with heights of 200–400 m above the seafloor, and consist of multiple stages of Pleistocene to Holocene reef growth stages separated by hiatuses (Collins et al., 2011). Termed ‘slope atolls’ by Wilson (2013), their morphologies include annular (e.g. Mermaid Reef at Rowley Shoals. Figure 19b), lunate (e.g. south Scott Reef. Figure 19a), and platform (e.g. Ashmore Reef. Figure 19c) types. Similar submerged forms, in which Halimeda is a significant sediment contributor, are termed ‘banks’ by Heyward et al. (1997) and Wilson (2013); these occur to the north on the Sahul Shelf. The OSS Bioregion is included here for comparative purposes, but the focus of this study is on the Kimberley Bioregion.

Kimberley Marine Research Program | Project 1.3.1 31

Reef Growth and Maintenance

(a) (b) (c)

North Scott Reef

(d) (e)

South Scott Reef

Figure 19. Morphology of slope atolls of the outer ramp and OSS Bioregion. Examples of slope atolls are: annular reefs (a) North Scott Reef and (b) Mermaid Reef; lunate reefs (a) South Scott Reef; platform reefs (c) Ashmore Reef and (e) Cartier Reef; banks (d) Heywood Shoal.

Kimberley Bioregion

The primary level of the geomorphological classification scheme shown in Figure 18 represents reef flat elevations against a tidal datum. The first category is high intertidal reef, where the reef flats are situated above the mean low water neap (MLWN) tidal level. The second category is low intertidal reef, where the reefs flats are situated between the level of MLWS and MLWN tidal levels. The third category is subtidal reef, which is situated below the mean low water spring level. The secondary level is based on descriptive morphological characteristics and include fringing, planar, patch and shoal reefs. The tertiary level divides each reef type into subclasses based on more detailed reef geometry and substrate type, these reef classes are described below.

Fringing reefs

Fringing reefs are well developed and widespread throughout the Kimberley and vary in shape and size, and can be found along both the mainland and islands coasts. The majority of fringing reefs in the Kimberley Bioregion can be further classified into five subclasses which include: (1) Headland, (2) Bayhead, (3) Narrow Beach Base, (4) Inter-Island and, (5) Circum-Island. Several of these classes, resemble the fringing reefs of the GBR that were identified by (Hopley et al.) 2007 with some modifications (Figure 20).

32 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Bathurst Is. (a) (b) (c) Cape Leveque

One Arm Point

Cygnet Maret Is. Bay

(d) Cape Londonderry

Irvine Is.

(e) Molema Is. (f) (g) Cockatoo Is.

Tallon Is.

Figure 20. Morphology of fringing reefs of the KIM Bioregion. Examples of fringing reefs: (1) high intertidal fringing reefs (reefs are indicated by red arrows) (a) reef between Bathurst and Irvine Islands, (f) Turtle reef north and south Molema Island and (g) reef on the eastern side of Tallon Island; (2) low intertidal headland fringing reefs (b) reefs on NW Maret Island and (d) Cape Londonderry; low intertidal bayhead fringing reefs (a) reefs on SE Irvine Island and (c) reefs on Cygnet Bay S One Arm Point; low intertidal headland fringing reefs on narrow beach base (c) reefs on sandy beach N One Arm Point; and low intertidal circum-island fringing reefs (e) reef around Cockatoo Island. Headland fringing reefs are found on many islands and mainland coasts. They are mainly developed on rocky intertidal shores that are often exposed to prevailing swell, e.g. reefs on NW Maret Island (Figure 20b) and reefs on Cape Londonderry (Figure 20d). 1. Bayhead fringing reefs also occur on many islands and mainland coasts. They are developed in embayments and advance toward the head of the bay, e.g. reef on SE Irvine Island, (Figure 20a); reef on Cygnet Bay, (Figure 20c); and reef on E Tallon Island, (Figure 20g). 2. Narrow beach-based fringing reefs are frequently found on island shores and occasionally on mainland shores. They are developed along extended sandy coasts, e.g. reefs on E Maret Island, (Figure 20b); and reef on N One Arm Point (Figure 20c). 3. Inter-Island fringing reefs are developed between two or more high islands and/or a peninsula. These islands are connected together by their fringing reefs. This type of fringing reef is largely found toward the southern end of the KIM, e.g. reef between Bathurst and Irvine Islands, (Figure 20a); and Turtle reef which connects Molema Island and other Islands to the south; and northward, Molema Island to a peninsula (Figure 20f). 4. Circum-Island fringing reefs are the most common type in the KIM. They are developed around high islands (e.g. Cockatoo Island. Figure 20e). Additionally, there are distinctive features of high intertidal fringing reefs that are common across these five subclasses. These is high intertidal flat-topped reefs with distinctive lithified algal terraces and coralline algae (rhodolith banks) which form coral filled pools during low tides, as well as Porites micro-atolls, e.g. reef between Bathurst and Irvine Islands, (Figure 20a); Turtle reef on north and south Molema Island, (Figure 20f); and reef on east Tallon Island, (Figure 20g).

Kimberley Marine Research Program | Project 1.3.1 33

Reef Growth and Maintenance

Planar reef The planar reefs of the Kimberley Bioregion are mainly characterised by their large area and are situated some distance offshore. These reefs are considered senile, where the lagoon is either completely infilled or semi infilled and a reef flat extends across the entire reef platform. The reef flats are usually emergent at low tides and can have a central low island and/or a sand cay. Planar reefs can be divided into two subclasses based on substrate type: (1) Sand Lagoon and (2) Coralgal (Figure 21).

(a) (b)

(c)

Figure 21. Low intertidal planar reefs (a) Adele Reef and (b) Long Reef; and high intertidal planar reef (c) Montgomery Reef. 1. Sand lagoon planar reefs are the more common reef type with large areas of the central lagoon dominated by mobile sand sheets and sand filled lagoons, though it is not known whether these sand infilled relatively shallow or deep lagoons, e.g. Long Reef, (Figure 21b). Sand cays or vegetated sand islands can also be a characteristic feature. 2. Coralgal planar reefs are mainly found on larger reef platforms that have a central high island. They are characterised by distinctive lithified algal terraces and coralline algae (rhodolith banks) predominantly towards the edge of the reef flat, e.g. Montgomery Reef, (Figure 21c). Some reef islands such as Adele Reef appear to show both coralgal and sand lagoon morphologie,s (Figure 21a).

Patch Reefs Patch reefs are mainly smaller in size, <3 km in length. They can be divided into two subclasses based on perimeter geometry: (1) Irregular margin and (2) Unbroken margin,

34 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

1. Irregular margin refers to patch reefs that have dentate or rugged edges. 2. Unbroken margin refers to patch reefs that have circular, elliptic or regular edges.

Shoals It is because shoals are subtidal features and difficult to discern using airborne or satellite remote sensing techniques that they were not catalogued as part of this study. However, they can be divided into three subclasses based on dominant substrates types. The most common substrates that have been identified include (1) Sand, (2) Reef, and (3) Proterozoic Basement. Reef distribution and size Coral reefs occur extensively throughout the KIM and display a remarkable geomorphological diversity. A total of 853 reefs were identified and mapped through the KIM with a combined reefal area of about 1950 km2. Each mapped reef was accurately recorded and classified according to the reef classification scheme described above (Figure 22).

Figure 22. Reef distribution in the Kimberley Bioregion by type: fringing; planar; patch; and shoals. (left plate) Distribution map of reefs in the Kimberley Bioregion with spring tidal range contours (calculated from the National Tidal Unit, Australian Bureau of Meteorology (right plate). The pie chart shows the percentage of area coverage of each reef type.

Fringing reefs are by far the most common type accounting for 687 reefs, covering more than 910 km2 of reefal area fringing almost a quarter of the total Kimberley mainland and island coastline. The majority of fringing reefs (70%) have developed around islands (Figure 22). There are approximately 20 planar reefs in the KIM, though their combined reefal area of > 909 km2 is almost equivalent to the fringing area (Table 8). Fringing and planar reefs together comprise approximately 93% of the total reef area in the Kimberley Bioregion. The remaining reefal area is covered by patch reefs (2%) and shoals (5%), (Figure 22).

Kimberley Marine Research Program | Project 1.3.1 35

Reef Growth and Maintenance

Despite their large number and abundant distribution, individual fringing reefs are relatively small in size having a modal area of 0.125 m2 and mean area of 1.3 km2, whereas shoal reefs are somewhat larger having a mean area of 7.1 km2 and planar reefs are significantly larger again having a mean area of 45.5 km2. Among all the reef types of the study area Patch reefs have the smallest size with a mean area of <1 km2 (Table 8). Table 8. Lengths and areas of reefs by type.

Length (km) Area (km2) Reef type Number Min Max Mean ± SD Total Min Max Mean ± SD Total Fringing 687 0.03 9.3 1.8 ± 14.0 2638.7 0.5x10-4 32.6 1.3 ± 3.3 910.7 Planar 20 3.2 22.1 10.9 ± 41.1 259 1.4 352.4 45.5 ± 90.1 909.4 Patch 133 0.04 2.9 0.85 ± 1.6 103 3.7x10-4 3.4 0.27 ± 0.5 35.4 Shoal 13 1.3 12.4 4.9 ± 10.2 76.1 0.6 18.7 7.1 ± 6.9 92.3 Total 853 3076.8 1948

Regarding spatial variability, the number of reefs declines markedly with distance from the mainland coast (Figure 23). Like Kimberley islands nearly half of the reefs are located in close proximity (<5 km) to the Kimberley coastline, a further 33% of reefs are found between 5 km and 20 km off shore. The number of reefs drops continuously at 40 km off shore, with a slight increase (3%) beyond 40 km, this pattern can be attributed to fringing reefs being the dominant reef type. More than half of the total reef area in the Kimberley (> 54%) is found within 10 km of the coastline. At distances greater than10 km from the coast, reef areas fall significantly to less than 4% at 25 km off shore, with a small increase in the reef area between 25 km and 35 km. An additional drop in reef area occurs at 40 km to reach its minimum value (<0.5%). However, the reef areas jump dramatically to approximately 15% more than 40 km from the coastline of the Kimberley which can be explained by a small number of large planar reefs located at this distance from the coast.

Figure 23. Number and size of reefs across the shelf. The horizontal axis represents distances from the Kimberley mainland coastline. The left vertical axis represents the number of reefs and the correlation between the reefs number and distance from the coastline is shown in the black solid line. The right vertical axis represents reef areas in (km2) and the correlation between reef areas and distance from the coastline is shown in the grey dashed line.

36 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Reef distribution by latitude

Fringing reefs are found throughout the Kimberley but have particularly high concentrations within the archipelago regions. These include the Bonaparte Archipelago in the North (14° S to 15° S) and the Buccaneer Archipelago in the South (16° S to 17° S), each of these regions account for about 30% of the total number of fringing reefs (Table 9). Planar reefs are mainly developed offshore. Although they are relatively few in number, they are characterised by a large reefal area. They are clustered in two main areas, with 38% of their area found between 13° 30’ S and 14° 30’ S and over 52% of their area situated between 15° S and 16° S, including Montgomery Reef, the Adele complex of 6 reefs, and other reefs. Planar reefs are rare or non-existent between 14° 30’ S and 15° 00’ S (Table 9). Patch reefs are smaller in size and are widespread across the Kimberley coast. They are often hard to identify. Approximately 133 reefs have been counted in the bioregion within a short distance of the mainland coast. Their number gradually increases from the north toward the south of the Kimberley coast. They make up the highest proportion (42%) of reefs between 15° 00’ S and 16° 00’ S, south of which their number decreases (Table 9). Submerged reefs or shoals are far less numerous. Approximately 13 reefs of this type have been mapped. Their number fluctuates from the north to the south of the bioregion. However, they reach their greatest density in the south between 16° 00’ S and 16° 30’ S, and are rare or non-existent between 14° 30’ S and 15° 00’ S and on the outer edge of the KIM bioregion (Table 9). Table 9. Frequency of reef types by latitude in the Kimberley Bioregion.

Latitude (°S) Fringing Planar Patch Shoal Total 13° 30’ - 14° 00’ 48 6 6 3 63 14° 00’ - 14° 30’ 140 2 9 2 153 14° 30’ - 15° 00’ 119 0 13 0 132 15° 00’ - 15° 30’ 74 4 56 1 135 15° 30’ - 16° 00’ 11 7 19 2 39 16° 00’ - 16° 30’ 259 1 29 5 294 16° 30’ - 17° 00’ 36 0 1 0 37 Total 687 20 133 13 853

4.4 Discussion Despite the early work of Teichert and Fairbridge (1948) and more recent investigations by Brooke (1997) and Wilson (2013), the Kimberley Bioregion was never considered to be major reef province. This view is mainly due to the low number of scientific studies which have investigated Kimberley reef systems, which can be attributed to its remoteness, lack of research infrastructure and settlements, challenging environmental conditions, and the focus of most Australian reef researchers towards the higher profile GBR. The results of this study show that the number of Kimberley reefs (853) and their area (1948 km2) are significantly greater (>60% greater) than the reef number and area found in other sources of information on reefs, such as ReefBase (2015) and the Millennium Coral Reef Mapping Project (IMaRS-USF and IRD, 2005). Compared with the GBR the Kimberley has a longer coastline, a greater number or islands, and although its total reefal area is 10 times smaller than the GBR (~2,000 km2 compared to 20,000 km2), this study has confirmed the Kimberley as one of the top 20 largest continental shelf reef provinces in the world. Reef distribution in the Kimberley Bioregion shows significant morphological complexity with clear regional patterns. Fringing reefs are the dominant reef type and are widely distributed throughout the Kimberley Bioregion. Coastal fringing reefs are more common in the northern Kimberley having developed intermittently along the extended mainland ria coast. These fringing reefs become more intermittent along the southern coast

Kimberley Marine Research Program | Project 1.3.1 37

Reef Growth and Maintenance

and areas proximal to major river mouths as a result of increased terrestrial runoff and higher water turbidity. Fringing island reefs have a far greater latitudinal spread and are more common than other reef types in the Kimberley Bioregion. As is expected fringing reef distribution correlates strongly with the distribution of islands. The most significant numbers of fringing reefs in the Kimberley Bioregion are found around nearshore high islands of the Buccaneer Archipelago. At this location the folded and faulted King Leopold Fold Belt has been dissected into many islands as a result of drowning of the continental during the Holocene. Fringing reefs are also abundant in the Bonaparte Archipelago in the north. Both these areas have been suggested as being important for fringing reef development (Wilson, 2013). Fringing reefs with a westerly aspect often have a wider, geomorphically mature reef flat with a steeply sloping reef front, these reefs are more often exposed to higher energy ground swells. Fringing reefs with an easterly aspect generally have a narrower gently sloping reef flat and tend to be sheltered from high swell energy. The asymmetric style of fringing reef development is particularly apparent around most islands in the Bonaparte Archipelago. This indicates that either exposed reefs grow faster than sheltered reefs, or they start growing earlier (Wilson, 2013). A notable subclass of fringing reef – inter-island – falls within this category. This type of reef is particularly common in the Buccaneer Archipelago (e.g. Molema Island, Bathurst-Irvine Islands, and Woninjaba Islands). The reef platform connects two or more islands seem to be formed by a coalescence of the fringing reefs that are attached to these islands. Where coalescence is incomplete deep pools can be observed, e.g., between Bathurst- Irvine Islands. High intertidal fringing reef is a newly defined reef morphotype and is characterised by reef flats elevations above the level of MHWS tides, coralline algal terraces, and rhodolith banks (Richards and O’Leary, 2015). This reef type is best developed in the Buccaneer Archipelago and Sunday Island group, where tidal ranges can exceed 10 m. However it is still unknown what process is ultimately driving the development of these high intertidal reefs. Planar reefs are isolated features usually located some distance from the mainland coast, beyond the fringing reef zone in water depths of 30–50 m. These reefs appear to have developed on shallow, pre-existing topographic highs, between palaeoriver channels (Collins et al., 2015). The majority of these reefs have reached sea level with the reef flat mostly blanketed by a sand sheet and surrounded by an intertidal reef platform or rampart. Based on Hopley’s evolutionary classification scheme these reefs can be classed as senile in that they have a central low, vegetated island (e.g. Adele Reef) or an unvegetated sand cay (e.g. Long Reef). Patch reefs are scattered over the Kimberley Bioregion. These reefs usually occur in low intertidal or subtidal zones. Due to their small size, their features are often difficult to detect using moderate spatial resolution RS images, and/or because of high water turbidity rendering statistics unreliable. Thus, patch reefs have not been mapped in great detail in this study. Due to the lack of adequate information on patch and shoal types in addition to their small size and depth under water, where they do not represent an area of great significance, these reefs have not been studied in as much detail as high intertidal and low intertidal fringing and planar reefs. Therefore they have been noted in this study in terms of basic information such as their distribution, number and area, based on available data. There is a steady correlation between island distributions and reef size because there are an abundance of reefs attached to islands near the coast. Thus, most Kimberley reefs can be found within 10 km of the mainland coast. However, some reefs can’t be presented clearly within less than 5 km from the shoreline where reef area drops significantly although the island’s area remains high. The reason is most of the reefs are located on steeply sloped forereefs and in a water depth >5 m where they cannot be detected using RS images due to high turbidity. This statistic is particularly significant in a management perspective as this close proximity to the coast means that there is a higher potential for reefs to be threatened or impacted by land use change. In summary, this study presents the most comprehensive dataset of reef typology and distribution for the

38 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Kimberley Bioregion, provides a firm scientific determination of the scale of reef habitats in the Kimberley and will provide a foundation for future biodiversity assessments and geomorphic studies. It was anticipated that the resulting data would provide some of the first detailed spatial analysis of the Kimberley Bioregion and would thus make available a reliable, spatially constrained data set for biodiversity assessment and reef structure comparisons. Furthermore, it would provide scientists and managers with quality information relevant to the monitoring, conservation and management of these vital natural resources. It would also pave the way for future studies addressing the major threats to reef ecosystems, such as the impact of tropical cyclones and wave exposure.

4.5 Acknowledgements This study is part of the Kimberley Reef Geomorphology Project 1.3.1 which has been funded by the Western Australian State Government through the Western Australian Marine Science Institution (WAMSI). We are grateful to the Traditional Owners of the Kimberley (the Bardi Jawi, Mayala and Dambimangari people) for their assistance, advice and consent to access their traditional lands. We also thank WA Museum for providing ground truth data through the WA Museum/Woodside Collection Project (Kimberley) 2008 – 2011. We also wish to thank the following: The Kimberley Marine Research Station for providing vessels and logistic support for marine operations and access to research facilities; Kimberley Media for providing quality site pictures; The Department of Parks and Wildlife (DPaW) for providing VHR orthophotographs and HR satellite images; Geoscience Australia (GA) for providing DEMs and GIS data; The United States Geological Survey (USGS) for providing Landsat imagery; The Geological Survey of Western Australia (GSWA) for providing digital geological maps. And last but not least, special thanks to valued members of the research team at Curtin University: Mr Tubagus Solihuddin and Ms Giada Bufarale.

4.6 References

Andréfouët, S., Muller-Karger, F., Robinson, J., Kranenburg, James, N. P., Bone, Y., Kyser, T. K., Dix, G. R., and Collins, L. C.,Torres-Pulliza, D., Spraggins, S. and Murch, B. B., (2004). The importance of changing oceanography (2006). Global assessment of modern coral reef in controlling late Quaternary carbonate extent and diversity for regional science and sedimentation on a high Centre..nservation management applications: A view from space. Monitoring Centre.g Centre.ervation Monitoring Proceedings of the 10th International Coral Reef Centre.ntre.a Part A: M Symposium, Okinawa, Japan, 1732-1745. Joubin, M. (1912). Carte des bancs et récifs de Coraux Brooke, B. (1997). Geomorphology of the north Kimberley (Madrépores). Paper presented at the Annales de coast, in Walker (ed) Marine biological survey of the Géographie. central Kimberley coast. Western Australia. University Jupp, D.L., Mayo, K.K. and Kuchler, D.A., (1985) Remote of Western Australia, Perth, unpublished report WA sensing for planning and managing the Great Barrier Museum Library No. UR377, 13-39 Reef of Australia. Photogrammetria, 41: 21-42. Collins, L. B. (2011). Controls on Morphology and Growth Purcell, S., (2002). Intertidal reefs under extreme tidal flux History of Coral Reefs of Australia’s Western Margin. in Buccaneer Archipelago, Western Australia. Coral Cenozoic Carbonate Systems of Australia (95), 195. Reefs, 21(2), 191-192. Collins, L. B., O'Leary, M., Stevens, A., Bufarale, G., Kordi, ReefBase: A Global Information System for Coral Reefs. M., and Solihuddin, T. (2015). Geomorphic patterns, March, 2015. http://www.reefbase.org internal architecture and reef growth in a macrotidal, high-turbidity setting of coral reefs from the Richards, Z. T., and O’Leary, M. J. (2015) The coralline algal Kimberley bioregion. Australian Journal of Maritime cascades of Tallon Island (Jalan) fringing reef, NW and Ocean Affairs, 7(1), 12-22. Australia. Coral Reefs, 34(2), 595-595. Collins, L.B., Testa, V., Zhao, J. and Qu, D., (2011). Holocene Rowlands, G., Purkis, S., and Bruckner, A., (2014). Diversity growth history of the Scott reef carbonate platform in the geomorphology of shallow-water carbonate and coral reef. Journal of the Royal Society of Western depositional systems in the Saudi Arabian Red Sea. Australia, 94(2):239-250. Geomorphology, 222, 3-13. Darwin, C.R., (1842). The Structure and Distribution of Coral Solihuddin, T., Collins, L.B., Blakeway, D., O’Leary, M.J., Reefs. Smith, Elder and Co., London, 214. 2015. Holocene Reef Growth and Sea Level in a Macrotidal, High Turbidity Setting: Cockatoo Island,

Kimberley Marine Research Program | Project 1.3.1 39

Reef Growth and Maintenance

DEC, Department of Environment and Conservation (2011). Kimberley Bioregion, Northwest Australia. Marine accessed 12/11/2011 http://www.dec. Geology, 359: 50 – 60. wa.gov.au/kimberleystrategy. Spalding M.D., Ravilious C., Green E.P., (2001). World Atlas Edwards, H., Farwell, D., Lee, J., and Fredericks, P. (2003). of Coral Reefs. Berkeley (California, USA): The Vibrational spectroscopic study of the contents of a UNIVERSITY OF CALIFORNIA Press. 436 pp. chest excavated from the wreck of the HMS Pandora. Teichert, C. and Fairbridge, R.W., (1948). Some coral reefs Spectrochimica Acta Part A: Molecular and of the Sahul Shelf. Geographical Review. 38(2):222- Biomolecular Spectroscopy, 59(10), 2311-2319. 249. Fairbridge, R., (1950). Recent and Pleistocene coral reefs of WA Museum Woodside Collection Project (Kimberley) Australia, The Journal of Geology 58: 330-401. 2008-2011 Fairbridge, R., (1967). Coral reefs of the Australian region. Wilkinson, C., (2008). Status of coral reefs of the world: In Jenings, J. N., and Mabbutt, J. A. (eds). Landform 2008. Global Coral Reef Monitoring Network and Reef Studies from Australia and New Guinea. Canberra: and Rainforest Research Centre, Townsville, Australia, Australian National University Press, 386-417. 296 p. Finkl, C.W., (2011). Reef Classification by Fairbridge (1950). Wilson, B. and Blake, S., (2011). Notes on the origins and In: Encyclopaedia of Modern Coral Reefs: structure, biogeomorphology of Montgomery Reef, Kimberley, form and process. Hopley D, editor, pp 846 – 850. Western Australia. Journal of the Royal Society of Green, J. N. (1975). The VOC ship Batavia wrecked in 1629 Western Australia, 94:107-119 on the Houtman Abrolhos, Western Australia. Wilson, B. R., (2013). The Biogeography of the Australian International Journal of Nautical Archaeology, 4(1), North West Shelf New York, USA: Elsevier. 43-63. Wilson, B., (1972). Western Australian coral reef with Heyward, A. A., Pinceratto, E. E., and Smith, L. L., (1997). Big preliminary notes on a study at Kendrew Island, Bank Shoals of the Timor Sea: an environmental Dampier Archipelago. Report of the Crown of Thorns resource atlas: Australian Institute of Marine Science Starfish Seminar. Brisbane 47-58. and BHP Petroleum. Wilson, B., Blake, S., Ryan, D. and Hacker, J., (2011). Hopley, D., (1982). The geomorphology of the Great Barrier Reconnaissance of species-rich coral reefs in a muddy, Reef: Quaternary development of coral reefs: Wiley macro-tidal, enclosed embayment, Talbot Bay, New York. Kimberley. Western Australia, Journal of the Royal Hopley, D., (2011) Encyclopaedia of Modern Coral Reefs: Society of Western Australia, 94:251-265 structure, form and process. Encyclopaedia of Earth Wolanski, E., and Spagnol, S., (2003). Dynamics of the Sciences. Springer, Dordrecht, The Netherlands. turbidity maximum in King Sound, tropical Western Hopley, D., Smithers, S. G., and Parnell, K. E., (2007). The Australia. Estuarine, Coastal and Shelf Science, 56(5- geomorphology of the Great Barrier Reef: 6), 877-890. development, diversity and change: Cambridge Wood, M., and Mills, D., (2008). A turning of the tide: University Press. science for decisions in the Kimberley-Browse marine IMaRS-USF and IRD (Institute de Recherche pour le region. A report prepared for the Western Australian Development) (2005). Millennium Coral Reef Marine Science Institution (WAMSI). Mapping Project. Validated maps. Cambridge (UK):

UNEP World Conservation Monitoring Centre.

40 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

5 Mapping Intra-reef Geomorphology and Associated Substrates and Habitats of the Kimberley Bioregion, North-western Australia.

Adapted from Kordi M.N., Collins, L.B., O’Leary, M.J., & Stevens A.M., in review. Mapping Intra-reef Geomorphology and Associated Substrates and Habitats of the Kimberley Bioregion, North-western Australia. Remote Sensing of the Environment.

5.1 Introduction The Kimberley, located in NW Australia, is subdivided into several marine bioregions including the Oceanic Shoals Bioregion (OSS), the North West Shelf Bioregion (NWS), and the Kimberley Bioregion (KIM) (see Thackway et al., 1988). The Kimberley Bioregion also contains the second largest reef system (after the Great Barrier Reef) in Australia spanning 4 degrees of Latitude between Cape Londonderry and Cape Leveque. The region is recognised as being globally significant as it represents one of the largest least impacted marine ecosystems on the planet, which can be attributed to the region’s geographic remoteness and inaccessibility. However this remoteness has also meant scientific studies into the Kimberley marine environments have at best been limited and infrequent. While threats from localised anthropogenic impacts are considered low (Collins, 2011), this may change, as the region has seen a recent increase in mining activities, particularly with the discovery of large gas fields in the offshore Browse and Bonaparte Basins, as well as a rise in the number of tourists visiting the region. In addition, the Kimberley’s geographic remoteness will not necessarily afford protection from climate related impacts resulting from changing sea surface temperature (SST), ocean acidification, and predicted increases in cyclone intensity. Accordingly, there is an urgent need to monitor the conditions of the reefs not only by developing a baseline map depicting their spatial distribution (Kordi et al., in review), but also by monitoring associated habitats and substrates to enable researchers and managers to document changing conditions over time. Until the late 1940s, there was a paucity of knowledge about habitats and substrates of the Kimberley reefs. This changed in 1948, when the first pioneering studies of the Kimberley coral reefs were conducted; in these studies, Admiralty charts and aerial photography were harnessed to map the major features of particular reef platforms (Teichert and Fairbridge, 1948; Fairbridge 1950, 1967). The following 50 years saw little scientific endeavour and only recently has the true size and nature of the Kimberley coral reefs been determined (Kordi et al., in prep). These more recent studies have found that the Kimberley reefs are home to a large number of species (DSEWPaC, 2012) and that there is a marked difference in faunal diversity between offshore and inshore reefs in this region (DSEWPaC, 2012; Richards et al., 2013). Nevertheless, limited remote sensing studies have been undertaken to map habitats and substrates from a geomorphological perspective at a reef scale in the Kimberley Bioregion. Recent work by Wilson (2013) highlighted the need for further geomorphic studies of the Kimberley reefs and their ecology.

Kimberley Marine Research Program | Project 1.3.1 41

Reef Growth and Maintenance

Figure 24. Map of the Kimberley Bioregion showing the location of the reefs in this study: (a) Cape Londonderry; (b) Long Reef; (c) Maret Island; (d) Montgomery Reef; (e) Adele Reef; (f) Tallon Island; (g) Molema Island (Turtle Reef); (h) Bathurst and Irvine Islands. Remote sensing The study and management of the Kimberley reefs presents a series of challenges, the foremost of these is the geographic remoteness and size of the region, which makes accessing the region prohibitively costly and time- consuming (Chin et al., 2008). A more cost effective approach in dealing with these kinds of reef mapping challenges has been for the application of remote sensing technologies in conjunction with limited ground truth surveys (Dahdouh-Guebas, 2002; Veron, 2008). Depending on the resolution of the imagery employed it is possible to map, with increasing resolution, reef geomorphology, substrate type, habitats and biological communities (Leon and Woodroffe, 2011). Remote sensing was employed to map reef systems soon after Landsat-1 was launched in the early 1970s (Smith et al., 1975; Jupp et al., 1985). Since then, the use of remote sensing has become a common reef-detection method in most coral reef studies around the world (Bina, 1982; Luczkovich et al., 1993; Andréfouët and Payri, 2000; Capolsini et al., 2003; Hochberg and Atkinson, 2003; Palandro et al., 2003; White et al., 2003) to mention a few. Remote sensing can be integrated into a Geographic Information System (GIS) environment, which has the ability to combine and store data of different types and from different sources, and manage and process these data. Moreover, GIS provides accurate mapping, which can be updated when new data become available, and reduces effort, cost, and time (Johnson, 2000). The models generated can be used to characterise the habitat types and substrates that exist on each reef. This knowledge enables estimation of likely types of habitats and substrates, which can be tested against available field data. This database leads to a better understanding of changes to the spatial distribution of reefs over time and paves the way for future studies of coral reefs in the region. However, no published works exist on the use of remote sensing data for monitoring coral reefs in the Kimberley Bioregion. The main goal of this study is to provide detailed geomorphological information on reefs of the Kimberley Bioregion to allow better understanding of these reefs. Landsat images and orthophotos were utilised in this

42 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

study to determine the boundaries of geomorphic zones and associated substrates and habitats on reefs. Ground truth, aerial photographs, and in situ images were also used to ensure that the image classes were correctly identified. The study is anticipated to provide a powerful management tool for coastal planners and managers involved in coastal conservation and resource assessment and will facilitate marine protected area planning and habitat mapping at a reefal scale.

5.2 Methodology Study area The Kimberley Bioregion stretches between Cape Londonderry in the north to Cape Leveque in the south (122°– 126°E; 13°–16°S). The presence of numerous bays, capes, and archipelagos has resulted in a very complex coastline (Figure 24). The straight-line distance between the two aforementioned capes is about 500 km, while the actual length of the mainland coastline is more than 5,000 km. The inner shelf region is characterised by numerous islands, which support extensive fringing reefs and associated coral habitats. The Kimberley Bioregion encompasses at least 2413 islands and more than 853 reefs (Kordi et al., in prep). The area is characterised by a macrotidal regime with spring tides exceeding 11 m in the southern Bioregion creating the highest tidal range of any coral reef system in the world (Purcell, 2002; Wolanski and Spagnol, 2003). Remote sensing datasets This study utilised Landsat imagery for mapping reef habitats as its sensors offer multispectral capability at a spatial resolution of 30 m and a panchromatic channel at 15 m; have a high return frequency and can be accessed free of charge. In this study, Landsat satellite images were used to study the following reefs: Montgomery Reef; Adele Reef; Turtle Reef; Cape Londonderry Reef; Tallon Island; Bathurst and Irvine Reefs; Long Reef; and Maret Island (Figure 25).

Kimberley Marine Research Program | Project 1.3.1 43

Reef Growth and Maintenance

(a) (b)

(g) (e) (f)

(h)

Figure 25. Satellite images used to map intra-reef geomorphology and associated substrates for the targeted reefs of this study: (a) Bathurst and Irvine Islands; (b) Cape Londonderry; (c) Montgomery Reef; (d) Maret Island; (e) Adele Reef; (f) Long Reef; (g) Molema Island; and (h) Tallon Island.

These reefs were selected because they are large enough to be detected by Landsat sensors, provide good quality satellite images during low tide (i.e. when the reef is exposed), and lastly but most importantly, have ground truth data and/or very high spatial resolution orthophotos available. Further, the reefs formed in a wide range of different settings including geographic location, tidal and hydrodynamic regimes. These particular reefs can be used as analogues for other reefs in the region and elsewhere where data availability is sparse. The fundamental data used in this study were multispectral satellite images from Landsat sensors Thematic Mapper (TM), Enhanced Thematic Mapper (ETM+), and the recently launched Operational Land Imager (OLI) (see Table 10).

44 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Table 10 – Landsat satellites and sensors and their specifications.

Wavelength Resolution Satellite Sensor Band (µm) (m)

Landsat 5 (TM) 1 Blue 0.45 - 0.52 30 2 Green 0.52 - 0.60 30 3 Red 0.63 - 0.69 30 4 Near Infrared (NIR) 0.76 - 0.90 30 5 Short-wave Infrared (SWIR) 1 1.55 - 1.75 30 6 Thermal Infrared (TIRS) 10.40 - 12.50 60 7 Short-wave Infrared (SWIR) 2 2.08 - 2.35 30

Landsat 7 (ETM+) 1 Blue 0.45 - 0.52 30 2 Green 0.52 - 0.60 30 3 Red 0.63 - 0.69 30 4 Near Infrared (NIR) 0.77 - 0.90 30 5 Short-wave Infrared (SWIR) 1 1.55 - 1.75 30 6 Thermal Infrared (TIRS) 10.40 - 12.50 60 7 Short-wave Infrared (SWIR) 2 2.09 - 2.35 30 8 Panchromatic 0.52 - 0.90 15

Landsat 8 (OLI) 1 Coastal aerosol 0.43 - 0.45 30 2 Blue 0.45 - 0.51 30 3 Green 0.53 - 0.59 30 4 Red 0.64 - 0.67 30 5 Near Infrared (NIR) 0.85 - 0.88 30 6 Short-wave Infrared (SWIR) 1 1.57 - 1.65 30 7 Short-wave Infrared (SWIR) 2 2.11 - 2.29 30 8 Panchromatic 0.50 - 0.68 15 9 Cirrus 1.36 - 1.38 30 10 Thermal Infrared (TIRS) 1 10.60 - 11.19 30 11 Thermal Infrared (TIRS) 2 11.50 - 12.51 30

More than 60 scenes covering the entire study area and acquired between 1996 and 2014 were sourced from the United States Geological Survey (USGS) and Earth Resources Observation and Science Centre (EROS). Landsat images are especially useful for this type of study, because the sensors collect scenes that cover nearly 170 km × 183 km in the north–south and east–west directions, respectively, with a ∼ 30 m spatial resolution and multispectral bands (Leon and Woodroff, 2011). Moreover, the images are archived over a reasonably long period, which suits the nature of this study area where cloud cover is frequent and macrotidal regimes dominate, often making high-resolution images unsuitable. In addition, 15 orthophotographs in true colour composite (RGB 321) with very high spatial resolution (<1 m) were used for verification and measuring purposes (Bouvet et al., 2003). Ground truth data Fieldwork data were collected from this study as well as from secondary sources including Western Australian Museum Woodside Collection Kimberley Project 2009–2012; Wilson and Blake, 2011; Wilson et al., 2011. Habitat and substrate information were recorded along with depth and coordinates and photos using a handheld global positioning system (GPS) with relatively high accuracy. The maximum surveyed water depth was 5 m due to limited visibility. In addition, georeferenced field photographs, aerial photography, and very high resolution orthophotos were used for validation purposes.

Kimberley Marine Research Program | Project 1.3.1 45

Reef Growth and Maintenance

Data processing A set of pre-processing procedures was applied to improve remote sensing image classification performance. Images were first converted to the Australian standard and GDA94 datum and geometrically corrected with ESRI’s ArcGIS 10 software package using ground-truth points with a root mean square error (RMSE) of <1.0 pixel. Then, images were calibrated and atmospherically corrected, using dark object subtraction to reduce the effects of atmospheric attenuation caused by degradation of the sensor and the zenith angle of the sun (Kawakubo et al., 2011). As images were mainly acquired at low tide, when water depths varied between 0 and <5 m at the forereef slope, the improvement from water column correction was marginal and can be neglected. The images were grouped into subsets and manually assigned to areas of interest (AOI) (i.e. shallow subtidal and intertidal reef platforms). Masking algorithms were applied using near infrared (NIR) bands to mask areas that were not of interest for mapping purposes such as land, deep and/or turbid water, clouds, and shadows, to reduce the variability of spectral classes (Phinn et al., 2008; Kaczmarek et al., 2010; Madden et al., 2013). Geomorphological zones usually have distinct boundaries and can be recognised using aerial photography or/and satellite images (Teichert and Fairbridge, 1948; Fairbridge, 1967; Hopley, 1982; Mumby and Harborne, 1999; Andréfouët et al., 2006; Hopley et al., 2007), in particular for intertidal and shallow reefs during low tides. The procedure that was followed to visualise these zones more clearly using ground truth data is summarised in Figure 26.

Figure 26. Schematic representation of typical geomorphological zones boundaries and associated habitats and substrate cover in the Kimberley Bioregion. (Photos courtesy of Kimberley Media).

Accordingly, geomorphological descriptions of zones based on substrates, habitats, and tidal zones were recorded for each reef. Table 11 shows an example of characteristic geomorphological zones and associated substrate and habitats found on Montgomery Reef.

46 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Table 11 – Methodology for defining the geomorphological zones and associated substrate and habitats on Montgomery Reef. Geomorphic zone Description Tidal zone Key substrates / habitats Land An area above the high tide water line Sand, mud, vegetation, rocks Supratidal that extends upland. Reef flat A wide shallow zone on top of the reef Mangroves, sand, coral rubble, platform usually exposed during low reef pavement, algal turf, tides. seagrass, coral communities Lagoon A combination between deeper Sand, coral communities antecedent topography and contemporary reef growth, major sinks of sediment. Reef crest The highest point of the reef between Sand, coral rubble, crustose reef flat and reef slope. It is mostly coralline algae, rhodolith exposed at low tide. Intertidal Forereef slope An area from the seaward edge of the Coral communities, seagrass, reef crest downwards (average steep Subtidal macro algae slopes 30°) or gently sloped (terraced).

A statistics-based unsupervised classification, Iterative Self-organizing Data Analysis (ISODATA) was applied, using ENVI 4.3 software. Unsupervised classification has been found to be very effective in identifying spectral clusters in remote sensing data (Campbell, 2002; Kakuta et al., 2010; Madden et al., 2013). Spectral classes were assessed and combined into seven classes based on expert-guided knowledge segmentation. Moreover, orthophotographs and ground-truth data were used to verify the classification. The thematic accuracy of image classification represents a quantitative estimation of the degree of correspondence between classified images and ground truth data. The validity of the classification can be determined based on the results of the accuracy assessment. The similarity of spectral reflection from many habitat and substrate types when using passive sensors with moderate spectral and spatial resolution make discriminating between these features difficult. Some of these habitats, however, usually occur in predictable geomorphological zones based on wave exposure and depth (Mumby et al., 1998). For instance, seagrass is most likely to be found in sheltered settings on reef flats. Therefore, if a feature was classified as seagrass on a forereef slope or reef crest zone, it is very likely to have been misclassified because these zones are often dominated by corals. Therefore, such seagrass dominated pixels have to be reassigned to the coral class. Although the habitat and substrate classification map was coupled to the geomorphological zonation map, the resulting maps were assigned two independent labels, one for habitats and substrates, and the other for the geomorphological zone. Classified images were vectorised to obtain a reefal-scale map of the geomorphological features of the study area. A digital elevation model (1 arc-second DEM) sourced from Geoscience Australia (GA) was used to identify reef geomorphic zones and to produce bathymetric maps. Reef maps were developed to present all related information of a reef platform on one map. The template used for these maps consists of the following: • a substrates and habitats classification map with a latitude and longitude grid and a scale bar; • a map legend showing map features based on geomorphic zones; • a geomorphic zones map showing the geomorphic units of the reef platform; • brief information about satellite imagery and ground truth points; and • a table of descriptive statistics of habitats and substrates based on statistics derived using GIS data analysis.

Kimberley Marine Research Program | Project 1.3.1 47

Reef Growth and Maintenance

Accuracy assessment The accuracy of the habitats and substrates map was assessed based on the computation of a confusion matrix process. The confusion matrix is a cross-tabulation which gives a quantitative estimation of the degree of correspondence between classified image in columns and the reality (i.e. ground truth) in rows. Ground truth data were compared to the predicted cover types on the habitat and substrates map to assess an overall accuracy of the map. The overall accuracy is a statistical value to measure the agreement between two different maps (e.g. classified and ground truth). It is calculated by adding the number of correctly classified pixels and dividing this number the total of all pixels, which is also useful for assessing the accuracy of the individual informational classes (Mumby et al., 1997). It is suggested that overall accuracies ranged between 60% and 80% is appropriate for reef management purposes (Green et al., 2000).

5.3 Results Seven classes of habitats and substrates were identified in five intra-reef geomorphic zones including two reef types (fringing and planar). Overall accuracies using a confusion matrix process ranged between 64% and 77% (Table 12). A standard template of reef map figures presenting all related information for each targeted reef platform was produced. The template consists of substrate and habitat classification maps with latitude and longitude grid and a scale bar, a map legend showing map features based on geomorphic zones, and a geomorphic zones map showing the intra-reef geomorphic units of the reef platform. Each geomorphic unit is presented in a different colour on the platform. Brief information is also provided on data sources, which include satellite imagery and ground truth data. Finally, a table of descriptive statistics of habitats and substrates is shown. Table 12 – The overall accuracy of each reef classification by classes. Overall accuracies have been broken down into individual accuracies of each habitat and substrate.

Classes (%) Overall Reef Man San SG & A CR RPAT CCA CC Accuracy Adele Reef 0.0 18.3 12.2 9.8 7.1 7.3 22.4 77.1% Bathurst & Irvine Islands 7.0 8.6 15.6 9.4 12.4 10.5 9.6 73.1% Cape Londonderry 8.1 11.2 9.3 7.4 15.7 0.0 12.7 64.4% Long Reef 0.0 18.3 8.7 10.4 14.5 7.6 9.4 68.9% Maret Island 0.0 20.3 11.5 0.0 14.0 0.0 20.6 66.4% Molema Island 10.3 9.3 9.0 0.0 14.4 9.6 16.7 69.3% Montgomery Reef 9.3 9.7 10.2 8.5 12.6 9.5 10.4 70.2% Tallon Island 9.5 8.2 11.2 7.6 18.3 7.2 13.6 75.6% Note: Description of substrate and habitat classes abbreviations: ‘Man’ = mangroves; ‘San’ = sand; ‘SG & A’ = seagrass and algae; ‘CR’ = coral rubble; ‘RPAT’ = reef pavement with algal turf; ‘CCA’ = crustose coralline algae; ‘CC’ = coral communities.

Fringing reefs

Maret Islands

The Maret Islands consist of two connected islands: North Maret and South Maret. These islands are situated in the Bonaparte Archipelago about 35 km off the coast of the Kimberley mainland (Figure 24c). The reef morphology of the Maret Islands is typical of fringing reefs along most of the Bonaparte Archipelago islands. There is a marked difference between windward and leeward reefs, generally, the reef is intertidal and shows a circum-island morphology. However, parts of the reefs on the north-western side of both islands can be considered as headland-attached. These reefs are exposed to the swell on the seaward shore and are dominated by domal faviids (Wilson, 2013). Reefs on the eastern side of both islands (leeward shore) can be classified as narrow beach and rock-base fringing reefs, and are protected from swells. Acropora dominates the coral

48 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

community. Figure 27 shows that the reef flat is wider towards the western, windward side of both islands. Sand covers about 22% of the reef area on both sides and more than 12% of the area is covered by reef pavement with algal turf on the western side. Seagrass and algae are also present, covering more than 13% of the reef flat. Coral communities are prolific on the reef flat on both sides of the reef and are prevailing on the forereef slope, covering >52% of the reef area.

Figure 27. (a) Habitats/substrate classification map of the Maret Islands Reef; (b) satellite image of Maret Islands with ground-truth points highlighted; (c) geomorphic zones map.

Cape Londonderry

The reefs around Cape Londonderry are the northern-most mainland fringing reefs in the Kimberley Bioregion (Figure 24a.). The reefs have developed on three rocky headlands and are exposed to the swell (Wilson, 2013). The reef flats are elongated, face northwest and extend between five and seven km seawards and are irregularly shaped (Figure 28). These reefs are considered intertidal and headland-attached fringing reefs. Based on classification verification using very high resolution aerial photography (<1 m spatial resolution), the habitats and substrate on the reef flats look very similar to other fringing reefs in this category. The reef flats are relatively level and reef pavement with algal turf dominates, covering about 36% of the reef flats. Mangroves have developed on mud flats behind sandy beaches on the three headlands. Sand and coral rubble are dispersed across the reef flats, covering more than 21% of the area. Seagrass and algae are prolific, covering about 12% of the reef flat. Coral communities are abundant on the forereef slope, covering nearly 24% of the reef.

Kimberley Marine Research Program | Project 1.3.1 49

Reef Growth and Maintenance

Figure 28. (a) Satellite image of Cape Londonderry—the classification map was verified using very high-resolution (<1 m spatial resolution) aerial photography; (b) geomorphic zones map; (c) habitats/substrate classification map of Cape Londonderry Reefs.

Molema Island

Molema Island is a high island situated in Talbot Bay on the northern Yampi Peninsula, adjacent to the mainland shore (Figure 24g). The island is connected on the northeastern side to a peninsula by a reef complex that extends approximately 3 km, and on the southwestern side to an island by another reef complex that extends about 2 km; these reefs are known as ‘Turtle Reef’ (Figure 29). Both reefs are protected from high wave energy as they are surrounded by a number of islands. The reef platforms are elevated and are exposed above sea level by about 7 m during extreme low spring tides (Wilson et al., 2011). These types of fringing reefs are classified as high intertidal and inter-island reefs. Mangroves have developed on muddy flats behind sandy beaches on the islands and the peninsula. Sand cays cover about 10% of the reef flats’ area and sand fans and rubble occurs on the reef edges. The reef flats are mainly (> 29%) covered by reef pavement with algal turf, while seagrass and algae cover about 12% of the reef flats, particularly on the northern reef. The reef edges are terraced with a veneer of crustose coralline algae with banks of rhodoliths in a landward direction. There are three wide mud flats: two on the northern and southern sides of the eastern part of Molema Island, and one on its western side. These mud flats are relatively large, occupying about 19% of the reef area. They are on a lower level than the reef flats, but are still exposed during tidal cycles. The coral communities show vigorous growth, covering nearly 18% of the reef flat margins and the forereef slope.

50 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 29. (a) Satellite image of Molema Island (Turtle Reef) with ground-truth points highlighted; (b) geomorphic zones map; (c) habitats/substrate classification map of Turtle Reef.

Bathurst and Irvine Islands

Bathurst and Irvine Islands are high islands situated in the Buccaneer Archipelago about 14 km off the Kimberley mainland shore (Figure 24h). The two islands, with 13 other small adjoining islands, are connected by fringing reefs (Figure 30).

Kimberley Marine Research Program | Project 1.3.1 51

Reef Growth and Maintenance

Figure 30. (a) Habitats/substrate classification map of the reefs between Bathurst and Irvine Islands; (b) satellite image of Bathurst and Irvine Islands with ground-truth points highlighted; (c) geomorphic zones map.

Reefs on the northwest side of Bathurst Island and on the west side of Irvine Island are headland reefs; they are exposed to the swell on the seaward shores and are mainly dominated by domal faviids (Wilson, 2013). The reefs on the northeast side of Bathurst Island and the south and southeast sides of Irvine Island can be classified as bayhead fringing reefs; they are protected from high-energy swells and Acropora dominates the coral community. The reefs on both sides appear to be similar to other fringing reefs on many islands in this region, in terms of geomorphic characteristics and surficial facies. However, an extensive reef platform between Bathurst and Irvine Islands appears slightly different from the other reefs, and this type of fringing reef is classified as an inter-island reef. Wilson (2013) suggested that this reef could have been formed through a process of coalescence between the fringing reefs of both islands. The reef flat is entirely exposed during low tide. It is characterised by a relatively elevated centre, with extensive coverage (> 23%) of the reef pavement with algal turf. Microatolls of between 30 and 100 cm in diameter are sparse on the reef flat. Mangroves have developed on the sandy beaches of both islands, particularly on the southern island. Carbonate sand banks are trapped behind edges made by encrusting coralline algae, mainly on the western side of the reef, with minor accumulation of rhodoliths. Many live coral communities, such as domal and branching Porites corals, are present in shallow lenticular pools that are scattered on the reef flat. There are also various living coral communities in and around three deep pools. Corals are prolific on the forereef slopes on both sides of the reef,

52 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

covering nearly 22% of the reef area.

Tallon Island

Tallon Island is a small, high island that has a curved (talon-like) shape. It is situated about 7 km to the east of One Arm Point, surrounded by numerous islands in a sheltered area (Figure 24f). There are two fringing reefs along the eastern and western sides of the island (Figure 31). Although both reefs are terraced and have some similarities in terms of habitats and substrates, their morphology is different. The reef on the eastern side of the island supports a dense growth of mangroves on its sandy beach and the reef flat is about 1 km wider than the flat of the western reef, and considerably higher. The eastern reef is protected from winds and swells and is considered a high intertidal and bayhead reef. The western reef is classified as an intertidal and headland reef. Seagrass and algae cover more than 30% of both reef flats. More than 14% of the reef flat is covered by coarse carbonate sand and coral rubble, some of which is trapped by terraces of crustose coralline algae along the edges, particularly on the eastern reef. A minor assemblage of rhodoliths is also present in this area. A large area of the reef flat (> 30%) is covered by reef pavement with algal turf. There are numerous shallow pools on both reef flats, which are filled with coral communities (e.g. domal and branching Porites and soft corals). Coral communities cover 13% of the reef, mostly on the forereef slope of the western reef.

Figure 31. (a) Satellite image of Tallon Island, with ground-truth points highlighted; (b) geomorphic zones map; (c) habitats/substrate classification map of Tallon Island reefs. Planar reefs

Kimberley Marine Research Program | Project 1.3.1 53

Reef Growth and Maintenance

Montgomery Reef

Montgomery Reef is located in Collier Bay about 25 km off the Kimberley mainland shore (Figure 24d). It is a large, terraced reef complex of about 380 km2 with a central Proterozoic island. The reef has an irregular oval shape with an elongated peninsula, or breakwater, extending from the western side of the reef edge towards the north (Figure 32). A small reef platform on the eastern side, which consists of five high islands, is called High Cliffy Reef. The small reef is separated from Montgomery Reef by a channel that is approximately 10 m deep. Montgomery Reef is a high reef platform; it emerges a few metres above the mean low water spring (MLWS) tide level during tidal cycles. Despite its unique structure (Wilson, 2013) it is considered a high intertidal and planar reef for classification purposes. At the centre of the reef flat are three low, vegetated islands that contain mud flats and mangroves surrounded by sand sheets. There are also four low, vegetated islands on the eastern edge of the reef flat. Sand covers about 12% of the reef flat, particularly near the reef centre, the southern breakwater, and the reef terraces. The reef flat is covered by a large area (> 42%) of reef pavement with algal turf. Coral rubble with seagrass and algae cover about 8.5% and 8% of the reef flat, respectively. The reef crest is dominated by crustose coralline algae with large areas of rhodolith banks occurring on the reef terraces. Coral communities are widespread and represent about 20% of the reef area, occurring in many geomorphic zones. Corals can be found in and near deep lagoons and channels towards the north-western centre of the reef flat, and fill the many shallow lenticular pools on the reef flat. They are also prolific on the forereef slopes. The terraced, elevated reef flat, with shallow pools, rhodolith banks, and crustose coralline algae near the reef crest, are all characteristics of a high intertidal reef (see Wilson, 2013).

Figure 32. (a) Habitats/substrate classification map of Montgomery Reef; (b) satellite image of Montgomery Reef with ground-truth points highlighted; (c) geomorphic zones map.

Adele Reef

Adele Reef is located near the edge of the inner ramp of the north-western shelf, about 90 km northwest of the Kimberley shore (Figure 24e). The reef covers a total area of 168 km2, is isolated and is nearly elliptical in shape, with its long axis oriented north-west–south-east. A wide, relatively deep channel on its north-western side is known as ‘Fraser Inlet’. Adele Reef has a low, vegetated island at its centre, surrounded by a large sand cay covering about 28% of the reef surface, which is submerged during high tide. The reef is considered intertidal

54 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

and planar. The reef flat is exposed during low tide and is mainly covered by coarse sand and coral rubble (> 24%) and minor algal turf (7%). Macroalgae (mostly Sargassum) occur near the edge of the reef flat as well as on the forereef slope. The reef has a terraced edge dominated by crustose coralline algae, with a wide area of rhodolith banks lying behind them. Coral communities are abundant on the reef, covering a considerable area (>23%) of the forereef slope. There are also numerous coral communities in shallow pools on the reef flat (Figure 33). In contrast to the inshore reefs Adele Reef is in a wave-exposed, open-ocean setting. It lacks a shallow Proterozoic basement; instead, it overlies previous generations of Pleistocene reefs and older limestone foundations, into which deep (> 100 m) lowstand channels are cut adjacent to the platform (Collins et al., 2015).

Figure 33. (a) Habitats/substrate classification map of Adele Reef; (b) satellite image of Adele Reef with ground-truth points highlighted; (c) geomorphic zones map.

Long Reef

Long Reef is situated to the north of the Bonaparte Archipelago about 27 km off the Kimberley coast (Figure 24b). The reef covers a total area of 195 km2 and is elongated northeast southwest. As shown in Figure 34, a vast area of the reef platform (about 40%) is covered by an intertidal sheet of carbonate sand fans and coral rubble, which is submerged during high tide.

Kimberley Marine Research Program | Project 1.3.1 55

Reef Growth and Maintenance

Figure 34. (a) Habitats/substrate classification map of Long Reef; (b) satellite image of Long Reef, with ground-truth points highlighted; (c) geomorphic zones map. There is no central vegetated island on the reef; however, a small sand cay is located on the north side of the reef, which remains exposed during high tide. The reef has a number of lagoons distributed on its flat, which vary in depth and width and are mostly filled with sand. Long Reef is considered an intertidal and planar reef. Algal turf on a reef pavement covers a considerable area (> 31%) of the reef flat. There are noticeable differences between the windward and leeward sides of the reef. The western (windward) side of the reef is exposed to high energy swells and has a terraced edge that is dominated by crustose coralline algae. The forereef slope on the windward side is narrow and steeply sloping; while the forereef slope of the leeward side is wider and more gently sloped. Macroalgae (Sargassum) cover about 14% of the reef area and is prolific in deeper areas of the reef flat as well as on the forereef slopes. Coral communities are abundant on the reef, covering more than 13% of the reef area, mainly on the forereef slopes, in and around the lagoons, and in shallow pools on the reef flat (Figure 34).

56 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

High intertidal reefs High intertidal reefs are a unique feature of macrotidal reefs (Wilson, 2013; Collins et al., 2015) that are characterised by a high, flat-topped reef surface that may be several metres above mean low water neap (MLWN) tide and are commonly exposed during the tidal cycle (Richards and O’Leary, 2015). The surfaces have distinctive lithified algal terraces and coralline algae (rhodoliths) banks (Figure 35), shallow coral-filled pools on reef flats, and Porites microatolls, which are often prolific. This type includes both planar reefs (e.g. Montgomery Reef) and fringing reefs such as Turtle Reef, Bathurst-Irvine Reef, and East Tallon Reef.

Figure 35. Common characteristics of high intertidal reefs: (a) An elevated reef flat of Montgomery Reef during low tide; (b) an exposed reef flat on East Bathurst Island with many shallow lenticular pools; (c) lithified algal terraces on Montgomery Reef; (d) a rhodolith bank on East Tallon Island. (Photos courtesy of Kimberley Media, Mick O’Leary and Tubagus Solihuddin).

Similarities and differences between reefs in the study area A total of seven substrates and habitats were identified on the major reef platforms in the study area. The distribution and coverage area of these substrates and habitats varied from one reef type to another. However, common factors found for all reefs were that they are characterised by prolific reef growth and have considerable areas covered by sand, reef pavement with algal turf, as well as seagrass and algae (see Table 13).

Kimberley Marine Research Program | Project 1.3.1 57

Reef Growth and Maintenance

) 2 (km 5.24 168.0 30.42 8.2 18.4 194.5 380.8 99.6 Total area Total 905.1 ubstrate and habitat habitat and ubstrate

0.7 (13.4%) 39.4 (23.5%) 6.7 (22.0%) 4.3 (52.4%) 4.0 (21.7%) 26.5 (13.6%) 74.6 (19.6%) 23.8 (23.9%) CC 180.0

0.2 (3.8%) 15.1 (9.0%) 0.9 (3.0%) 0 (0%) 0.6 (3.3%) 3.7 (1.9%) 24.3 (6.4%) 0.0 (0%) CCA 44.8

3 1.6 (30.5%) 11.5 (6.8%) 11.1 (36.5%) 1 (12.2%) 4. (23.4%) 61.5 (31.6%) 160.9 (42.3%) 35.8 (35.9%) RPAT 287.7

41.4 (24.6%) 0.2 (3.8%) 0 (0%) 0 (0%) 0.2 (1.1%) 11.7 (6.0%) 32.4 (8.5%) CR 4.0 (4.0%) 89.9 bble; ‘RPAT’ = reef pavement with algal turf; ‘CCA’ = crustose 1.6 (30.5%) 12.8 (7.6%) 4.6 (15.1%) 1.1 (13.8%) 5.1 (27.7%) 26.9 (13.8%) SG & A 12.3 (12.4%) 30.1 (7.9%) 94.5

0 (0%) 0 (0%) 0.02 (0.1%) 0 (0%) 0.4 (2.2%) 0 (0%) 0.5 Lag Lag 0.1 (0.1%) 0 (0%)

0.54 (10.3%) 47.8 (28.5%) 3.8 (12.5%) 1.8 (22.0%) 3.3 (17.9%) 64.2 (33.0%) 184.6 San 17.2 (17.3%) 46 (12.1%) 0.4 (7.6%) 0 (0%) 3.3 (10.8%) 0 (0%) 0.5 (2.7%) 0 (0%) 23.1 Man 6.4 (6.4%) 12.5 (3.3%) Substrates and Habitats and Substrates

) 2 High intertidal Low intertidal High intertidal Low intertidal High intertidal Low intertidal Low intertidal High intertidal Tide

Total area (km Total Fringing Planar Fringing Fringing Fringing Planar Fringing Planar Reef type

55 ′

16° 25 ′ 16° 15° 30 ′ 15° 16° 15 ′ 16° 14° 25 ′ 14° 16° 03 ′ 16° 13° 55 ′ 13° 15° 13° 43 ′ 13° Lat° (S)

Coverage area of substrates and habitats of selected reefs. .

Tallon Island Adele Reef. Molema Island Island Molema Maret Island. Bathurst & IrvineBathurst Islands Long Reef Montgomery Reef Cape Londonderry Reef Name Table s of Description types. tide and reef with (bottom) south to (top) north from ordered are Note: Latitudes Percentages brackets. in are of the coverage ru coral ‘CR’ = and algae; = & A’ seagrass lagoon; ‘SG = ‘lag’ = ‘San’ sand; = ‘Man’ mangroves; abbreviations: coralline algae; ‘CC’ coral = communities.

58 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Other substrates and habitats were found to be present on particular reef types. For example, mangroves were mainly found on fringing reefs, both attached to the mainland or islands. Molema Island (Turtle Reef), for instance, has > 10% mangrove coverage. Montgomery Reef is an exceptional case of a planar reef because it contains vegetated high islands, which are characterised by the presence of mangroves. Crustose coralline algae are usually found on the edge of planar reefs and high intertidal reefs, while they are either absent or rarely found on the other reef types. For example, Adele Reef has crustose coralline algae coverage of about 9% near its reef crest, followed by Montgomery Reef which has over 6% coverage.

5.4 Discussion and Conclusions The scarcity of reef research in the Kimberley Bioregion can mainly be attributed to its remoteness. Remote sensing can provide useful information for mapping reefs when used in conjunction with other relevant data sources such as aerial photography, bathymetric charts, and supporting ground truth information. Although reef mapping using Landsat images in such extreme water conditions is possible, it remains challenging because it is limited by light attenuation and spectral misclassifications. The results of this study indicate that significant improvements in accuracy can be achieved for mapping intra-reef geomorphic zones and associated habitats and substrates of many reefs in the Kimberley Bioregion. In this study, examples of Kimberley reefs were chosen to illustrate the properties of fringing and planar reefs. Special reference was also made to the high intertidal reefs (fringing and planar) of the southern Kimberley Bioregion. The methodology developed for this study was used to document the geomorphology of eight reef platforms in different geographical locations, with sizes that could be mapped using Landsat images, and sufficient available ground-truth information. The reef platforms were also mapped with regard to their biosedimentary substrates and sedimentary facies, including percent coverage for each substrate type. This study has shown that by applying image analysis techniques and using unsupervised classification, Landsat images can be effectively used to map the detailed geomorphology of coral reefs with an overall accuracy of between 64% and 77% using a confusion matrix. A further advantage of using unsupervised classification is that it can be replicated on other reefs in the region. Moreover, the resultant classification allows spatial analysis of the distribution and geometry of intra-reef components can be used to derive classification schemes that are consistent across other reefs in the Kimberley Bioregion. The majority of fringing reefs observed in the study area have grown around high islands or submerged rocky outcrops. Reefs attached to islands are often wider with more reef growth on the windward side, which is exposed to the northwest and has a more steeply sloping reef front. Reefs on the leeward side are more gently sloping and the reef flats tend to be narrowest where reefs are sheltered from high wave energy. The difference in width between the windward and leeward sides of the reefs appears to be consistent around most islands, particularly in the northern part of the Bonaparte Archipelago. This indicates either that exposed reefs grow faster than sheltered reefs, or start growing earlier (Wilson, 2013). Fringing reefs with a defined reef crest are best developed in the Buccaneer Archipelago such as Sunday Island, and Tallon Island; and are also present northward of the archipelago (Wilson, 2013). The area of fringing reefs is usually smaller than the area of the high island to which they are attached. However, the area of some reefs, such as Tallon Island, exceeds that of their island many times. Planar reefs are isolated features that are characterised by their large area and are mostly found offshore. According to Hopley’s classification, these reefs are considered to be in their senile stage as most of their lagoons are almost infilled and the reef flat extends across the entire reef platform. Their reef flats are often exposed at low tides. The majority of these reefs are covered by a massive sand sheet on the reef flat and are surrounded by an intertidal reef platform or rampart. However, some other senile reefs have a central low, vegetated island such as Adele Reef, or an unvegetated sand cay such as Long Reef. High intertidal reefs are a remarkable feature of the Kimberley Bioregion. Their reef flat is easily distinguished

Kimberley Marine Research Program | Project 1.3.1 59

Reef Growth and Maintenance

as the elevation is at the position of MSL rather than MLWS, which has allowed these reefs to be mapped in greater detail. High intertidal reefs are characterised by lithified algal terraces and coralline algae with rhodolith accumulations, in addition to Porites microatolls (Wilson, 2013). This type of reef includes both planar reefs (e.g. Montgomery Reef) and fringing reefs (e.g. Tallon Island) and is found particularly in the Buccaneer Archipelago, where the spring tide range exceeds nine metres (e.g. Molema Island and Bathurst-Irvine Islands). The reef platform between islands seems to be formed by a coalescence of fringing reefs that are attached to these islands. A recent seismic study conducted by Collins et al., (2015) suggests that these reefs have a long growth history, including Holocene and Late Pleistocene (last interglacial) reef-building events. The coral reefs of the Kimberley Bioregion occur in extreme environments, including areas influenced by macrotides, highly turbid water, and complex coastal morphology, which can limit accessibility and hinder the ability to map these coastal features. Despite these challenges, many reefs have been mapped successfully using remote sensing. As a result, this study produced a detailed analysis of reef geomorphology and key substrates and habitats, providing significant information about the distribution and extent of reef landforms, insights into reef growth and morphology patterns, and a description of the classification and distribution of reefs by type. This study recognised and mapped distinctive geomorphic reef features using remote sensing for the first time. One of the most common limitations of using remotely sensed images in mapping reefs is that it is not easy to discriminate between live coral and other habitats that consist of chlorophyll organisms such as macroalgae and seagrass due to similar light spectra reflectance characteristics (Zainal et al., 1993; Kaczmarek et al., 2010) which often result in misclassifications. However, Groom et al. (1996); Mumby et al. (1998); and Benfield et al. (2007) suggested using a manual editing process (contextual editing) when the habitat distribution is known. This process was applied in the current study to improve the accuracy of the maps. Another limitation is that there were time-lags between the ground-truth data collection and satellite image acquisition. Although the chronological gap in this study was insignificant, time convergence is preferable when changes on the reef are detected (Goodman et al., 2013). In conclusion, the results of this study provide a firm scientific foundation for biodiversity assessment and reef structure comparisons. Moreover, they facilitate monitoring and detection of changes in these vital natural resources for conservation and management purposes and pave the way for future studies to examine major threats to reef ecosystems such as the effects of climate change, tropical cyclones, and economic resource extraction.

5.5 Acknowledgements This study is part of the Kimberley Reef Geomorphology Project 1.3.1, which has been funded by the Western Australian State Government through the Western Australian Marine Science Institution (WAMSI). We are grateful to the Traditional Owners of the Kimberley (the Bardi Jawi, Mayala, and Dambimangari people) for their assistance, advice, and consent to access their traditional lands. We also thank the Western Australian Museum for providing ground truth data. We also wish to thank the following organisations: the Kimberley Marine Research Station; Kimberley Media; the Department of Parks and Wildlife (DPaW) for providing orthophotographs and satellite images; Geoscience Australia (GA) for providing DEMs and GIS data; The United States Geological Survey (USGS) for providing Landsat images; and special thanks to the valued members of the research team at Curtin University: Tubagus Solihuddin and Giada Bufarale. Finally, it must be noted that this research was completed in an area where the Traditional Owners have a rich cultural history of climate, land and environment based on thousands of years of habitation. It is important to consider that broad understanding alongside the modern science presented here.

60 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

5.6 References

Andréfouët, S., Muller-Karger, F., Robinson, J., Kranenburg, Kawakubo, F. S., Morato, R. G., Nader, R. S. and Luchiari, A. C.,Torres-Pulliza, D., Spraggins, S. and Murch, B. (2011). Mapping changes in coastline geomorphic (2006). Global assessment of modern coral reef extent features using Landsat TM and ETM+ imagery: and diversity for regional science and management Examples in south-eastern Brazil. International applications: A view from space. In Proceedings from Journal of Remote Sensing, 32, 2547–2562. the 10th International Coral Reef Symposium, 1732– Leon, J. and Woodroffe, C. D. (2011). Improving the 1745. synoptic mapping of coral reef geomorphology using Andréfouët, S. and Payri, C. (2000). Scaling-up carbon and object-based image analysis. International Journal of carbonate metabolism of coral reefs using in-situ data Geographical Information Science, 25, 949–969. and remote sensing. Coral Reefs, 19, 259–269. Luczkovich, J. J., Wagner, T. W., Michalek, J. L. and Stoffle, Benfield, S. L., Guzman, H. M., Mair J. M., and Young J.A.T. R. W. (1993). Discrimination of coral reefs, seagrass (2007). Mapping the distribution of coral reefs and meadows, and sand bottom types from space: A associated sublittoral habitats in Pacific Panama: a Dominican Republic case study. Photogrammetric comparison of optical satellite sensors and Engineering and Remote Sensing, 59, 385–389. classification methodologies, International Journal of Madden, R. H., Wilson, M. E. and O’Shea, M. (2013). Remote Sensing, 28(22), 5047-5070 Modern fringing reef carbonates from equatorial SE Bina, R. T. (1982). Application of Landsat data to coral reef Asia: An integrated environmental, sediment and management in the Philippines. In Proceedings of the satellite characterisation study. Marine Geology, 344, Great Barrier Reef Remote Sensing Workshop. 163–185. Townsville, QLD, 1–39. Masini, R. J., Sim, C. B, and Simpson, C. J. (2009). A synthesis Bouvet, G., Ferraris, J. and Andréfouët, S. (2003). Evaluation of scientific knowledge to support conservation of large-scale unsupervised classification of New management in the Kimberley region of Western Caledonia reef ecosystems using Landsat 7 ETM+ Australia. Western Australia: Department of imagery. Oceanologica Acta, 26, 281–290. Environment and Conservation. Campbell, J. B. (2002). Introduction to remote sensing. CRC Mumby, P., Green, E., Clark, C. and Edwards, A. (1998). Press. Digital analysis of multispectral airborne imagery of coral reefs. Coral Reefs, 17(1), 59–69. Capolsini, P., Andréfouët, S., Rion, C. and Payri, P. (2003). A comparison of Landsat ETM+, SPOT HRV, IKONOS, Mumby, P., Green, E., Edwards, A. and Clark, C. (1997). ASTER, and airborne MASTER data for coral reef. Coral reef habitat mapping: How much detail can Canadian Journal of Remote Sensing, 29, 187–200. remote sensing provide? Marine Biology, 130(2), 193–202. Chin, A., Sweatman, H., Forbes, S., Perks, H., Walker, R., Jones, G., Edgar, G. (2008). Status of the coral reefs in Mumby, P. J. and Harborne, A. R. (1999). Development of a Australia and Papua New Guinea, 159–176. systematic classification scheme of marine habitats to facilitate regional management and mapping of Collins, L. B. (2011). Geological setting, marine Caribbean coral reefs. Biological Conservation, 88(2), geomorphology, sediments and oceanic shoals 155–163. growth history of the Kimberley region. Journal of the Royal Society of Western Australia, 94(2), 89–105. Palandro, D., Andréfouët, S., Dustan, P. and Muller-Karger, F. (2003). Change detection in coral reef communities Collins, L. B., O’Leary, M., Stevens, A., Bufarale, G., Kordi, M. using the Ikonos sensor and historic aerial and Solihuddin, T. (2015). Geomorphic patterns, photographs. International Journal of Remote internal architecture and reef growth in a macrotidal, Sensing, 24, 873–878. high-turbidity setting of coral reefs from the Kimberley bioregion. Australian Journal of Maritime Phinn, S., Roelfsema, C., Dekker, A., Brando, V. and Anstee, and Ocean Affairs, 7(1), 12–22. J. (2008). Mapping seagrass species, cover and biomass in shallow waters: An assessment of satellite Dahdouh-Guebas, F. (2002). The use of remote sensing and multi-spectral and airborne hyper-spectral imaging GIS in the sustainable management of tropical coastal systems in Moreton Bay (Australia). Remote Sensing ecosystems. Environment, Development and of Environment, 112, 3413–3425. Sustainability, 4(2), 93–112. Purcell, S. (2002). Intertidal reefs under extreme tidal flux Department of Sustainability, Environment, Water, in Buccaneer Archipelago, Western Australia. Coral Population and Communities, DSEWPaC (2012). Reefs, 21, 191–192. Retrieved from http://www.environment.gov.au/coasts/mbp/north- Richards, Z., Bryce, M. and Bryce, C. (2013). New records of west/index.html atypical coral reef habitat in the Kimberley, Australia. Journal of Marine Biology. Fairbridge, R. (1950) Recent and Pleistocene coral reefs of Australia. The Journal of Geology, 58, 330–401.

Kimberley Marine Research Program | Project 1.3.1 61

Reef Growth and Maintenance

Fairbridge, R. (1967). Coral reefs of the Australian region. In Richards, Z. T., and O’Leary, M. J. (2015). The coralline algal J. N. Jenings and J. A. Mabbutt (Eds.), Landform cascades of Tallon Island (Jalan) fringing reef, NW studies from Australia and New Guinea (pp. 386–417). Australia. Coral Reefs, 34(2), 595-595. Canberra, ACT: Australian National University Press. Smith, V. E., Rogers, R. H. and Reed, L. E. (1975). Automated Green, E.P., Mumby, P.J., Edwards, A.J. and Clark, C.D. mapping and inventory of Great Barrier Reef zonation (2000) Remote Sensing Handbook for Tropical Coastal with Landsat. Oceans, 7, 775–780. Management (Paris: UNESCO). Teichert, C. and Fairbridge, R. W. (1948). Some coral reefs Goodman, J. A., Purkis, S. J., and Phinn, S. R. (2013). Coral of the Sahul Shelf. Geographical Review, 38, 222–249. reef remote sensing: A guide for mapping, monitoring Thackway, R., Cresswell, I. and Marine, I. (1998). Interim and management: Springer Science and Business marine and coastal regionalisation for Australia: An Media. pp. 66 ecosystem-based classification for marine and coastal Groom, G. B., Fuller, R. M. and Jones, A. R. (1996). environments: Environment Australia. Department of Contextual correction: Techniques for improving land the Environment. cover mapping from remotely sensed images. Veron, J. E. (2008). Mass extinctions and ocean International Journal of Remote Sensing, 17(1), 69– acidification: biological constraints on geological 89. dilemmas. Coral Reefs, 27(3), 459–472. Hochberg, E. J. and Atkinson, M. J. (2003). Capabilities of Western Australian Museum Woodside Collection Project remote sensors to classify coral, algae, and sand as (Kimberley) 2008–2012. pure and mixed spectra. Remote Sensing of Environment, 85(2), 174–189. White, W., Harborne, A., Sotheran, I., Walton, R. and Foster-Smith, R. (2003). Using an acoustic ground Hopley, D. (1982). The geomorphology of the Great Barrier discrimination system to map coral reef benthic Reef: Quaternary development of coral reefs. New classes. International Journal of Remote Sensing, 24, York, NY: Wiley. 2641–2660. Hopley, D., Smithers, S. G. and Parnell, K. E. (2007). The Wilson, B. and Blake, S. (2011). Notes on the origins and geomorphology of the Great Barrier Reef: biogeomorphology of Montgomery Reef, Kimberley, Development, diversity and change. Cambridge Western Australia. Journal of the Royal Society of University Press. Western Australia, 94, 107–119. Johnson, R. (2000). GIS technology for disasters and Wilson, B., Blake, S., Ryan, D. and Hacker, J. (2011). emergency management. An ESRI white paper. Reconnaissance of species-rich coral reefs in a Jupp, D. L., Mayo, K. K. and Kuchler, D. A. (1985) Remote muddy, macro-tidal, enclosed embayment, Talbot sensing for planning and managing the Great Barrier Bay, Kimberley, Western Australia. Journal of the Reef of Australia. Photogrammetria, 41, 21–42. Royal Society of Western Australia, 94, 251–265. Kaczmarek, S. E., Hicks, M. K., Fullmer, S. M., Steffen, K. L. Wilson, B. R. (2013). The biogeography of the Australian and Bachtel, S. L. (2010). Mapping facies distributions North West Shelf. New York, NY: Elsevier. on modern carbonate platforms through integration Wolanski, E. and Spagnol, S. (2003). Dynamics of the of multispectral Landsat data, statistics-based turbidity maximum in King Sound, tropical Western unsupervised classifications, and surface sediment Australia. Estuarine, Coastal and Shelf Science, 56, data. AAPG Bulletin, 94(10), 1581–1606. 877–890. Kakuta, S., Hiramatsu, T., Mitani, T., Numata, Y., Yamano, H. Zainal, A. J. M., Dalby, D. H. and Robinson, I. S. (1993). and Aramaki, M. (2010). Satellite-based mapping of Monitoring marine ecological changes on the east coral reefs in East Asia, Micronesia and Melanesia coast of Bahrain with Landsat TM. Photogrammetric Regions. International Archives of the Engineering and Remote Sensing (United States), Photogrammetry, Remote Sensing and Spatial 59(3) Information Science, XXXVIII, 534–537.

62 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

6 ReefKIM: an integrated geodatabase for sustainable management of the Kimberley Reefs, NW Australia

Adapted from Kordi M.N., Collins, L.B., O’Leary, M.J., & Stevens A.M., in review. ReefKIM: an integrated geodatabase for sustainable management of the Kimberley Reefs, NW Australia. Ocean and Coastal Management

6.1 Introduction The coral reefs of Australia’s North West Kimberley Bioregion have been recently recognised as being of international significance (Wilkinson 2008 and Chin et al. 2008). The Kimberley reefs are particularly unique in that they inhabit an environment with the largest tidal range in the southern hemisphere, up to 11 m, and appear to endure high levels of turbidity and frequent cyclones. These extreme environments have resulted in coral reefs with unique geomorphological attributes which merit further study (Wilson, 2013). The extreme remoteness of the Kimberley Bioregion has, for an inshore continental reef system, resulted in very low levels of direct anthropogenic impacts (Collins, 2011). While this remoteness has been an advantage in maintaining reef health, the cost of fieldwork and data acquisition can be prohibitively expensive, which has limited the ability of marine scientists and managers to access the region. The result is that compared to regions like the Great Barrier Reef, only a small pool of data has been collected from the Kimberley by a variety of sources including; universities, government agencies, tourism operators, commercial and recreational fishermen and most importantly, the traditional owners that inhabit the region. The result is a small amount of data spread across a large number of stakeholders making information on these reefs either rarely available or inaccessible. There is a critical need to know what kind of data has been collected from the Kimberley and if it is available to share or exchange. In the first instance greater accessibility of data should prevent duplication of previous work, reduce cost, increasing data quality (Chesnaux et al. 2011) and allow for multiuse datasets. In order to address these issues and more we have created a high-capacity, multi-source, easy-to-access and cost-effective geodatabase for Kimberley researchers, ReefKIM. It is envisaged that this database will be the go to resource for Kimberley marine researchers providing them with critical scientific issues and knowledge gaps that need to be addressed. Reef geodatabases A Geographic Information System (GIS)-based database, or geodatabase, is a computer-based system that can handle a variety of information, including both locational and attribute data on a particular feature. It not only displays and produces maps but can also record and analyse descriptive characteristics of map features. Geodatabases have been developed for many coral reef regions around the world, particularly in the areas of coral reef management and conservation with the efforts of the Great Barrier Reef Marine Park Authority (GBRMPA) in the early 1990s (Hartcher and Shearin 1996) being an early example. Since then, usage of GIS for data management has increased rapidly and expanded worldwide. The construction of the Global Coral Reef Database, also known as ReefBase or ReefGIS, was initiated in 1993 by the International Centre for Living Aquatic Resources Management (ICLARM), and is a good example of a global-scale GIS database (McManus 1994; McManus and Ablan 1997). Another global-scale assessment of total reef area was undertaken using GIS-based technology by Spalding and Grenfell (1997). Similarly, map-based indicator Reefs at Risk was developed by the World Resources Institute (WRI) and uses GIS to assess potential threats to coral reef ecosystems around the world (Bryant et al. 1998). Dahdouh-Guebas (2002) applied an environmental GIS database to the sustainable development and management of tropical coastal ecosystems by collecting and integrating data from different disciplines. More recently, a regional-scale GIS database named ReefBahia was developed to assist in managing and conserving the coral reefs of Bahia in Brazil (Carvalho and de Kikuchi 2013), and another was created to improve management of the Coral Triangle (Beare 2014; Cros et al. 2014). Rapid developments in information and communication technology, particularly over the last decade, have led to an exponential increase in computational and networking efficiency, which facilitates the aggregation of vast

Kimberley Marine Research Program | Project 1.3.1 63

Reef Growth and Maintenance

amounts of data through sophisticated, relatively affordable, and highly accurate devices, such as smart phones, tablets and portable computers (Heipke 2010). Moreover, these devices are often equipped with high-definition digital cameras, built-in Gobal Positioning Systems (GPS), internet connectivity and high-capacity memory storage, which allow for recording of data and capturing images and videos associated with essential metadata, such as location (geo-referenced), date and time. Other optional information can be entered manually or through a software application menu designed for a particular type of data in order to avoid entry errors (Briner et al. 1999). Crowdsourcing data New technologies have changed the way researchers and managers receive information from the field. The crowdsourcing geospatial data approach for information aggregation has been implemented worldwide for a variety of purposes, including creating and sharing geographic information volunteered by individuals through common and freely available platforms such as Wikimapia and OpenStreetMap (Goodchild 2007; Hacklay 2008) and assisting people during crises through programs such as Ushahidi (Okolloh 2009). Furthermore, the crowdsourcing approach has been employed in many scientific endeavours, including Geo-Wiki, a global network of volunteers helping to improve the quality of global land cover maps (Fritz et al. 2009; Comber et al. 2013). In a recent study, Franzoni and Sauermann (2014) thoroughly discussed benefits of scientific research in open collaborative projects using the ‘crowd science’ or ‘citizen science’ approach. Rovere et al. (2012) also conveyed the advantages of crowdsourcing for the creation of databases of sea levels from the mid-Pliocene warm period. According to Lewis et al. (2003), community participation played a significant role in re-zoning the Great Barrier Reef Marine Park Authority (GBRMPA). Volunteer involvement in the monitoring program Reef Watch, coordinated by the Conservation Council of South Australia for the sustainable management of marine ecosystems, has helped increase knowledge about the status of temperate reefs in South Australia (CCSA 2009). Additionally, volunteer divers and snorkelers recorded about 180 marine species in Victoria, Australia, through the monitoring initiative Reef Watch Victoria, developed by the Victoria National Parks Association and Museum Victoria to protect Victoria’s marine environment (VNPA 2014). Another remarkable monitoring program called Eye on the Reef, managed by GBRMPA, in partnership with the Queensland Parks and Wildlife Service, enables Great Barrier Reef visitors to report reef observations though a web map or/and smartphone or tablet application. The data provide Marine Park managers and researchers with up-to-date information on current reef status and other related information (GBRMPA 2014). The main aim of this study was to establish a geodatabase of information on the Kimberley reefs (ReefKIM) in order to provide a regional picture of the reef numbers, types and status. The work included compiling existing spatial and non-spatial data and collecting new data through online GIS sources. Furthermore, the database allows for crowdsourcing of future data to fill in information gaps. ReefKIM has been developed through a partnership with the Western Australian Marine Science Institution (WAMSI) for the Kimberley Marine Research Program (KMRP) and implemented with the sponsorship of the Government of Western Australian. The outcomes of this geodatabase should improve our knowledge and provide decision makers with helpful information for the sustainable management of the vital reef ecosystem.

6.2 Methodology Constructing a useful geodatabase requires a considerable amount of different types of data distributed across both time and space. ArcGIS 10 software developed by the Environmental Systems Research Institute (ESRI) was selected for developing the database due to its high performance and wide recognition. Procedures used in this study are shown in Figure 36.

64 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 36. Methodological scheme of data acquisition, processing, integration and storage for the Kimberley reef geodatabase (ReefKIM) Data sources The information related to the study area was acquired from various sources (Table 14). Most of the datasets used in this study were available in digital formats (i.e. raster and vector); otherwise, they were digitised. The compiled data covered the entire study area with satellite images, maps and charts. The database comprises more than 90 satellite image scenes acquired between 1998 and 2014 from Landsat Thematic Mapper (TM), Enhanced Thematic Mapper Plus (ETM+) and Operational Land Imager (OLI) sensors. ReefKIM also contains 18 geological maps, 12 bathymetric charts and a set of 24 very high resolution orthophotographs covering the entire Kimberley coastline and nearshore islands. Moreover, there are over 300 ground truth points, about 300 km of seismic lines, over 50 core points and more than 400 site images. Other related data were extracted from a wide range of secondary and tertiary sources, such as reports, publications, atlases, books, maps and encyclopaedias. The geodatabase was designed to allow modification of current data and addition of new information and photos, which can be aggregated or crowdsourced through a web-map platform (e.g. Google Earth) using laptops, smartphones or tablets.

Table 14. Datasets used in this study and their sources Data Dataset Source of data* format Satellite images (USGS), orthophotos (Landgate, Western Kimberley coastline Raster Australia and DPaW). Satellite images (USGS), orthophotos (Landgate, Western Islands Australia and DPaW), bathymetric charts (AHO), geological Raster maps (GSWA) Satellite images (USGS), orthophotos (Landgate, Western Australia and DPaW), bathymetric charts (AHO), geological Raster Coral reef maps (GSWA) Points WAM/Woodside Collection Kimberley Project 2008-2011; Brooke 1995, 1997; Wilson 2011, 2013 Seabed geomorphology Geoscience Australia Polygon Sea surface temp. NOAA Polygon Bathymetric contours GA, AHO Polyline

Kimberley Marine Research Program | Project 1.3.1 65

Reef Growth and Maintenance

Sub-bottom profiles Collins et al. 2015 Polyline WAM/Woodside Collection Kimberley Project 2008-2012; Ground-truth Wilson and Blake 2011; Wilson et al. 2011; WAMSI 1.3.1 Reef Points Geomorphology Project 2012-2015 Reef coring Solihuddin et al. 2015 Points Weather and tide BOM Various *Note: The data were sourced from national and international government agencies, including the Department of Parks and Wildlife (DPaW), Geoscience Australia (GA), the Western Australia Marine Science Institute (WAMSI), the Geological Survey of Western Australia (GSWA), Western Australian Museum (WAM), Australian Hydrographical Service (AHO), Australian Bureau of Meteorology (BOM), the United States Geological Survey (USGS) and the National Oceanic and Atmospheric Administration (NOAA). Data processing Initial preparation of the data (pre-processing) is important to ensure accuracy and reliability. The satellite imagery was processed using ENVI 4.3 software prior to data extraction. ArcGIS was then used to integrate various images with other data sources. Maps and charts were geo-referenced and projected according to the Geocentric Datum of Australia (GDA 94) for consistency and homogeneity. All remotely sensed images were calibrated and atmospherically corrected. Moreover, some images were pre-processed to offset problems with band data and to recalculate DN values. Subsequently, a range of map features, including coastline, islands, reefs and areas of shoaling, was precisely digitised, and geomorphological zones and associated habitats and substrates were determined. Data integration Because a single dataset usually does not contain all the necessary information, data integration was employed to fill in the missing information. Integration allows for data verification, detection of changes, and data updating (Gösseln and Sester 2005). It has also been used to combine multiple sources of data for the same object or feature in order to extract information or add value. As a result, a consistent and accurate dataset that is more informative than the original can be produced. The resulting features were stored in vector format so that they could be presented using different colours and symbols for easy differentiation. Each feature was linked to its attributes, such as the feature name, type, area, location, date of survey, source of data and any other related information. All this data was then saved in an attribute table enabling further data analysis to improve understanding of linkages between geological substrate, reef geomorphology and reef classification and distribution. Subsequently, all data were compiled into a data library, and information relevant to this project was extracted, rectified and entered into a database. The selected information was represented as feature classes and raster- based datasets in ArcGIS as usable data layers. This geodatabase also contains the most significant elements and conditions that influence reef growth in the region. Data accessibility A GIS-based database enables an administrator to control access to avoid accidental loss of data and to restrict access to authorised users only. It also supports a wide range of database queries and operations. For example, users can search for information within the database using Structured Query Language (SQL). A copy of reef and island data layers was converted to an open web based format using Keyhole Markup Language (KML), in this case to enable visualisation in Google Earth. Dissemination of data though a web-based platform allows aggregating new information though consistent crowdsourcing. These outsourced data must be rectified before being compiled into the geodatabase.

66 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

6.3 Results A geodatabase of the Kimberley reefs, termed ReefKIM, was constructed. A variety of data sources fostered using data fusion to maximise extraction and interpretation of information. As a result, six feature classes were derived from these datasets and saved consistently in the geodatabase. The feature classes (described in Table 15) include reefs, coastline, islands, geomorphological zones, habitats and substrates and other studies and work. Table 15. Resultant feature classes included in ReefKIM

Feature class Contents Format

Reefs 853 reef features including sizes, shapes and types were recorded. Vector

Coastline A >5,300 km line of the mainland coast was mapped using the mean low Vector water neap (MLWN) level to ensure that mangroves and reef flats were not included.

Islands Over 2,400 island features (i.e. islets, exposed rocks and exposed cays) as Vector defined by Geoscience Australia were recorded.

Geomorphological zones Five reef geomorphologic zones (i.e. land, reef flat, lagoon, reef crest and Vector fore-reef slope) were recorded for 30 reefs.

Habitats and substrates Seven key habitats and substrates (i.e. mangroves, sand, seagrass and algae, Vector coral rubble, reef pavement with algal turf, crustose coralline algae and coral communities) were recorded.

Other studies and work This class encompasses geographic locations of all studies, work and Vector information related to the Kimberley reefs that have been encountered and accessed (e.g. ground truth points, survey, sample collections, coring sites, sub-bottom profiles, images, etc.)

Each resulting feature class can be displayed in ArcGIS either as a solo data layer or in conjunction with other data layers according to the information that needs to be illustrated and/or calculated. For example, three feature classes (reefs, islands and coastline) are displayed simultaneously in Figure 37, showing the extent of the Kimberley coastline and the distribution of reefs and islands on a regional-scale map. 853 reefs and 2,413 islands were mapped along the 5,300 km coastline, for which amount, areas, geographic distribution and distance from coastline can be calculated. The numbered reefs highlighted in orange indicate the locations of 30 reefs that have been mapped in detail.

Kimberley Marine Research Program | Project 1.3.1 67

Reef Growth and Maintenance

Figure 37. Spatial distribution map of the Kimberley reefs and islands compiled in ReefKIM. The arrow on the left-hand side of the map points toward reef number 25, Scott Reef. The 30 reefs are listed alphabetically in Table 16 along with sources of information used to map them in more detail. Landsat images, geological maps and bathymetric charts were used for all these reefs, while ground truth data were available for 90% of the studied reefs. However, sub-bottom profiling and coring data were only available for fewer than 23% of the reefs. Adele Reef and Cockatoo Island had the largest number of data sources. On the other hand, Beagle Reef, Cape Londonderry, Mavis Reef, and King Island had the fewest available data sources.

68 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Table 16. Data sources used to produce geomorphic, substrate and facies maps for targeted reefs (listed in alphabetical order) Data sources* Reef O H G G B P C LS SBP SI P R M T C S O Adele Reef          Albert Reef      Bathurst and Irvine islands         Beagle Reef     Browse Island        Brue Reef       Cape Londonderry     Cassini Island       Champagny Island      Churchill Reef      Cockatoo Island          Colbert Island       Condillac Island      De Freycinet Island      Hedley Island       King Island     Long Reef        Maret Island        Mavis Reef     Molema Island        E. Montalivet Island         W. Montalivet Island         Montgomery Reef         Robroy Reef      Scott Reef        Sunday Island        Tallon Island        White Reef        Wildcat Reef       Woninjaba Island      *Note: Data source acronyms are LS = Landsat images, OP = orthophotos, HR = higher resolution satellite images, GM = geological maps, GT = ground truth, SBP = sub-bottom profiling, BC = bathymetric charts, PS = previous studies, CO = coring and SI = site images. The geodatabase can also present information at the reef scale to display further details on individual reefs. For instance, the geomorphic zone feature class can include information on geomorphology, including shapes, areas and distributions, for specific reefs. Moreover, further information on associated habitats and substrate coverage for these geomorphic zones were stored in the habitats and substrates feature class. Figure 38 illustrates an example of geomorphic zones and habitat and substrate maps for Adele Reef. Geomorphic zones

Kimberley Marine Research Program | Project 1.3.1 69

Reef Growth and Maintenance

are presented in different colours to delineate the boundaries of each zone, while related information, such as the zone’s name, area, percentage of coverage and other values, is recorded in an attribute table and linked to a certain map feature. For example, in Figure 38a, a land zone is highlighted on the map, and information related to this selected feature is shown in the bottom attribute table, revealing that land area on Adele Reef is 0.28 km2, covering approximately 0.2% of the total reef. Similarly, Figure 38b shows the distribution of habitats and substrates on Adele Reef with distinguishable boundaries for each feature. The highlighted feature represents the coarse sand and coral rubble class. The statistical information of this selected feature is presented in the top attribute table, revealing that the coarse sand and coral rubble class covers over 41 km2, which represents more than 24% of the total reef coverage.

Figure 38. Two maps of Adele reef:(a) intra-reef geomorphic zones and (b) distribution of habitats and substrates on the reef platform. Each map feature is connected to an attribute table in the geodatabase. Information on previous studies and work on the Kimberley reefs can also be presented on a reef-scale map. Figure 39, for example, shows the locations of some available information sourced from a variety of studies and work on Adele Reef. Each source of information is represented by a distinct symbol on the map. On Adele Reef, for instance, seven sub-bottom profiles have been surveyed, and nine habitat sites have been studied. Additionally, 13 monitoring stations have been established, and eight sites have been cored. All the information is stored in the other studies and work feature class with links to its origin in the attribute table.

70 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 39. Satellite image of Adele Reef showing locations of previous studies and work with links to their origins in attribute tables Reef data was converted to KML for visualisation in Google Earth. Figure 40 depicts reef distribution in vivid orange across the entire Kimberley coast in Google Earth. It also demonstrates that some reefs can have map pins and name labels as visualisation tools. The layers panel on the left-hand side of the map allows users to choose (by turning on or off) the information to be displayed over the map in the map view.

Kimberley Marine Research Program | Project 1.3.1 71

Reef Growth and Maintenance

Figure 40. Screenshot of an oblique view of the Kimberley Bioregion. Reefs are shown in vivid orange in Google Earth. Moreover, additional crowdsourced information was examined in the map. Figure 41 shows an example of photos taken using smartphone cameras from 10 different locations and added into the map. After the photos were uploaded, their locations were marked on the map by yellow map pins, which can be clicked for further information. For instance, the pop-up window shown in Figure 41 points to one of these yellow map pins, located on a reef flat between Bathurst and Irvine islands. The window presents the reef picture and essential information, such as the name of that site, date, time and coordinates.

Figure 41. Screenshot showing photo locations (as yellow map pins). A pop-up window displays a reef photo with relevant information.

72 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

6.4 Discussion Prior to the gazetting of the KMRP, there had only been a handful of studies that investigated the coral reefs and marine environments in the Kimberley bioregion. While additional information has been collected by scientists and managers over the course of decades, much of this information remained unpublished, or housed as internal reports making access to information difficult. ReefKIM is the first attempt to collate and integrate this information into one comprehensive geodatabase. Accordingly, the preliminary focus in this stage of ReefKIM is mainly on obtaining essential information, such as reef occurrence, spatial distribution, geomorphic classification and typology and key habitats and substrates. ReefKIM users are consequently able to explore any reef system as well as individual reefs through reef-scale maps, to obtain details on reef dimension, habitats and substrate cover and other information related to management efforts. Acquiring further information, however, will necessitate additional literature searches and the involvement of a wide range of stakeholders and other individuals. Our study developed a process for constructing an integrated geodatabase using various types and sources of datasets, fostering data fusion and maximising accessibility of important information for a better understanding of the Kimberley’s coral reefs. This data integration approach has resulted in significant improvements in reef mapping. It showed, for example, that the number and area of Kimberley reefs are (over 60%) greater than described by previous studies, as many areas that may comprise reefs were unmapped due to lack of information. Reef number and area are anticipated to increase considerably as more information becomes available. The data integration approach also led to detailed reef-scale maps of 30 reefs in different geographical locations within the Kimberley Bioregion. Intra-reef geomorphic zones and associated biosedimentary substrates were mapped. During field work in the Kimberley Bioregion, it was noted that many people connected with the marine environment (e.g. rangers, commercial and recreational fishers, pearl farmers, traditional owners, nature photographers and tourists) possess valuable information, such as site images, underwater videos and photos and aerial photos of marine fauna and flora, including reefs. This information has had a significant role in the verification of satellite images when reef habitats and substrates have been mapped (Kordi et al., in prep b). Optimistically, these people were willing to share their knowledge for the sake of conservation of this vital ecosystem. Some information in the geodatabase can be made accessible to the public through a web-based interactive map allowing the selection, query and addition of information. This reef map can be used as a platform for crowdsourcing information from other participants through the implementation of reef monitoring. ReefKIM, as with many newly constructed geodatabases, requires frequent updates and insertion of new information to increase efficiency and effectiveness at long-term monitoring. Google Earth was employed to examine reef data visualisation due to its capabilities in handling various types of imagery and other geographic information and viewing and searching for specific locations. Furthermore, this platform enables crowdsourcing of information, through which users can easily contribute to the development of this geodatabase by adding and sharing their knowledge. Crowdsourcing is a promising approach for filling knowledge gaps and enhancing the understanding of such complex reef ecosystems. Both scientists and managers need a tremendous number of data points to be collected, which may not be feasible with small teams. However, the task may be more achievable if a large number of people are involved. Each data entry can be verified, labelled according to its reliability, referenced and acknowledged. Crowdsourcing will help produce more reliable information as the project progresses. The management efforts of the GBRMPA and its associated ecosystems are considered a model example of the contemporary reef management process. ReefKIM will allow the best practices from these efforts to be replicated for the Kimberley. ReefKIM’s data is expected to enable the detection of changes and quantification of current reef conditions at the regional and national levels, helping keep managers and conservation organisations informed and equipped

Kimberley Marine Research Program | Project 1.3.1 73

Reef Growth and Maintenance to implement the required policy changes. In addition to being a management tool, ReefKIM can be also useful in pinpointing future research priorities (cf. GBRMPA).

6.5 Conclusions ReefKIM is a GIS-based database constructed upon previous and current work, incorporating a wide range of datasets, including remote sensing images, bathymetric charts, site photos and many geological and biological datasets into one inclusive geodatabase. It is intended to provide researchers with an overview of essential information on the Kimberley reefs, where many reefs have yet to be mapped. It also constitutes a significant decision-making tool, as it can provide managers with practical support for the implementation of their plans and will help shape the direction of future management policies of coral reefs in the Kimberley region. The database was designed to be developed in collaboration with other regional and national institutions as well as individuals through a web-based map. The crowdsourcing approach allows many people already in the field, such as researchers, rangers, fishermen, tourists and traditional owners, to become involved in mapping and share their valuable knowledge.

6.6 Acknowledgements The authors are grateful to the Traditional Owners of the Kimberley (the Bardi Jawi, Mayala and Dambimangari people) for their assistance, advice and consent to access their traditional lands. The Western Australia (WA) Museum is appreciated for providing ground truth data through the WA Museum/Woodside Collection Project (Kimberley) 2008-2011. Many thanks also go to the Kimberley Marine Research Station, Kimberley Media, the Department of Parks and Wildlife (DPaW), Geoscience Australia (GA), the Geological Survey of Western Australia (GSWA) and the United States Geological Survey (USGS). Finally, special thanks are given to valued members of the research team at Curtin University: Mr Tubagus Solihuddin and Ms Giada Bufarale.

6.7 References

Australian Hydrographic Office, AHO (2009). Goodchild, M. (2007). Citizens as sensors: the world of http://www.hydro.gov.au. Accessed 06/02/2012. volunteered geography. GeoJournal 69 (4), 211–221. Beare, D. (2014) The Coral Triangle Atlas: An Integrated Gösseln, G. V., Sester, M. (2005). Change detection and Online Spatial Database System for Improving Coral integration of topographic updates from ATKIS to Reef Management. PLoS ONE 9(6): e96332. geoscientific datasets. In: Agouris, P., Croitoru, A. doi:10.1371/journal.pone.0096332. (Eds.), Next Generation Geospatial Information. In: ISPRS Book Series, Taylor & Francis Group, London, pp. Briner, A. P., Kronenberg, H., Mazurek, M., Horn, H., Engi, 69–80. M. & Peters, T. (1999). FieldBook and GeoDatabase: tools for field data acquisition and analysis. Great Barrier Reef Marine Park Authority, GBRMPA (2014). Computers & Geosciences, 25(10), 1101-1111. Eye on the Reef program. http://www.gbrmpa.gov.au/managing-the-reef/how- Brooke, B. (1997). Geomorphology of the north Kimberley the-reefs-managed/eye-on-the-reef. Accessed coast, in: Walker D. (Ed.), Marine biological survey of 15/11/2014. the central Kimberley coast. Western Australia Museum Library No. UR377 3–39. Hacklay, M., 2008. How good is volunteered geographical information? A comparative study of OpenStreetMap Bryant, D., Burke, L., McManus, J. & Spaulding, M. (1998). and Ordnace Survey datasets. In: Environment and Reefs at Risk: A Map-Based Indicator of Threats to the Planning B: Planning and Design (in press) Pre-print World’s Coral Reefs. WRI/ ICLARM/WCMC/UNEP. available at: http//: www. ucl.ac.uk /~ ucfamha /OSM World Resources Institute, Washington, DC, Available % 20data%20analysis%20070808_web.pdf. Accessed via WRI. http://www.wri.org/wri/indictrs/ 22/10/2014. reefrisk.htm. Accessed 10/08/2014. Hartcher, M. & Shearin, J. (1996). Developing a corporate Bureau of Meteorology, BOM (2012) wide network for GIS. Available via Reef http://www.bom.gov.au/wa. Accessed 10/03/2012. Research.http://kurrawa.gbrmpa.gov.au/corp_site/in Carvalho, R. C. & de Kikuchi, R. K. P. (2013). ReefBahia, an fo_services/publications/reef_research. Accessed integrated GIS approach for coral reef conservation in 22/08/2014. Bahia, Brazil. Journal of Coastal Conservation, 1-14.

74 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Chesnaux, R., Lambert, M., Walter, J., Fillastre, U., Hay, M., Heipke, C. (2010). Crowdsourcing geospatial data. ISPRS Rouleau, A., Germaneau, D. (2011). Building a Journal of Photogrammetry and Remote Sensing, geodatabase for mapping hydrogeological features 65(6), 550-557. and 3D modeling of groundwater systems: Application Lewis, A., Slegers, S., Lowe, D., Muller, L., Fernandes, L. & to the Saguenay–Lac-St.-Jean region, Canada. Day, J. (2003). Use of spatial analysis and GIS Computers & Geosciences, 37(11), 1870-1882. techniques to rezone the Great Barrier Reef Marine Chin, A., Sweatman, H., Forbes, S., Perks, H., Walker, R., Park. Paper presented at the Coastal GIS Workshop. Jones, G., Williamson, D., Evans, R., Hartley, F., McManus, J. (1994). Reefbase—A global database of coral Armstrong, S., Malcolm, H. & Edgar, G. (2008). Status reef systems and their resources. Marine pollution of the Coral Reefs in Australia and Papua New Guinea bulletin, 28(3), 133. 159-176 McManus, J. W. & Ablan, M. C. A. (1997) Reefbase: a global Collins, L. B. (2011). Controls on Morphology and Growth database of coral reefs and their resources. Proc 8th History of Coral Reefs of Australia’s Western Margin. Int Coral Reef Symp 2:1541–1544. Cenozoic Carbonate Systems of Australia (95), 195. National Oceanic and Atmospheric Administration, NOAA Collins, L. B., O’Leary, M., Stevens, A., Bufarale, G., Kordi, M. (2012). http://www.noaa.gov. Accessed 16/03/2012. & Solihuddin, T. (2015). Geomorphic patterns, internal architecture and reef growth in a macrotidal, high- Okolloh, O. (2009) Ushahidi or 'testimony': Web 2.0 tools turbidity setting of coral reefs from the Kimberley for crowdsourcing crisis information. Participatory bioregion. Australian Journal of Maritime & Ocean learning and action, 59(1), 65-70. Affairs, 7(1), 12-22. Rovere, A., Raymo, M. E., O’Leary, M. & Hearty, P. (2012). Comber, A., See, L., Fritz, S., Van der Velde, M., Perger, C. & Crowdsourcing in the Quaternary sea level Foody, G. (2013). Using control data to determine the community: insights from the Pliocene. Quaternary reliability of volunteered geographic information Science Reviews, 56, 164-166. about land cover. International Journal of Applied Solihuddin, T., Collins, L. B., Blakeway, D. & O’Leary, M.J. Earth Observation and Geoinformation, 23, 37-48. (2015). Holocene Reef Growth and Sea Level in a Conservation Council of South Australia, CCSA (2009). Reef Macrotidal, High Turbidity Setting: Cockatoo Island, Watch South Australia: The first decade of community Kimberley Bioregion, Northwest Australia. Marine reef monitoring SA, Australia: Conservation Council of Geology, 359: 50 – 60. SA. Spalding, M. & Grenfell, A. (1997). New estimates of global Cros, A., Ahmad Fatan, N., White, A., Teoh, S., Tan, S., and regional coral reef areas. Coral Reefs, 16(4), 225- Handayani, C., Huang, C., Peterson, N., Li, R., Siry, H., 230. Fitriana, R., Gove, J., Acoba, T., Knight, M., Acosta, R., Swart, P. K. (2013). Coral Reefs: Canaries of the Sea, Andrew, N., and Rainforests of the Oceans. Nature Education Dahdouh-Guebas, F. (2002). The use of remote sensing and Knowledge 4(3):5. GIS in the sustainable management of tropical coastal United States Geological Survey, USGS (2012). Earth ecosystems. Environment, Development and Resources Observation and Science (EROS) Centre Sustainability, 4(2), 93-112. http://eros.usgs.gov/. Accessed 20/03/2012. Department of Parks and Wildlife, DPaW (2013). The Victoria National Parks Association, VNPA (2014). Reef Kimberley Science and Conservation Strategy Watch Victoria. http://www. Reefwatchvic. asn.au/ http://www.dec. wa.gov.au/kimberleystrategy. Home.htm. Accessed 20/05/2014. Accessed 12/02/2013. Western Australia Marine Science Institute, WAMSI (2012). Franzoni, C. & Sauermann, H. (2014). Crowd science: The Kimberley Marine Research Program. organization of scientific research in open http://www.wamsi.org.au/research- collaborative projects. Research Policy, 43(1), 1-20. category/research-programs-kimberley-0. Accessed Fritz, S., McCallum, I., Schill, C., Perger, C., Grillmayer, R., 20/10/2013. Achard, F., Obersteiner, M. (2009). Geo-Wiki. Org: The Wilkinson, C. (Ed.). (2008). Status of coral reefs of the world: use of crowdsourcing to improve global land cover. 2008. Townsville, Australia: Global Coral Reef Remote Sensing, 1(3), 345-354. Monitoring Network and Reef and Rainforest Geological Survey of Western Australia, GSWA. Data and Research Centre. Software Centre 2013 Available: Wilson, B. R. (2013). The Biogeography of the http://geodownloads.dmp.wa.gov.au/datacentre/dat acentreDb.asp Australian North West Shelf New York, USA: Elsevier

Kimberley Marine Research Program | Project 1.3.1 75

Reef Growth and Maintenance

7 Quaternary onset and evolution of Kimberley coral reefs revealed by high resolution seismic imaging

Adapted from Bufarale, G., Collins, L.B., O’Leary, M.J., Stevens, A. M., Kordi, M., Solihuddin, T., submitted. Quaternary onset and evolution of Kimberley coral reefs (Northwest Australia) revealed by high resolution seismic imaging. Continental Shelf Research. In review

7.1 Introduction The remote Kimberley region (Figure 42) is located on Australia’s north-western continental margin and is characterised by complex coastal and marine habitats, rich in biodiversity (Chin et al., 2008; Collins, 2011; Department of the Environment, 2014; Pepper and Scott Keogh, 2014), which endure macrotidal conditions of up to 12 m in range (Cresswell and Badcock, 2000), high turbidity and frequent high energy cyclonic events (Brocx and Semeniuk, 2011). The Kimberley is also characterised by ancient folded and faulted Proterozoic metasedimets and volcanics resulting in a complex coastal geomorphology (Tyler et al., 2012). The Kimberley has recently been recognised as a major coral reef province of international significance (Chin et al., 2008; Wilson, 2013), however there is still a lack of scientific understanding of the evolution and development of Kimberley reefs (Wilkinson, 2008; Wilson, 2013). While the continental shelf edge reefs have been investigated throughout (e.g. Scott Reef; Collins et al., 2011), the extensive onshore coral reef systems are very poorly studied. Until recently it was not known whether reefs are thin veneers over rock platforms or significant long-lived accretionary structures (Solihuddin et al., 2015). Whilst the linkages between present geomorphology, Holocene sea level rise, reef growth history and coastal processes have been documented (e.g. Wilson, 2013), these settings and depositional environments are still poorly described and little understood, and requires an intensive study. With support from the Western Australian State Government through the Western Australian Marine Science Institution (WAMSI), this study attempts to establish how Kimberley reefs have developed over the time. As part of the Kimberley Marine Research Program (KMRP) Science Plan (Project 1.3.1) this work also aims to improve the understanding of the stratigraphic and geomorphic evolution and distribution of several Kimberley reefs and determine their interaction with different substrates, morphological patterns.

76 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 42. Map of the southern Kimberley showing the geology of the region (after Tyler et al., 2012) and the marine bioregions (Integrated Marine and Coastal Regionalisation of Australia, IMCRA, v4.0. Commonwealth of Australia, 2006). The reefs seismically surveyed in this study are labelled. Due to the remote and pristine conditions, macrotidal settings, marine hazards (crocodiles) and high turbidity of this environment, geophysical studies represent the most effective method to determine architecture and reef growth history, with negligible impact on the environment (i.e. Hutchinson and Ferrero, 2011). During this project the first detailed seismic study in this region has been carried out on various reef morphotypes in order to better constrain reef morphology, internal reef architecture and provide information on the past reef grow history.

7.2 Environmental Setting and Geology The Kimberley occupies a land area of about 424,000 km2 (O'Faircheallaigh, 2013) and the highly indented ria coastline extends from Cape Keraudren (southern end of Eighty Mile Beach, 19°58′S 119°46°E) to the border of Western Australia / Northern Territory (14°53′S 129°00′E), for a length of about 8.000 km (Brocx and Semeniuk, 2011, Figure 42). The coastal and marine habitats include extensive archipelagos, bays, capes, tidal plains, mangroves and a broad variety of inshore and offshore coral reefs. These complex environments are strongly influenced by macrotidal conditions, tropical monsoonal climate and proximity to the Indo – Pacific Throughflow (ITF) and the Leeuwin Current (e.g., Wolanski and Spagnol, 2003; Condie and Andrewartha, 2008; Collins and Testa, 2010; Brocx and Semeniuk, 2011; Wilson, 2013). Situated in a semiarid to subhumid climate zone, the Kimberley experiences seasonal rainfall (between November and March) that influences local freshwater and sediment drainage from the hinterland to the coast (Parkinson, 1986; Brocx and Semeniuk, 2011). Turbidity is another important factor that controls ecosystem

Kimberley Marine Research Program | Project 1.3.1 77

Reef Growth and Maintenance

distribution. Tides, ranging from 12 m during springs and less than 3 m during some neaps, with tidal currents of 2 m/s, have resulted in extensive intertidal zones (Cresswell and Badcock, 2000). These extreme tidal motions remobilise and resuspend the fine particles and sediments with the result being at high tide coral reefs become invisible from the surface and at low tide, the reefs appear covered to some extent by mud (Wilson et al., 2011). The turbidity increases in wet season when the rates of river runoff are higher (Wolanski and Spagnol, 2003). Combined with these metocean processes, the Kimberley marine environment is strongly influenced by a complex and ancient geological history. After an initial extensional tectonic event along the North Australian Craton margin (1910 – 1880 Ma), the subsequent plate collision (1870 – 1790 Ma) produced the crystalline basement that constitutes the Hooper and Lamboo provinces (Figure 42. Tyler et al., 2012). The suturing phase was followed by the development of a retroarc foreland basin (Speewah Basin, 1835 Ma) and a post-orogenic basin, formed in shallow marine to fluvial environments (Kimberley Basin, 1800 Ma. Jones, 1973; Tyler et al., 2012). An intense Palaeozoic orogenic deformation and associated sets of transcurrent faults resulted in the formation of the Canning, Ord and Southern Bonaparte basins (Tyler et al., 2012). Mesozoic successions of transgressive marine shale and prograding deltaic to fluvial sandstone cover much of the Canning Basin (Tyler et al., 2012). Since the Miocene, tectonism has determined much of the modern landforms and shore marine biota; in particular, during subsidence (Sandiford, 2007), the Kimberley margin progressively drowned determining the development of coral reef systems along the inner shelf of the region, accentuated by sea level rise after the Last Glacial Maximum (Tyler et al., 2012; Wilson, 2013).

7.3 Materials and Methods A preliminary Kimberley reef geomorphic classification scheme (Figure 44) was developed using GIS Software (Geographic Information System; ArcGIS® by Esri) and ENVI 4.3 software to process satellite images from Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper (ETM+) sourced from the USGS Earth Resources Observation and Science Centre (EROS). Existing literature on Kimberley reefs (particularly from Brooke, 1997; Wilson and Blake, 2011; Wilson et al., 2011; Richards et al., 2013 and WA Museum Woodside Collection, 2008 – 2011) provided the necessary ground truth information. The surveyed reef sites (Figure 42) were chosen in order to capture a broad range of Kimberley reef morphotypes described in our reef classification scheme (Figure 44) and included almost 300 km of SBP lines. Cockatoo Island fringing reef was examined in detail, serving as the base of the seismic calibration (see further in Section 7.4.2). The consistency of this correlation was verified by surveying the neighbouring fringing reefs of Irvine and Bathurst islands. Montgomery and Turtle Reefs were chosen as special types of, respectively, planar and fringing reefs, as recognised by Wilson (2013) and Wilson et al. (2011). The mid shelf reefs of the Adele complex were targeted representing planar reefs in an offshore environment and correlated with petroleum well log data (see further in Section 7.4.4.3.2). Sunday and Tallon islands were selected based on their strategic position and the variety of fringing reef types present. Geophysical survey lines were planned to target the orientation, internal architecture and morphology of the reefs. Profiles perpendicular to the reef allowed capture of the reef growth axis, transects parallel to the reef crest and crossing tie lines permitted a correlation during the interpretation and creation a three-dimensional perspective of the acoustic units’ framework. Sub-bottom profiling (SBP) surveys provided information on reef architecture, Holocene thickness and foundations, as well as earlier Pleistocene reef growth events. The surveys were carried out using an AA201 boomer system, mounted on a CAT100 surface tow catamaran (Applied Acoustic Engineering Limited, Great Yarmouth, UK) with a variable power setting, and a trigger rate and sweep time set accordingly with the water depth. In addition, an Applied Acoustic Engineering AH360/8, hydrophone streamer (sound receiver) was also towed behind the vessel. Accurate positioning (decimetric accuracy) was obtained with a dual frequency Differential Global Positioning System (DGPS) Fugro SeaSTAR 8200 XP/HP with Trimble Antenna. Data were digitally recorded in SegY format (Rev 1), using SonarWiz 5 (Chesapeake Technology Inc., Mountain View, CA) as acquisition and post‐processing software. The acquisition software also simultaneously recorded the NMEA 0183 GGA and VTG position and heading into the header file. The boomer profiles were post-processed to

78 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

improve the signal to noise ratio (S/N) by tracking the bottom and applying standard signal processing procedures such as bandpass filter and user defined gain/attenuation. The acoustic reflectors were then digitised, considering a two way travel (TWT) time seismic velocity of 1500 m/s. This setup saw signal penetration through carbonate reef framework of up to 40 m, however the acoustic signal did vary across the survey areas, depending on the surface and subsurface substrate type.

7.4 Results and Discussion Primarily due to its challenging geographical settings, the published scientific investigations of the Kimberley region are scarce compared to other marine and coastal ecosystems in Australia. (Wilkinson, 2008; McKenzie et al., 2009; Wilson, 2013). This research, as part of the KMRP Science Plan, aims to fill a gap in our understanding of key reef processes including the Holocene evolution and development of Kimberley Reefs. Seismic Facies Analysis Here we present seismic data on the surface and shallow geology of the Kimberley reefs, with the goal of understanding the stratigraphic evolution of these systems and determining their interaction with different substrates, morphological patterns and distribution. Two significant seismic reflectors R1 and RF were identified and defined on the basis of their relative position, acoustic reflection and architectural characters (Table 17 and seismic profiles in Sections 7.4.2 to 0). Between R1 and the seabed, many profiles show also further minor acoustic horizons, named H1, H2 and H3. On Adele Island, two additional deeper acoustic reflectors were identified and named R2 and R3 (Table 17 and seismic profiles in Section 7.4.4.3.2). The interpretation of seismic reflectors from inshore islands was ground truthed using a stratigraphically surveyed geological section from the inshore Cockatoo Island (see Solihuddin et al., 2015), and petroleum well log data from Adele 1 was used to interpret offshore seismic data.

Inner Shelf Reefs

R1 Reflector. R1 (Table 17 and depicted in green colour in the seismic profiles in Sections 7.4.2 to 0) is a high energy reflector, which represents the top of the Pleistocene calcretised reef limestone unit recognised by Solihuddin et al. (2015). Occasionally R1 is masked by seabed multiple echoes or by a thick and hard overlying substrate. The R1 follows a similar topographic profile to the modern reef flat seabed in exhibiting a quasi- horizontal reef flat that steeply dips at the fore reef slope matching the modern forereef slope. The seismic unit bounded by the sea floor/modern reef flat and R1 represents the Holocene reef/sediment buildup. H Reflectors. The Holocene unit is characterised by a series of internal discontinuous, subparallel reflectors (H1, H2 and H3, represented in shades of orange colour in the seismic profiles in Sections 7.4.2 to 0) of moderate to low amplitude. H1 is the shallowest reflector, found in the first 5 metres from the seafloor surface. In some profiles, it is not recognisable, due to very high impedance of the seafloor reflector masking the first few metres of substrate. H2 can be usually found between 5 and 8 metres from the seabed and H3, where identified, is generally 3 to 7 metres above the R1. RF Reflector. The post-processing and interpretation of the acoustic profiles across the area define the RF reflector (Table 17 and depicted in blue colour in the seismic profiles in Sections 7.4.2 to 0) as the deepest seismic horizon from the inshore reefs and locally forms deep valley-like depressions and ridges. The RF caps a chaotic, low amplitude unit that can be considered on the basis of the profile analysis and correlation with the data derived from a mine pit in Cockatoo Island (Solihuddin et al., 2015) as the acoustic basement of the inshore reefs, and represents the Proterozoic rock foundation. The terrestrial expression of this unit is represented by folded metamorphics, conglomerates and sandstones of the Kimberley Group. The seismic unit bounded by the R1 and RF represents an older Pleistocene reef unit, which belongs to the last interglacial sea level highstand (see further, Section 7.4.2).

Kimberley Marine Research Program | Project 1.3.1 79

Reef Growth and Maintenance

Mid Shelf Reefs

R2 Reflector. R2 (Table 17 and depicted in yellow colour in the seismic profiles in Figure 55) is a low to medium amplitude reflector, mostly acoustically transparent, that can be observed in the northern portion of Adele Reef. It covers a low amplitude unit, at least 25 m thick, where no internal architecture can be recognised. R3 Reflector. The low amplitude reflector R3 (Table 17 and depicted in pink colour in the seismic profiles in Figure 55) has been detected only within the Fraser Inlet, at a depth of about 65 m below the sea level. Since the entrance of this channel is relatively deep (~ 30 m), the substrate that overlies this reflector is thinner than elsewhere, allowing the penetration of the acoustic signal. Reflectors R2 and R3 are only found in Adele Reef. Table 17. Characteristics and acoustic features of seismic units identified in the profiles.

80 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Seismic Calibration The stratigraphy and geochronology of the fringing circum-island reef at Cockatoo Island is described in Solihuddin et al. (2015). Both the cross-sectional and distal marine parts of the south eastern reef close to the mine pit sections were seismically profiled and provided a basis for calibration of key seismic horizons (Figure 43). The reliability of this correlation was tested by laterally tracing the key seismic surfaces onto measured stratigraphic profiles reported by Solihuddin et al. (2015) and demonstrated the regional consistency of the stratigraphic pattern of two stages of reef growth (Holocene and Pleistocene).

Figure 43. A) Cockatoo Island reef digital terrain model (DTM) showing a 3D representation of the seafloor bathymetry. Note the sediment mound in front of the reef flat. Landgate aerial photography provided by the Department of Parks and Wildlife (DPaW); SBP lines are marked in red. SOL: start of line; EOL: end of line. Depth values are in metres, below the sea level. B) Intersecting cross-sections (Profiles 1 and 2) and distal longitudinal section (Profile 3) adjacent to mapped mine pit sections of Solihuddin et al., (2015) established position of Proterozoic (RF, blue), Last Interglacial (R1, green) and overlying Holocene reef intervals within seismic profiles collected across the fringing reef, SE Cockatoo Island (see inset for location of sections). The Cockatoo Island reef consists of stacked Last Interglacial (LIG) and Holocene reefs (Solihuddin et al., 2015). The Holocene reef growth commenced some 18 m below sea level (bsl), at 8.3 14C ky BP. The reef was in catch up mode during most of the Holocene, with reef fabric consisting of muddy branching corals with minor plate

Kimberley Marine Research Program | Project 1.3.1 81

Reef Growth and Maintenance

and massive corals which differ from contemporary communities on the reef flat (Solihuddin et al., 2015). Branching corals dominate the modern forereef slope and persist to depths of 10 m below mean sea level (Solihuddin et al., 2015). Within the Holocene reef, there are two minor seismic reflectors, which on the basis of their stratigraphic position are identified as H2 and H3. These reflectors could be interpreted as hiatuses or a change of the coral framework (possibly from domal to branching). The LIG section has been diagenetically altered and consists of calcretised, muddy branching coral framestone containing recrystallised corals, and directly overlies Proterozoic rocks (Solihuddin et al., 2015). Reef Classification Scheme Most of the coral reefs of the Kimberley are found in two bioregions: the Oceanic Shoals Bioregion, which includes reefs that rise from deep water along the edge of the continental shelf, and the inshore Kimberley Bioregion, where this geophysical investigation was performed (Figure 42). In order to select targets for this study, the Reef Geomorphology node 1.3.1 of the WAMSI developed a reef geomorphic classification scheme specific for the Kimberley (Figure 44) after considering previous studies, particularly the early work of Fairbridge (1950, 1967), together with Wilson (2013) who adapted the Great Barrier Reef (GBR) classification scheme of Hopley et al. (2007).

Figure 44. Reef classification scheme. See text for explanation Most Kimberley reef flats are exposed during low tide and are, therefore, intertidal reefs, the exceptions being where reefs have not yet grown to sea level and remain permanently subtidal. A specialised type of high intertidal reef recognised by Wilson (2013) and Wilson et al. (2011) is characterised by a high, flat topped surface that may be several metres above mean low water spring (MLWS) tide level, and experiences significant subaerial exposure during the tidal cycle. The surfaces have distinctive lithified algal terraces and coralline algae (rhodolith banks), as well as Porites microatolls which are often prolific. Small reef flat pools with healthy corals

82 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

are frequently present. These high intertidal reefs can occur as planar (e.g., Montgomery Reef), inter-island (e.g., Bathurst-Irvine Reef) and other fringing reefs (e.g., Tallon Island). Fringing reefs are widespread in the Kimberley and utilise the complex embayment and island coastal morphology, both as mainland and island associated features. In this study of the southern Kimberley we essentially described five main types of fringing reefs: Bay Head, Inter-island, Circum-island, Headland and Narrow Beach-Base. Wilson (2013) compared the GBR fringing reef scheme of Hopley et al. (2007) with the Kimberley reefs, noting two common types (fringing reefs on rock platforms and fringing reefs without rock platforms) and less common bayhead fringing reefs, with patch reefs recognised separately (Table 18). This analysis by Wilson adapts the Hopley classes to represent the most common reef types he observed, and additionally considers the degree of exposure of reefs, along with the most common coral communities found during his biodiversity studies. Table 18. Classification of Kimberley Fringing Reefs. Terminology after Hopley et al. (2007) and Wilson (2013).

Hopley et al. (2007) Wilson (2013)

Great Barrier Reef Kimberley Reefs 1 Headland attached fringing reefs, Fringing reefs on rock Exposed north or north west developed on rocky headlands platforms facing shores. Domal Faviid, massive robust corals on reef front

2 Bayhead fringing reefs that develop in Rare bayhead fringing reefs embayments and prograde out from the head of the bays

3 Narrow beach based fringing reefs, Fringing reefs without rock Sheltered leeward shores. developed along sandy coasts platforms Acropora dominated, minor clusters of massive corals and Fungiids 4 Nearshore shoals, not directly attached to Patch reefs the shoreline

Planar (or platform) reefs are large isolated features characterised by flat topped platforms that are usually emergent only at low tide. Examples are the Adele group of reefs (Adele and Churchill) and Brue Reef (see Figure 42 for location). Of the major reef types in the Kimberley, the numerically dominant Fringing Reefs and a really significant Planar Reefs were targeted for seismic surveys. High Intertidal, Intertidal and Subtidal Reefs High intertidal reefs are elevated reefs whose surface is several metres above MLWS tides. These reefs have unique surface characteristics, including shallow reef flat pools containing active corals, rhodolith banks, lithified algal ridges, terraced reef platforms, and fields of Porites microatolls. These reefs include some types of fringing and planar reefs in the southern Kimberley. Wilson et al. (2011) and Wilson (2013) previously reported such reefs, describing them as “intertidal platforms”. Most reefs of the Kimberley Bioregion are intertidal in character due to the macrotidal conditions, and are usually exposed during low tides, but these reefs lack the elevation of the high intertidal reefs, and lack their specialised substrates and communities. Reefs which have not yet grown to sea level are permanently subtidal, and many small patch reefs and most shoals are of this type.

Fringing, Inter-island Reef

Kimberley Marine Research Program | Project 1.3.1 83

Reef Growth and Maintenance

Bathurst Island – Irvine Island

The Buccaneer Archipelago (see Figure 42 for location) is characterised by a highly discordant coastline resulting from the partial submergence of a structurally controlled geological landscape with anticlinal features forming the region’s islands and coastal headlands. Reefs are commonly found fringing the islands of the Buccaneer Archipelago and some appear to coalesce to form inter-island reefs where two islands are located in close proximity. One such inter-island reef is located between Bathurst and Irvine Islands (Figure 45). This inter-island fringing reef is characterised by a deep elongate depression which cuts across the platform (Figure 45F), and it has been suggested by Wilson (2013) that this indicates the partial coalescence of two fringing reefs, which left a residual gap in the platform as a function of incomplete reef growth, as opposed to a single platform with an elongate karst depression. The platform surface is frequently emergent, with a transverse sand sheet (Figure 45B), shallow reef flat pools containing corals (Figure 45C and 4D) and fields of Porites microatolls present (Figure 45E), features in common with other high intertidal reefs such as Montgomery and Turtle Reefs.

84 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 45. Bathurst – Irvine Reef. A) Landgate aerial photography provided by DPaW; SBP lines are marked in red. B) Sand sheet on platform surface. C) High density branching corals in southern shallow pool. D) Exposed platform in low tide, with the pools containing corals. E) Dead Porites microatolls. F) Oblique view of the interisland fringing reef between Irvine Island (foreground) and Bathurst Island (background) during low tide. The central deep “pool” is clearly recognisable, together with a second minor pool (south east corner). Note the exposed platform on the north east side of Bathurst Island (Image courtesy of Kimberley Media, 2014).

Kimberley Marine Research Program | Project 1.3.1 85

Reef Growth and Maintenance

The internal structure of the Bathurst – Irvine Reef in west – east section (Figure 46) reveals its growth on Proterozoic bedrock 30 – 40 m bsl, a surface which is relatively flat. The LIG sequence is present only within the platform, where it is approximately horizontal and about 10 – 12 m thick. Within the Holocene reef buildup (~ 15 m), there are 3 minor acoustic reflectors (H1, H2, and H3) which are interpreted as hiatuses or temporary pauses in reef growth. The central “pool” in the reef, where traversed, has only a thin (5 m) cover of LIG reef, overlying the Proterozoic substrate. Holocene pinnacles are developed and up to 10 m high. Here, both the two stages of reef growth (LIG and Holocene) can be considered as incipient reefs which grew and became drowned by rising sea levels in successive transgressions. Both margins of the reef flat have well-developed bedded sediment lobes, up to 20 m thick. Longitudinal and transversal seismic profiles have shown that the bodies have a complex internal architecture (Figure 46 and Figure 47), composed of surficial seaward prograding layers, in discordant relationship with the deeper and more horizontal ones. Similarly for the Holocene discontinuities (H1, H2, and H3), the internal beds could also be linked to possible non-deposition (with minor erosion). Based on a qualitative interpretation of the seismic data, the drapes are composed by fine to medium grained sediments and, according to the local geomorphology and the collected samples, the sediment supply could be a combination of clastic influx coming from the surrounding islands and from suspension (from the mainland), and bioclastic carbonate deposits derived from the reefs. The western sediment mound (Figure 46A) is likely a result of ebb tides flowing off the reef platforms and bottom current dynamics, resulting in filling of pre-existing accommodation space. A north – south profile (Figure 47) of the platform and adjacent small lagoon shows a similar stratigraphy for the reef platform, with 25 m of Holocene sediment infilling the small lagoon to the south of the platform. Holocene and LIG thicknesses are similar to those already recorded in the W – E profile of the platform. Topographic changes in the Proterozoic surface as it rises to the north have influenced the position of the platform by providing suitable elevated substrate which was colonised by LIG reef growth. Within the sediment body, the internal geological pattern results almost completely obscured by an acoustic turbidity anomaly. This seismic signature, named curtain, is associated with fluid escape through the sediment (Baltzer et al., 2005). It is noteworthy that Traditional Owners have reported freshwater springs in the general area. The presence of the two shallow pools along north east Irvine Island could be linked to upward migration of fresh water into the seawater. The mixed brackish waters could have been a significant control on reef growing processes, limiting the Holocene (and possibly LIG) buildup in the area.

86 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 46. A) W – E section of the Bathurst – Irvine high intertidal fringing reef (width 7 km), showing two stages of platform growth, marginal sediment bodies, drowned reefs (insert, B) in the central elongate pool and basement topography. Depth values are in metres, below the sea level.

Figure 47. N – S section of Bathurst – Irvine Reef and adjacent embayment substrate. Note influence of platform elevation in the location of platform building; 2 stages (LIG and Holocene) of reef growth, and the 25 m thick bedded sediment pile filling the embayment to the S, with an internal signal probably representing fluid escape. Depth values are in metres, below the sea level.

Molema Island and Turtle Reef Molema Island is located less than 2 km west of the mainland coastline (Figure 42). Turtle Reef is a species rich, high intertidal, inter-island fringing reef in a muddy, enclosed embayment, within a terrain of islands of Palaeoproterozoic metamorphic and igneous rocks at the western end of the King Leopold Orogen (Wilson et al., 2011, Figure 48). Turtle Reef consists of two island-attached reefs separated by high Proterozoic cliffs, strongly folded and trending NW – SE. Turtle Reef is a high intertidal fringing reef within elongated fjord-like estuarine embayments. Both reef platforms are flanked by muddy channel bank complexes consisting of prominent sediment bodies crossed by tidal channel drainage with multiple trends. Tides of up to 12 m provide

Kimberley Marine Research Program | Project 1.3.1 87

Reef Growth and Maintenance

the main physical energy in this otherwise sheltered environment. The extreme shallowness of the platform caused difficulty in acquiring seismic data and the survey was confined to platform marginal areas and marginal mudflats (Figure 48A, survey lines marked in red). Reef platform substrates consist of a reef crest dominated by rhodolith banks and live corals, heavily smothered by soft mud and reef terraces surrounded by sand bodies with reef top coral rubble and sands (Wilson et al., 2011). There is a general similarity with the Montgomery platform including shallow pools containing small corals and algae. Porites microatolls are common on the high platform surfaces (Wilson et al., 2011).

Figure 48. Molema Island and Turtle Reef.A) Landgate aerial photography provided by DPaW; SBP lines are marked in red. B) Rhodolith bank at reef crest, note the channel draining tidal water off the reef. C) Rhodolith bank at reef crest with drainage channel, looking towards inner platform. D) Platform surface showing scattered corals and mud veneer. E) Oblique view of Proterozoic islands, northern Turtle Reef crest and muddy channel bank complex in foreground. F) Oblique view of southern Turtle Reef, with sand bank in foreground. Images courtesy of Kimberley Media, 2014. The northern reef’s eastern platform margin (Figure 49) reveals the structure of the reef platform overlying Proterozoic rocks (below RF horizon, in blue colour). An initial reef building phase of LIG reefs (below R1 horizon, in green colour) of 5 – 10 m thickness overlies the Proterozoic. Thereafter Holocene reef comprises the reef platform and is 15 m thick near the platform margin, confirming that the major phase of reef buildup occurred during the Holocene. Profile 1 (Figure 49A) crossed a platform lobe separated from the main platform. Near the south-western end (Start of the Line, SOL), the strong multiples indicate a lithified substrate. This is in contrast to the sedimentary terrain intersected at the north eastern end of the profile, where muddy channel bank sediments are present. Vast mudflats adjacent to the reef platforms, dissected by tidal channels are present (Wilson et al., 2011), and these may become exposed at low tide. Profile 1 (Figure 49A) shows a zone of acoustic wipe-out on the SBP record, possibly caused by absorption of the seismic signal in fluid-charged sediments. Up to 25 m of well-laminated, probably muddy, sediment, with intervening tidal channels, characterises the channel bank complex situated at the north western margin of the North Turtle Reef platform (Figure 49B).

88 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

These bedded sediment bodies may directly overlie Proterozoic rocks or, in some cases, overlie a reef building phase. This reef system, consisting of linked pinnacle reefs up to 15 m thick, is likely to have been active during the LIG and would have been expressed as a pinnacle reef complex, which became exposed during the post LIG regression. Small Holocene pinnacle reefs are also present, growing on some channel floors. The stratigraphy of the channel bank complex and that of the reef platform confirm the clear pattern of an initial Proterozoic terrain, followed by a reef building phase, most likely during the LIG, then a later Holocene reef building event which generated either mature reef platforms, or, in the channel bank complex areas, was expressed as small pinnacle reefs in the second reef building episode.

Figure 49. SBP images of the NW channel bank complex associated with North Turtle Reef. A) The Proterozoic basal surface (RF, blue) at ca.30 m bsl, rising in the SE where it emerges as an island surface, is overlain by initial LIG pinnacle reefs (R1, green) with small cappings of Holocene reef growth. Bedded sediments of the channel bank complex are regularly cut by tidal channels, with a pinnacle reef growing from a channel floor in the SE. B) In a similar profile at 90 degrees only the LIG stage of reef building is represented. Note the pinnacle reef at the base of the reef platform in the SE.

Fringing, Bay Head Reefs

South Sunday Island Fringing reefs are mainly developed in the southern and south-eastern sectors of the Sunday Islands (Figure 50). The reef flats are sheltered and elongate along the coast line, with an approximate width between 500 m and 1500 m. The acoustic profiles cover only the external edges of the platform, due to the tidal conditions during the survey. The data available reveal that the pre-existing Proterozoic topography (RF) rises from 30 m to 10 –

Kimberley Marine Research Program | Project 1.3.1 89

Reef Growth and Maintenance

15 m below the seafloor, significantly attenuating the reef development. Both the Last Interglacial and the Holocene reefs are relatively thin, with a similar internal architecture. Whereas the Proterozoic rock foundation presents irregularities, such as channels or depressions, the LIG reef growth tends to level the topography (see Figure 50). For this reason, it is problematic to establish an average thickness of this unit. The overlying Holocene reef buildup is about 7 m thick on a mostly flat pre-existing LIG surface.

Figure 50. Sunday Islands. A) Landgate aerial photography provided by DPaW; track plot of seismic profiles are marked with thin red lines. B) Seismic profile. Depth values are in metres, below the sea level. These Proterozoic islands are separated by deep depressions (probably structurally controlled), with both Holocene and LIG reef growth as fringing reefs. Irregular topography of Kimberley Group is blanketed by 2 stages of reef growth.

Planar, Coralgal Reefs

Montgomery Island

Montgomery Island is located approximately 23 km west of the coastline (Figure 42). It is bordered by a large high intertidal reef which is a planar coralgal reef, which is relatively isolated, and contains a central Proterozoic island, other associated islands such as High Cliffy Island and 3 small platform top islands (Wilson, 2013, Figure 51). Montgomery Reef is known to have a unique set of reef substrates on the platform, in particular rhodolith dominated substrates and associated crustose coralline algae forming a distinctive reef crest, and at least 2 terraces described by Wilson (2013) as upper and lower lagoons. Shallow pools with internal coral growth characterise parts of the reef flat. The well-developed Proterozoic central island is surrounded by sand cays and vegetated by mangroves and grasses (Wilson, 2013). Reef flats are dominated by sand and coral rubble with small living corals in shallow pools. The platform is very shallow and emergent at low tide with distinctive waterfall cascades across the reef crest (Figure 51C). Spring tidal range is 12 m, and the platform margin is exposed by a few metres at low tide. A prominent north – south trending spine protruding from the main platform on its western side is called “The Breakwater”.

90 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 51. A) Montgomery Reef, showing morphology (central Proterozoic island and surrounding sand sheet, upper and lower lagoon with rhodolith banks, and reef front ramp with crustose coralline algae). Note sand bodies and channels marginal to the platform and the Breakwater, which has shallow sandy substrates to the east and deeper water to the west. Landsat satellite image sourced from USGS-EROS; SBP lines are marked in red. B – D) View of the reef crest with pools of corals (Acropora) and sand patches. C – E) Aerial view of rhodolith banks (dark), showing complex channelling. F) Oblique view of Proterozoic central island, with surrounding sand sheets and two level terrace lagoon (upper and lower). Images courtesy of Kimberley Media, 2014. Due to the extreme shallowness of water on the platform in all parts of the tidal cycle, except during high water spring tides, seismic transects could only be obtained at the platform margins (Figure 51A), in particular both eastern and western side of the Breakwater, north west corner (forereef slope and outer lagoon), south west corner (Station Reef to main platform forereef slope and outer lagoon), and south east corner, a platform parallel profile of the distal forereef slope.

The shallow Breakwater area could not be crossed due to the tidal conditions; however, good quality profiles were collected on east and west sides of the spine. There are many pinnacle reefs and significant sediment cover in places along with occasional palaeochannels. Two stages of pinnacle growth can be frequently recognised, with a thicker LIG buildup under a thin Holocene reef substrate. In some cases pinnacle reefs occur above a deep basal unconformity, interpreted as the Proterozoic surface, capped by sediment cover terminating at 10 m bsl on the eastern side of the breakwater (Figure 52). An axial section across the northern tip of the Breakwater (Profile 1, Figure 52A) confirms an initial ridge of LIG coalescent pinnacle reefs that was followed by sediment cover during the Holocene. The N – S axis of the Breakwater apparently follows a bathymetric ridge with the same trend which controlled reef initiation and provided a favourable template for initial reef growth. A similar west – east profile, situated just to the north eastern extremity of the Breakwater (Profile 2, Figure 52B) endorses the pattern of LIG coalescent pinnacle growth on the Proterozoic, then pinnacle reef recolonization during the Holocene accompanied by infill of depressions by active sedimentation. On the south western margin of the Breakwater, in a N – S section, a series of pinnacle reefs occurs above the basal unconformity with a sediment blanket between pinnacles. The pinnacles are buried by the sediment and, in some cases, protrude through the

Kimberley Marine Research Program | Project 1.3.1 91

Reef Growth and Maintenance sediment cover demonstrating contemporaneous infill and active Holocene reef building. These pinnacles usually colonised an earlier (probably LIG) stage of pinnacle growth (see Profile 3, Figure 52C).

Figure 52. History of reef growth for the Breakwater from seismic profiles. A) Profile 1. Cross section of the Breakwater along its distal submerged northern margin reveals an initial LIG ridge composed of marginal pinnacles and central coalescent pinnacle architecture, overlain by Holocene coral reef with small surficial pinnacle reefs; note thickening of sediment drapes at margins of the ridge structure. B) Profile 2. A similar history is shown by this profile of the east margin of the Breakwater at its northern extremity. Note incipient colonisation of LIG pinnacle reefs by Holocene reefs, and influx of Holocene sediments. C) Profile 3. N – S view of the south western seaward margin of the Breakwater. Proterozoic surface (RF, blue) with two stacked generations of pinnacle reef development (LIG and Holocene), with bedded sediment infilling the reef terrain and overwhelming reef growth, proximal to and near the point of attachment to the Montgomery platform. Depth values are in metres, below the sea level. Together these profiles establish the pattern of development of the Breakwater, first as a ridge of LIG coalescent pinnacles, overlain by Holocene reef, then incipient regrowth on LIG pinnacles by Holocene reefs, and finally continued pinnacle growth and coalescence to reach near emergence as a continuous reef flat, forming a linear

92 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

N – S trending reef, and establishment of the current surface morphology as a spine-like protuberance northward from the Montgomery platform.

At its south western corner, the Montgomery Reef platform margin has distinctive morphology of a subtidal near vertical, terraced, descending to the seafloor forereef (Figure 53); a barren zone lacking spur and groove, with occasional sand sheets; a lower littoral reef front ramp, with slope 5 – 10 degrees, high energy tidal flow, encrusted by coralline algal ridges; a midlittoral reef crest with 100 m wide rhodolith banks and lacking a boulder zone; and finally a lower and upper reef flat dominated by rhodoliths, and small pools containing corals (Wilson, 2013). Two reef buildups are present overlying Proterozoic basement; a 10 m thick LIG reef and a 22 m thick Holocene reef, immediately underlying the platform margin (Figure 53). The distinctive terracing of the steep forereef is controlled by the boundary between these two reef building events.

Figure 53. Seismic sections showing structure of the SW margin of Montgomery Reef. Note terraced morphology of forereef is controlled by the boundary of LIG and Holocene reef buildup events. The lower level of basement substrate (RF, blue) under the crest of Montgomery Reef allowed for 2 stages of reef growth, LIG, and Holocene separated by an unconformity (R1, green). Depth values are in metres, below the sea level.

Reef building processes in and around Montgomery platform include both phases of reef building and buried sedimentary sequences, however it is clear that the Holocene buildup phase of the main platform is significant. In shallow platform conditions the internal structure of the platform is difficult to determine due to prominent ringing in the profiles and lack of penetration. It is likely that reef thickness declines toward the central island as the Proterozoic topography rises.

Kimberley Marine Research Program | Project 1.3.1 93

Reef Growth and Maintenance

Adele Island The planar, coralgal reef of Adele Island is located at the seaward extremity of the Kimberley Bioregion (Figure 42 and Figure 54) in the southern Kimberley, about 90 km from the coast.

Figure 54. A) Adele Reef. Landsat satellite image sourced from USGS-EROS; SBP lines are marked in red. B) Low tide view of reef flat showing rubble surface. C – D) Rhodoliths on reef flat, with an average diameter of 10 cm. E) View of dense community of corals (including Porites) covering the southern reef platform.

The clusters of platforms near Adele Reef are separated by deep channels cut into older rocks on the shelf. The Adele Reef platform is 17 km long by 9 km wide, has a central vegetated island surrounded by sand flats with N – S trend. Fraser Inlet, an incised channel with radiating tributaries, cuts into the north eastern platform margin. A well-developed intertidal reef platform or rampart, etched with small drainage channels, slopes into a sublittoral forereef with an active coral community (Wilson, 2013; Richards et al., 2013). Westerly winds and swell, together with diurnal tides (spring tides of 5 m) have shaped the platform, particularly the intertidal reef platform, where there are unique communities of rhodoliths and coralliths (see Richards et al., 2013). The Adele Island 1 well, drilled in 1982 on the northern tip of Adele Island (Figure 54A), intersected a 156 m limestone sequence and demonstrated that the Proterozoic basement is at 798 m, indicating that the Adele platform, which rests on a domal structure, is likely to consist of multiple stacked platforms or reefs (Unpublished Well Completion Reports: Ingram, 1982; Marshall, 1995). This setting contrasts with the shallow (ca 30 m) depth to Proterozoic basement of the inshore reefs. The limestone package consists of Holocene and

94 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Pleistocene to Neogene shallow water carbonates (Ingram, 1982; Marshall, 1995). The height of the reef platform is up to 60 m but the platform margin has a relatively low gradient characterised by groups of seabed features which rise from the platform surface to heights of 15 m along the edge of the forereef slope, in the seaward position of platform margins (Figure 55A). They appear to be reef pinnacles that stand at variable depths (from 10 m up to 25 m) below sea level and may indicate drowned features from repeated failures of catch up reef growth during sea level transgressions. These drowning events apparently postdate major reef building phases and suggest that the major platform growth took place during the Last Interglacial and was followed by subsequent drowning events such as during early to mid-Holocene time. Similarly to Scott Reef (Collins et al., 2011), seismic profiles of the Adele platform (Figure 55) identified, within equipment limitations, a minimum of 3 stages of reef platform buildup, separated by unconformities. Since the remains of the Adele Island 1 core (ditch cuttings) are not suitable for dating, it is possible to estimate the age of the stages through a comparison with the well-studied Scott Reef (Collins et al, 2011). As found in Scott Reef (Collins et al., 2011) and in the inshore reefs (Solihuddin et al., 2015 and this study), the shallowest reef package is represented by the Holocene growth stage. The Holocene reef is usually 20 – 30 m thick with a basal unconformity usually 25 – 35 m bsl (R1, in green, Figure 55). Reflector R2 (in yellow, Figure 55) marks the top of MIS 7 and indicates that the LIG unit that lays between the reflectors R1 and R2, there is a thickness of 3 – 10 m, which is relatively thin, when compared to Scott Reef and Rowley Shoals of the Oceanic Shoals Bioregion to seaward (Collins, 2002; Collins et al., 2011). The lowermost unconformity (R3, in pink colour) recognised is at 65 m bsl (see Figure 55) can be tentatively considered to be the top MIS 9 (or MIS 11), and could indicate that the MIS 7 package is about 25 m thick.

Kimberley Marine Research Program | Project 1.3.1 95

Reef Growth and Maintenance

Figure 55. Cross-sections showing multiple stages of reef buildup (MIS 1, 5, 7, 9 or 11 respectively). A) Profile 1, oriented E – W, intersects the northern portion of Adele Reef; B) Profile 3 runs longitudinally to Fraser Inlet and Profile 2 cuts the southern branch of the Inlet. Holocene reef is 20 m thick, with drowned pinnacle reefs on the western forereef slope (A; right corner insert). Along Fraser Inlet, a series of buried pinnacles can be recognised, covered by a thick layer of muddy sediments (left corner insert). Location of profiles is in the insert in B at lower right.

Samples collected with a pipe dredge along the Fraser Inlet show that the channel has a muddy substrate, indicating it is a pre-existing feature, probably cut during a past lowstand erosion of the platform. Holocene pinnacle reefs growing in the proximal part of the channel upon the LIG substrate were overwhelmed and buried by sediment filling the channel (Figure 55B). The channel was likely maintained during the Holocene due to tidal flushing, high sediment loads and limited coral growth on the channel floor.

Patch Reefs and Shoals

There are many patch reefs and shoals scattered throughout the Kimberley Bioregion (see Wilson, 2013 for examples). These reefs usually grow on isolated topographic highs suitable for colonisation by coral growth and are common on exposed margins of fringing reefs. They are usually intertidal to subtidal features and are often small. These features were not studied in detail in this project which focused on the larger and more diverse reef platforms described above.

96 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Station Reef Station Reef is a small (about 2 km in length; Figure 56) patch reef adjacent to Montgomery Reef, with a Proterozoic topographic high related to similar rocks on which the Montgomery Reef platform has developed. Station Reef is an isolated feature whose outer edge is 44 m bsl. It has an internal unconformity indicating a single stage Holocene reef built on a topographic high. Also identified is the LIG surface which underlies Montgomery Reef (Figure 53), but no LIG reef buildup can be identified overlying the basement high which provides the foundation for Station Reef.

Figure 56. Seismic sections showing the structure of the SW margin of Montgomery Reef, and continuity of seismic horizons between an adjacent patch reef. The Proterozoic nucleus in Station Reef is overlain by Holocene reef growth only, as the LI reef surface (R1, green) is at a significantly lower elevation (Profiles 2 and 3). Refer to Figure 53 for the explanation related to Profile 1.Depth values are in metres, below the sea level. Sea Level Changes and Subsidence During the Late Tertiary and Quaternary, periodic exchanges of mass between the ice sheets and oceans have greatly affected the global temperature and precipitations and controlled the amplitude of the sea level changes, relative to the land (Miller, 2009; Chappell, 2009). When ice sheets melted, the sea level has rose and, vice versa, as ice sheets retreated, the submerged shoreline features have become exposed (Chappell, 2009). Evidence of Quaternary sea level changes can be recognised in fossil coral reefs and sedimentary formations along the entire Western Australian coast, for example (selected references) in Perth area (Teichert, 1950; Murray-Wallace and Kimber, 1989), at the Abrolhos (Fairbridge, 1948; Eisenhauer et al., 1993; Collins et al., 2006), in Shark Bay (O’Leary et al., 2008a; Bufarale and Collins, 2015), Quobba Ridge (O’Leary et al., 2008b; 2013) Ningaloo Reef (Kendrick et at., 1991; Collins et al., 2003), Scott Reef and Rowley Shoals (Collins, 2002; Collins et al., 2011; Collins, 2011) and in the Kimberley coast (Wilson, 2013; Solihuddin et al., 2015 and this study). Evidences of sea level oscillations are present on the shelf adjacent to the Adele Group of platform reefs (Figure 57). Palaeochannels dissecting the submarine landscape have been sculptured by active rivers during sea level lowstands. Adele platform reef is a multistage reef buildup reflecting events of reef growth during interglacial periods of rising sea level, and erosional processes during glacial periods of low sea levels, probably on a ~100,000 year Milankovitch periodicity as expected from the Marine Isotope Curve and now demonstrated by

Kimberley Marine Research Program | Project 1.3.1 97

Reef Growth and Maintenance

SBP surveys. A cross section (Figure 57) through the reefs displays deep channels cut between the platforms as part of river erosion during sea level lowstands, and the superposition of reef building events which followed repeated transgressions and highstands during the late Pleistocene – Holocene (Marine Isotope Stages 1, 5, 7, 9 or 11). Such complex sculpturing is important for providing rocky habitats of deformed Proterozoic and other rocks suitable for coral reef colonisation as antecedent substrate throughout the repeated drowning events of Late Quaternary eustatic sea level fluctuations, with reef growth occurring during sea level highstands. Antecedent substrate is important, both for provision of surfaces for colonisation and for determining the elevation at which such colonisation may occur, relative to ultimate sea level, which controls the amount of accommodation available for reef growth and stacking of reef building events. However, subsidence of the LIG substrates must also be taken into consideration (Miller, 2009). The Kimberley coast has many islands and estuaries and is considered to represent a subsiding landscape (Sandiford, 2007). Solihuddin et al. (2015) provided the first evaluation of the post-LIG reef subsidence for the Kimberley region. Through stratigraphic data, seismic sections and geochronology from Cockatoo transects, Solihuddin et al. (2015) suggested that a linear subsidence has occurred nearshore, with a rate of around 0.11 m/ky since the LIG, providing about 13–20 m of accommodation for the Holocene reef accretion upon the older Last Interglacial reef.

Figure 57. Bathymetric and seismic profile across Adele group of platforms (see inset for location). Note repeated deep incision by rivers cut during sea level lowstands, which separates the platforms. Reef building events are vertically stacked within the platforms as repeated platform growth has occurred during SL highstands (MIS 1, 5, 7, 9 or11), as interpolated from seismic data. Palaeochannel positions in inset were derived from analysis of Geoscience Australia’s 250 m bathymetry dataset using ESRI’s Arc Hydro Tools Terrain Preprocessing toolset.

From the seismic profiles, in the Adele Group, the unconformity between Holocene and LIG reefs (R1, in green colour) is about 10 m deeper than in the inshore reefs, suggesting that these offshore reefs could have

98 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

experienced higher subsidence. This hypothesis is supported by the subsidence rates calculated in Scott Reef and Rowley Shoals (Collins, 2011; Collins et al., 2011). Since the last glacial, these offshore reefs experienced a subsidence of 0.37 m/ky (average between North and South Reef. Collins et al., 2011) and 0.16 m/ky, respectively. Hence, it is likely that the position of Adele Reef, closer to the continental margin, has influenced the subsidence rates and the related available accommodation space, being an important control of the overall Holocene morphology.

7.5 Conclusions In a review of Kimberley coral reef biogeography, Wilson (2013) noted the lack of information on the geomorphology and evolution of reefs of the Kimberley Bioregion, in contrast to the Western Australian coast elsewhere, including the Oceanic Shoals Bioregion of the offshore Kimberley, which has been well studied (see reviews in Montaggioni, 2005; Collins, 2011). The first detailed geological account of Holocene reef growth in the Kimberley Bioregion was accomplished in Cockatoo Island by Solihuddin et al. (2015.), as part of Project 1.3.1. of the Western Australian Marine Science Institution. As part of the same WAMSI project, the study presented here has achieved the first regional geological study of the Southern Kimberley coast. Whilst Wilson et al. (2011) suggested that the intertidal reef platforms might be Holocene veneers overlying pre-existing Pleistocene reef growth with a veneer of corals and calcareous algae, the Cockatoo Island chronology and seismic analysis demonstrated that the reef fringing this island consists of 18 m of vertical Holocene coral growth, commenced in Late Holocene, overlying a LIG reef, then the Proterozoic basement at depth. By developing an understanding of seismostratigraphic events, correlated to the reef chronology determined by Solihuddin et al. (2015), it has been possible to document the subsurface evolution and growth history of diverse reef systems for a range of reef types mapped in the southern Kimberley (see Figure 42), as reported here, at the scale of multiple reef building stages correlated to the Marine Isotope Curve. Seismically, all surveyed fringing reefs exhibit LIG and Holocene reef building events overlying Proterozoic basement at varying depths. High intertidal reefs (Bathurst – Irvine, Montgomery and Turtle Reefs) all consist of stacked LIG and Holocene reef building phases, with the limitation that some shallow platform interiors could not be profiled. Whereas reefs such as Montgomery Reef have a morphology controlled by the template provided by pre-existing Proterozoic terrain, inter-island fringing reefs appear in some cases to have formed by progradation and coalescence of fringing reefs attached to adjacent islands. Where coalescence has not gone to completion, deep inter reef channels or elongate depressions still remain between adjacent fringing reefs or within intertidal platform reefs (refer to Section 7.4.4.1). Within the intertidal reefs, like at the Sunday Islands, there is only 5 – 7 m of Holocene section due to higher basement topography. Planar reefs further offshore such as those of the Adele Group exhibit a longer term growth history, and consist of Holocene, LIG and earlier reef building events, most likely representing MIS 1, 5, 7 (Holocene, Last Interglacial and Penultimate Interglacial) and perhaps 9 or 11 in their stratigraphy, as would be predicted by the Marine Isotope Curve. Rather than overlying Proterozoic rocks, these reefs have Neogene limestone foundations, and older clastic sequences overlie Proterozoic rocks at depth. Some of the key interacting factors in the longer term evolutionary patterns of coral reef development and growth include morphological settings, physical processes, antecedent topography and sea level change (see recent reviews by Montaggioni, 2005; Hopley, 2011; Macintyre and Neumann, 2011; Finkl, 2011; Collins, 2011 and Flood, 2011). The interaction between the complex, subsiding Kimberley landscape and Quaternary global sea level fluctuations has been of primary importance in controlling available accommodation for and timing of multiple events of reef growth over geological timescales. Shorter term processes including macrotidal conditions, high turbidity and warm ocean temperatures have acted in concert with longer term events to generate habitat complexity and species diversity in what is a “biodiversity hotspot” (Department of the Environment, 2014; Pepper and Scott Keogh, 2014). As coral dominated Kimberley macrotidal reefs grow through catch up to keep up stages and reach the tidal window, either exposure-adapted corals become dominant on the reef flat (Solihuddin et al., 2015), or coralgal communities and facies, including algal ridges,

Kimberley Marine Research Program | Project 1.3.1 99

Reef Growth and Maintenance rhodolith and/or corallith fields, and Porites microatolls may succeed the reef building corals (see Wilson, 2013 and this paper). The seismic data discussed here demonstrate the long term resilience of the Kimberley reefs, their capacity to thrive in challenging environmental circumstances, and their persistence through oscillating sea levels of the late Quaternary. Ongoing work will investigate the Holocene sequences and their facies, particularly in relation to the onset and generation of the macrotidal substrates known to develop on platform surfaces.

7.6 Acknowledgements The Kimberley Reef Geomorphology Project 1.3.1 is funded by the Western Australian State Government through the Western Australian Marine Science Institution. This research was assisted by the Bardi Jawi, Mayala and Dambimangari people, the Traditional Owners of these lands, through essential assistance and guidance in part of the field work. The authors wish to thank: Cygnet Bay Marine Research Station staff (in particular James Brown and Dr Erin McGinty) that provided vessel support for marine operations and access to research facilities at Cygnet Bay; Mark Hardman (Fugro Satellite Positioning Pty Ltd) for supplying the DGPS; Neil MacDonald (Applied Acoustic Engineering Ltd) and Western Advance (Malaga, Western Autralia) for the equipment support; Giovanni De Vita for his technical advice; Richard Costin and Annabelle Sandes (Kimberley Media). It must be noted that the Aboriginal history of climate, land and environment is based on thousands of years of residence and rich culture: it is important to consider this broad understanding alongside the modern science completed here.

7.7 References

Baltzer, A., Tessier, B., Nouzé, H., Bates, R., Moore, C., & Hopley, D., Smithers, S., Parnell, K., 2007. The Menier, D., 2005. Seistec seismic profiles: a tool to Geomorphology of the Great Barrier Reef: differentiate gas signatures. Marine Geophysical development, diversity, change. Cambridge. Researches, 26(2-4), 235-245. Hopley, D., 2011. Encyclopedia of modern coral reefs. Brocx, M., and Semeniuk, V., 2011. The global geoheritage Structure, form and process. Hopley D, editor. significance of the Kimberley coast, Western Hutchinson, D. R., & Ferrero, R. C., 2011. Marine mammals Australia. Journal of the Royal Society of Western and anthropogenic noise. In: An Evaluation of the Australia 94, 57-88. Science Needs to Inform Decisions on Outer Brooke, B., 1997. Geomorphology of the north Kimberley Continental Shelf Energy Development in the Chukchi coast, in: Walker (Ed.), Marine biological survey of the and Beaufort Seas. Holland-Bartels, L. and Pierce, B. central Kimberley coast, Western Australia. University (Eds). USGS Alaska, 292. of Western Australia, Crawley, WA. Ingram, B.S., 1982. Palynological examination of samples Chappell, J. and Shackleton, N.J., 1986. Oxygen isotopes from Adele Island No. 1 from Well Completion Report, and sea level. Nature 324, 137 – 140. Brunswick Oil N.L. Chappell, J., 2009. Sea level change, Quaternary. In: Macintyre, I.G., Neumann, A.C., 2011. Reef classification, Encyclopedia of Paleoclimatology and Ancient response to sea level rise, in Hopley, D. (Ed), Environments, V. Gornitz, Ed., 893-899. Springer Encyclopedia of Modern Coral Reefs Structure, Form Netherlands. and Process, pp. 855 – 856. Chin, A., Sweatman, H., Forbes, S., Perks, H., Walker, R., Marshall, N.G., 1995. Adele Island No. 1 Palynological Jones, G., Williamson, D., Evans, R., Hartley, F., Report. Woodside Offshore Petroleum. Armstrong, S., Malcolm, H., Edgar, G., 2008. Status of McKenzie, N.L., Start, A.N., Burbidge, A.A., Kenneally, K.F., the coral reefs in Australia and Papua New Guinea, in: Burrows, N.D., 2009. Protecting the Kimberley: a Wilkinson, C. (Ed), Status of Coral Reefs of the World. synthesis of scientific knowledge to support Global coral reef monitoring network. Reef and conservation management in the Kimberley region of Rainforest Research Centre, pp 159 – 176. Western Australia. Department of Environment and Condie, S.A., Andrewartha, J., 2008. Circulation and Conservation, Perth, WA. connectivity on the Australian North West Shelf. Miller, K. G., 2009. Sea level change, last 250 million years. Continental Shelf Research 28, 14: 1724 – 1739. In: Encyclopedia of Paleoclimatology and Ancient

100 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Collins, L.B., 2002. Tertiary Foundations and Quaternary Environments, V. Gornitz, Ed., 879-887. Springer, Evolution of Coral Reef Systems of Australia's North Netherlands. West Shelf, in Moss, S. J. and Keep, M (Ed), The Montaggioni, L. F., 2005. History of Indo-Pacific coral reef sedimentary basins of Western Australia 3, October systems since the last glaciation: Development 20-23 2002, pp. 129-152. Perth, Western Australia: patterns and controlling factors. Earth – Science Petroleum Exploration Society of Australia. Reviews 71 (1 – 2), 1 – 75. Collins, L. B., Zhu, Z. R., Wyrwoll, K. H., & Eisenhauer, A., Murray-Wallace, C.V., Kimber, R.W.L., 1989. Quaternary 2003. Late Quaternary structure and development of marine aminostratigraphy: Perth Basin, Western the northern Ningaloo Reef, Australia. Sedimentary Australia Australian Journal of Earth Sciences, 36, Geology, 159(1), 81-94. 553–568. Collins, L. B., Zhao, J. X., & Freeman, H., 2006. A high- O'Faircheallaigh, C., 2013. Extractive industries and precision record of mid–late Holocene sea-level Indigenous peoples: a changing dynamic?. Journal of events from emergent coral pavements in the Rural Studies 30, 20 – 30. Houtman Abrolhos Islands, southwest Australia. Quaternary International, 145, 78-85.Collins, L. B., O'Leary, M.J, Hearty, P.J. & McCulloch, M.T., 2008a. U- Testa V., 2010. Quaternary development of resilient series evidence for widespread reef development in reefs on the subsiding Kimberley continental margin, Shark Bay during the last interglacial. northwest Australia. Brazilian Journal of Palaeogeography, Palaeoclimatology, Palaeoecology, Oceanography 58, 1 – 13. 259(4), pp.424–435. Collins, L.B., Testa, V., Zhao, J., Qu, D., 2011. Holocene O'Leary, M.J, Hearty, P.J. & McCulloch, M.T., 2008b. growth history of the Scott reef carbonate platform Geomorphic evidence of major sea-level fluctuations and coral reef. Journal of the Royal Society of Western during marine isotope substage-5e, Cape Cuvier, Australia 94(2), 239 – 250. Western Australia. Geomorphology 102(3-4), pp.595– 602. Collins, L. B., 2011. Geological setting, marine geomorphology, sediments and oceanic shoals O'Leary, M.J., Hearty, P.J., Thompson, W.G., Raymo, M.E., growth history of the Kimberley Region. Journal of the Mitrovica, J.X., Webster, J.M., 2013. Ice sheet collapse Royal Society of Western Australia 94(2), 89 – 105. following a prolonged period of stable sea level during the last interglacial. Nature Geoscience, 6 (9), pp. 796- Commonwealth of Australia, 2006. A guide to the 800. integrated marine and coastal regionalisation of Australia Version 4.0. Department of the Environment Parkinson, G., 1986. Atlas of Australian Resources, Third and Heritage, Canberra, Australia. Series, Volume 4: Climate. Division of National Mapping, Canberra. Cresswell, G. R., Badcock, K. A., 2000. Tidal mixing near the Kimberley coast of NW Australia. Marine Freshwater Pepper, M. and Scott Keogh, J., 2014. Biogeography of the Research 51, 641 – 6. Kimberley, Western Australia: a review of landscape evolution and biotic response in an ancient refugium. Department of the Environment, 2014. Australia's 15 Journal of Biogeography, 41(8), 1443-1455. National Biodiversity Hotspots. Australian Government, Canberra. Available at: Richards, Z. T., Bryce, M. and Bryce, C., 2013. New Records http://www.environment.gov.au/node/13909 of Atypical Coral Reef Habitat in the Kimberley, Australia. Journal of Marine Biology, Eisenhauer, A., Wasserburg, G. J., Chen, J. H., Bonani, G., http://dx.doi.org/10.1155/2013/363894. Collins, L. B., Zhu, Z. R., & Wyrwoll, K., 1993. Holocene sea-level determination relative to the Australian Sandiford, M., 2007. The tilting continent: a new constraint continent: U/Th (TIMS) and 14C (AMS) dating of coral on the dynamic topographic field from Australia. cores from the Abrolhos Islands. Earth and Planetary Earth and Planetary Science Letters 261 (1 – 2), 152– Science Letters, 114(4), 529-547. 163. Fairbridge, R. W. 1948. Notes on the geomorphology of the Solihuddin, T., Collins, L.B., Blakeway, D., O’Leary, M.J., Pelsart Group of the Houtman’s Abrolhos Islands. 2015. Holocene Reef Growth and Sea Level in a Journal of the Royal Society of Western Australia 33: Macrotidal, High Turbidity Setting: Cockatoo Island, 1–36. Kimberley Bioregion, Northwest Australia. Marine Geology 359, 50–60. Fairbridge, R.W., 1950. Recent and Pleistocene coral reefs of Australia, Journal of Geology 58, 330 – 401. Teichert, C. 1950. Late Quaternary changes of sea-level at Rottnest Island, We stern Australia. Proceedings of Fairbridge, R.W., 1967. Coral reefs of the Australian region, the Royal Society of Victoria 59: 63–79. in: Jennings, J.N. and Mabbutt, J.A. (Eds.), Landform Studies from Australia and New Guinea. Canberra, Tyler, I. M., Hocking, R. M., & Haines, P. W., 2012. Australia: Australian National University, pp. 386– Geological evolution of the Kimberley region of 451. Western Australia. Episodes – Newsmagazine of the International Union of Geological Sciences 35(1), 298.

Kimberley Marine Research Program | Project 1.3.1 101

Reef Growth and Maintenance

Finkl, C.W, 2011. Reef classification by Fairbridge (1950), in: Wilson, B., S. Blake, D. Ryan, J. Hacker., 2011. Hopley, D. (Ed), Encyclopedia of Modern Coral Reefs Reconnaissance of species-rich coral reefs in a muddy, Structure, Form and Process, pp. 846 – 850. macro-tidal, enclosed embayment, Talbot Bay, Kimberley, Western Australia. Journal of the Royal Flood, P., 2011. Reef classification by Maxwell (1968), in: Society of Western Australia 94, 251 – 265. Hopley, D. (Ed), Encyclopedia of Modern Coral Reefs Structure, Form and Process, pp. 854 – 855. Wilson, B. and Blake, S., 2011. Notes on the origins and biogeomorphology of Montgomery Reef, Kimberley, Jones, H.A., 1973. Marine geology of the Northwest Western Australia. Journal of the Royal Society of Australian Continental Shelf. Bureau of Mineral Western Australia 94, 107 – 119. Resources, Geology and Geophysics 136, 1–102. Wilson, B., 2013. The Biogeography of the Australian North Kendrick, G. W., Wyrwoll, K. H., & Szabo, B. J., 1991. West Shelf, environmental change and life's Pliocene-Pleistocene coastal events and history along response. Elsevier, 1st Edition. the western margin of Australia. Quaternary Science Reviews, 10(5), 419-439. Wolanski, E., Spagnol, S., 2003. Dynamics of the Hopley, D., 1982. Geomorphology of the Great Barrier Reef: turbidity maximum in King Sound, tropical Western Quaternary development of coral reefs. John Wiley – Australia Estuarine, Coastal and Shelf Science 56, 5 – Interscience, New York. 6

102 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

8 Cockatoo Island Reef Study

Adapted from Solihuddin, T., Collins, L.B., Blakeway, D., O’Leary, M.J., 2015. Holocene Reef Growth and Sea Level in a Macrotidal, High Turbidity Setting: Cockatoo Island, Kimberley Bioregion, Northwest Australia. Marine Geology Volume 359, pp 50–60.

8.1 Introduction The Kimberley coast is a remote sparsely populated and poorly studied region located in the NW of Western Australia. However, the discovery of a major hydrocarbon province in the offshore Browse Basin, and an increase in petroleum exploration in the region, has led to a heightened awareness of the region’s rich biodiversity (Collins, 2002; Chin et al., 2008; Collins et al., 2011). While the presence of coral reefs have been broadly documented, occurring as fringing reefs in coastal settings, platform reefs in mid-ramp settings and atoll-like reefs along the shelf margin (Wilson et al., 2011), there have been few investigations into their biogeography, diversity and developmental history. Coral reefs are particularly ubiquitous in the complex drowned landscape of the Kimberley coast, which provides abundant Proterozoic deformed rocky substrate for fringing reef development. These inshore fringing reefs occur in sheltered and exposed settings and endure in seemingly extreme environment conditions including; high turbidity and sediment input, elevated water temperatures (av. 28.5°C), a 10 m macrotidal range, significant subaerial exposure during low tides, and frequent cyclones. Despite these extreme conditions, the coral biodiversity in the Kimberley is far richer than that of the inner GBR fringing reefs and a little richer than those of the Pilbara to the south (Wilson, 2013). Critically, our understanding of the development of the Kimberley reefs still remains a gap in our knowledge. For example it is not known whether reefs are thin veneers over rock platforms or significant long-lived accretionary structures. Additionally, the linkages between present reef geomorphology, Holocene sea level rise, reef growth history, and coastal processes have been recognized (e.g. Wilson 2013) but are yet to be explored in any detail. Despite this lack of knowledge, Kimberley reefs have been identified as having international significance and are in need of comprehensive study (Chin et al., 2008; Wilson, 2013). This study will for the first time investigate the Holocene development and evolution of an inshore Kimberley coral reef located at Cockatoo Island (Figure 58). Cockatoo is unique in that iron ore mining on the Island has exposed a complete vertical section of the inner part of a Holocene fringing reef. This has allowed for detailed stratigraphic, palaeoecological and geochronological analysis spanning the entire reef growth history, thus enabling an investigation into how these reefs were able to persist under extreme environmental conditions as well as respond to Holocene sea level change.

Kimberley Marine Research Program | Project 1.3.1 103

Reef Growth and Maintenance

Figure 58. Map showing Bioregions modified from the Integrated Marine and Coastal Regionalisation of Australia (IMCRA) and geology of the Kimberley Region (After Griffin and Grey, 1990). The addition of seismic reconnaissance data has allowed for the broader reef architecture and structure to be assessed. Lastly this study develops a Holocene reef growth rate curve for the Kimberley, so that the shelf drowning history can be reconstructed. The initiative to protect these marine areas through the Kimberley Science and Conservation Strategy (Government of Western Australia, 2011) also requires a fundamental baseline understanding of how the environment formed for effective long term protection. Oceanography of the Kimberley Region The region is tidally dominated, with coastal mean spring range of 9.2 m in King Sound (see Figure 58; Harris et al., 1991), the highest tide range in Australia (Short, 2011) and the second highest tide in the world after the Bay of Fundy in Canada (Purcell, 2002; Wolanski and Spagnol, 2003). This macro-tidal system generates an extensive intertidal zone and strong tidal currents, which in turn cause high turbidity in coastal waters (Brooke, 1997). The region lies in the monsoonal belt with prevailing westerly or north-westerly rain-bearing winds from November- March, and dry south-easterly or easterly trade winds from May to September. It is cyclone-influenced (average 3 per year, Lough, 1998) and has southwest prevailing swell. The local oceanography of the inshore Kimberley Bioregion is influenced by the Holloway current, which is driven by the Indonesian Throughflow waters through a southward flow over the shallow Timor Shelf (DEWHA, 2008). This current flows seasonally from March to July and is closely associated with the northwest monsoon by which a south-westerly flow of surface water mass is released along the shelf margin. During the summer months (December to March), the Throughflow is deflected eastward, releasing a weak Holloway current along the inner shelf of the Kimberley. The upwelling generated by the Holloway current along with in situ planktonic production are believed to be contributing to the reef development in this region due to their roles in delivering nutrients where measures of N (0.05/12.8 µM), P (0.11/0.85 µM) and carrying planktic biota including planktotrophic larvae and reef animals (Wilson, 2013). Nutrient concentrations in the nearshore coastal Kimberley waters are

104 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

relatively high here, influenced by riverine inputs from the adjacent terrestrial, high runoff environment (Wilson, 2013).

8.2 Location and Methodology Cockatoo Island (Figure 58) is located in the Buccaneer Archipelago (123.6˚E; -16.1˚S), approximately 7 km north of the Yampi Peninsula. The geomorphology of the Buccaneer Archipelago and Yampi Peninsula are structurally controlled, reflecting the deformation geometry of the King Leopold Orogen (Wright, 1964). Cockatoo is an elongate island approximately 6 km long and 1.5 km wide and oriented along a NW/SE axis; with the NE side exposed to higher wave energies while the SW side is largely sheltered. Geophysical Surveys Sub-bottom profiling was performed using an Applied Acoustics Boomer SBP system CSP-P300; with SBP Sound Source, AA201 Boomer Plate, mounted on a CAT100 surface tow catamaran. The receiver streamer had 8 element hydrophones, and A/D Interface Box from NI (National Instruments) Device Monitor V5.3.1. Data acquisition was made using Chesapeake Technology Inc SonarWiz 5 software. Position acquisition was made using a Fugro Seastar 8200XP/HP DGPS receiver. Survey lines were run across the modern reef flat up to a retaining seawall, which runs the length of the mining pit. Two survey lines were also run along the length of the reef flat, parallel to the seawall and close to sections measured onshore. Reef Mapping The distribution of living coral and associated substrates was based on aerial photography acquired on 5 May 2010. The image was corrected geometrically using a geographic coordinate system and WGS 84 ellipsoid reference. The three bands of aerial photography (RGB 321), along with ground truth information, were employed to extract reef facies information. Discrimination of substrates into distinct classes was recognized from reflectance spectra grouped using unsupervised classification digital processing in ArcMap’s Image Analysis toolkit. The spectral signature of each pixel in the aerial photography was determined based on a grouping of the spectra of each individual band. The habitat classifications were ground truthed using towed camera transects over the reef flat and direct observations of the reef flat made on foot during spring low tides. Ore Pit Mapping Excavation of a 50 m deep, 750 m long open cut pit on Cockatoo Island has exposed a complete vertical section through the Holocene reef. Four stratigraphic sections were established along the face of the pit to log the reef exposures. At each site a vertical 0.7 m wide section from the base of the sequence to the top was logged, sampled and photographed. Information recorded includes: (i) the ratio of reef framework to matrix (following Embry and Klovan 1971); (ii) sediment textural characteristics (using the Udden-Wentworth nomenclature, as well as a visual assessment of sediment composition); and (iii) coral generic identification. Overlapping photographs (0.7 m x 0.5 m) were also taken up each profile. Reef framework analysis and facies descriptions followed the terminology suggested by Montaggioni (2005), which highlights the growth forms of the dominant coral reef builders and environmental indicators. The carbonate content of matrix sediment was determined by carbonate bombe (weight % loss after treatment with 50% HCL) following guidelines from Müller & Gastner (1971). Position fixing was by DGPS tied to mine site datum. Elevations from each transect were plotted relative to the Australian Height Datum (AHD), which is 3.987 m below the Cockatoo mine grid datum (mine survey data). Geochronology In situ corals were collected from levels along each vertical section for accelerator mass spectrometry (AMS) radiocarbon dating in order to establish a geochronologic record of reef accretion. Radiocarbon dates were recalibrated using CALIB Version 5.0.2 and calibration curve Marine04 (http://calib.qub.ac.uk/marine; accessed November 2012). A weighted mean ∆R value of +58 (average calculation from 3 nearest points) was used as the best current estimate of variance in the local open water marine reservoir effect for Cockatoo Island and adjacent areas. Dates discussed in the text are calibrated in years Before Present (cal y BP) with the 68.2% (2σ)

Kimberley Marine Research Program | Project 1.3.1 105

Reef Growth and Maintenance

probability range for all dated samples.

8.3 Results Living coral community zonation A map of reef habitats and substrates around Cockatoo Island, based on aerial photographic interpretation (Figure 59) provides a broad characterisation of the reef flat and its coral and macroalgal cover. Aerial photographic mapping linked to on ground observations and video transects delineates four gradational intertidal habitat zones. From landward to seaward these are the (1) intertidal beach and boulder rubble zone (2) inner sand flat and reef platform, (3) outer reef platform, and (4) forereef slope.

Figure 59. Map showing Cockatoo Island geomorphic and substrate classification, based on aerial photography interpretation (RGB 321).Numbers in legend correlate with habitats identified by on ground and towed video observation: (1) intertidal beach and boulder rubble zone, (2) inner sand flat and reef platform, (3) outer reef platform. The narrow and steep forereef slope is not mappable here. The inset map (A) shows sub bottom profile (white line) and towed camera (yellow line) transects and (B) location of the mine-pit measured sections. The upper intertidal zone consisting of steep sloping beaches of coarse, well-sorted siliciclastic sand is found in more embayed sections of the Island. Along sections of the Island consisting of steep plunging cliffs, hematite rich, sandstone boulder rubble deposits dominate the upper intertidal and supratidal zones. The lower intertidal zone consists of sand flats of poorly sorted, predominantly calcareous, sand with a matrix of pale grey-green calcareous mud. This transition into the inner reef platform includes sparsely distributed small coral colonies, mainly Porites and faviids but including many other genera. Macroalgae, predominantly Sargassum, are seasonally abundant on coral rubble substrate in this zone. Small thickets of the branching hydroid Millepora and the branching coral Porites cylindrica are common on the mid reef flat about 50–150 m from shore, occurring within extensive shallow pools (Figure 60A).

106 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 60. Living coral communities on the SW Cockatoo Island fringing reefs. (A) The extensive shallow pools on reef flat at low tide showing small thickets of the branching hydroid Millepora and the branching coral Porites cylindrica. (B) Turbinaria and branching Porites in the outer reef flat. (C) High coral cover of branching Acropora on the forereef slope.

The outer reef platform zone occurs seaward of these pools where the reef flat rises slightly, and is emergent for several hours on extreme spring low tides. Corals here occur as numerous close-packed colonies. Porites predominates but an extensive range of other genera occur, including, in estimated order of abundance: Turbinaria (Figure 60B), Favia, Favites, Goniopora, Lobophyllia, Astreopora, Montipora, Merulina, Pectinia, Goniastrea, Cyphastrea, Galaxea and Acropora (tabular and fine aborescent forms) P. cylindrica is the most common Porites but massive species are also common, usually as relatively small (< 0.5 m) colonies, often with microatoll architecture due to low tide exposure. The forereef slope is well defined and slopes seaward at approximately 30°; however, there is no raised reef crest along the reef edge. Video observations on the reef slope show a high cover (50 – 100%) of live branching Acropora (Figure 60C) and, occasionally, Seriatopora, generally as large colonies, from the reef edge to approximately 10 m below AHD. Downslope, dead coral branches are sparsely colonised by encrusting Montipora, encrusting to plating pectiniid corals and small fine branching Acropora colonies to approximately 20 m below AHD. The lower slope consists of coral rubble, colonised by gorgonians and sponges, in a mud matrix. The base of the lower slope grades gently into burrowed mud at approximately 30 m below AHD.

Kimberley Marine Research Program | Project 1.3.1 107

Reef Growth and Maintenance

Stratigraphy and palaeoecology Stratigraphic and palaeoecological data from the Cockatoo Island fringing reef is shown in Figure 61. Each of the four measured reef sections are located approximately 50 m seaward of the original steep rocky shoreline, and represent the inner reef platform environment.

Figure 61. Lithostratigraphic and chronostratigraphic summary of measured sections of Cockatoo mine-pit.

108 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

The basement is the Proterozoic Elgee Siltstone. Overlying the Elgee Siltstone is a 1 to 2 m thick sedimentary talus breccia containing subangular haematite boulders. Above the breccia is a 3 to 5 m thick pale yellow coral- rich mudstone with occasional weakly-bedded sand horizons and a calcretised, iron-stained upper surface. Corals in this unit are mainly domal, with a few encrusting and branching forms. Colonies are generally small (<30 cm diameter) and invariably recrystallised. Faviids are predominant, including Favia, Goniastrea, Cyphastrea and Platygyra. Overlying this unit is a second 0.3 to 1 m thick haematite boulder breccia, followed by the Holocene reef, which is 8.4 to 12.7 m thick in the measured profiles. This is a minimum thickness because the upper contact is partly obscured by the rock overburden of the overlying seawall, and because there may have been some compaction of the Holocene by the seawall, a 13 m high structure comprising a clay core and rock armour. All stratigraphic sections are dominated by single coral rudstone reef facies dominated by fragmented Acropora (typical diameter approximately 10 – 15 mm and length up to 150 mm) but also containing fragments and whole colonies of many other coral genera as well as abundant and diverse molluscan fauna. Additionally, the preservation of fine surface skeletal detail on most of the coral fragments indicates that they have undergone little transport; many appear to have collapsed where they grew. All interstices in the reef are completely filled with a poorly sorted grey-green calcareous mud containing sand to gravel-sized calcareous fragments and foraminifera tests. Coralline algae are rare throughout the sequence and there is virtually no inorganic cementation. There are several non-acroporid dominant coral horizons in some sections but they do not appear to be laterally continuous. Porites is the most abundant genus but is not dominant to the extent that it is on the modern reef flat, and only occurs as massive forms, P. cylindrica is conspicuously absent from the Holocene. Additional taxonomic differences between the Holocene and the living reef are the absence of branching Millepora from the Holocene and the relative scarcity of Turbinaria and tabular Acropora in the Holocene (Figure 62). Analysis of matrix sediments (Figure 61) showed carbonate content increased up section from 4 – 18% carbonate in the basal 2 – 3 m of the measured sections to 43 – 49% for the uppermost 2 m of the sections. The non- carbonate components include terrigenous clays and minor organics.

Kimberley Marine Research Program | Project 1.3.1 109

Reef Growth and Maintenance

Figure 62. Idealised stratigraphic column of Cockatoo mine-pit section (S2_P1). Photo (A) showing the domal coral framestone with a muddy matrix, photo (B) showing the prevailing branching coral and coral rubble, photo (C) showing the contact between Holocene reef and Last Interglacial reef with a haematitic breccia as pre-transgressive deposits, and photo (D) showing the calcretised Last Interglacial reef exposure. Note 14C ages and % carbonate in matrix fraction are also shown.

Reef Geochronology and Growth History Sixteen radiometric dates were made on coral clasts collected from reef sections (NW and SE) exposed along the seaward mine-pit wall. All colonies were interpreted as in situ on taphonomic and orientation criteria. Results indicate earliest Holocene reef growth initiated directly on the underlying pre-transgressive haematitic breccias along the NW reef section (S2) at a depth of 18.1 m below AHD (MSL) by ∼8,970cal y BP. While the upper 1 m of reef surface is obscured by the sea wall, it would appear the reef flat had reached the approximate level of mean low water (∼5.5 m below present AHD) in the NW reef section (S2) as early as ∼3,105cal y BP, while the SE reef section (S3) reached mean low water level at around ∼1,905cal y BP. (Table 19).

110 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Table 19. Radiocarbon dates from selected samples across Cockatoo mine-pit transects

SAMPLE CALIBRATED CALIBRATED NUMBER/ LAB. MEASURED CONVENTIONAL (68% (95% MATERIAL 13C/12C ELEVATION CODE AGE AGE Probability) Probability) (m) cal BP cal BP

S3_P2-31/ - Beta- 7490 +/- 30 +1.3 Goniastrea 7920 +/- 30 BP 8360 - 8300 8390 - 8270 14.794 364241 BP o/oo

S3_P2-28/ - Beta- 1980 +/- 30 -1.7 Galaxea 2360 +/- 30 BP 1950 - 1860 1990 - 1820 6.387 361585 BP o/oo

S3_P2-27/ - Beta- 2430 +/- 30 -0.6 Porites 2830 +/- 30 BP 2590 - 2580 2660 - 2340 7.337 361584 BP o/oo

S3_P2-26/ - Beta- 2920 +/- 30 +0.2 Porites 3330 +/- 30 BP 3160 - 3050 3210 - 2980 8.757 361583 BP o/oo

S3_P225/ - Beta- 3490 +/- 30 -2.4 Porites 3860 +/- 30 BP 3810 - 3680 3840 - 3630 10.667 361582 BP o/oo

S3_P2-24/ - Beta- 4430 +/- 30 -1.5 Porites 4820 +/- 30 BP 5040 - 4950 5190 - 5140 12.177 361581 BP o/oo

S2_P1-13/ - Beta- 2530 +/- 30 -0.7 Porites 2930 +/- 30 BP 2700 - 2640 2720 - 2510 5.737 361580 BP o/oo

S2_P1-12/ - Beta- 2920 +/- 30 -0.1 Porites 3330 +/- 30 BP 3160 - 3050 3210 - 2980 5.377 361579 BP o/oo

S2_P1-11/ - Beta- 2980 +/- 30 -1.2 Montastrea 3370 +/- 30 BP 3210 - 3110 3260 - 3050 6.447 361578 BP o/oo

S2_P1-10/ - Beta- 3170 +/- 30 -1.4 Cyphastrea 3560 +/- 30 BP 3420 - 3340 3450 - 3310 8.357 361577 BP o/oo

S2_P2-09/ - Beta- 4780 +/- 30 -1.0 Porites 4010 +/- 30 BP 3970 - 3880 4060 - 3830 9.956 364239 BP o/oo

S2_P2-08/ - Beta- 3172 +/- 30 -1.9 Porites 5160 +/- 30 BP 5520 - 5430 5560 - 5320 12.23 364238 BP o/oo

S2_P1-07/ - Beta- 7150 +/- 30 -3.1 Porites 7510 +/- 30 BP 7940 - 7870 7970 - 7830 14.257 361576 BP o/oo

S2_P1-06/ - Beta- 7320 +/- 30 -3.4 Porites 7670 +/- 30 BP 8120 - 8000 8150 - 7970 16.087 361575 BP o/oo

S2_P1-05/ - Beta- 7650 +/- 40 -1.0 Lepastrea 8040 +/- 40 BP 8490 - 8380 8540 - 8350 16.617 361574 BP o/oo

S2_P1-04/ - Beta- 8040 +/- 50 -1.1 Goniastrea 8430 +/- 50 BP 9000 - 8940 9070 - 8840 18.087 361573 BP o/oo

S2_P1-03/ - Beta- -4.4 Coral > 43500 BP 22.088 385986 o/oo

S3_P2-30/ - Beta- -4.9 Coral > 43500 BP 18.163 385987 o/oo

All samples were dated at the Beta Analytic Inc. Miami, Florida USA

Kimberley Marine Research Program | Project 1.3.1 111

Reef Growth and Maintenance

Vertical accretion rates varied across NW and SW reef sites (Figure 63). The NW section shows initial rapid vertical reef aggradation on the order of 3.8 mm/yr between 9 and 8 kyr BP, which occurred at some time post the initial flooding. There was an abrupt slowing after 7.9 kyr BP to around 1.2 mm/yr until around 5.5 kyr BP, where the reef showed gradual but sustained increase in accretion rates of up to 27 mm/yr as the reef flat reached near sea level with increasing light and energy. The SE section, unlike the NW section, did not show an initial rapid accretion but instead exhibited a more uniform accretion across its entire growth history. Unlike the NW section the SE section exhibited a slight reduction in vertical accretion after about 4 kyr BP from 3.6 mm/r to 2.0 mm/yr. The study was not able to establish the exact timing of when the SE section reached base level due to the top of the section being covered by mine waste.

Figure 63. Holocene vertical reef accretion and growth history curve for Cockatoo Island sections S2_P1, S2_P2, and S3_P2.

112 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Reef accretion records from the Abrolhos, Scott Reef, Middle Reef (GBR) and now Cockatoo Island are shown in Figure 64. While the Abrolhos record exhibits a keep-up growth history closely following sea level, reef accretion data from Cockatoo suggest it grew in a catch-up phase for most of its growth history. A lack of subsidence since the LIG and presence of a late Holocene highstand in SW Australian reefs (below 22 degrees south) contrasts with the Kimberley where no late Holocene highstand has been recorded to date, whilst there has been significant post-LIG subsidence at Cockatoo and Scott Reefs (Collins et al., 2011).

Figure 64. Composite growth history records for Kimberley region and GBR derived from coral sections at Cockatoo Island (this study), coral in core from Middle Reef GBR (Perry et al., 2012), coral in core from North Scott Reef (Collins et al., 2011), cemented coral shingle pavement from Abrolhos reefs (Collins et al., 2006), and corals in core from Morley Island Abrolhos (Eisenhauer et al., 1993). Microtidal Abrolhos reefs from SW Australia lack subsidence since the LIG, in contrast with macrotidal Kimberley reefs with post-LIG subsidence. This is reflected in presence or absence of a Late Holocene highstand in the contrasting SL records. Abrolhos data represents a keep up reef; Kimberley data represents a catch up reefs.

Kimberley Marine Research Program | Project 1.3.1 113

Reef Growth and Maintenance

Reef Architecture and Seismic Structure Boomer profiles immediately seaward of the logged pit profiles show the modern reef flat surface terminating in a steep (± 30°) forereef slope with a 20 m thick-bedded sediment mound to seaward (Figure 65).Three subsurface units are apparent, separated by reflectors at 37 m below sea level (bsl) and 40–45 m bsl.

Figure 65. Southwest Cockatoo Island SBP cross section showing two stages of reef development; Holocene and Last Interglacial, with a clear correlation to the measured island sections. Unconformities are coloured. (Blue = top Proterozoic; Green = top Last Interglacial Reflector).

Based on correlation with the logged sections, the seismic units correspond to the Holocene reef, the LIG reef and the Proterozoic bedrock. The Holocene reef is 15-20 m thick, with a flat upper surface at 8 m bsl, shallowing slightly to landward. The surface of the underlying LIG reef is irregular and has a distinct reef crest at 30 – 40 m (Figure 66). A deeper unconformity with an average depth of 40 – 50 m marks the position of the Proterozoic rock foundation.

114 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 66. SBP line with superimposed core log from adjacent mine pit, showing correlation between SBP units and lithological units in the subsurface.

8.4 Discussion Evidence of neotectonic subsidence along the Kimberley Coast and offshore shelf As a trailing, intraplate continental margin, coastal WA should be relatively tectonically stable. However, recent studies have identified seismic activity, and structural and geomorphological features consistent with ongoing crustal deformation (Sandiford 2003; Quigley et al. 2006, 2007). On the continental scale Sandiford (2007) and DeCaprio et al. (2009) have identified tilting of the Australian plate driven by changes in mantle convection, whereby eastern and north western Australia is experiencing subsidence and southern and south western Australia is experiencing uplift. Evidence supporting this theory can be found in the relative elevation of LIG shorelines along the WA coast. Presently LIG reefs outcrop intermittently above present sea level along a stretch of coast from Foul Bay (34˚S) to Cape Keraudran (20˚ S) (O’Leary et al., 2008; 2013 and Collins et al., 1996). North of this line the only occurrence of LIG reefs that has been observed is in bore holes from the shelf edge Scott Reef at an elevation of 36 to 60 m below AHD (Collins et al., 2011). Here we are able to provide the first evidence of a LIG reef from the Kimberley region, which outcrops at an elevation of 18 m below AHD. Although U-series analysis has not been carried out on this basal reef unit, the fact that corals exhibit a high degree of diagenetic alteration in contrast to the Holocene reef, and the Holocene and LIG reef are separated by a hematite breccia layer supports this interpretation. Stratigraphic data from Cockatoo transects confirm that the LIG reef surface underlies the Holocene reef more broadly and exhibits reef terrace morphology. Assuming the LIG reef terrace reached base level the elevation would suggest a linear subsidence rate of around 0.1 m/ky since the LIG, this compares to a recent study which estimated subsidence rates at South and North Scott Reef on the Rowley Shelf margin as 0.45 m/ka and 0.29 m/y respectively (Collins et al., 2011), where the shallowest LIG reef is at 36 m below AHD. Based on the proximal Cockatoo and distal Scott Reef subsidence data a subsidence with tilting offshore appears likely for the Kimberley coast and adjacent shelf.

Kimberley Marine Research Program | Project 1.3.1 115

Reef Growth and Maintenance

Holocene reef growth Reef growth on Cockatoo Island initiated around 9 kyr BP at -18 m below AHD as evidenced from the radiocarbon ages from the basal Holocene corals. The timing of this event would suggest that corals became established not long after flooding of the antecedent substrate had occurred with sea level transgressing the -20 m elevation at around 10 kyr BP (Lambeck et al., 2010). Accretion curves plotted in Fig.6 indicate the reef initially grew in a Keep-Up phase as shown by the high vertical accretion rates for this period. A reduction in reef accretion rate after 7.5 kyr BP and a shift from Keep-Up to Catch-Up growth phase likely resulted from the rapid deepening the reef which became submerged to >10 m below AHD by ~7 kyr BP before eventually reaching MLW (∼4 m below AHD) at ∼3,000cal y BP. Palaeoecological and sedimentological evidence for a Catch-Up reef growth includes the abundance of branching Acropora throughout the sequence and the predominance of a sandy mud matrix that becomes dominated by sand at the top, only becoming sandy at the top of the sequence. The absence of present-day intertidal indicators such as P. cylindrica, branching Millepora, tabular Acropora and coral micro atolls can also be considered to support a subtidal characterisation of the reef facies. However these indicators are also absent from the top of the logged profiles, when the reef had reached the intertidal zone. We conclude either that environmental or ecological conditions at the time did not suit the modern intertidal taxa, or that the modern coral community and zonation have only developed after prolonged intertidal exposure, and may therefore be restricted to a thin zone at the very top of the logged profiles which may not always show up in cross-section, or which we may have missed due to the seawall overburden. As we currently lack any cross-reef age isochrons, we can only propose a reef growth style, based on field observations supported by seismic profiles. The interpretation outlined above, of the reef initiating as a blanket and growing upward, is one possibility; another is that the reef initiated close to shore and prograded seaward. These alternatives correspond to models A and B respectively in Kennedy and Woodroffe’s (2002) review of fringing reef growth and morphology. We consider Kennedy and Woodroffe’s more complex models C to F to be less likely analogs, due mainly to the apparently uniform internal structure of the Cockatoo reef in seismic profiles. While we believe the vertical accretion model best represents our logged profiles, the reef further seaward may have become submerged too deeply to allow significant vertical accretion. On the modern reef, live coral cover is sharply reduced below about 10 m AHD and virtually absent below 20 m AHD, probably due to rapid light attenuation in the turbid water (Blakeway 1997; Wilson et al., 2011). If this depth limit applied throughout the Holocene, only those areas submerged to less than ∼10–15 m AHD are likely to have accreted vertically to sea level, while deeper areas may have drowned before eventually being prograded over by the shallow reef (c.f. Hopley and Partain, 1986). Live coral cover on the shallow reef was probably relatively high throughout the vertical accretion phase. The baffling effect of the coral framework would have trapped suspended sediment, depositing the grey-green interstitial mud seen in cross-section. Tropical cyclones must have intermittently affected the reef, and are perhaps the main agent responsible for its detrital reef fabric (c.f. Braithwaite et al., 2000). When the reef reached sea level, which probably first occurred at or near the rocky shore, surface sediments would have become coarser due to wave winnowing, the zone of high coral density would have retracted seaward, and the characteristic community makeup of the current intertidal reef flat would have begun to develop. Eventually the reef attained its present state, with vertical accretion essentially ceased but progradation continuing through in-situ growth of branching Acropora colonies on the upper slope and deposition of Acropora fragments on the lower slope. The typical sub-bottom profile of the reef shows a steep forereef slope unconformably overlying the LIG reef surface at 37 m. A further unconformity marks the boundaries between LIG reef and Proterozoic rock foundation underneath at a depth of 40 –50 m. A 15 – 20 m Holocene reef build-up overlies consistently LIG reef at 37 –40 m. This is in accordance with the stratigraphy examined at the adjacent mine pit with the exception that the Holocene reef is thicker to seaward.

116 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Comparisons with other Holocene Reef Systems The Cockatoo Island reef conforms reasonably well to the ‘arborescent coral facies’ description in Montaggioni’s (2005) Indo-Pacific reef classification scheme. It differs from the typical arborescent coral reef in a couple of respects which probably reflect the reef’s environmental setting, particularly the high turbidity and high terrigenous input. Firstly, the mud matrix is more pervasive at Cockatoo Island than in most arborescent coral reefs. As mentioned earlier we attribute this to high suspended sediment loads being baffled by the coral framework. Secondly, live branching coral (virtually all Acropora) on the modern Cockatoo Island reef occurs solely on the upper fore-reef between 4 and 10 m below AHD, whereas on other Indo-Pacific reefs it typically occurs on mid and lower fore-reef slopes to approximately 20 m, and also on shallow inner reef flats and back- reef slopes (Montaggioni, 2005; Hopley et al., 2007). The compressed shallow subtidal distribution of branching Acropora at Cockatoo Island is probably a function of a) the low hydrodynamic energy, which allows it to grow undamaged in shallow water (at least during fair-weather conditions), b) the high turbidity, which sets the lower depth limit through light attenuation and possibly also sedimentation stress, and c) their low tolerance to subaerial exposure, which excludes them from the reef flat. Within their optimal 4 to 10 m depth range at Cockatoo Island, branching corals are prolific and seem able to competitively exclude (i.e. outgrow) all other coral types. The restriction of Porites cylindrica and branching Millepora to the intertidal zone may be a consequence of this competitive exclusion, as both would normally extend into subtidal habitat (Veron 2000; Glynn and de Weerdt, 1991). The Cockatoo fringing reef shares several similarities with the nearshore turbid zone reefs of the GBR. Browne et al. (2012) describe reefs within 20 km of the Queensland coast influenced by episodic terrigenous sediment input, and fluctuating salinities (24–36 ppt). Table 20 compares the inshore GBR systems with the Cockatoo fringing reef. With the exception of the greater tidal range at Cockatoo, both have a similar oceanographic setting with significant sediment influx, sediment-tolerant corals, high coral cover, a significant proportion of in situ coral framework, and a mud-dominated matrix (coarsening at the top of the sequence). Differences include the hard foundation of the Cockatoo fringing reef versus the sand and gravel substrates of the GBR reefs, the greater age and thickness of the Cockatoo fringing reef, the higher proportion of detrital branching Acropora in the Cockatoo reef, and an apparently more sharply-defined coral depth zonation on the modern Cockatoo Island reef. The overall mean accretion rate of the Cockatoo Island reef is ∼2 mm/year, significantly lower than most turbid zone GBR reefs. However, the mean rate at Cockatoo Island includes approximately 2,000 years of slow accretion when we infer the reef was deeply submerged. Accretion rates in the lowermost and uppermost few metres of the Cockatoo section, when the reef was probably submerged to the depth at which the GBR reefs have developed (≤5 m), are approximately equivalent to the mean rate of 8.4 mm/year recorded at Middle Reef, the fastest-growing of the cored turbid zone GBR reefs. These accretion rates are high in a global context and are only exceeded by a handful of reefs, generally within the arborescent or detrital categories (Perry et al., 2012; Montaggioni, 2005). Perry et al. (2012) attribute Middle Reef’s rapid accretion to the deposition of fine-grained terrigenous sediments that are suspended from the surrounding seafloor or introduced during major floods. The fine-grained sediment infills the open reef fabric and inhibits post depositional skeletal destruction, thus aiding primary framework preservation and reef accretion. The Cockatoo fringing reef appears to have developed in a very similar way, with only a few deviations from Perry et al.’s model, including the predominance of detrital fabric instead of in-situ framework.

Kimberley Marine Research Program | Project 1.3.1 117

Reef Growth and Maintenance

Table 20. Summary of characteristics of turbid reefs from the GBR (after Browne et al., 2012) and the Cockatoo fringing reef (this study).

Environmenta Inshore turbid reefs on the GBR (Browne et al., Inshore turbid reefs at Cockatoo Island (this l controls 2012) study)

Typical Description Typical Description condition condition

Oceanography Locally wind Lack of coralline algae and Strong tidal Domal and robust-branching driven waves robust Acropora communities, currents coral, coral rubble coralline lack of oceanic swell generated by high algae exist on the tidal range contemporary reef surface Low swell waves

Fine-grained High levels of Sedimentation rate > 10 Intermittent high Background turbidity is 1-2 sediment load sedimentation mg/cm2/day and turbidity turbidity NTU, and high turbidity is and turbidity levels from <10 mg/L to >50 event-related mg/L

Reef initiation Holocene 6000 - 1000 y BP Holocene 9000 y BP

Substrate mixed Sediment deposits on shallow Lithified pre- Most reefs develop on the LI availability coastal embayment are the existing reef, or reef with/without pre- most common substrates for Proterozoic transgressive deposits reef growth metasediments are common substrates.

Average rate of Variable 2-7 mm/year Variable 2.16 - 7.4 mm/year reef growth

Mode of growth Mixed Reefal foundations, terrigenous Mixed Reefal foundations, sand/silt sand and rubble matrix, coral rubble

Surrounding Shallow water Reef growth restricted by Shallow water Reefs occur down to 15 m bathymetry (<15 m) shallow, turbid waters (<20-40 m) depth with a sharp fore reef slope and sand/silt sediment mound to seaward

Sea level Strongly influence high level Rising Holocene Plays a major role in reef wind driven suspended sea levels have development sediment, substrate availability influenced reef and reef morphology growth

Composition Variable to low 100-150 species Moderate 17 coral genera are identified coral diversity Many inshore reefs have biodiversity with the most common diverse coral communities, but community and facies many are also dominated by association being of physiologically robust corals branching corals (especially which may be more tolerant to of the genera Acropora) and change domal corals including Porites

Community age Low coralline <1% cover on the GBR Low coralline Specialised fauna capable of structure algae cover algae cover tolerating high sediment Many inshore reef loads Mixed to older characterised by large, older coral colonies which are

118 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

capable of tolerating high sediment loads Low Less suitable substrate recruitment and availability due to high survival rates sediment cover and high level of algal competition. High sediment loads can affect the survival of coral juveniles

There are strong contrasts between the nearshore Cockatoo reef and the offshore reefs of the Oceanic Shoals Biozone located ∼300 km to the Northwest including Scott Reef (Figure 58) which rise from depths of 400 m, in cyclone and swell wave exposed and clear water settings. These reefs are atoll like, dominated by coral framework with sandy calcareous intercalations, with earlier Holocene reef onset ages (e.g. 11.5 ka for Scott Reef; Collins et al., 2011) and higher subsidence rates.

8.5 Conclusions This study provides the first information on Holocene reef growth for a nearshore reef of the Kimberley Biozone. • The Cockatoo fringing reef initiated on lithified Pleistocene substrates at approximately 9,000 cal y BP and accreted in a catch-up mode, first reaching sea level at approximately 3,000 cal y BP. The predominance of fast-growing branching Acropora, and the high rate of mud deposition, produced very rapid reef accretion in the upper and lower few metres of the sequence, when the reef surface was probably ≤ 5 m deep. • Knowledge of the reef’s foundations has leads to a calculated subsidence rate of the coastal Kimberley since Last Interglacial time of 0.11 m/ka. • Like Scott Reef of the Oceanic Shoals Biozone to seaward, the Holocene reef at Cockatoo Island lacks evidence of the Late Holocene highstand that characterises coral reefs of the tectonically stable microtidal southwest Australian coast. Macrotidal conditions and coastal subsidence may have obscured such records, although further data from the Kimberley are needed to evaluate this conclusion. Whilst growth rates of the Cockatoo reef fall into the Montaggioni (2005) “fast to moderate growing reef” category of Indo-Pacific reefs, a direct comparison with turbid reefs of the GBR is more appropriate given both systems are muddy in character, and similar growth rates are recorded for the two systems which contrast with the mainly clear water reefs studied by Montaggioni. • Whilst the GBR and Cockatoo turbid reefs show broad similarities, one contrast is between palaeoecological and contemporary reef communities, which are similar in the GBR but distinctly different at Cockatoo Island. A likely explanation is that the modern Cockatoo community needs to adapt to manage the extreme conditions of currents, mud from slack water suspension fallout, and prolonged exposure due to macrotidal conditions, in contrast to more stable reef growth in catch up mode earlier in the Holocene. • A potential analogue for the Holocene palaeoenvironment of muddy branching Acropora is provided by the contemporary shallow subtidal reef slope, which is the only habitat where live branching Acropora occur at present. The contemporary intertidal environment is characterised by abundant Porites cylindrica and Millepora intricata, both absent from the Holocene. • The muddy framework facies recorded at Cockatoo Island are in keeping with the turbid macrotidal conditions, whilst contrasting with the off shore sand Kimberley reefs. The Cockatoo reef growth pattern suggests that the Kimberley near-shore reefs may be more resilient to periodic disturbance and the effects from terrestrial runoff than reefs elsewhere. Further, the study shows that reef response to climate is resilience through time. As such the information presented here should be taken into consideration when planning and siting Marine Parks.

Kimberley Marine Research Program | Project 1.3.1 119

Reef Growth and Maintenance

8.6 Acknowledgements The Kimberley Reef Geomorphology Project 1.3.1 is funded by the Western Australian State Government and partners of the Western Australian Marine Science Institution. This research was assisted by the Dambimangari people through their advice and consent to access their traditional lands. Pluton Resources (particularly Jeremy Bower and Anson Griffith) are thanked for providing access to parts of their Cockatoo Island Mining Tenement and for logistic support during the study. The Cygnet Bay Marine Research Station provided vessel support for marine operations and access to research facilities at Cygnet Bay. James Brown assisted in the planning stage of the project, and Erin McGinty capably managed marine operations. MScience is thanked for providing access to marine video of the reef. Giada Bufarale (seismic data acquisition), Moataz Kordi (remote sensing) and Alexandra Stevens (editing and improvement of drafts) at Curtin University were valued members of the research team. Finally, it must be noted that this research was completed in an area where the Traditional Owners have a rich cultural history of climate, land and environment based on thousands of years of habitation. It is important to consider that broad understanding alongside the modern science completed here.

8.7 References

Blakeway D., 1997. Scleractinian corals and reef Wilson, B. R., 2013. The Biogeography of the Australian development, part 9, in: Walker D. (Ed.), Marine North West Shelf: Environmental Change and life’s biological survey of the central Kimberley coast, response. Elsevier, Burlington MA, USA. Western Australia. Perth: University of Western Wolanski, E. and S. Spagnol., 2003. Dynamics of the Australia. Unpublished report, W.A. Museum Library turbidity maximum in King Sound, tropical Western No. UR377; pp. 77–85. Australia. Estuarine, Coastal and Shelf Science 56(5– Braithwaite, C. J. R., Montaggioni, L.F., Camoin, G.F., 6): 877–890. Dalmasso, H., Dullo, W.C., Mangini, A., 2000. Origins Wright, R. L., 1964. Geomorphology of the West Kimberley and development of Holocene coral reefs: a revisited Area. CSIRO Land Research, Series 9, 103– model based on reef boreholes in the Seychelles, 118.Department of the Environment, Water, Heritage Indian Ocean. International Journal of Earth Science and the Arts (DEWHA), 2008. A Characterisation of 89, 431–445. the Marine Environment of the North-west Marine Brooke, B., 1997. Geomorphology of the north Kimberley Region: Perth Workshop Report, A Summary of an coast, in: Walker D. (Ed.), Marine biological survey of Expert Workshop Convened in Perth, Western the central Kimberley coast. Western Australia. Australia, 5-6 September 2007, Commonwealth of University of Western Australia, Perth, unpublished Australia, Hobart. report, W.A. Museum Library No. UR377, pp. 13–39 Embry, A. F. and J. E. Klovan. 1971. A late Devonian reef Browne, N. K., S. G. Smithers, 2012. Coral reefs of the turbid tract on northeastern Banks Island, N.W.T. Bulletin of inner-shelf of the Great Barrier Reef, Australia: An Canadian Petroleum Geology 19(4): 730–781. environmental and geomorphic perspective on their Eisenhauer, Wasserburg, A. G. J., Eisenhauer, A., Chen, J.H., occurrence, composition and growth. Earth Science Bonani, G., Collins, L.B., Zhu, Z. R., Wyrwoll, K. H., Reviews 115(1–2): 1–20. 1993. Holocene sea-level determination relative to Chin, A., Sweatman, H., Forbes, S., Perks, H., Walker, R., the Australian continent: U/Th (TIMS) and 14C (AMS) Jones, G., Williamson, D., Evans, R., Hartley, F., dating of coral cores from the Abrolhos Islands. Earth Armstrong, S., Malcolm, H. & Edgar, G., 2008. Status and Planetary Science Letters 114(4): 529–547 of the Coral Reefs in Australia and Papua New Guinea. Government of Western Australia, 2011. Kimberley Science In: Wilkinson, C., (Ed) Status of Coral Reefs of the and Conservation Strategy. Department of World. Global Coral Reef Monitoring Network. Reef Environment and Conservation, Kensington, Perth. and Rainforest Research Centre, pp. 159–176. Glynn, P. & W.H. de Weerdt. 1991. Elimination of two reef Collins, L. B., Zhu, Z. R., and Wyrwoll, K. H., 1996. The building hydrocorals following the 1982-83 El Niño structure of the Easter Platform, Houtman Abrolhos warming event. Science 253, 69-71. reefs: Pleistocene foundations and Holocene reef growth. Marine Geology 135(1–4), 1-13. Griffin, T. J., & Grey, K., 1990. Kimberley Basin. In: Memoir 3, Geology and Mineral Resources of Western Collins, L. B., 2002. Tertiary Foundation and Quartenary Australia. Perth, Geological Survey of Western Evolution of Coral Reef Systems of Australia's North Australia, pp.293–304. West Shelf. In: Keep, M., Moss, S.J. (Eds) Proceeding of the Petroleum Exploration Society of Australia Harris, P. J., Baker, E. K., and Cole, A. R., 1991. Physical Symposium, Perth, WA, pp. 129–152. sedimentology of the Australian continental shelf with emphasis on Late Quaternary deposits in major Collins, L. B., Zhao, J. X., Freeman, H., 2006. A high-precision shipping channels, port approaches and choke points. record of mid–late Holocene sea-level events from

120 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

emergent coral pavements in the Houtman Abrolhos University of Sydney, Ocean Sciences Institute, Report Islands, southwest Australia. Quaternary 51. International 145–146(0): 78–85. Hopley, D. & Partain, B., 1986. The structure and Collins, L. B., Testa, V., Zhao, J., and Qu,D., 2011. Holocene development of fringing reefs of the Great Barrier growth history and evolution of the Scott Reef Reef Province, in: Baldwin, C. L., (Ed.), Fringing Reef carbonate platform and coral reef. Journal of the Workshop: Science, Industry and Management. Great Royal Society of Western Australia 94(2), 239-250. Barrier Reef Marine Park Authority, Townsville, pp. 13–33. DeCaprio, L., Gurnis, M. & Mueller, R.D., 2009. Long- wavelength tilting of the Australian continent since Hopley, D., Smithers, S., Parnell, K., 2007. The the Late Cretaceous. Earth and Planetary Science Geomorphology of the Great Barrier Reef: Letters 278(3-4), 175–185. Development, diversity, change. Cambridge University Press, Cambridge, UK. Perry, C. T., S. G. Smithers, P. Gulliver, and N. K. Browne., 2012.Evidence of very rapid reef accretion and reef Kennedy, D.M., Woodroffe, C.D., 2002. Fringing reef growth under high turbidity and terrigenous growth and morphology: a review. Earth Science sedimentation. Geology 40(8): 719–722. Review 57, 255–277. Purcell, S. P., 2002. Intertidal reefs under extreme tidal flux Lough, J. M., 1998. Coastal climate of northwest Australia in Buccaneer Archipelago, Western Australia. Coral and comparisons with the Great Barrier Reef: 1960 to Reefs 21(2), 191–192. 1992. Coral Reefs 17(4): 351–367. Quigley, M.C., Cupper, M.L. & Sandiford, M., 2006. Lambeck, K., Woodroffe, C. D., Antonioli, F., Anzidei, M., Quaternary faults of south-central Australia: Gehrels, W.R., Laborel, J., Wright, A. J., 2010. Palaeoseismicity, slip rates and origin. Australian Paleoenvironmental records, geophysical modelling, Journal of Earth Sciences 53(2), 285–301. and reconstruction of sea-level trends and variability on centennial and longer timescales, in: Church, J.A., Quigley, M.C., Sandiford, M. & Cupper, M.L., 2007. Woodworth, P.L., Aarup, T., Wilson, W.S. (Eds.), Distinguishing tectonic from climatic controls on Understanding Sea-level Rise and Variability. range-front sedimentation. Basin Research 19(4), Blackwell Publishing Ltd, pp. 61-121. pp.491–505. Montaggioni, L. F., 2005. History of Indo-Pacific coral reef Sandiford, M., 2003. Neotectonics of southeastern systems since the last glaciation: Development Australia: linking the Quaternary faulting record with patterns and controlling factors. Earth-Science seismicity and in situ stress, in: Hillis, R.R., Muller, D. Reviews 71(1-2), 1–75. (Eds.), Evolution and Dynamics of the Australian Plate. Geological Society of Australia Special Publication, vol Müller, G., & Gastner, M., 1971.The'Karbonat-Bombe', a 22, pp. 101–113. simple device for the determination of carbonate content in sediment, soils, and other materials. Neues Sandiford, M., 2007. The tilting continent: A new constraint Jahrbuch für Mineralogie-Monatshefte 10, 466–469. on the dynamic topographic field from Australia. Earth and Planetary Science Letters, 261(1-2), 152– O’Leary, M.J., Hearty, P.J., McCulloch, M.T., 2008. U-series 163. evidence for widespread reef development in Shark Bay during the last interglacial. Palaeogeography. Short, A. D., 2011. Kimberley beach and barrier systems: An Palaeoclimatology. Palaeoecology 259, 424. overview. Journal of the Royal Society of Western Australia 94(2), 121–132. O'Leary, M., Hearty, P. J., Thompson, W. G., Raymo, M. E., Mitrovica, J. X., & Webster, J.M., 2013. Ice sheet Veron, J. E. N., 2000. Corals of the world. Australian collapse following a prolonged period of stable sea Institute of Marine Science, vols. 1-3, pp.1,382. level during the last interglacial. Nature Geoscience Wilson, B. R., S. Blake, D. Ryan, J. Hacker., 2011. 6(9), 796–800. "Reconnaissance of species-rich coral reefs in a muddy, macro-tidal, enclosed embayment, - Talbot Bay, Kimberley, Western Australia." Journal of the Royal Society of Western Australia 94: 251–265.

Kimberley Marine Research Program | Project 1.3.1 121

Reef Growth and Maintenance

9 Holocene reef accretion styles in the Buccaneer Archipelago.

Adapted from Solihuddin, T., Collins, L.B., Blakeway, D., O’Leary, M.J. In Progress. Holocene Reef Accretion Styles in the Buccaneer Archipelago, Kimberley Bioregion, Northwest Australia. In Progress and subject to change. See section 9.10 for preliminary analysis of Adele Island cores.

9.1 Introduction Buccaneer Archipelago is situated in the inshore Kimberley coast and the reefs are exposed to macrotidal conditions, frequent cyclones and elevated turbidity. Such environmental conditions would generally be considered inhospitable for reef development and inhibit the growth of reef build-up. However, evidences from Kimberley reefs (Wilson, 2013) and reefs from Great Barrier Reef (GBR; Perry, 2003; Smithers and Larcombe, 2003; Perry et al., 2008), indicate that many reefs endure and have high diversity of coral species, suggesting great conditions for reef growth. The extreme tides and remote conditions have limited coral reef research to date, including gaining long core records. Previous studies on intertidal platform reefs under extreme tidal flux in Buccaneer Archipelago (Purcel, 2002; Wilson and Blake, 2011; Wilson et al., 2011) have demonstrated the specialised nature of Kimberley reefs and their coral communities. Also, the linkages between reef growth history, coastal and reef geomorphology have been recognized in recently collected data on biogeography by Wilson (2013). Solihuddin et al. (2015) recently described the Holocene reef growth history of Cockatoo Island. In this study, we investigate selected inshore reefs in the Buccaneer Archipelago including Bathurst-Irvine Island, Sunday Island, and Tallon Island to obtain a Holocene record of reef building communities, chronology and growth patterns. This involved addressing the following issues: 1) How do Kimberley reefs vary regionally in response to coastal substrate controls and terrigenous sediment inputs? What are the contrasts in reef substrates and oceanographic controls between coastal, mid shelf and offshore reefs? 2) How have environmental conditions in the inshore Kimberley region changed over the last several thousand years and what has been the associated response of coral communities and growth patterns? 3) How have the interactions between substrates, sea levels, extreme macrotidal conditions, high turbidity and subsidence controlled the inshore reefs, and how are these factors reflected in reef building communities, “turn on –turn off “ history and ecology and substrates? These palaeoecological and geological studies provide a context for understanding current community dynamics and indicate that turbid zone reefs are robust and resilient.

9.2 Field Settings Biogeography The Buccaneer Archipelago is a series of coastal islands in the southern part of Kimberley Bioregion, stretching from One Arm Point in the south to the Yampi Peninsula in the north. It comprises over 700 islands (Kordi et al., 2015 in review), many of which are uninhabited and unnamed. These islands are the remnants of the Pleistocene Kimberley plateau which was exposed during the LGM - Holocene sea-level transgression (Brooke, 1997). The mainland coast is characterised by incised rias and broad embayments as a result of deformation of the King Leopold Orogen (Wright, 1964). Reefs grew and developed in this archipelago both on the mainland and the islands in the form of either fringing or platform reefs.

9.3 Geology The hinterland geology of the Kimberley Bioregion is dominated by the Proterozoic Kimberley Basin which consists of sandstones, siltstones, volcanics and intrusive igneous rocks (Griffin and Grey, 1990). The Buccaneer

122 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Archipelago occurs within the King Leopold Orogen, which is characterised by highly metamorphosed, strongly folded and faulted rocks with a northwest–southeast foliation (Wright, 1964).

9.4 Oceanography The regional oceanography of the Kimberley coast is significantly influenced by the south equatorial current and the Indonesian Throughflow (ITF) which transports warm, low salinity water to northwest Australia (Pearce and Griffiths, 1991). The southward flowing Leeuwin current is driven by the waters of the ITF, and is believed to contribute to reef development in this region by delivering nutrients and planktic biota including planktotrophic larvae and reef animals (Pearce and Griffiths, 1991; Fang and Morrow, 2003; Wilson, 2013). Tides are semi-diurnal, with a mean spring range of 9.2 m in King Sound (Figure 67). The ocean water surface temperatures are 22 to 28°C nearshore and 17 to 27°C offshore, whilst salinity ranges from 34.5 to 35.7 PSU (Pearce and Griffiths, 1991). Tropical storms during the southern monsoon induce intense rainfall with an annual average of 300 mm between November and March, the dry season occurs from May to September (Brooke, 1997). This region is also prone to cyclones, experiencing approximately 3 per year (Lough, 1998) and has a prevailing southwest swell.

Figure 67. Map of research location and coring sites distribution

Kimberley Marine Research Program | Project 1.3.1 123

Reef Growth and Maintenance

9.5 Methods Reef coring Forty two cores were collected from three different islands, including Tallon Island, Sunday Island, and Bathurst- Irvine Island. Coring sites were selected to support interpretation of sub-bottom profiles from geophysical surveys, also located to ensure a spatially representative record of reef growth. Cores were obtained using percussion and rotary drilling core techniques. Unconsolidated reef and sediments were sampled by percussion coring, using either a manual slide hammer or a hydraulic post driver to drive lengths of 80 mm diameter aluminium pipe. Consolidated, coralline algal-cemented reef was sampled by rotary coring, using a handheld hydraulic core drill to drive a 50 cm long, 80 mm diameter diamond core with a 1 m extension rod. Core site elevations were corrected to mean sea-level (MSL) for comparison with palaeo sea level data.

Core logging and sampling

Core logging and sampling documented characteristics including (i) the ratio of reef framework to matrix (following Embry and Klovan 1971); (ii) sediment textural characteristics (using the Udden-Wentworth nomenclature, as well as a visual assessment of sediment composition); and (iii) generic coral identification. Reef framework analysis and facies descriptions followed the terminology suggested by Montaggioni (2005), which highlights the growth forms of the dominant coral reef builders and environmental indicators. Matrix sediments were collected for carbonate content analysis determined by the carbonate bombe technique (weight % loss after treatment with 50% HCL) following guidelines from Müller & Gastner (1971). Positions from each core were fixed by a GPS. All elevations from each core were plotted relative to mean sea-level (MSL). The core logs were corrected for compaction.

14C coral dating

Coral specimens from distinct facies in each core were selected for accelerator mass spectrometry (AMS) and radiometric carbon dating to establish a geochronologic record of reef accretion. All specimens were in situ, based on orientation and skeletal condition, and free of boring encrustation and submarine cementation. Radiocarbon dates were recalibrated using CALIB Version 5.0.2 and calibration curve Marine04 (http://calib.qub.ac.uk/marine; accessed November 2014). A weighted mean ∆R value of +58 (average calculation from 3 nearest points) was used as the best current estimate of variance in the local open water marine reservoir effect for Buccaneer Archipelago and adjacent areas. Dates discussed in the text are calibrated in years Before Present (cal y BP) with the 68.2% (2σ) probability range for all dated samples. To provide a neutral sea level datum across all sites, transect data from each site were replotted relative to the mean sea-level (MSL).

9.6 Results Reef Stratigraphy

Bathurst-Irvine Island

Sixteen percussion cores were collected around Bathurst and Irvine Islands, comprising three cores east of Bathurst Island, ten cores from the inter-island fringing reefs and three cores around the deep pools east of Irvine Island (Figure 68). Eleven cores (B04, B05, B06, B07, B08, B08b, B09, B10, B12, B13, B14, and B15) were sub- sampled and radiocarbon dated. Penetration ranged from 0.6 m (B01) to 6.3 m (B05), compaction was 17% (B09) to 38% (B04) and twelve reef facies were distinguished (Table 21).

124 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Irvine core locations. - Irvine Island core logs. Inset map shows Bathurst Bathurst - 68. ure Fig

Kimberley Marine Research Program | Project 1.3.1 125

Reef Growth and Maintenance

Table 21. Reef Facies of Bathurst – Irvine Islands.

1 unconsolidated sand unit with encrusting coralline algae, coral fragments and rhodoliths

2 unconsolidated branching coral floatstone unit with sandy matrix

3 unconsolidated domal coral floatstone unit with sandy matrix

4 unconsolidated branching coral floatstone unit with muddy sand matrix

5 unconsolidated domal coral floatstone unit with muddy sand matrix

6 unconsolidated branching coral floatstone unit with sandy mud matrix

7 unconsolidated branching coral floatstone unit with muddy matrix

8 unconsolidated domal coral floatstone unit with muddy matrix

9 unconsolidated robust branching coral floatstone unit with muddy sand matrix

10 unconsolidated mud unit with minor branching coral

11 coralline algal bindstone unit with minor coral

12 domal coral framestone unit

The reef facies are dominated by a branching coral unit, especially the core sites around the deep pools. Domal coral dominant horizons were also recovered in some sections, especially in the B04 and B05 cores. In estimated order of abundance, coral clasts identified within cores are: Porites cylindrica, Acropora, Porites, Galaxea, Favia, Cyphastrea, Fungia, Montastrea, Psammocora, Goniopora, Lobophyllia, Pavona, Goniastrea, Seriatopora, and Montipora. Matrix sediment is carbonate mud to very coarse sand composed of broken coral, coralline algae fragments, gastropods, shells, and echinoid debris with carbonate contents range from a minimum of 46% (B05) to a maximum of 88% (B15) (Table 24). All cores were Holocene in age; Pleistocene limestone was not recovered due to limitation of the core length (max 6.5 m). However, reef transects at Cockatoo mine-pit, three km east of Bathurst-Irvine Island, recorded a Pleistocene limestone at depths of 16–18 m below mean sea level, dated by AMS to more than 43k y BP, indicating a likely Last Interglacial (LIG) age. The unconformity between this reef and the overlying 15 m thick Holocene reef is marked by a 0.3–1 m layer of haematitic breccia deposited by rockfalls from the cliff face above (Solihuddin et al., 2015).

126 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Sunday Island

Thirteen cores were collected at Sunday Island, comprising 9 percussion cores and one rotary core in the inter- island fringing reef (pool), and two percussion cores and one rotary core from the southern bayhead fringing reef (Hancock Reef) (Figure 69). Six cores (PN1, PN3, PN5, PS1, PS3, and PS4) were sub-sampled and radiocarbon dated. Penetration ranged from 0.35 m (SS1) to 4.5 m (PN5) and compaction of the percussion cores was 17% (PS3) to 44% (PN5). Six distinct reef facies were recorded (Table 22). Table 22. Reef Facies of Sunday Island.

1 unconsolidated sand unit with coralline algae fragments, coral (mainly branching) and rhodoliths

2 unconsolidated branching coral floatstone unit with sandy mud matrix

3 unconsolidated branching coral floatstone unit with sandy matrix

4 unconsolidated domal coral floatstone unit with muddy sand matrix

5 unconsolidated domal coral floatstone unit with sandy matrix

6 rhodoliths dominant coralline algal bindstone with minor coral

The reef facies throughout the sequences share similar percentage composition between branching and domal coral facies unit. Coral clasts identified from cores including, in estimated order of abundance: Acropora, Porites, Cyphastrea, Goniopora, Platygra, Montipora, Porites cylindrica, Favia, Galaxea, Seriatopora, Goniastrea, Lobophyllia, Merulina ampliata, and Pocillopora. Matrix sediment is yellow to brown carbonate mud to coarse sand with abundant shells, Amphiroa algal fragments, coral fragments gastropods, echinoid debris, and benthonic foraminifera, Carbonate content ranged from 74% (PN5) to 86% (PS2), averaging 82% throughout the cores (Table 24).

Kimberley Marine Research Program | Project 1.3.1 127

Reef Growth and Maintenance

Sunday Island core logs. Inset map shows Sunday Island core locations. 69. Figure

128 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Tallon Island

Fifteen cores were collected at Tallon Island, comprising two percussion cores from the western headland fringing reef, four percussion cores, and nine rotary cores in the eastern bayhead fringing reef (Figure 70). Nine cores (WT1, WT2, ET1, ET2, ET7, ET8, ET11, ET12, and ET13) were sub-sampled and radiocarbon dated. Penetration ranged from 0.22 m (ET6) to 6.3 m (WT1) and compactions of the percussion cores was 20 % (WT2) to 34.5 % (WT1). Analysis of the core data revealed 6 distinct reef facies (Table 23) Table 23. Reef Facies of Sunday Island.

1 unconsolidated branching coral floatstone unit with sandy matrix

2 unconsolidated branching coral floatstone unit with muddy sand matrix

3 unconsolidated branching coral floatstone unit with sandy mud matrix

4 unconsolidated branching coral floatstone unit with muddy matrix

5 unconsolidated sand unit with encrusting coralline algae fragments

6 rhodoliths dominant coralline algal bindstone unit with minor coral

The overall reef facies is dominated by a branching coral facies unit of the genera Acropora with muddy matrix at the base to sandy matrix at the top of the sequences (coarsening upward). This characteristics reef facies occurs at both West and East Tallon throughout the sequences. Coral clasts in cores are generally well preserved; permitting identification of 8 coral genera within cores, including: Acropora, Porites cylindrica, Goniopora, Seriatopora, Lobophyllia, Chyphastrea, Pocillopora, and Stylophora. Matrix sediment is carbonate mud to sand, poorly sorted, and composed of coral debris, coralline algae fragments, Amphiroa fragments, broken shells, gastropods, and echinoid with carbonate contents range from 55 % (ET7) to 89% (ET1) (Table 24). Massive coralline algal build-ups occur along the reef margin of the eastern side of Tallon Island forming terraces on the reef crest as a response to macrotidal conditions (Purcell, 2002; Wilson, 2013). This coralline reef framework is limited to the eastern side of Tallon Reef. The framework is predominantly constructed of thick coralline algal crusts with minor coral and is dense and well-lithified, with some primary cavities. Component clasts identified include coralline red algae, vermetid gastropods, encrusting foraminifera, molluscs, corals, and cemented sediments.

Kimberley Marine Research Program | Project 1.3.1 129

Reef Growth and Maintenance . . Tallon Island core logs. Inset map shows Tallon Island core locations core Island Tallon shows map Inset logs. core Island Tallon 70 – 70 –

Figure

130 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Table 24. Average carbonate content percentage in sediment samples.

Sample Depth in Average Average CaCO Average CaCO Location 3 3 code core (m) CO2 weight (gr) percentage (%) Bathurst- B04-20 -0.2 108.25 0.508 80.7897 Irvine B04-90 -0.9 95 1.7849 72.4322 B04-240 -2.4 81.5 0.3534 59.7141 B05-25 -0.25 105.5 0.49465 79.7804 B05-125 -1.25 107.25 0.50315 80.7897 B05-435 -4.35 65.25 0.299675 46.7135 B06-28 -0.28 75.5 0.349325 55.1114 B06-110 -1.1 84.75 0.39415 62.6615 B06-187 -1.87 74 0.340925 53.3349 B06-266 -2.66 33 0.1434 57.3643 B07-50 -0.5 86.75 0.403825 63.8727 B07-165 -1.65 85.75 0.398975 62.823 B07-240 -2.4 76 0.35175 55.2326 B07-365 -3.65 78 0.361425 57.009 B07-450 -4.5 70.5 0.3251 52.2852 B09-20 -0.2 114.5 0.538275 84.262 B09-50 -0.5 105.25 0.49345 78.125 B10-10 -0.1 104 0.487375 76.4696 B10-60 -0.6 117 0.5504 88.4205 B11-20 -0.2 123.75 0.5831 93.5078 B11-115 -1.1 110.5 0.518775 83.6563 B11-170 -1.7 108.5 0.508975 81.7587 B12-110 -1.1 70.75 0.3263 52.4063 B12-220 -2.2 75.25 0.3481 55.1922 B12-490 -4.9 78.5 0.36385 57.4935 B13-40 -0.4 85.75 0.398975 62.6211 B13-93 -0.93 109 0.511625 81.5568 B13-167 -1.67 79.25 0.3675 57.6954 B13-242 -2.42 76.5 0.35415 56.0804 B13-315 -3.15 79.75 0.369925 57.0494 B13-405 -4.05 78.25 0.36265 57.4935 B14-40 -0.4 76.5 0.342075 54.0213 B14-165 -1.65 72.5 0.3348 53.3349 B14-340 -3.4 72.25 0.333575 52.1641 B15-20 -0.2 119 0.55985 88.905 B15-160 -1.6 107.25 0.50315 80.6282

Sunday PN1-3 -0.03 107.5 0.504375 81.1935 Island PN1-42 -0.42 110.25 0.517675 82.4047 PN1-90 -0.9 114.75 0.5395 85.6751 PN1-150 -1.5 107.5 0.50435 80.386 PN1-270 -2.7 113.25 0.532225 84.4234 PN2-14 -0.14 107.5 0.50435 80.4264 PN2-50 -0.5 102 0.4777 75.2584 PN2-90 -0.9 112.75 0.5298 84.3427 PN3-30 -0.3 125.5625 0.508 78.8921 PN3-100 -1 113.75 0.534625 84.9887 PN3-185 -1.85 107.5 0.52935 80.7494 PN4-40 -0.4 103.5 0.484975 76.7926 PN4-76 -0.76 116 0.54555 87.0882 PN5-30 -0.3 100.5 0.470425 74.0875 PN5-140 -1.4 112.75 0.5298 84.9887 PN5-220 -2.2 110.75 0.5201 82.3643 PS1-5 -0.05 115.25 0.541925 86.4826 PS1-40 -0.4 126.125 0.514075 81.1531 PS1-105 -1.05 115.5 0.543125 86.3614 PS2-25 -0.25 115.5 0.54315 86.8863 PS3-23 -0.23 104.75 0.491 77.9231 PS3-70 -0.7 114 0.53585 85.4328

Tallon ET1-20 -0.2 118.25 0.55645 89.4703 Island ET1-84 -0.84 119.75 0.5637 88.8647 ET7-12 -0.12 76.25 0.35295 55.3537 ET7-38 -0.38 100.5 0.47045 74.249 ET7-210 -2.1 108.75 0.510425 80.8705 ET7-358 -3.58 86 0.4002 63.469 ET12-5 -0.05 94.75 0.442575 70.3731 ET12-93 -0.93 114.25 0.537075 85.5943

Kimberley Marine Research Program | Project 1.3.1 131

Reef Growth and Maintenance

Reef geochronology and Growth History It should be noted that the oldest dates recorded in the reefs studied represents the age of the preserved coral and are not necessarily indicative of reef initiation due to the limitations of core barrel length. The Pleistocene foundation was not cored, but can be inferred from seismic records.

Bathurst-Irvine Island

The oldest 14C reef encountered was at ∼3.5k cal y BP at a depth of 1.7 m below MSL (B15). All cores dated near surface horizons (upper 50 cm of cores) returned modern or near modern dates with the exception of the top section of B15, which gave a 14C age of ∼2.2–2k cal y BP at 0.4 m deep below MSL. This indicates that the reef in B15 represents one of the oldest parts of the reef. Vertical accretion rates vary from a maximum of 54.4 mm/year (B12) to a minimum of 0.4 mm/year (B09) with commonly range from 8 to 10 mm/year (Table 25). The highest rate of accretion corresponds to the robust branching coral facies with a muddy matrix at the base of B12 (see Figure 68). The growth history of the reef can be divided into three stages due to differences in the timing, location, and rates of vertical reef growth (Figure 71), including: 1) oldest stage, 2) younger stage, and 3) youngest stage. The oldest stage can be found in B15 on the western side of the inter-island reef. Here, the reef has been dated at ∼3.5k cal y BP at a depth of 1.7 m below MSL with accretion rates up to 1.9 mm/year, suggesting that the Holocene reef in this area grew in shallow water and can be classified as slow growing. Meanwhile, the reef on the eastern side exhibits a younger age, as shown in B05, with an age of ∼3.35k cal y BP at a depth of 6.1 m below MSL (B05) and an accretion of rate up to 4.9 mm/year. This continued until modern time and can be classified as moderate growth. Samples collected around the deep tidal pools show the latest stage of reef establishment and can be classified as rapid growth, occurred at ∼1k cal y BP at a depth of 6.3 m below MSL (B07) with accretion rates up to 54.4 mm/year. The reef is prograding at a steep angle here so the apparent vertical accretion rate will be significantly higher than the actual rate. These records support the hypothesis from Wilson (2013) that the deep tidal pools existing in the inter-island reef flat are spaces left by coalescing reef growth (Figure 72).

Figure 71. Holocene vertical reef accretion curves for Bathurst-Irvine Island.

132 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Irvine Island Irvine island fringing reef of Bathurst - - Full cross section with isochrones and summary of Holocene reef facies for the inter – – 72 Figure

Kimberley Marine Research Program | Project 1.3.1 133

Reef Growth and Maintenance

Sunday Island

Radiocarbon dating indicates that the oldest 14C dates of reef deposition occurred at 6.7k cal y BP at a depth of - 4.7 m below MSL (PN5), dates from close to the reef surface returned modern or near modern dates, ranging from ∼120– >0 cal y BP (PN1, PS3) to ∼545–465 cal y BP (PS1) (Table 25). Accretion rates show some variation of vertical reef growth during Holocene time, giving a range from a maximum of 2.16 mm/year (PN1) to a minimum of 0.28 mm/year (PN1) with an average of 0.99 mm/year (Figure 73). The highest rate of accretion was found at the base of PN1 core which is the nearest core site to the island; containing mainly domal coral in a muddy matrix sediment (see Figure 69).

Figure 73. Holocene vertical reef accretion curve and polynomial curve of best-fit for Sunday Island.

The growth pattern of reefs at Sunday Island did not show an initial rapid accretion but instead exhibited a more uniform accretion across their entire growth history. Also, it appears that the reefs appear to have lagged behind sea level, which is assumed to have been close to present sea level at around 6k cal y BP. Here, the reefs established at about 6k cal y BP at a depth of 4 m and subsequently grew linear uniformly to sea level (Figure

134 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

island reef platform of Sunday Island - Full cross section with isochrones and summary of Holocene reef facies for (A) Northern and (B) Southern inter 74. Figure 74).

Kimberley Marine Research Program | Project 1.3.1 135

Reef Growth and Maintenance

Tallon Island

Nineteen radiometric dates collected from Tallon Island indicate that the oldest date was from the base of WT1 core sequence, giving an age of 7.79k cal y BP at a depth of 8.3 m below MSL (Table 25). Coral dates from the surface horizons are modern or nearly modern in age with the exception of the ET7 core top sequence, which returned a 14C age of ∼4.075–3.87k cal y BP at a depth of 0.4 m below MSL. Vertical accretion rates vary throughout the sequences, ranging from a maximum of 7.02 mm/year (WT1) to a minimum of 0.23 mm/year (ET12) with an average of 1.58 mm/year (Figure 75). The highest rate of accretion occurred in the most landward (WT1 and ET7) core sites at the base of the core sequence which was composed of mainly branching Acropora within a muddy matrix sediment (see Figure 70).

Figure 75. Holocene vertical reef accretion curve and polynomial curve of best-fit for the Tallon Island reefs. The reefs appear to have accreted rapidly following initiation as sea level rose considerably from 7.8–5k cal y BP, producing reef from 8.3 m to 1.0 m deep below MSL with accretion rate of up to 7 mm/year. Following that, the reefs experienced a trend of slowing growth as indicated by dates from 5–2k cal y BP at depths of 1.0–0.75 m below MSL indicating accretion rates of only 0.2–0.4 mm/year. Ultimately, when the reef had almost reached sea-level, coralline algae overtook the reef growth and shifted from vertical (aggradational) to lateral (progradational) growth, moving off the island until modern time. This can be clearly seen in the East Tallon reef with prevailing coralline algae pavement at the edge of the reef platform as shown in Figure 76. The progradation rate of the reef is indicated by a horizontal core (ET11) of coralline algae pavement at the vertical or undercut wall margin of the reef platform, giving an age from 1.32k cal y BP to modern time with an accretion rate of 0.6 mm/year (see Figure 70).

136 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Island reefs.

Full cross section with isochrones and summary of Holocene reef facies for the Tallon 76. Figure

Kimberley Marine Research Program | Project 1.3.1 137

Reef Growth and Maintenance

Table 25. Radiocarbon dates from selected samples across Tallon, Sunday, and Bathurst-Irvine reef.

Sample Conventional Calibrated (68% Accretion Location Depth (m) Lab. Code Material Method Measured Age 13C/12C Name Age (BP) Probability) cal BP (mm/year) Tallon I. ET7-30-0.3 Beta-385991 Acropora AMS 3610 +/- 30 +0.4 o/oo 4030 +/- 30 4075 - 3870 ET7-313 -3.13 Beta-385992 Seriatophora? AMS 5730 +/- 30 -0.3 o/oo 6140 +/- 30 6605 - 6410 1.12 ET7-573 -5.73 Beta-385993 Porites Cylindrica AMS 6360 +/- 30 0.0 o/oo 6770 +/- 30 7305 - 7165 3.57 ET1-24.5 -0.245 Beta-385988 Acropora AMS 170 +/- 30 +0.5 o/oo 590 +/- 30 265 - 55 ET1-74 -0.74 Beta-385989 Porites Cylindrica AMS 1840 +/- 30 -0.4 o/oo 2240 +/- 30 1865 - 1690 0.31 ET1-123 -1.23 Beta-385990 Coralline algae AMS 2630 +/- 30 +1.0 o/oo 3060 +/- 30 2840 - 2715 0.49 ET12-23 -0.23 Beta-389407 Coral AMS 370 +/- 30 0.0 o/oo 780 +/- 30 450 - 285 ET12-132 -1.32 Beta-389408 Coral AMS 4470 +/- 30 +0.7 o/oo 4890 +/- 30 5265 - 5005 0.23 ET12-202 -2.02 Beta-389409 Coral AMS 4760 +/- 30 -0.3 o/oo 5170 +/- 30 5570 - 5420 1.94 ET2-8 -0.08 Beta-391788 Coralline algae AMS 650 +/- 30 +1.8 o/oo 1090 +/- 30 660 - 540 ET2-100 -1 Beta-391789 Coral AMS 2460 +/- 30 -0.1 o/oo 2870 +/- 30 2690 - 2440 0.47 WTI-30 -0.3 Beta-389404 Acropora AMS 140 +/- 30 -1.3 o/oo 530 +/- 30 230 - post 0 WTI-294 -2.94 Beta-389405 Coral AMS 6410 +/- 30 +1.0 o/oo 6840 +/- 30 7395 - 7245 0.37 WTI-573 -5.73 Beta-389406 Coral AMS 6890 +/- 30 +0.9 o/oo 7310 +/- 30 7790 - 7645 7.02 WT2-87 -0.87 Beta-389469 Porites Cylindrica Radiometric 4790 +/- 30 +0.5 o/oo 5210 +/- 30 5805 - 5445 2.32 ET8-12 -0.12 Beta-397764 Rhodolith AMS 1300 +/- 30 -3.2 o/oo 1660 +/- 30 1250 - 1070 ET8-115 -1.15 Beta-397765 Coralline algae AMS 2880 +/- 30 +1.0 o/oo 3310 +/- 30 3175 - 2960 0.54 ET11-10 -0.1 Beta-397766 Coralline algae AMS 0.4 +/- 0.4 pMC +0.4 o/oo 390 +/- 30 Post 0 BP ET11-90 -0.9 Beta-397767 Coralline algae AMS 1330 +/- 30 +1.8 o/oo 1770 +/- 30 1320 - 1220 0.63

Sunday I. PN3-13 -0.13 Beta-389410 Acropora AMS 130 +/- 30 +0.3 o/oo 540 +/- 30 240 - post 0 PN3-120 -1.2 Beta-385994 Acropora AMS 1670 +/- 30 +0.3 o/oo 2080 +/- 30 1685 - 1515 0.72 PN3-319 -3.19 Beta-385995 Platygra AMS 2710 +/- 30 -3.0 o/oo 3070 +/- 30 2850 - 2725 1.68 PN1-25 -0.25 Beta-389470 Acropora Radiometric 90 +/- 30 -1.2 o/oo 480 +/- 30 120 - post 0 PN1-122 -1.22 Beta-389471 Pavia Radiometric 3220 +/- 30 +1.3 o/oo 3650 +/- 30 3570 - 3395 0.28 PN1-308 -3.08 Beta-389472 Coral Radiometric 4040 +/- 30 -2.9 o/oo 4400 +/- 30 4565 - 4395 1.86 PN1-388 -3.88 Beta-389473 Rhidoliths Radiometric 4300 +/- 30 -2.1 o/oo 4680 +/- 30 4905 - 4795 2.16 PN5-169 -1.69 Beta-389474 Porites Radiometric 1630 +/- 30 -1.1 o/oo 2020 +/- 30 1595 - 1415 PN5-271 -2.71 Beta-389475 Galaxea Radiometric 4420 +/- 30 -1.2 o/oo 4810 +/- 30 5115 - 4865 0.29 PN5-410 -4.1 Beta-389476 Porites Radiometric 5900 +/- 30 -0.2 o/oo 6310 +/- 30 6780 - 6625 0.81 PS1-20 -0.2 Beta-385996 Coral AMS 540 +/- 30 -0.1 o/oo 950 +/- 30 545 - 465 PS1-184 -1.84 Beta-385997 Coralline algae AMS 2800 +/- 30 +1.7 o/oo 3240 +/- 30 3070 - 2870 0.67 PS3-9 -0.09 Beta-389411 Acropora? AMS 80 +/- 30 -0.7 o/oo 480 +/- 30 120 - post 0 PS3-90 -0.9 Beta-391787 Acropora? AMS 1570 +/- 30 +1.0 o/oo 2000 +/- 30 1565 - 1400 0.39 PS4-100 -1 Beta-385998 Coral AMS 4160 +/- 30 +1.3 o/oo 4590 +/- 30 4825 - 4625

Bathurst- B04-30 -0.3 Beta-389477 Coralline algae Radiometric 140 +/- 30 +1.1 o/oo 570 +/- 30 255 - post 0 Irvine I. B04-130 -1.3 Beta-389479 Favia Radiometric 490 +/- 30 -1.0 o/oo 880 +/- 30 505 - 415 2.42 B04-427 -4.27 Beta-389480 Porites? Radiometric 2370 +/- 30 +4.4 o/oo 2850 +/- 30 2680 - 2395 1.43 B05-16 -0.16 Beta-389413 Galaxea AMS 160 +/- 30 -2.2 o/oo 530 +/- 30 230 - post 0 B05-160 -1.6 Beta-389414 Galaxea AMS 740 +/- 30 -1.0 o/oo 1130 +/- 30 680 - 565 2.84 B05-385 -3.85 Beta-389415 Galaxea AMS 2690 +/- 30 +0.5 o/oo 3110 +/- 30 2900 - 2740 1.02 B05-600 -6 Beta-389416 Porites Cylindrica AMS 3070 +/- 30 -1.7 o/oo 3450 +/- 30 3350 - 3165 4.91 B08-12 -0.12 Beta-389417 Acropora AMS 140 +/- 30 -0.3 o/oo 550 +/- 30 245 - post 0 B08-81 -0.81 Beta-389418 Coralline algae AMS 270 +/- 30 +0.5 o/oo 690 +/- 30 320-240 4.38 B08b-35 -0.35 Beta-389481 Coralline algae Radiometric 360 +/- 30 -1.7 o/oo 740 +/- 30 420-265 B08b-64 -0.64 Beta-389482 Coralline algae Radiometric 390 +/- 30 +2.2 o/oo 840 +/- 30 490-325 4.46 B09-12 -0.12 Beta-389483 Porites Radiometric 70 +/- 30 +0.3 o/oo 480 +/- 30 120 - post 0 B09-30 -0.3 Beta-389484 Coralline algae Radiometric 580 +/- 30 +0.4 o/oo 1000 +/- 30 615-415 0.40 B09-114 -1.14 Beta-389485 Porites Radiometric 2270 +/- 30 +0.4 o/oo 2690 +/- 30 2365 - 2275 0.47 B12-35 -0.35 Beta-389419 Acropora AMS 90 +/- 30 -0.5 o/oo 490 +/- 30 130 - post 0 B12-238 -2.38 Beta-389420 Acropora AMS 220 +/- 30 -0.2 o/oo 630 +/- 30 285 - 120 14.76 B12-367 -3.67 Beta-389421 Acropora AMS 360 +/- 30 -1.0 o/oo 750 +/- 30 425 - 270 8.90 B12-571 -5.71 Beta-389422 Coral AMS 430 +/- 30 -2.1 o/oo 810 +/- 30 470 - 300 54.40 B14-16 -0.16 Beta-389423 Favia AMS 110 +/- 30 +2.6 o/oo 560 +/- 30 250 - post 0 B14-154 -1.54 Beta-389424 Coral AMS 280 +/- 30 -1.0 o/oo 670 +/- 30 305 - 190 11.50 B14-325 -3.25 Beta-389425 Coral AMS 340 +/- 30 +1.4 o/oo 770 +/- 30 440 - 280 14.87 B14-554 -5.54 Beta-389426 Coral AMS 500 +/- 30 -1.1 o/oo 890 +/- 30 510 - 420 21.81 B15-54 -0.54 Beta-391046 Coral AMS 2060 +/- 30 +3.3 o/oo 2520 +/- 30 2200 - 2005 B15-101 -1.01 Beta-391047 Coral AMS 2830 +/- 30 -0.7 o/oo 3230 +/- 30 3060 - 2860 0.55 B15-189 -1.89 Beta-391048 Coral Radiometric 3190 +/- 30 -0.4 o/oo 3590 +/- 30 3500 - 3345 1.90 B13-18 -0.18 Beta-391043 Coral AMS 120 +/- 30 -2.5 o/oo 490 +/- 30 130 - post 0 B13-321 -3.21 Beta-391044 Coral AMS 560 +/- 30 -3.4 o/oo 910 +/- 30 520 - 435 7.35 B13-597 -5.97 Beta-391045 Coral AMS 700 +/- 30 -1.5 o/oo 1090 +/- 30 660 - 540 22.53 B07-32 -0.32 Beta-391040 Coral AMS 110 +/- 30 -1.8 o/oo 490 +/- 30 130 - post 0 B07-316 -3.16 Beta-391041 Coral AMS 520 +/- 30 -0.2 o/oo 930 +/- 30 530 - 450 6.68 B07-620 -6.2 Beta-391042 Coral AMS 1100 +/- 30 -1.1 o/oo 1490 +/- 30 1050 - 910 6.20 B06-43 -0.43 Beta-391037 Coral AMS 80 +/- 30 -1.6 o/oo 460 +/- 30 75 - post 0 B06-294 -2.94 Beta-391038 Coral AMS 640 +/- 30 -0.8 o/oo 1040 +/- 30 635 - 510 4.71 B06-434 -4.34 Beta-391039 Coral Radiometric 880 +/- 30 -0.6 o/oo 1280 +/- 30 855 - 680 7.18 All samples were dated at the Beta Analytic Inc, Miami, Florida USA

138 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

9.7 Discussion Reconstruction of Holocene sea-level and growth history The Kimberley Bioregion of the Northwest Continental Shelf of Australia is considered to have been relatively tectonically stable during the Quaternary (Collins et al., 2006) and has been categorized as a “far-field” location (far away from a former ice sheet region), providing an excellent location for observing sea level events and undertaking palaeo-shoreline reconstruction. In such environments, the effects of seafloor compaction, subsidence and hydro-isostatic (melt water release from the ice sheets) compensation are negligible during the relatively short time interval of thousands of years (Lambeck et al., 2002; Wong et al., 2003). Studies on Holocene sea level and reef growth history near the Kimberley Shelf have been carried out on reef core data from Morley Island in the Abrolhos, which resembles a ‘keep-up’ reef (Collins et al, 1993; Eisenhauer et al., 1993), cemented coral shingle pavement from the Abrolhos reefs (Collins et al., 2006), reef core data from Scott Reef (Collins et al., 2011). The most relatively recent studies (Solihuddin et al., 2015) demonstrated recent dataset from a number of coral samples in the Cockatoo Island mine-pit which, for the first time, revealed Holocene coral reef growth, including chronology and growth history in the inshore Kimberley Bioregion. The data provided Holocene sea-level records characterised by moderate rates of reef growth and suggested tectonic subsidence of 0.1 mm/year. Incorporating new data obtained from this study with existing reef records from and adjacent to the Kimberley region incorporating new data obtained from this study provide a composite sea-level record with a modified Houtman Abrolhos sea-level curve (Figure 77). The radiocarbon age of each dated core sample is plotted against depth below present sea-level (bpl). The curve indicates that the Holocene sea-level transgression and reef growth history in the inshore Kimberley bioregion was initiated at ∼10k cal y BP when sea level was estimated to be around 20 m bpl. At this time, the rising sea-level drowned the Kimberley Shelf and provided different substrates and foundation types for reef growth, marking the initiation of reef deposits and accumulation. Late Holocene highstands at ∼6k y BP are recorded at the microtidal Abrolhos Reefs in southwest Australia, where there has been little or no subsidence. There is no similar highstand recorded at Scott Reef in the oceanic shoals or the reefs in the inshore Kimberley such as Tallon reef, Sunday reef, Bathurst-Irvine reef, and Cockatoo reef, suggesting ongoing Holocene subsidence. Finally, sea level gradually fell returning to its present-day level at nearly modern time (approx. 1k y BP) (Figure 77).

Kimberley Marine Research Program | Project 1.3.1 139

Reef Growth and Maintenance

Figure 77. Composite sea level records for Kimberley Shelf from the last 10,000 years BP derived from coral in core from Tallon reef, Sunday reef, Bathurst-Irvine reef (this study), coral exposure from Cockatoo Island mine-pit (Solihuddin et al., 2015), coral in core from Scott Reef (Collins et al., 2011), cemented coral shingle pavement from Abrolhos reefs (Collins et al., 2006), and corals in core from Morley Island Abrolhos (Eisenhauer et al., 1993). Chronology of Holocene reef build-up Our dataset indicates two discrete periods of reef initiation and growth in the inshore Kimberley bioregion: 1) during the Holocene transgression – Early highstand between 8k and 5k y BP, and 2) as sea level stabilised between 2k y BP and the present (Twiggs and Collins 2010; Perry and Smithers, 2010; 2011). Lambeck et al. (2010) and Solihuddin et al. (2015) reported that the establishment of coral reefs in the inshore Kimberley bioregion closely followed flooding of the antecedent substrate by sea level transgression at around 10k y BP at -20 m elevation. However, the timing of reef initiation and rate of vertical reef growth vary within reef variation depending mainly on the availability of accommodation, substrate, internal structure and composition of the reef. At Bathurst-Irvine Island, the reef growth style is based on cross-reef age isochrons profiles and field observations (see Figure 72), revealing that the reefs accreted vertically until near sea level and subsequently prograded. The ages of coral material preserved around the vertical or overhanging wall reef margin show modern or near modern time dates (<1k y BP). Compared to other reef systems investigated in this study such as Tallon and Sunday Island, the reef at Bathurst-Irvine Island shows the highest rate of accretion, and slightly higher accretion than the reef at Cockatoo Island (Solihuddin et al., 2015). At Sunday Island, there are distinct morphological features in the reef between the south and north of the island reef platform which is characterised by terraced coralline algal pavement over or near the reef crest at the south and gently sloping or no reef crest at the north. Also, the internal structure and composition of reef assemblages is different between the coral structure dominant facies in the north and the unconsolidated carbonate sand dominant in the south. The chronology of the Holocene reef build-up in the north inter-island reef began at ∼6k cal y BP at about 4 m below MSL (PN5). Following this, the reefs aggraded vertically catching up to sea-level with

140 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance fairly linear accretion rates until their current position. Similarly, reefs in the southern part also share this growth pattern with the north; with vertical aggradation starting from ∼4.8k cal y BP at a depth of 1.8 m below MSL (PS4). However, the reef profile is not perpendicular to the shoreline. Such reef growth corresponds to the catch- up reef growth mode described by Neumann and Macintyre (1985). The chronology of Holocene reef build-up at Tallon Island started at ∼7.79k cal y BP at a depth of 8.7 m below MSL and is represented by the west Tallon (WT1) reef core. Similarly, dated coral material in the eastern side has an age of ∼7.3k cal y BP at a depth of 5.1 m below MSL, indicating that the reef build-up at both west and east Tallon Island developed at approximately the same time and elevation. Subsequently, both reefs aggraded vertically to near sea level and reached an elevation of 3.8 m below MSL at ∼5.8k y BP and 1.4 m below MSL at 5.6k y BP for the west and east side respectively. Finally, both west and east reefs experienced slowing growth as the intertidal was exposed and ceased growing vertically at an elevation of 3.3 m below MSL at ∼230 y BP– modern time and <0.5 m below MSL at ∼4k y BP respectively, suggesting a considerable separation in vertical reef cessation time and elevation between them (west and east). This was interpreted as different environmental conditions controlling reef growth between west and east Tallon reef. As we currently lack water quality data including sea surface temperature, nutrient levels, turbidity, sea surface salinity as well as tectonic and antecedent reef platform, we can only propose hydrodynamic energy, particularly tidal currents, as a major control on different Holocene reef accretion (Larcombe and Carter, 2004). This hypothesis is supported by the existence of a massive terraced coralline algal pavement over or near the reef crest of the eastern side of Tallon Reef which is strongly controlled by hydrodynamic energy (Gherardi and Bosence, 2001) and has grown laterally until modern time. Also, a strong tidal current which is coming from the southeast, has accelerated coralline algal development such that is has overtaken vertical reef growth and cessation. The framework is predominantly constructed of thick coralline algal crusts with minor coral and is dense and well-lithified, with some primary cavities. The rapid growth at Tallon Reef in its initial stage can be classified as a “keep-up” phase following the description of Neumann and Macintyre (1985). In the following stage, the reef experienced a slowing growth that can be classified as a “catch-up” phase. When the reef reached sea-level, the reef growth shifted from vertical to lateral growth with a low reef progradation rate. A shift from keep-up to catch-up reef growth has been interpreted as the reef reaching MLW at 1.3 m below the surface due to sea level stabilisation during the Mid Holocene–Late Highstand before gradually regressing until its recent level. Supporting data that characterises catch-up reef growth is the abundance of branching Acropora throughout the sequence and the coarsening upward (muddy to sandy) of the matrix sediments of the sequence (Solihuddin et al., 2015). Rate and mode of Holocene reef accretion Montaggioni (2005) suggested that the accretion rates in reef environments are controlled by the nature of the framework facies. An estimation of average reef accretion rates within each stage has been determined using reef accumulation (vertical reef accretion) and isochrones data. The data shows that the Holocene reef communities experienced rapid growth where they correspond with branching Acropora in muddy matrix. This can be clearly seen at Bathurst and Tallon Reef both in the reef stratigraphy and isochrones profiles, exhibiting a Holocene reef predominantly constructed by Acropora, Porites and Goniopora, all of which are tolerant to high sedimentation and turbidity (Done 1982; Smithers et al., 2006). In contrast, the modern reef communities on the reef flat, where exposed on low spring tides, are typically dominated by corals including Goniastrea aspera, Favia, and Montipora digitata. While the subtidal areas, such as the forereef slope, tend to be dominated by fast- growing branching and plate corals, such as Acropora and Montipora which are tolerant of higher wave energy (Done et al., 2007; Brown et al., 2010). Cross-reef age isochrons suggest that the reef initiated as a blanket close to the shore and grew upward before prograding seaward when the reef approached sea level under high, stable and then slowly falling sea levels, before ‘turning off’. This growth style corresponds to the type A and B respectively of fringing reef growth and morphology from Kennedy and Woodroffe’s (2002).

Kimberley Marine Research Program | Project 1.3.1 141

Reef Growth and Maintenance

During the Holocene highstand at around 6k y BP, the sea level was stable and bioclastic sands, with a high proportion of terrestrial detritus, began to accumulate and prograde seaward over the reef framework. This is likely to have led to an increase in coastal turbidity and decline in water quality as identified in the Kimberley regions (Macintyre, 1988; Kleypas, 1996; Smithers and Larcombe, 2003). Also, a high proportion of alluvial material fills interstices in the framework, suggest that severe storm/cyclone events resulted in episodic terrigenous sediment accumulation and periods of high and prolonged turbidity similar to those that occur today. This would have contributed to the demise of an already stressed community. In areas of macrotidal influence, which generally generates high levels of turbidity, coral reef growth and initiation are usually constrained by light penetration into the water column, so that reefs may often only survive in a very shallow marine environment (<5 m). However, the growth morphologies and accretionary histories of the reefs investigated reveal that these coral assemblages have an ability to rapidly cope with the effects of sediment ‘stress’ (Solihuddin et al., 2015). In addition, considering the mud-rich environment, reefs in the inshore Kimberley bioregion have remarkable accretion rates.

9.8 Conclusion This core study provides the first subsurface sedimentary samples for the key types of reef found in the southern Kimberley and sheds considerable light on their growth history. It has shown that the reefs are muddy in character, similar to the Cockatoo island fringing reefs. Adele Reef offshore is distinctly sandier in character. Geochronology demonstrates the recent Holocene age of reef top cores. The massive coralline algal structure is the most striking feature of the reef in this area and terraces reef flat near or over the reef crest is common.

Early reef buildups are likely to be muddy with branching Acropora and are corresponding to the rapid growth on Holocene reef communities. In contrast, the modern reef communities where exposed on low spring tides are typically dominated by Goniastrea aspera, Favia, and Montipora digitate, while the subtidal areas tend to be dominated by fast-growing branching and plate corals, such as Acropora and Montipora. The data shows that at Bathurst-Irvine, the reef growth, but not initiation, occurred at ∼3.4k years BP at a depth of 1.9 m. Sunday Island at ∼5k years BP at a depth of 4 m and at Tallon Island at ∼7.8k years BP at the depth of 6 m. By comparison, Cockatoo fringing reef was initiated by ∼8.9k years BP at a depth of 18 m while reef initiation at the deeper oceanic shoals, more distal Scott Reef was at ∼11.3k years BP, characterized by moderate rates of rise of 10 mm/year. All available sea level proxies are limited by the macrotidal (up to 11 m) tides, and a propensity to represent catch-up rather than keep-up conditions.

9.9 References

Adey W.H., 1978. Coral reef morphogenesis: a Lambeck, K., Woodroffe, C. D., Antonioli, F., Anzidei, M., multidimensional model. Science 202: 831-837. Gehrels, W.R., Laborel, J., Wright, A. J., 2010. Paleoenvironmental records, geophysical modelling, Adey W.H., 1986. Coralline algae as indicators of sea level. and reconstruction of sea-level trends and variability In: van de Plassche O (ed) Sea-level research: a manual on centennial and longer timescales, in: Church, J.A., for the collection and evaluation of data. Geo Books, Woodworth, P.L., Aarup, T., Wilson, W.S., (ed), Amsterdam, pp. 229-280. Understanding sea level rise and variability, John Bosence, D.W.J., 1983. Description and Classification of Wiley & Sons. Copyright, pp. 61–105. Rhodoliths (Rhodoids, Rhodolites). In Peryt, T.M. (Ed.) Larcombe, P., Carter, R.M., 2004. Cyclone pumping, Coated Grains. Springer-Verlag, Berlin, pp. 218–224. sediment partitioning and the development of the Brooke, B., 1997. Geomorphology of the north Kimberley Great Barrier Reef shelf system: a review. Quaternary coast, in: Walker D. (Ed.), Marine biological survey of Science Reviews 23, 105–135. the central Kimberley coast. Western Australia. Lough, J. M., 1998. Coastal climate of northwest Australia University of Western Australia, Perth, unpublished and comparisons with the Great Barrier Reef: 1960 to report, W.A. Museum Library No. UR377, pp. 13–39. 1992. Coral Reefs 17(4): 351–367. Browne, N.K., Smithers, S.G., Perry, C.T., 2010. Geomorphology and community structure of Middle

142 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Reef, central Great Barrier Reef, Australia: an inner- Macintyre, I.G., 1988. Modern coral reefs of western shelf turbid zone reef subjected to episodic mortality Atlantic new geological perspective. American events. Coral Reefs 29, 683–689. Association of Petroleum Geologists Bulletin 72, 1360–1369. Collins, L. B., Zhao, J. X., Freeman, H., 2006. A high-precision record of mid–late Holocene sea-level events from Montaggioni, L.F., 2005. History of Indo-Pacific coral reef emergent coral pavements in the Houtman Abrolhos systems since the last glaciation: Development Islands, southwest Australia. Quaternary International patterns and controlling factors. Earth-Science 145–146(0): 78–85. Reviews 71(1-2), 1–75. Collins, L. B., Zhu, Z. R., Wyrwoll, K. H., Hatcher, B. G., Müller, G., & Gastner, M., 1971.The 'Karbonat-Bombe', a Playford, P. E., Eisenhauer, A., Chen, J. H., Wasserburg, simple device for the determination of carbonate G. J. & Bonani, G., 1993. Holocene growth history of a content in sediment, soils, and other materials. Neues reef complex on a cool-water carbonate margin: Jahrbuch für Mineralogie-Monatshefte 10, 466–469. Easter Group of the Houtman Abrolhos, Eastern Indian Neumann, A.C., Macintyre, I., 1985. Reef response to sea Ocean. Marine Geology, 115(1-2), 29–46. level rise: keep-up, catch-up or give-up. Proceedings Collins, L. B., Testa, V., Zhao, J., and Qu,D., 2011. Holocene of the 5th International Coral Reef Congress, pp. 105– growth history and evolution of the Scott Reef 110. carbonate platform and coral reef. Journal of the Pearce, A.F., Griffiths, R.W., 1991. The Mesoscale Structure Royal Society of Western Australia 94(2), 239–250. of the Leeuwin Current: A Comparison of Laboratory Davies, P.J., Marshall, J. F., & Hopley, D., 1985. Relationship Models and Satellite Imagery. J. Geophys. Res., between reef growth and sea level in the Great Barrier 96(C9), 16739-16757. Reef. Proceedings Fifth International Coral Reef Perry, C. T., 2003. Coral reefs in a high-latitude, siliciclastic Congress Tahiti, 3, 95-103. barrier island setting: reef framework and sediment Done, T.J., 1982. Patterns in the distribution of coral production at Inhaca Island, southern Mozambique. communities across the central Great Barrier Reef. Coral Reefs 22, 485–497. Coral Reefs 1, 95–107. Perry, C. T., Smithers, S.G., Palmer, S.E., Larcombe, P., Done, T.J., Turak, E., Wakeford, M., DeVantier, L., Johnson K.G., 2008. 1200 year paleoecological record McDonald, A., Fisk, D., 2007. Decadal changes in of coral community development from the turbid-water coral communities at Pandora Reef: loss terrigenous inner shelf of the Great Barrier Reef. of resilience or too soon to tell? Coral Reefs 26, 789– Geology 36(9): 691–694. 805. Perry, C.T., Smithers, S.G., 2010. Evidence for the episodic Embry, A. F., Klovan J. E., 1971. A late Devonian reef tract “turn on” and “turn off” of turbid-zone coral reefs on northeastern Banks Island, N.W.T. Bulletin of during the late Holocene sea-level highstand. Geology Canadian Petroleum Geology 19(4): 730–781. 38, 119–122. Eisenhauer, A., Wasserburg, G., Eisenhauer J. A., Chen, Perry, C.T., Smithers, S.G., 2011. Cycles of coral reef ‘turn- J.H., Bonani, G., Collins, L.B., Zhu, Z.R., Wyrwoll, K.H., on’, rapid reef growth and ‘turn-off’ over the past 1993. Holocene sea-level determination relative to 8500 years: a context for understanding modern the Australian continent: U/Th (TIMS) and 14C (AMS) ecological states and trajectories. Global Change dating of coral cores from the Abrolhos Islands. Earth Biology 17, 76–86. and Planetary Science Letters 114(4): 529–547. Purcell, S. 2002. Intertidal reefs under extreme tidal flux in Twiggs, E.J., Collins, L. B., 2010. Development and demise of Buccaneer Archipelago, Western Australia. Coral a fringing coral reef during Holocene environmental Reefs 21(2): 191-192. change, eastern Ningaloo Reef, Western Australia. Smithers, S., Larcombe, P., 2003. Late Holocene initiation Marine Geology 275, 20–36. and growth of a nearshore turbid-zone coral reef: Fang, F., Morrow, R., 2003. Evolution, movement and decay Paluma Shoals, Central Great Barrier Reef, Australia. of warm-core Leeuwin Current eddies. Deep Sea Coral Reefs 22, 499–505. Research Part II: Topical Studies in Oceanography Smithers, S.G., Hopley, D., Parnell, K.E., 2006. "Fringing and 50(12–13): 2245–2261. Nearshore Coral Reefs of the Great Barrier Reef: Gherardi D.F.M., 1996. Recent carbonate sedimentation on Episodic Holocene Development and Future the coralline-algal Atol das Rocas, equatorial South Prospects." Journal of Coastal Research: 175–187. Atlantic, Brazil. PhD Diss, Royal Holloway University of Solihuddin, T., Collins, L. B., Blakeway, D., O”Leary, M.J., London, Engham, Surrey, pp 1-353. 2015. Holocene coral reef growth and sea level in a Gherardi, D.F.M., Bosence, D.W.J., 1999. Modeling of the macrotidal, high turbidity setting: Cockatoo Island, ecological succession of encrusting organisms in Kimberley Bioregion, northwest Australia. Marine Recent coralline-algal frameworks from Atol das Geology 359: 50–60. Rocas, Brazil. Palaios 14: 145-158. Wilson, B. R., Blake, S., Ryan, D., Hacker, J., 2011. Gherardi D.F.M, Bosence D.W.J., 2001. Composition and “Reconnaissance of species-rich coral reefs in a community structure of the coralline-algal reefs from muddy, macro-tidal, enclosed embayment, Talbot

Kimberley Marine Research Program | Project 1.3.1 143

Reef Growth and Maintenance

Atol das Rocas, South Atlantic, Brazil. Coral Reefs 19: Bay, Kimberley, Western Australia.” Journal of the 205–219. Royal Society of Western Australia 94: 251–265. Griffin, T. J., Grey, K., 1990. Kimberley Basin. In: Memoir 3, Wilson, B. R., S. Blake., 2011. "Notes on the origin and Geology and Mineral Resources of Western Australia. biogeomorphology of Montgomery Reef, Kimberley, Perth, Geological Survey of Western Australia, pp. Western Australia." Journal of the Royal Society of 293–304. Western Australia 94: 107–119. Hubbard, D. K., 2011. Coral drilling. In: Hopley, D., (ed) Wilson, B.R., 2013. The Biogeography of the Australian Encyclopedia of Modern Coral Reefs: structure, form North West Shelf: Environmental Change and life’s and process. Encyclopedia of Earth Sciences. Springer, response. Elsevier, Burlington MA, USA. Dordrecht, The Netherlands, 856-869. Wong, H.K., Haft, C., Paulsen, A.M., Lüdmann, T., Hübscher, Kennedy, D.M., Woodroffe, C.D., 2002. Fringing reef growth C., Geng, J., 2003. Late quaternary sedimentation and and morphology: a review. Earth Science Review 57, sea level fluctuations on the northern Sunda Shelf, 255–277. southern South China Sea. In: Sidi, F.H., Nummedal, D., Imbert, P., Darman, H., Posamentier, H.W. (Eds.), Kleypas, J.A., 1996. Coral reef development under naturally Tropical Deltas of Southeast Asia – Sedimentology, turbid conditions: fringing reefs near Broad Sound, Stratigraphy, and Petroleum Geology: Society Australia. Coral Reefs 15, 153–167. Economical Palaeontologists Mineralogists Special Kordi, Moataz, Collins, L. B., O’Leary, M., and Stevens, A., (in Publication, 76, pp. 200–234. Review). Mapping and Geomorphic Classification of Wright, R.L., 1964. Geomorphology of the West Kimberley the Kimberley Reefs, North West Australia. Area. CSIRO Land Research, Series 9, 103–118. Lambeck, K., Yokoyama, Y., Purcel, A., 2002. Into and out of the Last Glacial Maximum: sea-level change during Oxygen Isotope Stages 3 and 2. Quaternary Science Reviews, 21(1–3): 343-360.

9.10 Adele Island First described by Teichert and Fairbridge (1948), Adele Reef is an isolated mid-shelf platform reef with more recent study by Richards et al (2013) and Wilson (2013). Wilson (2013) describes the reef platform as, “a bioherm . . . [whose] top is a cap of Quaternary reefal limestone, built on an inundated rocky hill of the dissected Kimberley Basin margin over a Proterozoic basement.” Prevailing westerly wind, swell and diurnal tides, with up to 11 m range on high-water springs, have shaped the intertidal reef platform (rampart) that is heavily etched with small drainage channels and low, widely spaced ridges running parallel to the reef edge that resemble a series of long low and wide corrugations. Beyond the rampart, the reef slopes into sublittoral fore-reef and back-reef zones featuring hard corals, soft corals, hydroids, bryozoans and macroalgae, such as Sargassum (Richards et al. 2013). The terraced nature of the reef platform is striking, from Adele Island and the surrounding sand spit, along the long axis of the platform, inner and outer terraced reef flat, to low gradient forereef slope. The inner reef flat becomes fully exposed at low tide. A prominent 6 km long channel in the NE is deeply incised into the reef flat coral substrate and is up to 30 m deep, with an entrenched network of tributaries. The height of the reef platform is up to 60 m but the platform margin has a relatively low gradient characterised by numerous pinnacle reefs which rise from the platform surface to heights of 15 m. In summary Adele Reef is a multi-stage reef build-up reflecting periods of reef growth during interglacial periods of rising sea level, and erosional processes during glacial periods of low sea levels, probably on a 100,000 year periodicity as expected from the Marine Isotope Curve and now demonstrated by SBP surveys.

144 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Figure 78. Adele Island showing planned coring locations. SBP survey in red.

Kimberley Marine Research Program | Project 1.3.1 145

Reef Growth and Maintenance

Figure 79. Adele Island showing actual coring locations. Table 26. Table of Adele Island Coring locations.

Coordinates (deg min sec) Core Id Penetration (m) Latitude Longitude

A03 15 28 36.084 123 7 46.05 5.06

A04 15 26 14.862 123 6 56.982 3.1

A05 15 26 40.608 123 6 36.732 2.27

A06 15 28 37.344 123 7 18.522 4.14

A07 15 30 20.244 123 9 22.65 3.47

A08 15 33 33.03n 123 10 49.5 4.27

A09 15 34 16.692 123 10 56.202 1.84

A10 15 34 54.474 123 10 56.712 2.25 Note: site A01 and A02 were too deep to core

146 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Ecological Characteristics The large NW-SE planar reef has a beach ridge island in the centre surrounded by a sand cay in the middle of the reef platform. The reef flat formed terraces as a response to the macrotidal regime (± 6 m) and these are mostly occupied by rhodolith banks. In some particular areas, the rhodoliths are trapped on sand banks and/or overlie encrusting coralline algae. The reef flats are covered by bioclastic sand, macroalgae (Sargassum), live coral (Goniastrea Aspera dominant) and encrusting coralline algae. In the southern part, coral rubble is abundant on reef flat and is exposed during low tide. Macroalgae, predominantly Sargassum, are abundant on crustose algae and coral rubble substrates. In the northern part, the reef flat is submerged even at the low tide with the exception of a mounded sand cay which is exposed at all times. Core Summary Eight percussion cores were taken at Adele Island (Figure 80). Core locations are shown in Table 26 and Figure 79 above. Generally, Adele cores contain abundant coral clasts dominated by domal coral of the genera of Porites and Faviids, but including many other genera. Coral clasts are mostly coated by encrusting coralline algae and are generally not in place/in situ based on orientation criteria. Sediments throughout the section are dominated by sand with upward coarsening

Kimberley Marine Research Program | Project 1.3.1 147

Reef Growth and Maintenance structure from muddy sand to coarse sand.

Adele Island Core Logs Core Island Adele 80. Figure

148 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

10 Management Implications and Recommendations

See discussion in Chapters 4-9 for detailed information. • A scheme of reef classification, with GIS database of habitats, (ReefKIM) prepared from RS data which includes georeferenced data and location of every significant reef in the Kimberley (Landsat resolution =30 m). • The ReefKIM database can be used as a primary platform for further reef census data • ReefKIM already includes detailed substrate & geomorphology maps for 30 Kimberley reefs providing preliminary data on key habitats and the relative significance on the Kimberley Reefs • Many Kimberley reefs are atypical and the ReefKIM database can be used to highlight those with a high conservation value • The database was designed to be developed in collaboration with other regional and national institutions as well as individuals through a web-based map. The crowdsourcing approach allows many people already in the field, such as researchers, rangers, fishermen, tourists and traditional owners, to become involved in mapping and share their valuable knowledge. • First detailed shallow seismic and shallow coring undertaken on reefs the Western Kimberley • Holocene sea level rise curve from Cockatoo, low stand channels, and drowning history model. • Short cores in 2014 assessed reef community changes as reefs grew up into tidal zone, ~last 1-8,000 years • Seismic and reef core data show Kimberley reefs to be robust on geological timescales i.e., record reefs surviving and evolving within ever changing environments • Findings Suggests that despite extreme environmental conditions including; (1) high turbidity and sediment input, (2) elevated water temperatures, (3) extreme tides, (4) significant subaerial exposure during low tides, and (5) frequent cyclones, reefs can endure, survive and thrive in the Kimberley

11 Conclusions

During 2015 work on the GIS geodatabase ‘ReefKIM’ was finalised, analysis of the SBP data was completed and analysis of the findings of the coring fieldwork was ongoing. Over the life of the project 8 Milestones; GIS development commenced and in progress, Complete Project Planning and Fieldwork Logistics, Complete Fieldwork for 2013 (seismic), Annual Data Analysis Completed (GIS Stage 1), Complete Fieldwork (coring), Annual Data Management (GIS Stage 2), Finalise Data Analysis, Final Report (GIS Stage 3) were achieved. See conclusions in chapters 4-9 for detailed conclusions.

11.1 Reef Mapping and Geomorphology • Kimberley reefs are geographically and geomorphically complex and atypical when compared to other reef systems • The Kimberley reef bioregion, although 1/3 geographic size of the GBR, is almost equivalent in magnitude in terms of the total number of reef and islands • Kimberley reef morphology data provides a firm scientific foundation for biodiversity assessment and reef structure comparisons. • ReefKIM provides a dynamic geodatabase of Kimberley Reefs with 30 reefs mapped in detail • ReefKIM constitutes a significant decision-making and management tool.

11.2 Kimberley Reef Seismic Stratigraphy • First Seismic study in the southern Kimberley • Supports reef core data - Kimberley reefs are significant geological structures

Kimberley Marine Research Program | Project 1.3.1 149

Reef Growth and Maintenance

• Holocene reef growth initiated over antecedent highs particularly older fossil reefs • Evidence reef growth occurred during past interglacial periods and therefore not unique to the Holocene – at least 3 stages of growth offshore, 2 onshore. • Long term resilience – success over challenging environmental conditions and persistence through oscillating sea levels.

11.3 Reef geology, stratigraphy and evolution • The Cockatoo Reef study provides the first information on Holocene reef growth for a nearshore reef of the Kimberley Biozone. • Core study collected the first subsurface sedimentary samples for the key types of reef found in the southern Kimberley & sheds considerable light on the growth history of these muddy reefs • Initiation of coral reef growth in the Kimberley occurred very soon after post glacial flooding of the continental shelf, between 7,000 and 8,500 years ago • Inshore reefs were able to sustain growth despite highly turbid waters, as indicated by muddy matrix in reef core samples – indicating greater resilience than reefs elsewhere • Early reef assemblages were dominated by branching coral species, switching to massive and finally coralline algal assemblages as reefs attained sea level • While high reefs show continual vertical aggradation throughout the Holocene, low reefs appear to have “turned off” with vertical growth stopping between around 5 to 6,000 years ago 2013 and 2014 fieldwork could only have been completed with the vessel support available through the Kimberley Marine Research Station, the marine expertise of Erin McGinty, and the interest and in kind support provided by James Browne of KMRS. Traditional Owners of the Bardi and Jawi, Dambimangari and Mayala communities are thanked for access to Country and assistance with fieldwork. The failure to gain access to Montgomery and Turtle Reefs for coring was disappointing, impeding knowledge of the transition between rhodolith dominated intertidal substrates and the potentially coral dominated reefs below. Good results were obtained from the substituted Adele Reef.

150 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

12 Communication

12.1 Publications

Solihuddin, T., Bufarale, G., Blakeway, D., O'Leary, M.J., Kordi M.N., Collins, L.B. and Stevens A. (2015). A Large Scale 2016. Geomorphology and Late Holocene Accretion Geomorphological and Surficial Cover Map of History of Adele Reef: a Northwest Australian Mid- Nearshore Reefs in the Kimberley Coast, WA. In Shelf Platform Reef, Geo-Marine Letters, 2016, 1–15. Proceedings from Coast to Coast Conference 2014, DOI: 10.1007/s00367-016-0465-3 Mandurah, Western Australia. ISBN-10: 0994357206 pp 15–20 Bufarale G, Collins LB, O’Leary MJ, Stevens A, Kordi M, Solihuddin T (2016). Quaternary onset and evolution Collins L.B., O’Leary M.J., Stevens A. M., Bufarale G., Kordi of Kimberley coral reefs (Northwest Australia) M., Solihuddin T, 2015. Geomorphic Patterns, internal revealed by high-resolution seismic imaging. Journal architecture and Reef Growth in a macrotidal, high of Continental Shelf Research, 123, 80–88 turbidity setting of coral reefs from the Kimberley doi:10.1016/j.csr.2016.04.002 Bioregion. Australian Journal of Maritime & Ocean Affairs, Volume 7, Issue 1, pp 12-22. (open access from Kordi, M.N., and O’Leary, M. (2016). Geomorphic Nov 2017) classification of reefs in the north western Australia Shelf. Regional Studies in Marine Science, DOI Kordi M.N., Collins L.B. and Stevens, A. (2015). Geomorphic 10.1016/j.rsma.2016.05.012. Patterns, Habitats and Substrates of Macrotidal Reefs from the Kimberley, North West Australia. In Kordi, M.N., and O’Leary, M. (2016) A Spatial Approach to Proceedings from 2015 WAMSI Research Conference, Improve Coastal Bioregion Management of the North Perth, Western Australia pp 72 Western Australia. Ocean & Coastal Management, 127, 26-42 doi:10.1016/j.ocecoaman.2016.04.004 Z. T. Richards, M. J. O’Leary, The coralline algal cascades of Tallon Island (Jalan) fringing reef, NW Australia. Coral Solihuddin T, O’Leary M, Blakeway D, Parnum I, Kordi M, Reefs June 2015, Volume 34, Issue 2, pp 595-595 First Collins L (March 2016) Holocene reef evolution in a online: 04 February 2015 doi: 10.1007/s00338-015- macrotidal setting: Buccaneer Archipelago, Kimberley 1262-6 Bioregion, Northwest Australia Coral Reefs DOI 10.1007/s00338-016-1424-1 Giada Bufarale · Lindsay B. Collins Stratigraphic architecture and evolution of a barrier seagrass bank in the mid- Ryan J. Lowe, Arturo S. Leon, Graham Symonds, James L. late Holocene, Shark Bay, Australia. Marine Geology Falter, and Renee Gruber The intertidal hydraulics of 11/2014; 359. DOI:10.1016/j.margeo.2014.11.010 tide-dominated reef platforms Journal of Geophysical Presentations and Meetings Research: Oceans Volume 120, Issue 7 July 2015 Pages 4845–4868 DOI: 10.1002/2015JC010701 Kordi M.N., Collins, L.B., O'Leary M, Stevens A (November 2015) ReefKIM: An integrated geodatabase for sustainable management of the Kimberley Reefs, North West Australia Ocean & Coastal Management doi:10.1016/j.ocecoaman.2015.11.004

Dambimangari TO meeting – Friday 17th August 2013 2 presentations at KMRS on initial findings of SBP and coring 2 presentations at Royal Society of Western Australia’s Centenary Postgraduate Symposium, Cockatoo Island presentation won an award. Applied Geology Departmental Seminar 2 presentations and 2 posters at Coast to Coast 2014: Solihuddin T - Holocene Reef Growth and Sea Level in a Macrotidal, High Turbidity Setting: Cockatoo Island, Kimberley Bioregion, Northwest Australia. Presentation Kordi M - Geomorphic Patterns, Habitats and Substrates of Macrotidal Reefs from the Kimberley, NW Australia. Presentation Bufarale G - Geomorphology, internal architecture and growth patterns of macrotidal reefs of the Kimberley Biozone, Northwest Australia. Poster – Won best Poster at the conference

Kimberley Marine Research Program | Project 1.3.1 151

Reef Growth and Maintenance

Stevens A - ReefKIM, an integrated geodatabase of nearshore reef along the Kimberley coast, WA. Poster Trimble Dimension, Las Vegas, 2014. O’Leary M.J., Advancing paleosea level and climate reconstructions through DGPS technologies. Presentation Research Colloquium Marine and Coastal Resources - Research & Development Center for Marine & Coastal Resources Indonesia. January 2015: Solihuddin T - Geomorphology and Holocene Reef Growth of the Inshore Kimberley Bioregion and Response to Sea- Level Change and Climate Impacts. Presentation Kimberley Research Conference Perth, 2015, 1 presentation and 4 Posters: Kimberley Reef Growth and Maintenance. Presentation Solihuddin T - Holocene Reef Growth and Sea Level in a Macrotidal, High Turbidity Setting: Cockatoo Island, Kimberley Bioregion, Northwest Australia. Poster Kordi M - Geomorphic Patterns, Habitats and Substrates of Macrotidal Reefs from the Kimberley, NW Australia. Poster Bufarale G - Geomorphology, internal architecture and growth patterns of macrotidal reefs of the Kimberley Biozone, Northwest Australia. Poster Stevens A - ReefKIM, an integrated geodatabase of nearshore reefs along the Kimberley coast, WA. Poster 6th International Geosciences Student Conference, 13 – 16 July 2015, Prague: Bufarale G - Late Quaternary sea level effects on coral reef evolution and growth in Southern Kimberley, North West Australia. Presentation AAPG/SEG 2015 International Conference and Exhibition, 13 – 16 September, 2015, Melbourne.. Bufarale G - Late Pleistocene and Holocene Reef Growth in Southern Kimberley, Northwest Australia

12.2 Media and internet Communication Activity Total to date Peer reviewed publication 12 Popular publication (i.e. Landscope, newsletter, etc.) 6 Conference Presentation 9 Presentations/Meetings with DPAW managers 2 Presentations/Meetings with Traditional Owners 1 Presentations/Meetings with other stakeholders (i.e. industry, 2 (KMRS) tourism) Presentations to general public 3 Media releases 1 Radio interviews 1

152 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

13 Appendices

Appendix I

Definitions http://support.esri.com/en/knowledgebase/GISDictionary http://www.geoproject.com.au/gda.faq.html attribute

Nonspatial information about a geographic feature in a GIS, usually stored in a table and linked to the feature by a unique identifier. For example, attributes of a river might include its name, length, and sediment load at a gauging station.

In raster datasets, information associated with each unique value of a raster cell. attribute table

A database or tabular file containing information about a set of geographic features, usually arranged so that each row represents a feature and each column represents one feature attribute. In raster datasets, each row of an attribute table corresponds to a certain zone of cells having the same value. In a GIS, attribute tables are often joined or related to spatial data layers, and the attribute values they contain can be used to find, query, and symbolize features or raster cells. bathymetric map

A map representing the topography of a seafloor or lake bed, using contour lines to indicate depth. coordinate system

A reference framework consisting of a set of points, lines, and/or surfaces, and a set of rules, used to define the positions of points in space in either two or three dimensions. The Cartesian coordinate system and the geographic coordinate system used on the earth's surface are common examples of coordinate systems. dataset

Any collection of related data, usually grouped or stored together.

GDA94

The Geocentric Datum of Australia (usually referred to as GDA94, or just GDA) is a coordinate system for Australia. That is, it's a system of latitudes and longitudes, or east and north coordinates, which we can use to keep track of locations.

GDA is GEOCENTRIC - it is a system of coordinates centred at the centre of the earth's mass.

GDA94 is compatible with modern positioning techniques such as the Global Positioning System (GPS). It supersedes the existing Australian Geodetic Datum 1984 (AGD84) and older coordinate systems. GDA94 is based on a global framework, the IERS Terrestrial Reference Frame (ITRF), but is fixed to a number of reference points in Australia.

Kimberley Marine Research Program | Project 1.3.1 153

Reef Growth and Maintenance geodatabase

A database or file structure used primarily to store, query, and manipulate spatial data. Geodatabases store geometry, a spatial reference system, attributes, and behavioral rules for data. Various types of geographic datasets can be collected within a geodatabase, including feature classes, attribute tables, raster datasets, network datasets, topologies, and many others. Geodatabases can be stored in IBM DB2, IBM Informix, Oracle, Microsoft Access, Microsoft SQL Server, and PostgreSQL relational database management systems, or in a system of files, such as a file geodatabase.

GIS

Acronym for geographic information system. An integrated collection of computer software and data used to view and manage information about geographic places, analyze spatial relationships, and model spatial processes. A GIS provides a framework for gathering and organizing spatial data and related information so that it can be displayed and analyzed.

GIScience

Abbreviation for geographic information science. The field of research that studies the theory and concepts that underpin GIS. It seeks to establish a theoretical basis for the technology and use of GIS, study how concepts from cognitive science and information science might apply to GIS, and investigate how GIS interacts with society.

Landsat

Multispectral, earth-orbiting satellites developed by NASA (National Aeronautics and Space Administration) that gather imagery for land-use inventory, geological and mineralogical exploration, crop and forestry assessment, and cartography. legend

The description of the types of features included in a map, usually displayed in the map layout. Legends often use graphics of symbols or examples of features from the map with a written description of what each symbol or graphic represents. multispectral

Related to two or more frequencies or wavelengths in the electromagnetic spectrum. multispectral image

An image created from several narrow spectral bands. point

A geometric element defined by a pair of x,y coordinates. polygon

On a map, a closed shape defined by a connected sequence of x,y coordinate pairs, where the first and last coordinate pair are the same and all other pairs are unique

154 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance polyline

In ArcGIS software, a shape defined by one or more paths, in which a path is a series of connected segments. If a polyline has more than one path (a multipart polyline), the paths may either branch or be discontinuous. query

A request to select features or records from a database. A query is often written as a statement or logical expression raster

A spatial data model that defines space as an array of equally sized cells arranged in rows and columns, and composed of single or multiple bands. Each cell contains an attribute value and location coordinates. Unlike a vector structure, which stores coordinates explicitly, raster coordinates are contained in the ordering of the matrix. Groups of cells that share the same value represent the same type of geographic feature. remote sensing

Collecting and interpreting information about the environment and the surface of the earth from a distance, primarily by sensing radiation that is naturally emitted or reflected by the earth's surface or from the atmosphere, or by sensing signals transmitted from a device and reflected back to it. Examples of remote-sensing methods include aerial photography, radar, and satellite imaging. remotely sensed imagery

Imagery acquired from satellites and aircraft, including panchromatic, radar, microwave, and multispectral satellite imagery. resolution

The dimensions represented by each cell or pixel in a raster table

A set of data elements arranged in rows and columns. Each row represents a single record. Each column represents a field of the record. Rows and columns intersect to form cells, which contain a specific value for one field in a record. shapefile

A vector data storage format for storing the location, shape, and attributes of geographic features. A shapefile is stored in a set of related files and contains one feature class.

SQL

Acronym for Structured Query Language. A syntax for retrieving and manipulating data from a relational database. SQL has become an industry standard query language in most relational database management systems.

Kimberley Marine Research Program | Project 1.3.1 155

Reef Growth and Maintenance vector

A coordinate-based data model that represents geographic features as points, lines, and polygons. Each point feature is represented as a single coordinate pair, while line and polygon features are represented as ordered lists of vertices. Attributes are associated with each vector feature, as opposed to a raster data model, which associates attributes with grid cells.

WGS84

Acronym for World Geodetic System 1984. The most widely used geocentric datum and geographic coordinate system today, designed by the U.S. Department of Defense to replace WGS72. GPS measurements are based on WGS84

156 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Appendix II

WAMSI Metadata Form

Please answer the following questions about your WAMSI research datasets and any other datasets you would like to store at iVEC. If you have other documentation that describes WAMSI Node the data you collected please also attach with this spreadsheet or just attach that if it Project X.X.X answers all these questions. What What is the title of the study? (e.g. what would like to be the title of the metadata

record) What would be some key words for searching for this data? What constraints would you place on the data (e.g. legal, usage - purposes that

shouldn't use the data)

Who Who did the research? Please list names and the contact details. Who is point of contact in case of questions? Please list their contact details - is there

a generic contact that could be used to ensure longivity? Who else should be acknowledged? Any links to journal articles?

Why Why was the research done? This is the abstract - a brief summary of the content of

the research is required including the research intentions

How How was the research done? (e.g. instrumentation, brief description of procedure)? How often were measurements taken? Were they aggregated into a specific unit of

time (e.g. day, multi-day, week, month, multi-month, year, multi-year)? How is the data currently stored, that is what format is the data? (e.g. GIS shapefiles,

compressed AVI etc.) Please provide as much information as possible.

When When was the research carried out? When were the start and end dates?

Where Where was the research done? As a minimum please indicate the 'bounding box' in latitude/longitude (decimal degrees) (e.g. North bound latitude -22.00; West bound longitude 113.00; East bound longitude 114.00; South bound latitude -23.00) Where are any other related publications/information about the research published - if

any? (e.g. url ) Where in the vertical column of the ocean was the research undertaken? (e.g. minimum

and maximum depth)

Supplementary information - Please attach any further information you think would

be useful for future researchers

Image - If you have one handy please also attach a picture (JPEG preferable) that best describes your research. This will be used as the thumbnail image next to the metadata records in the MEST

Kimberley Marine Research Program | Project 1.3.1 157

Reef Growth and Maintenance

Research Type and Category Record the type and category of research from the lists below

Type Inventory – what is there and where is it found? Baseline – Quantifying the value, status, variability and trends Process – How does the system function? Cause/effect pathways? Prediction – Modelling/scenarios/ what ifs?

Category Social Ecological Physical

Objectives (What is the project doing?) List the key objectives of the research

Management Questions (Why?) List the management questions that were used to guide and frame the research question, It is expected that the final report will provide answers to these questions. Thus, note for each question where the research project will not fully answer the question, but will provide information towards answering it.

Key Stakeholders/End-users (Who will use this?) List the individuals in as much detail as possible who will have a use for this study and whether this is through a decision-making capacity or operational role.

Outputs (What do they want?) List the outputs expected from the research, including the format in which these will be presented.

Links to other projects (How will the science be integrated?) List the projects within the KMRP that will provide additional information in the reporting and interpretation of findings for this project. Also list projects that will be similarly informed by the outcomes of this project. Include information on how this project will interact with the linked projects to ensure information sharing.

Synthesis reports that will require input from this project (How will the science be integrated?) List the key KMRP synthesis reports that will require input from this project.

158 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Appendix III

SBP Surveyed lines

Kimberley Marine Research Program | Project 1.3.1 159

Reef Growth and Maintenance

160 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Kimberley Marine Research Program | Project 1.3.1 161

Reef Growth and Maintenance

162 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Kimberley Marine Research Program | Project 1.3.1 163

Reef Growth and Maintenance

164 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Appendix IV

Core Descriptions Location and core no: North Pool Sunday Island (PN2) Core length: 102 cm Compaction: 74 cm Penetration: 176 cm

Photo Description

(0 – 8 cm depth): coarse to very coarse grain sediment, shelly fragments, pale grey with bivalve fragments and some live coralline algae, fine to gravel grained coral and branching Acropora fragments.

(8 – 60 cm depth): unconsolidated green – grey sandy mud sand with brown- stained gravel with coralline algae concretion. Branching coral fragments mainly of the genera of Acropora and coralline algae concretion. Rhodolith nodules, bivalve fragments and scallops.

(60 – 102 cm depth): unconsolidated very coarse pale grey sand with abundant branching Acropora fragments and encrusted coralline algae. Some brown-stained unidentified coral and a small clam.

Dating samples: PN2-A (5 cm): branching Acropora PN2-B (12 cm): coralline algae concretion PN2-C (40 cm): branching coral PN2-D(68 cm): unidentified coral PN2-E (84 cm): encrusted coralline algae

Sediment samples: PN2 – 14 cm PN2 – 50 cm

PN2 – 90 cm

Kimberley Marine Research Program | Project 1.3.1 165

Reef Growth and Maintenance

Location and core no: North Pool Sunday Island (PN3) Core length: 248 cm Compaction: 114 cm Penetration: 362 cm

Photo Description

(0 – 20 cm depth): living macroalgae (sargassum), encrusting coralline algae, bivalves, urchins, branching coral and brown-staining on coral fragments in unconsolidated brownish sandy matrix.

(20 – 64 cm depth): unconsolidated brownish sandy mud matrix with coralline algae, rhodoliths nodule, serpulids, encrusting coral fragments and coralline algae, branching coral coated and cemented. Matrix approx. 35% corals.

(64 – 136 cm depth): unconsolidated pale grey sandy mud with branching coral, mainly Acropora fragments. Coralline algae concretion and rhodoliths. Foram tests, bivalves, amphiora, and abundant coral fragments coated and cemented by coralline algae. Matrix approx. 25% coral.

(136 – 248 cm depth): unconsolidated coarse – very coarse sand grey sand with abundant amphiora fragments and molluscs. Domal coral, mainly Platygra and little Porites at 148 cm depth. Occasional massive coral colonies, no branching coral. Corals are not oriented in growth position. Grey-staining on Platygra surface and serpulids, urchin tube.

Dating samples: PN3-A (9 cm): shell PN3-B (83 cm): branching Acropora PN3-C (147 cm): coralline algae concretion PN3-D (220 cm): Platygra

Sediment samples: PN3 18.5 cm PN3 30 cm PN3 100 cm

166 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: North Pool Sunday Island (PN4) Core length: 87 cm Compaction: 66 cm Penetration: 153 cm

Photo Description

(0 – 18 cm depth): unconsolidated dark grey muddy matrix with branching Acropora fragments coated and cemented by coralline algae. Porites 3-5 cm. Matrix approx. 20% corals.

(18 – 71 cm depth): unconsolidated grey sandy mud matrix with coralline algae, rhodoliths, branching Acropora fragments, branching coral concretions, bivalve fragments, foram tests, Seriatipora, gastropod operculums.

(71 – 87 cm depth): unconsolidated coarse pale grey sand with brown-stained branching Acropora fragments. Coralline algae concretions.

Dating samples: PN4-A (10 cm): Porites PN4-B (50 cm): branching Acropora PN4-C (86 cm): coralline algae concretion

Sediment samples:

PN4 – 10 cm (muddy sand) PN4 – 40 cm (muddy sand) PN4 – 76 cm (coarse sand)

Kimberley Marine Research Program | Project 1.3.1 167

Reef Growth and Maintenance

Location and core no: North Pool Sunday Island (PN4B) Core length: 120 cm Compaction: 43 cm Penetration: 163 cm

Photo Description

(0 – 16 cm depth): uppermost surface of the unit containing live coralline algae and branching Acropora fragments in the sandy matrix

(16 – 104 cm depth): unconsolidated green – grey sandy mud with abundant branching Acropora fragments. Occasional rhodolith nodules, crab claw at 43 cm depth, bivalve fragments, foram tests, stained gravel, coralline algae concretions.

(104 – 120 cm depth): unconsolidated very coarse pale grey sand with some brown-stained branching coral fragments, foram tests, bivalve fragments.

Dating samples: PN4B-A (3 cm): bioeroded branching Acropora stick PN4B-B (34 cm): Acropora fragment PN4B-C (80 cm): branching Acropora PN4B-D (119 cm): coralline algae concretion

Sediment samples: PN4B – 25 cm (muddy sand) PN4B – 70 cm (muddy sand) PN4B – 110 cm (coarse sand)

168 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: North Pool Sunday Island (PN1) Core length: 278 cm Compaction: 130 cm Penetration: 408 cm

Photo Description

(0 – 21 cm depth): unconsolidated brown-grey coarse sand, shell fragments, coralline algae, domal coral (Montipora?).

(21 – 37 cm depth): unconsolidated green-grey sandy mud with branching Acropora fragments, coralline algae, bivalves, foram tests.

(37 – 54 cm depth): unconsolidated brown-grey coarse sand with branching Acropora fragments, shell fragments, coralline algae, rhodolith nodules.

(54 – 75 cm depth): unconsolidated green-grey sandy mud with branching Acropora fragments (Aspira at 55 cm depth), coralline algae, shell fragments.

(75 – 120 cm depth): unconsolidated grey coarse sand with branching coral fragments and domal coral (Platygra?), coralline algae, bivalves, shell fragments.

(120 – 197 cm depth): unconsolidated grey coarse sand, coralline algae, bivalves, gastropods.

(197 – 259 cm depth): unconsolidated grey coarse sand with domal coral fragments (Platygra?), coralline algae, rhodolith nodules, shell (±3cm), bivalves.

(259 – 278 cm depth): unconsolidated very coarse brown-grey sand with coralline algae concretion, rhodolith nodules, shell fragments, domal coral (Porites at 220 cm depth and Cypstrea at 247 cm depth), bivalves, foraminifera.

Dating samples: PN1-A (17 cm): Montipora? PN1-B (83 cm): Platygra? PN1-C (210 cm): Merulina Ampiata?

PN1-D (264 cm): Coralline algae concretion

Kimberley Marine Research Program | Project 1.3.1 169

Reef Growth and Maintenance

Sediment samples: PN1 – 3 cm PN1 – 30 cm PN1 – 42 cm PN1 – 68 cm PN1 – 90 cm PN1 – 150 cm

170 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: North Pool Sunday Island (PN5) Core length: 247 cm Compaction: 193 cm Penetration: 440 cm

Photo Description

(0 – 15 cm depth): unconsolidated grey sandy matrix with branching Acropora fragments, coralline algae.

(15 – 65 cm depth): unconsolidated brownish-grey sandy mud matrix with branching Acropora fragments, domal coral (Goniopora?) at 18 cm depth, coralline algae, coral rubbles, shell fragments.

(65 – 200 cm depth): unconsolidated grey coarse sand with branching coral mainly Acropora fragments at 63, 68, 156-157 cm, Porites at 90 cm, Chypastrea at 150- 155 cm, Goniastrea at 183-189 cm, coralline algae at 115, 120-130, 136 cm, shell fragments, rhodolith nodules at 167 cm, bivalves.

(200 – 247 cm depth): unconsolidated very coarse dark-grey sandy matrix with branching Acropora fragments at 201-213 cm, Porites at 226-240 cm, Pocylopra at 216-220 cm, shell fragments.

Dating samples: PN5-A (18 cm): Goniopora PN5-B (90 cm): Porites PN5-C (152 cm): Chypastrea PN5-D (230 cm): Porites

Kimberley Marine Research Program | Project 1.3.1 171

Reef Growth and Maintenance

Location and core no: South Pool Sunday Island (PS1) Core length: 129 cm Compaction: 59 cm Penetration: 188 cm

Photo Description

(0 – 11 cm depth): coarse sand sediment, shelly fragments, some anoxic mud? Less rhodoliths.

(11 – 56 cm depth): unconsolidated dark grey coarse sand with branching Acropora coral colonies coated and cemented in coralline algae, bivalve fragments, foram tests, brown-stained branching coral colonies. Matrix approx. 25% coral.

(56 – 129 cm depth): coarse sand with cemented rhodoliths. Brown-stained coralline algae concretion.

Dating samples: PS1-A (14 cm): coralline algae PS1-B (84 cm): rhodoliths PS1-C (127 cm): coralline algae concretion

Sediment samples: PS1 – 5 cm PS1 – 40 cm PS1 – 105 cm

172 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: South Pool Sunday Island (PS2) Core length: 52 cm Compaction: 14 cm Penetration: 66 cm

Photo Description

(0 – 52 cm depth): unconsolidated very coarse skeletal sand, brown-grey, bivalve fragments, rhodolith fragments, foram test, no coral recovered.

Dating samples: PS2-A (2 cm): concretion PS2-B (48 cm): coralline algae concretion

Sediment samples: PS2 25 cm

Kimberley Marine Research Program | Project 1.3.1 173

Reef Growth and Maintenance

Location and core no: South Pool Sunday Island (PS2B) Core length: 34 cm Compaction: 7 cm Penetration: 41 cm

Photo Description

(0 – 10 cm depth): unconsolidated coarse skeletal sand, foram tests, coralline algae fragments, live rhodoliths, rhodoliths coated and cemented by coralline algae.

(10 – 34 cm depth): unconsolidated pale sandy mud with branching coral cemented by coralline algae, coralline algae concretion (±3x5 cm). Coral cover very poor (<10%), encrusted coralline algae dominant. Coral concretion.

Dating samples: PS2B-A (8 cm): cemented rhodoliths PS2B B (28 cm): branching coral fragments

Sediment samples: PS2B – 9 cm

PS2B – 23 cm

174 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: South Pool Sunday Island (PS3) Core length: 80 cm Compaction: 21cm Penetration: 101 cm

Photo Description

(0 – 48 cm depth): unconsolidated brown-orange sandy mud, branching coral fragments along the unit, coralline algae at 40 cm, bivalves.

(48 – 80 cm depth): unconsolidated brown-orange sand with coralline algae concretions, cemented together shells, mollusc, rhodoliths at 50 cm.

Dating samples: PS3-A (7 cm): encrusted coral? PS3-B (71 cm): coralline algae

Sediment samples: PS3 23 cm PS3 70 cm

Kimberley Marine Research Program | Project 1.3.1 175

Reef Growth and Maintenance

Location and core no: West Tallon (WT1) Core length: 410 cm Compaction: 216 cm Penetration: 626 cm

Photo Description

(0–50 cm depth): Unconsolidated branching coral floatstone with grey sand matrix, encrusted coral (coral fragment encrusted by coralline algae) at 20 cm and domal coral (fungia?) at 45 cm.

(50–120 cm depth): Unconsolidated branching coral floatstone with muddy sand matrix, rhodoliths nodule at 55 and 62 cm, encrusted coral at 68, 75 and 95 cm, encrusting coralline algae at 100 cm, bivalves at 105 cm.

(120–200 cm depth): Unconsolidated branching coral floatstone mixed with encrusting coralline algae lying in the bright-grey sand matrix. Branching coral at 13, 155, 165-180, 190-200 cm, encrusted coralline algae at 136-158 cm, 180-190 cm. Bivalves at 158 cm.

(200–280 cm depth): Unconsolidated branching coral floatstone with muddy sand matrix, encrusting coralline algae at 228 and 235 cm, bivalves, foram.

(280–380 cm depth): Unconsolidated branching coral floatstone with sandy mud matrix, encrusting coralline algae at 326 cm, bivalves, foram.

(380–410 cm depth): Unconsolidated branching coral floatstone with dark- grey muddy clay matrix. Branching coral at 382, 396, 402 cm.

Dating samples: WT1-20 (30) cm: branching coral WT1-80 cm: branching coral WT1-164 cm: coral WT1-193 (294) cm: encrusting coral WT1-230 cm: encrusting coral WT1-255 cm: domal coral WT1-376 (573) cm: branching coral

176 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Matrix samples: WT1-30 cm WT1-103 cm WT1-170 cm WT1-240 cm WT1-320 cm WT1-400 cm

Kimberley Marine Research Program | Project 1.3.1 177

Reef Growth and Maintenance

Location and core no: West Tallon (WT2) Core length: 107 cm Compaction: 28 cm Penetration: 135 cm

Photo Description

(0–50 cm depth): Unconsolidated branching coral floatstone with bright-grey very coarse sand matrix, rhodoliths at 26 and 44 cm, branching 9, 20 and 33 cm, encrusting coralline algae fragment at 3-6 cm, and encrusted coral at 45- 50 cm.

(50–107 cm depth): Unconsolidated branching coral floatstone with dark-grey coarse sand matrix and encrusting coralline algae fragments at 81, 87, 91 cm, shell at 90 cm, and Lobophyllia at 105cm.

Dating samples: WT2-25 cm (rhodoliths) WT2-70 cm (branching coral)

Sediment samples: WT2-10 cm WT2-70 cm

178 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: East Tallon (ET1) Core length: 92 cm Compaction: 34 cm Penetration: 126 cm

Photo Description

(0 – 64 cm depth): unconsolidated brown-orange coarse sand with branching coral fragments, Acropora at 8-19 cm, Goniopora at 22 and 30 cm, and Porites Cylindrica at 37 - 56 cm. Gastropods at 61 cm, bivalves.

(64 – 92 cm depth): brown-orange coarse sand with coralline algae fragments at 66, 78, 90 cm.

Dating samples: ET1-A (18cm): Branching Acropora ET1-B (54 cm): Branching Porites Cylindrica ET1-C (90 cm): Coralline algae

Sediment samples: ET1 – 20 cm ≈ 27 cm --> coarse sand, coral gravels ET1 – 84 cm ≈ 115 cm --> coarse sand, coral gravels, bio-eroded grains

Kimberley Marine Research Program | Project 1.3.1 179

Reef Growth and Maintenance

Location and core no: East Tallon (ET2) Core length: 100 cm

Photo Description

Coralline algae bindstone with brown-stained coral fragments at 68, 70, 74, 76, 88, 95, 99 cm, gastropods at 50 and 63 cm, cavities infilled by cemented coarse sand, mollusc.

180 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: East Tallon (ET3) Core length: 68 cm

Photo Description

Brown-orange coralline algae bindstone with branching coral fragments at 65 cm, gastropods at 53, 60, and 62 cm, cavities, cemented coarse sand, shell fragments.

Kimberley Marine Research Program | Project 1.3.1 181

Reef Growth and Maintenance

Location and core no: East Tallon (ET4) Core length: 88 cm

Photo Description

Coralline algae bindstone, reticulate coralline predominant from 15-50 cm, cavities (± 2 cm) with brown-stained on roof, coral and mollusc fragments, cemented lime mud.

182 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: East Tallon (ET5) Core length: 100 cm

Photo Description

Coralline algae bindstone with some cavities infilled by sand, some are open, partly cemented sand in cavities.

Kimberley Marine Research Program | Project 1.3.1 183

Reef Growth and Maintenance

Location and core no: East Tallon (ET6) Core length: 22 cm

Photo Description

Coralline algae bindstone with cavities infilled by cemented sand and dark-brown-stained on roof of the cavities.

184 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: East Tallon (ET7) Core length: 380 cm Compaction: 200 cm Penetration: 580 cm

Photo Description

(0 – 20 cm depth): dark-grey sandy mud with mollusc at 10 cm, gastropods at 13 and 17 cm, root of seagrass, shell fragment.

(20 – 152 cm depth): grey sandy mud with branching coral fragments, Goniopora at 33 and 77 cm, Acropora at 40-153 cm, mollusc at 63 and 141 cm, spiky rhodoliths at 153 cm, and coralline algae at 149 cm.

(152 – 380 cm depth): branching coral fragments with muddy sand matrix, coralline red algae fragments along the unit, Seriatopora, Pocilopora at 320 cm, Acropora, Ampiora, mollusc at 195 and 245 cm, abundant bivalve fragments, couple of urchins, occasionally fungi at 302 cm, gastropods at 234 and 326 cm, and Porites Cylindrica at 376 cm.

Dating samples: ET7-A (20 cm): ET7-B (118 cm): ET7-C (207 cm): ET7-D (288cm): ET7-E (378cm):

Sediment samples: ET7-A 12 cm ≈ 18 cm --> dark sandy mud ET7-B 68 cm ≈ 103 cm --> grey sandy mud ET7-C 210 cm ≈ 318 cm --> coarse sand, coral gravels, rock fragments ET7-D 312 cm ET7-E 358cm ≈ 542 cm --> dark muddy sand

Kimberley Marine Research Program | Project 1.3.1 185

Reef Growth and Maintenance

Location and core no: East Tallon (ET8) Core length: 121 cm

Photo Description

Coralline algae bindstone with dark-brown-stained on roof of the cavities, infilled by mud partly are cemented, mollusc at 44-49 cm, gastropods at 70 cm, shells at 15-17 cm.

186 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: East Tallon (ET9) Core length: 105 cm

Photo Description

Coralline algae bindstone with some encrusted coral fragments at 30, 47, 64 cm, branching coral fragments at 97 cm, shells at 42, 58, 76-77 cm, cemented lime mud, coralline algae fragments, cavities with brown-stained on roof of the cavities.

Kimberley Marine Research Program | Project 1.3.1 187

Reef Growth and Maintenance

Location and core no: East Tallon (ET10) Core length: 66 cm

Photo Description

Coralline algae bindstone, shells at 5, 11, 22, 35, 38, 57, and 62 cm, dark- brown-stained on the roof of cavities infilled by cemented sand, no coral fragment recovery, cemented lime mud infilled the pores of coralline algae.

188 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: East Tallon (ET11) Core length: 90 cm

Photo Description

Coralline algae bindstone with cavities infilled by sand, partly cemented, dark- brown-stained on the roof of cavities, no coral fragment recovery, shells at 2, 62, and 70 cm. coralline algae fragments.

Kimberley Marine Research Program | Project 1.3.1 189

Reef Growth and Maintenance

Location and core no: East Tallon (ET12) Core length: 159 cm Compaction: 56 cm Penetration: 215 cm

Photo Description

(0 – 14 cm depth): dark-grey sandy mud with seagrass (Thalasia?) grow on it, coral fragment (chypastrea?) at 8-12 cm, coralline algae fragments, shell fragments.

(14 - 159 cm depth): grey muddy sand with dark-grey coarse sand horizon at 140 - 145 cm, branching coral fragments along the core mainly Acropora, Stylophora? At 17 cm with encrusted coral at 29 cm and coral fragments coated/cemented together by coralline algae, shell fragments at 35 cm, root of seagrass at 45 cm attached to the coral fragments, gastropods.

Dating samples: ET12-A (17 cm) ET12-B (98 cm) ET12-C (150 cm)

Sediment samples: ET12 – 5 cm ≈ 7 cm --> dark-grey sandy mud ET12 – 93cm ≈ 126 cm --> dark coarse sand

190 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: East Tallon (ET13) Core length: 25 cm Compaction: - Penetration: -

Photo Description

Grey coarse muddy sand matrix, encrusted coralline algae at 6 and 12 cm, big coralline algae concretion at 20 cm with grey-stained, coral and seagrass at the surface.

Kimberley Marine Research Program | Project 1.3.1 191

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B01) Core length: 59 cm Compaction: 1 cm Penetration: 60 cm

Photo Description

(0–15 cm depth): Unconsolidated brownish-grey coarse sand with encrusting coralline algae fragments well-lithified together with forams, shells.

(15–45 cm depth): Unconsolidated domal coral floatstone with grey coarse muddy-sand matrix.

(45–59 cm depth): Unconsolidated branching coral floatstone with brownish coarse sand matrix containing shell fragments, foram, bio-eroded grains.

Dating samples: B01-3 cm B01-30 cm B01-43 cm

Sediment samples: B01-10 cm B01-40 cm B01-50 cm

IDs: B01-24 cm

192 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B02) Core length: 99 cm Compaction: 35 cm Penetration: 134 cm

Photo Description

(0–30 cm depth): Unconsolidated brownish very coarse sand with gravelly coral rubble and encrusting coralline algae (at 25 cm). matrix contains abundant coralline algae fragments (rhodolits), foram and molluscs.

(30–76 cm depth): Unconsolidated domal coral floatstone with brownish-grey very coarse sand matrix containing bio-eroded grains mostly coralline algae fragments. Some domal coral covered by encrusting coralline algae (at 53, 61, 70 cm), encrusting coralline algae (at 40 cm), rhodoliths (at 30 cm).

(76–99 cm depth): Unconsolidated branching coral floatstone with brownish- grey coarse sand matrix containing shell fragments, foram, bivalves (at 73 cm), bio-eroded grains. Branching coral covered by encrusting coralline algae (at 80, 85, 88, 90 cm). encrusting coralline algae (at 85, 97, 105 cm)

Dating samples: B02-25 cm B02-61 cm B02-97 cm

Sediment samples: B02-8 cm B01-45 cm B01-90 cm

IDs: B02-40 cm: encr. Coralline algae B02-30 cm: rhodolits, B02-68 cm: coral covered by encr. Coralline algae B02-71 cm: bivalves B02-76 cm: branching coral

Kimberley Marine Research Program | Project 1.3.1 193

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B03) Core length: 70 cm Compaction: 10 cm Penetration: 80 cm

Photo Description

(0–40 cm depth): Unconsolidated brownish-grey very coarse sand with encrusting coralline algae. Matrix contains shell fragments, foram, bivalves and bio-eroded grains mostly from coralline algae fragments.

(40–70 cm depth): Unconsolidated grey sand with encrusting coralline algae. Corals covered by encrusting coralline algae, domal coral of Porites? (at 70 cm).

Dating samples: B03-5 cm B03-70 cm

Sediment samples: B03-15 cm B03-60 cm

IDs: B03-30 cm B03-44 cm

194 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B04) Core length: 275 cm Compaction: 172 cm Penetration: 447 cm

Photo Description

(top sequence - rotary drill): coralline algae bindstone with domal coral of Porites.

(0–200 cm depth): Unconsolidated domal coral floatstone with brownish coarse sand matrix.

(200–275 cm depth): Unconsolidated domal coral floatstone with muddy clay matrix.

Dating samples: B04- 2 cm B04- 50 cm B04- 150 cm B04- 265 cm

Sediment samples: B04- 20 cm B04- 90 cm B04- 240 cm

IDs: B04- 15 cm B04- 30 cm B04- 50 cm B04- 130 cm B04- 250 cm

Kimberley Marine Research Program | Project 1.3.1 195

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B05) Core length: 470 cm Compaction: 162 cm Penetration: 632 cm

Photo Description

(top sequence - rotary drill): domal coral of Porites and tridacna.

(0–10 cm depth): Unconsolidated brownish coarse sand with coralline algae fragments and shells.

(10–40 cm depth): unconsolidated domal coral floatstone with grey coarse sand matrix containing shells, bio-eroded grains. Domal coral recovered (at 12, 18, 33 cm).

(40–340 cm depth): unconsolidated domal coral floatstone with grey muddy sand matrix. Branching coral of Acropora (at 316 cm), encrusting coralline algae (at 310 cm).

(340-470 cm depth): unconsolidated branching coral floatstone with grey muddy clay matrix.

Dating samples: B05- 12 cm B05- 120 cm B05- 285 cm B05- 450 cm

Sediment samples: B05- 35 cm B05- 130 cm B05- 310 cm B05- 435 cm

196 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B08) Core length: 85 cm Compaction: 14 cm Penetration: 99 cm

Photo Description

(0–85 cm depth): Unconsolidated branching coral floatstone with brownish-grey muddy sand matrix. Some branching corals are covered by encrusting coralline algae. Encrusting coralline algae are well- lithified together with branching coral, shells. Encrusted rhodoliths are also recovered at 32-45 cm.

Dating samples: B08-10 cm B08-70 cm

Sediment samples: B08- cm B08- cm

ID samples: B08-15 cm B08-30 cm B08-60 cm

Kimberley Marine Research Program | Project 1.3.1 197

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B08b) Core length: 70 cm Compaction: - cm Penetration: 70 cm

Photo Description

(0–30 cm depth): domal coral framestone.

(31–70 cm depth): encrusting coralline algae bindstone with some branching corals, shells, bivalves, and gastropods lithified together.

Dating samples: B08b-15 cm B08b-35 cm B08b-64 cm

198 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B09) Core length: 103 cm Compaction: 22 cm Penetration: 125 cm

Photo Description

(0–18 cm depth): coralline algal bindstone

(18–33 cm depth): Unconsolidated brownish sand unit with some coralline algae fragments (at 30 cm).

(33–103 cm depth): Unconsolidated branching coral floatstone with dark-grey sand matrix. Gastropods recovered at 38, 41, 76 cm, domal coral of Porites recovered at 80 cm, rhodoliths recovered at 70 and 74 cm.

Dating samples: B09-10 cm B09-80 cm

Sediment samples: B09-20 cm B09-50 cm

ID samples: B09-30 cm B09-55 cm B09-100 cm

Kimberley Marine Research Program | Project 1.3.1 199

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B10) Core length: 80 cm Compaction: 58 cm Penetration: 138 cm

Photo Description

(0–25 cm depth): Unconsolidated brownish coarse sand with well- lithified encrusting coralline algae together with branching coral fragments, shells.

(25–80 cm depth): Unconsolidated branching coral floatstone with brownish-grey muddy sand matrix. Encrusting coralline algae, rhodoliths at 27 cm, domal coral at 55 cm, branching coral covered by encrusting coralline algae at 44 cm, branching coral recovered at 37, 42 and 68-80 cm.

Dating samples: B10-8 cm B10-55 cm

Sediment samples: B10-10 cm B10-60 cm

ID samples: B10-20 cm B10-40 cm B10-70 cm

200 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B11) Core length: 214 cm Compaction: 172 cm Penetration: 386 cm

Photo Description

(0–16 cm depth – rotary drill): domal coral of Porites and encrusting coralline algae.

(16–90 cm depth) Unconsolidated brownish coarse sand unit with encrusting coralline algae fragments, shells, coral fragments.

(90–130 cm depth): Unconsolidated branching coral floatstone with brownish grey sand matrix. Some corals are covered by encrusting coralline algae.

(130–214 cm depth): Unconsolidated branching coral floatstone with grey muddy sand matrix. Most corals are covered by encrusting coralline algae. Matrix contains articulated coralline algae fragments, shells, bivalves, foram and gastropods.

Dating samples: B11- 5 cm B11- 55 cm B11- 120 cm B11- 190 cm

Sediment samples: B11- 20 cm B11- 115 cm B11- 170 cm

ID samples: B11- 60 cm B11- 95 cm B11- 128 cm

B11- 205 cm

Kimberley Marine Research Program | Project 1.3.1 201

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B12) Core length: 525 cm Compaction: 100 cm Penetration: 625 cm

Photo Description

(0–120 cm depth) Unconsolidated robust branching coral floatstone with grey muddy matrix. Domal coral of Faviidae? Recovered at the surface.

(120–290 cm depth): Unconsolidated delicate branching coral floatstone with grey muddy matrix.

(290–425 cm depth): grey muddy matrix with minor branching coral fragments.

(425–525 cm depth): Unconsolidated robust branching coral floatstone with grey muddy matrix.

Dating samples: B12- 30 cm B12- 109 cm B12- 200 cm B12- 308 cm B12- 415 cm B12- 480 cm

Sediment samples: B12- 110 cm B12- 220 cm B12- 356 cm B12- 490 cm

ID samples: B12- 20 cm B12- 35 cm

202 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

B12- 92 cm B12- 230 cm B12- 334 cm B12- 410 cm B12- 455 cm

Kimberley Marine Research Program | Project 1.3.1 203

Reef Growth and Maintenance

Location and core no: Bathurst-Irvine (B14) Core length: 455 cm Compaction: 118 cm Penetration: 573 cm

Photo Description

(0–25 cm depth) Unconsolidated domal branching coral rudstone with grey muddy matrix and minor branching coral fragments.

(25–55 cm depth): grey muddy unit with minor branching coral fragments.

(55–124 cm depth): Unconsolidated robust (tabulate) branching coral floatstone with grey muddy matrix and minor delicate branching coral fragments.

(124–199 cm depth): Unconsolidated delicate branching coral floatstone with grey muddy matrix and minor robust (tabulate) branching coral.

(199–300 cm depth): Unconsolidated robust branching coral floatstone with grey muddy matrix and minor delicate branching coral fragments.

(300–455 cm depth): Unconsolidated delicate branching coral floatstone with grey muddy matrix. Dating samples: B13- 13 cm B13- 70 cm B13- 123 cm B13- 260 cm B13- 346 cm B13- 443 cm

Sediment samples: B13- 40 cm B13- 100 cm B13- 165 cm B13- 216 cm B13- 340 cm

204 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

ID samples: B13- 5 cm B13- 95 cm B13- 178 cm B13- 224 cm B13- 273 cm B13- 322 cm

Kimberley Marine Research Program | Project 1.3.1 205

Reef Growth and Maintenance

Location and core no: Adele Island (A03) Core length: 350 cm Compaction: 156 cm Penetration: 506 cm

(0–180 cm): Unconsolidated domal coral floatstone (mostly Paviid) with grey coarse sand matrix, containing gastropods, shells, foraminifera test and other bio-clastic materials. Most coral are coated by encrusting coralline algae.

(180–350 cm): Unconsolidated domal coral floatstone with grey-green muddy coralline sand matrix, containing mostly coral and encrusting coralline algae fragments

Dating samples:

Matrix samples:

206 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: Adele Island (A04) Core length: 210 cm Compaction: 95 cm Penetration: 305 cm

Photo Description

(0–12 cm): brownish-grey coarse sand unit.

(12–140 cm): Unconsolidated domal coral floatstone (mostly Porites, minor branching Acropora) with grey coralline sand matrix. Coral colonies: Fungia, Porites, Acropora.

(140–210 cm): Unconsolidated domal coral floatstone (dominated by Porites) with grey muddy sand matrix. Coral colonies: Porites, Heliopora, Galaxea.

Dating samples:

Matrix samples:

Kimberley Marine Research Program | Project 1.3.1 207

Reef Growth and Maintenance

Location and core no: Adele Island (A05) Core length: 160 cm Compaction: 67 cm Penetration: 227 cm

Photo Description

(0–45 cm): Unconsolidated brownish-grey coarse bio-clastic sand unit with minor rhodoliths, coralline algae fragments and coral (Chypastrea).

(45–110 cm): Unconsolidated branching coral (Acropora dominant) floatstone with grey coarse coralline muddy sand matrix.

(110–160 cm): Unconsolidated domal coral (mostly Porites, minor branching Acropora) floatstone with grey muddy sand matrix.

Dating samples:

Matrix samples:

208 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: Adele Island (A06) Core length: 311 cm Compaction: 103 cm Penetration: 414 cm

Photo Description

(0–12 cm): Unconsolidated brownish-orange coarse bio-clastic sand unit.

(12–100 cm): Unconsolidated domal coral floatstone with grey coarse bio-clastic sand matrix. Coral colonies: Paviid, Astreopora, Porites, Branching Acropora.

(100–205 cm): Unconsolidated grey coarse bio-clastic sand unit with abundant shells, gastropods, benthic forams.

(205–311 cm): Unconsolidated branching coral floatstone with grey coarse bio-clastic sand matrix. Most corals are coated by coralline algae. Coral colonies: Pavona, Goniopora, Galaxea, Acropora, Fungia.

Dating samples:

Matrix samples:

Kimberley Marine Research Program | Project 1.3.1 209

Reef Growth and Maintenance

Location and core no: Adele Island (A07) Core length: 250 cm Compaction: 97 cm Penetration: 347 cm

Photo Description

(0–100 cm): Unconsolidated grey medium-coarse (fining upward) sand unit.

(100–250 cm): Unconsolidated domal coral floatstone with grey muddy sand matrix and minor branching coral. Coral colonies: Lobophyllia, Favia, Goniopora, Galaxea, Acropora.

Dating samples:

Matrix samples:

210 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: Adele Island (A08) Core length: 340 cm Compaction: 87 cm Penetration: 427 cm

Photo Description

(0–90 cm): Unconsolidated branching coral floatstone with grey coarse sand matrix. Most corals are coated by encrusting coralline algae. Corals are not really in situ/ in place, transported from other places clearly seen from the orientation and preservation. Coral colonies: 50% branching Acropora, 30% Faviid, 20% mixed Lobophyllia and other corals. Matrix: 90% sand, 10% mud.

(90–154 cm): Unconsolidated branching coral floatstone with grey muddy sand matrix. Most corals are coated by encrusting coralline algae. Coral colonies: Porites, Lobophyllia, Galaxea, Acropora.

(154–345 cm): Unconsolidated robust branching coral floatstone with grey-green sandy- mud matrix. Most corals are coated by encrusting coralline algae. Coral colonies dominated by Lobophyllia.

Dating samples:

Matrix samples:

Kimberley Marine Research Program | Project 1.3.1 211

Reef Growth and Maintenance

Location and core no: Adele Island (A09) Core length: 137 cm Compaction: 47 cm Penetration: 184 cm

Photo Description

(0–25 cm): Unconsolidated domal coral floatstone (mostly Chypastrea) with grey coarse bio-clastic sand matrix (coral and encrusting coralline algae fragments).

(25–137 cm): Unconsolidated domal coral floatstone (mostly Porites) with grey muddy bio-clastic sand matrix.

Dating samples:

Matrix samples:

212 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Location and core no: Adele Island (A10) Core length: 192 cm Compaction: 33 cm Penetration: 225 cm

Photo Description

(0–25 cm): Unconsolidated oxidized brownish coarse bio-clastic sand unit with abundant rhodoliths and molluscs, foram, gastropods, coral and coralline algae fragments.

(55–192 cm): Unconsolidated domal coral floatstone with reduced (anoxic) grey sand matrix. Coral colonies: Chypastrea, Porites, Favia.

Dating samples:

Matrix samples:

Kimberley Marine Research Program | Project 1.3.1 213

Reef Growth and Maintenance

Appendix V

Core Details

68 88 66 41 440 100 188 100 362 101 153 112 163 126 176 408 (cm) Penetration

7 59 14 21 66 43 34 74 193 114 130 N/A N/A N/A N/A N/A (cm) Compaction

68 88 52 34 80 87 92 247 100 129 100 248 112 120 102 278 (cm) Core Length

Method Drilling core Drilling core Drilling core Drilling core Drilling core ercussion core P Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core

Lat

16° 24' 24' 16° 52.62" 16° 24' 24' 16° 36.57" 16° 23' 23' 16° 41.25" 16° 23' 23' 16° 43.476" 16° 24' 24' 16° 38.706" 16° 24' 24' 16° 50.844" 16° 24' 24' 16° 38.664" 16° 24' 24' 16° 40.602" 16° 24' 24' 16° 52.476" 16° 24' 24' 16° 41.076" 16° 23' 23' 16° 47.658" 16° 23' 23' 16° 49.614" 16° 24' 24' 16° 59.417" 16° 24' 24' 16° 57.522" 16° 23' 23' 16° 50.244" 16° 23' 23' 16° 46.290"

GPS position

11.046"

Long 123° 9' 9' 123° 55.11" 123° 123° 8' 8' 123° 18.042" 123° 10'123° 0.246" 123° 8' 8' 123° 18.828" 123° 8' 8' 123° 20.766" 123° 10'123° 2.292" 123° 8' 8' 123° 20.658" 123° 10'123° 3.432" 123° 10'123° 2.160" 123° 10'123° 1.908" 123° 9' 9' 123° 53.652" 123° 10'123° 1.800" 123° 8' 8' 123° 18.906" 123° 10'123° 6.918" 123° 10'123°

10' 9.356"

Location ast Tallon Island Tallon ast E East Tallon Island Tallon East East Tallon Island Tallon East East Tallon Island Tallon East East Tallon Island Tallon East S. Passage Sunday I. Sunday Passage S. S. Passage Sunday I. Sunday Passage S. S. Passage Sunday I. Sunday Passage S. S. Passage Sunday I. Sunday Passage S. S. Passage Sunday I. Sunday Passage S. N. Passage I. Sunday Passage N. N. Passage I. Sunday Passage N. N. Passage I. Sunday Passage N. N. Passage I. Sunday Passage N. N. Passage I. Sunday Passage N. N. Passage I. Sunday Passage N.

Date collected 7/07/2014 4/07/2014 7/07/2014 7/07/2014 3/07/2014 4/07/2014 7/07/2014 4/07/2014 3/07/2014 6/07/2014 6/07/2014 3/07/2014 7/07/2014 3/07/2014 6/07/2014 6/07/2014

E PS1 ET3 ET4 PS2 ET5 PS3 PS4 ET1 PN3 PN4 PN2 PN1 PN5 PS2B PN4B Core ID

7 1 8 9 2 3 4 5 6 14 15 16 10 11 12 13 No

214 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

25 60 90 70 80 22 66 99 135 215 134 626 125 580 121 105 138 447 632 N/A N/A (cm) Penetration

1 28 56 35 10 22 58 14 216 200 172 162 143 130 N/A N/A N/A N/A N/A N/A N/A (cm) Compaction

25 59 99 90 70 70 22 66 80 85 107 410 159 103 380 275 121 105 470 N/A N/A (cm) Core Length

illing core Method Dr Drilling core Drilling core Drilling core Drilling core Drilling core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Drilling/percussion Drilling/percussion Drilling/percussion Drilling/percussion

Lat

24' 24' 10.638" 16°3'26.55" 16°24'11.13" 16°24'12.75" 16°2'17.742" 16°2'17.886" 16°3'23.244" 16°3'16.662" 16°2'19.344" 16°3'26.232" 16°2'43.548" 16°3'26.292" 16°3'15.864" 16°3'23.238" ------16° 24' 24' 16° 37.02" 16° 24' 24' 16° 10.35" 16° 24' 24' 16° 31.722" 16° 24' 24' 16° 35.094" 16° 16° 24' 24' 16° 25.668" 16° 24' 24' 16° 10.872" 16° 24' 24' 16° 10.938"

GPS position

Long 123°7'1.89" 123°7'6.672" 123° 7' 7' 123° 59.88" 123° 7' 7' 123° 44.82" 123°32'9.864" 123°31'8.148" 123°32'47.64" 123°32'31.71" 123°32'42.57" 123°32'9.738" 123° 8' 8' 123° 11.238" 123° 8' 8' 123° 18.108" 123° 8' 8' 123° 19.752" 8' 123° 17.646" 123° 8' 8' 123° 17.538" 123° 8' 8' 123° 17.784" 123°32'54.096" 123°32'54.359" 123°32'57.864" 123°32'52.926" 123°31'19.608"

Irv Irv - -

I Island Pool Pool Island Island Pool Pool Island Island Pool Island Pool - - - Island Reef Edge - Island Reef Edge Location Island Bath Island Island Bath Island - - - st Bathurst Island Bathurst st East Tallon Island Tallon East Island Tallon East East Tallon Island Tallon East East Tallon Island Tallon East East Tallon Island Tallon East Island Tallon East East Tallon Island Tallon East East Tallon Island Tallon East West West Tallon Island West West Tallon Island S. Inter - S. N. Inter Pool sland W. Inter W. Inter Inter Inter West Bathurst Island Bathurst West West Bathurst Island Bathurst West We W. Inter NW. Inter

Date collected 1/08/2014 1/08/2014 9/07/2014 9/07/2014 3/08/2014 3/08/2014 6/08/2014 4/08/2014 3/08/2014 8/07/2014 8/07/2014 6/08/2014 7/07/2014 8/07/2014 8/07/2014 8/07/2014 4/08/2014 5/08/2014 5/08/2014 5/08/2014 5/08/2014

-b ID ET7 ET6 ET8 ET9 B01 B02 B09 B0 B04 B10 B05 B06 B07 B08 WT1 WT2 Core ET12 ET13 ET11 ET10 B08

27 25 28 23 24 26 36 29 30 3 37 18 22 17 19 20 21 31 32 33 34 35 No

Kimberley Marine Research Program | Project 1.3.1 215

Reef Growth and Maintenance

11080

- 35 225 506 310 227 184 414 347 427 165 386 194 625 573 N/A (cm) Penetration

33 95 67 97 87 47 35 54 156 103 172 100 178 118 157 N/A (cm) Compaction

- 8187 35 350 192 215 160 311 250 340 137 214 130 140 525 455 N/A (cm) Core Length

Method

Drilling core ercussion core P Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Percussion core Drilling/percussion Total Total

Lat

16.06621 16°3'46.5" - - 16°3'38.586" 16°3'56.454" 16°3'22.752" - - - 16° 26' 26' 16° 25.65" 15° 33' 33' 15° 33.03" 16° 26' 26' 16° 5.904" 15° 28' 28' 15° 36.084" 15° 34' 34' 15° 54.474" 15° 26' 26' 15° 14.862" 26' 15° 40.608" 28' 15° 37.344" 15° 30' 30' 15° 20.244" 34' 15° 16.692" 16° 26' 26' 16° 19.662"

GPS position

Long 123.55001 123° 7' 7' 123° 46.05" 123° 9' 9' 123° 22.65" 10'123° 49.5" 10'123° 56.2" 123° 10'123° 56.71" 123° 6' 6' 123° 56.982" 6' 123° 36.732" 7' 123° 18.522" 123° 9' 9' 123° 40.644" 123° 9' 9' 123° 41.166" 123° 9' 9' 123° 38.232" 123°31'21.888" 123°32'52.182" 123°32'56.346" 123°31'36.714"

Island Reef - Edge and cay Location S Sand cay Sandbank Irvine I. Pool Irvine Irvine I. Pool Irvine Irvine I. Pool Irvine

outh Adele Island Adele outh S Island Adele North Island Adele North South Adele Island Adele South Island Adele South South Sunday Island Sunday South South Sunday Island Sunday South South Sunday Island Sunday South SW. Inter SW. Sand cay (near island) (near cay Sand

Date collected 7/08/2014 8/08/2014 8/08/2014 8/08/2014 9/08/2014 18/10/2014 21/10/2014 18/10/2014 17/10/2014 17/10/2014 17/10/2014 17/10/2014 18/10/2014 18/10/2014 21/10/2014 21/10/2014

ID B11 B12 B13 B14 B15 A10 A04 A05 A06 A07 A08 A09 A03 Core Sun01 Sun02 Sun03

51 44 45 46 50 47 48 49 38 52 53 39 40 41 42 43 No

216 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

PN = North Passage of Sunday Island PS = South Passage of Sunday Island ET = East Tallon Island WT = West Tallon Island B = Bathurst and Irvine Island N/A = not available

Kimberley Marine Research Program | Project 1.3.1 217

Reef Growth and Maintenance

Appendix VI

Core Geochronology

218 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

3.57 0.63 0.49 1.12 0.37 0.31 0.54 7.02 0.72 0.47 1.94 0.23 2.32 1.68 Accretion (mm/year)

55 7165 3870 1220 2715 6410 7245 285 1690 2960 7645 1515 2440 540 5420 5005 1070 5445 2725 - -

post 0 post 0 post 0 ------(68% (68%

cal BP 265 Post 0 BP BP 0 Post 450 660 Calibrated Probability) 240 230 120 7305 4075 1320 2840 6605 7395 1865 3175 7790 1685 2690 5570 5265 1250 5805 2850

30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 ------

Age (BP)

390 +/ 590 +/ 540 +/ 780 +/ 530 +/ 480 +/

6770 +/ 4030 +/ 1770 +/ 3060 + 6140 +/ 2240 +/ 3310 +/ 2080 +/ 2870 +/ 1090 +/ 1660 +/ 3070 +/ 6840 +/ 7310 +/ 5170 +/ 4890 +/ 5210 +/

Conventional

C 12

C/ 13 0.3 o/oo 0.4 o/oo 0.1 o/oo 0.3 o/oo 3.2 o/oo 3.0 o/oo 1.3 o/oo 1.2 o/oo 0.0 o/oo 0.0 o/oo ------+0.4 o/oo+0.4 +1.8 o/oo+1.8 +1.0 o/oo+1.0 +0.3 o/oo+0.3 +0.4 o/oo+0.4 +0.9 o/oo+0.9 +0.7 o/oo+0.7 +1.0 o/oo+1.0 +0.5 o/oo+0.5 +1.0 o/oo+1.0 +0.5 o/oo+0.5 +0.3 o/oo+0.3 +1.8 o/oo+1.8

30 30 30 30 30 30 30 30 30 30 30 30 30 30 0.4 30 30 30 30 30 30 30 ------30 ------

------/ - -

Age pMC 90 +/ 370 +/ 170 +/ 650 +/ 140 +/ Measured 130 +/

6410 + 5730 +/ 6890 +/ 4470 +/ 2630 +/ 1840 +/ 4790 +/ 1670 +/ 2460 +/ 4760 +/ 2710 +/ 6360 +/ 3610 +/ 1330 +/ 2880 +/ 1300 +/

100.4 +/

AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS diometric Method Radiometric Ra

Coral Coral Coral Coral Coral Coral Platygra Material Acropora Acropora Acropora Acropora Acropora Acropora Rhodolith Seriatophora? Coralline algae Coralline algae Coralline algae Coralline algae Coralline algae Porites Cylindrica Porites Cylindrica Porites Cylindrica

385992 385991 389410 389407 385988 385993 397767 389406 385994 389408 385990 385989 397766 389469 385995 391789 391788 389409 389404 397765 397764 389470 389405 ------Lab. Code eta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta B Beta Beta Beta Beta

-1 0.3 0.9 1.2 0.1 0.3 (m) 3.13 0.13 0.23 5.73 5.73 1.32 1.23 0.74 0.87 3.19 0.08 2.02 1.15 0.12 0.25 2.94 - - - - - 0.245 ------Depth -

23 90 10 87 13 25 30 132 202 30 74 12 -8 - - - - 120 319 573 294 - - - 313 573 123 100 115 - - - - - 24.5 - -

------ET2 Name ET7 ET1 ET8 WTI PN3 Sample PN1 WT2 ET7 ET12 ET7 ET11 ET1 ET11 ET2 ET8 WTI WTI PN3 PN3 ET1 ET12 ET12

unday I. unday Tallon I. S Location

Kimberley Marine Research Program | Project 1.3.1 219

Reef Growth and Maintenance

80 0.28 1.86 2.16 4.91 0.29 0.81 2.42 0.67 1.43 0.39 2.84 1.02 Accretion (mm/year)

3395 4395 4795 1415 3165 6 4865 465 415 2870 2395 1400 4625 565 2740

p post 0 post 0 post 0 ------cal BP 545 505 680 Calibrated ost 0 255 260 120 230 3570 4565 4905 1595 3350 6780 5115 625 3070 2680 1565 4825 2900 (68% Probability)(68%

30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 ------

Age (BP)

950 +/ 580 +/ 880 +/ 480 +/ 530 +/ 570 +/

3240 +/ 4590 +/ 3650 +/ 4400 +/ 4680 +/ 2020 +/ 3450 +/ 4810 +/ 6310 +/ 2850 +/ 2000 +/ 1130 +/ 3110 +/

Conventional

C 12 o/oo

C/ 13 2.9 o/oo 2.1 o/oo 1.1 o/oo 1. 1.2 o/oo 0.1 o/oo 0.2 o/oo 1.0 o/oo 0.7 2.2 o/oo 1.0 o/oo - - - 7 o/oo ------+1.3 o/oo+1.3 +0.5 o/oo+0.5 +1.7 o/oo+1.7 +4.4 o/oo+4.4 +1.0 o/oo+1.0 +1.3 o/oo+1.3 +1.1 o/oo+1.1 +0.5 o/oo+0.5

30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 ------30 ------

------

Age 80 +/ 540 +/ 490 +/ Measured 160 +/ 160 +/ 140 +/ 740 +/

4420 +/ 5900 +/ 2800 +/ 2370 +/ 4160 +/ 4040 +/ 4300 +/ 1630 +/ 3220 +/ 1570 +/ 3070 +/ 2690 +/

AMS AMS AMS AMS AMS AMS AMS AMS AMS Method Radiometric Radiometric Radiometric Radiometric Radiometric Radiometric Radiometric Radiometric Radiometric Radiometric

Favia Coral Coral Coral Pavia Porites Porites Porites Galaxea Galaxea Galaxea Galaxea Porites? Material Rhidoliths Acropora? Acropora? Coralline algae Coralline algae Porites Cylindrica

389472 389473 389474 389471 389475 389476 385996 389478 385997 389479 389480 389411 391787 389413 385998 389414 389477 389415 389416 ------Lab. Code Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta

-1 -6 4.1 0.2 0.5 1.3 0.9 1.6 0.3 (m) 3.08 3.88 1.69 1.22 2.71 1.84 4.27 0.09 0.16 3.85 ------Depth

50 16 30 20 90 -9 308 388 169 122 271 410 130 427 160 385 600 184 100 ------

------PS3 Name PS1 PS3 B04 B05 B04 Sample PS1 PS4 B04 B04 B05 B05 B05 PN1 PN1 PN5 PN1 PN5 PN5

- -

athurst Irvine I. Irvine Location Sunday I. Sunday B

220 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

0.4 8.9 1.9 11.5 0.47 4.38 4.46 0.55 54.4 14.87 14.76 21.81 Accretion (mm/year)

190 2275 280 120 420 270 2005 2860 300 3345 4 240 265 325 post 0 post 0 post 0 post 0 post 0 ------cal BP 15 615 - 320 - 420 - 490 - 305 440 285 510 425 470 Calibrated 245 130 120 250 2365 130 2200 3060 3500 (68% Probability)(68%

30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 ------+/ Age (BP)

550 +/ 670 +/ 690 +/ 770 +/ 490 +/ 740 +/ 630 890 +/ 840 +/ 750 +/ 480 +/ 810 +/ 560 +/ 490 +/ 2520 +/ 3230 +/ 3590 +/ 2690 +/ 1000 +/

Conventional

C 12 o/oo C/ 13 0.3 o/oo 1. 0.5 o/oo 1.7 o/oo 0.2 o/oo 1.1 o/oo 1.0 o/oo 2.1 o/oo 0.7 o/oo 0.4 o/oo 2.5 o/oo - - 0 o/oo ------+0.4 o/oo+0.4 +0.5 o/oo+0.5 +1.4 o/oo+1.4 +2.2 o/oo+2.2 +3.3 o/oo+3.3 +0.3 o/oo+0.3 +0.4 o/oo+0.4 +2.6 +2.6

30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 ------30 30

------Age 90 +/ 70 +/ 580 +/ 120 +/ Measured 140 +/ 270 +/ 340 +/ 360 +/ 220 +/ 500 +/ 360 +/ 390 +/ 430 +/ 110 +/ 280 +/

2270 +/ 2060 +/ 2830 +/ 3190 +/

AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS AMS Method Radiometric Radiometric Radiometric Radiometric Radiometric Radiometric

Coral Coral Coral Coral Coral Coral Favia Coral Coral Porites Porites Material Acropora Acropora Acropora Acropora Coralline algae Coralline algae Coralline algae Coralline algae

389417 389418 389485 389425 389419 389481 389426 389420 389482 389421 391046 389483 389422 391047 389484 391048 389423 391043 389424 ------Lab. Code eta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta Beta B Beta Beta

0.3 (m) 0.12 0.81 1.14 3.25 0.35 0.35 5.54 2.38 0.64 3.67 0.54 0.12 5.71 1.01 1.89 0.16 0.18 1.54 ------Depth

35 64 12 81 35 54 12 30 16 18 - - 114 325 554 238 367 571 101 189 154 ------Name B08 B08 B12 B15 B09 B09 B14 B13 Sample B09 B14 B14 B12 B12 B12 B15 B15 B14 B08b B08b

- -

Irvine I. Irvine Location Bathurst

Kimberley Marine Research Program | Project 1.3.1 221

Reef Growth and Maintenance

Appendix VII

Dating Samples

Pre-treatment Post-treatment Explanation Sample ID: 3 Easting: 564585 Northing: 8220316 Coral colony: PL calcrete Elevation: -18.1 m Status: dated Dating technique: AMS

Sample ID: 30 Easting: 565124 Northing: 8219976 Coral colony: Faviid Elevation: -14.263 m Status: dated Dating technique: AMS

Sample ID: ET7-A Long: 123° 7' 44.82" Lat: -16° 24' 25.668" Elevation: -0.3 m Colony: Acropora Status: dated Dating technique: AMS

Sample ID: ET7-C Long: 123° 7' 44.82" Lat: -16° 24' 25.668" Elevation: -3.13 m Colony: Acropora Status: dated Dating technique: AMS

222 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Sample ID: ET7-E Long: 123° 7' 44.82" Lat: -16° 24' 25.668" Elevation: -5.73 m Colony: Acropora Status: dated Dating technique: AMS

Sample ID: ET1-A Long: 123° 8' 18.906" Lat: - 16° 24' 36.57" Elevation: -0.245 m Colony: Acropora Status: dated Dating technique: AMS

Sample ID: ET1-B Long: 123° 8' 18.906" Lat: - 16° 24' 36.57" Elevation: -0.74 m Colony: Acropora Status: dated Dating technique: AMS

Sample ID: ET1-C Long: 123° 8' 18.906" Lat: - 16° 24' 36.57" Elevation: -1.23 m Colony: Acropora Status: dated Dating technique: AMS

Sample ID: PN3-B Long: 123° 10' 3.432" Lat: - 16° 23' 47.658" Elevation: -1.2 m

Kimberley Marine Research Program | Project 1.3.1 223

Reef Growth and Maintenance

Colony: Acropora Status: dated Dating technique: AMS

Sample ID: PN3-D Long: 123° 10' 3.432" Lat: -16° 23' 47.658" Elevation: -3.19 m Colony: Acropora Status: dated Dating technique: AMS

Sample ID: PS1-A Long: 123° 10' 0.246" Lat: -16° 24' 50.844" Elevation: -0.2 m Colony: Acropora Status: dated Dating technique: AMS

Sample ID: PS1-C Long: 123° 10' 0.246" Lat: -16° 24' 50.844" Elevation: -1.84 m Colony: Acropora Status: dated Dating technique: AMS

Sample ID: PS4 Long: 123° 9' 53.652" Lat: -16° 24' 57.522" Elevation: -1m Colony: Acropora Status: dated Dating technique: AMS

224 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Sample ID: WT1-30 Long: 123°7'6.672" Lat: -16°24'12.75 Elevation: -0.3 m Colony: Acropora Status: Sent Dating technique: AMS

Sample ID: WT1-250 Long: 123°7'6.672" Lat: -16°24'12.75 Elevation: -2.5 m Colony: Acropora covered by CA Status: unprepared

Sample ID: WT1-294 Long: 123°7'6.672" Lat: -16°24'12.75 Elevation: -2.94 m

Colony: coral Status: Sent Dating technique: AMS

Sample ID: WT1-573 Long: 123°7'6.672" Lat: -16°24'12.75 Elevation: -5.73 m

Colony: Porites Cylindrica Status: Sent Dating technique: AMS

Sample ID: WT2-87 Long: 123°7'1.89" Lat: -16°24'11.13" Elevation: -0.87 m

Colony: Porites Cylindrica

Kimberley Marine Research Program | Project 1.3.1 225

Reef Growth and Maintenance

Status: Sent Dating technique: Radiometric

Sample ID: ET12-23 Long: 123° 7' 59.88" Lat: 16° 24' 31.722" Elevation: -0.23 m

Colony: coral Status: Sent Dating technique: AMS

Sample ID: ET12-132 Long: 123° 7' 59.88" Lat: 16° 24' 31.722" Elevation: -1.32 m Colony: Acropora? Status: Sent Dating technique: AMS

Sample ID: ET12-202 Long: 123° 7' 59.88" Lat: 16° 24' 31.722" Elevation: -2.02 m

Colony: coral Status: Sent Dating technique: AMS

Sample ID: PN1-25 Long: 123° 10' 11.046" Lat: 16° 23' 41.25" Elevation: -0.25 m

Colony: coral Status: Sent Dating technique: Radiometric

226 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Sample ID: PN1-122 Long: 123° 10' 11.046" Lat: 16° 23' 41.25" Elevation: -1.22 m

Colony: Favia? Status: Sent Dating technique: Radiometric

Sample ID: PN1-308 Long: 123° 10' 11.046" Lat: 16° 23' 41.25" Elevation: -3.08 m

Colony: coral Status: Sent Dating technique: Radiometric

Sample ID: PN1-388 Long: 123° 10' 11.046" Lat: 16° 23' 41.25" Elevation: -3.88 m Colony: Rhodoliths

Status: Sent Dating technique: Radiometric

Sample ID: PN3-13 Long: 123° 10' 3.432" Lat: 16° 23' 47.658" Elevation: -0.13 m

Colony: Rhodoliths Status: Sent Dating technique: AMS

Kimberley Marine Research Program | Project 1.3.1 227

Reef Growth and Maintenance

Sample ID: PN5-32 Long: 123° 10' 9.356" Lat: 16° 23' 43.476" Elevation: -0.32 m

Colony: coral Status: Ready

Sample ID: PN5-169 Long: 123° 10' 9.356" Lat: 16° 23' 43.476" Elevation: -1.69 m

Colony: Porites Status: Sent Dating technique: Radiometric

Sample ID: PN5-271 Long: 123° 10' 9.356" Lat: 16° 23' 43.476" Elevation: -2.71 m Colony: Galaxea

Status: Sent Dating technique: Radiometric

Sample ID: PN5-410 Long: 123° 10' 9.356" Lat: 16° 23' 43.476" Elevation: -4.1 m Colony: Porites

Status: Sent Dating technique: Radiometric

Sample ID: PS3-9 Long: 123° 9' 55.11" Lat: 16° 24' 59.417" Elevation: -0.09 m Colony: Acropora covered by CA

228 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Status: Sent Dating technique: AMS

Sample ID: PS3-90 Long: 123° 9' 55.11" Lat: 16° 24' 59.417" Elevation: -0.9 m Colony: Acropora covered by CA

Status: Sent Dating technique: AMS

Sample ID: B04-30 Long: 123°32'52.926" Lat: -16°3'26.232" Elevation: -0.3 m Colony: Encrusting Coralline Algae

Status: Sent Dating technique: Radiometric

Sample ID: B04-50 Long: 123°32'52.926" Lat: -16°3'26.232" Elevation: -0.5 m

Colony: Porites Status: Sent Dating technique: Radiometric

Sample ID: B04-53 Long: 123°32'52.926" Lat: -16°3'26.232" Elevation: -0.53 m

Colony: coral Status: unprepared

Kimberley Marine Research Program | Project 1.3.1 229

Reef Growth and Maintenance

Sample ID: B04-130 Long: 123°32'52.926" Lat: -16°3'26.232" Elevation: -1.3 m Colony: Favia

Status: Sent Dating technique: Radiometric

Sample ID: B04-292 Long: 123°32'52.926" Lat: -16°3'26.232" Elevation: -2.92 m Colony: Encrusting CA

Status: unprepared

Sample ID: B04-427 Long: 123°32'52.926" Lat: -16°3'26.232" Elevation: -4.27 m Colony: coral

Status: Sent Dating technique: Radiometric

Sample ID: B05-16 Long: 123°32'47.64" Lat: -16°3'26.292" Elevation: -0.16 m

Colony: Galaxea Status: Sent Dating technique: AMS

Sample ID: B05-160 Long: 123°32'47.64" Lat: -16°3'26.292" Elevation: -1.6 m Colony: Galaxea

230 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Status: Sent Dating technique: AMS

Sample ID: B05-385 Long: 123°32'47.64" Lat: -16°3'26.292" Elevation: -3.85 m Colony: Galaxea

Status: Sent Dating technique: AMS

Sample ID: B05-600 Long: 123°32'47.64" Lat: -16°3'26.292" Elevation: -6 m

Colony: Porites Cylindrica Status: Sent Dating technique: AMS

Sample ID: B08-12 Long: 123°32'9.738" Lat: -16°3'23.238" Elevation: -0.12 m

Colony: Acropora covered by encrusting CA Status: Sent Dating technique: AMS

Sample ID: B08-81 Long: 123°32'9.738" Lat: -16°3'23.238" Elevation: -0.81 m Colony: Acropora covered by encrusting CA Status: Sent Dating technique: AMS

Kimberley Marine Research Program | Project 1.3.1 231

Reef Growth and Maintenance

Sample ID: B08b-35 Long: 123°32'9.864" Lat: -16°3'23.244" Elevation: -0.35 m Colony: Encrusting CA

Status: Sent Dating technique: Radiometric

Sample ID: B08b-64 Long: 123°32'9.864" Lat: -16°3'23.244" Elevation: -0.64 m Colony: Porites + Encrusting CA

Status: Sent Dating technique: Radiometric

Sample ID: B09-12 Long: 123°31'8.148" Lat: -16°3'16.662" Elevation: -0.12 m Colony: Porites Status: Sent Dating technique: Radiometric

Sample ID: B09-30 Long: 123°31'8.148" Lat: -16°3'16.662" Elevation: -0.3 m

Colony: Encrusting CA Status: Sent Dating technique: Radiometric

232 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Sample ID: B09-114 Long: 123°31'8.148" Lat: -16°3'16.662" Elevation: -1.14 m Colony: Porites

Status: Sent Dating technique: Radiometric

Sample ID: B12-35 Long: 123°32'52.182" Lat: -16°3'56.454" Elevation: -0.35 m Colony: Acropora

Status: Sent Dating technique: AMS

Sample ID: B12-238 Long: 123°32'52.182" Lat: -16°3'56.454" Elevation: -2.38 m

Colony: Acropora Status: Sent Dating technique: AMS

Sample ID: B12-367 Long: 123°32'52.182" Lat: -16°3'56.454" Elevation: -3.67 m Colony: Acropora Status: Sent Dating technique: AMS

Kimberley Marine Research Program | Project 1.3.1 233

Reef Growth and Maintenance

Sample ID: B12-571 Long: 123°32'52.182" Lat: -16°3'56.454" Elevation: -5.71 m Colony: coral

Status: Sent Dating technique: AMS

Sample ID: B14-16 Long: 123.55001 Lat: -16.06621 Elevation: -0.16 m Colony: Favia

Status: Sent Dating technique: AMS

Sample ID: B14-154 Long: 123.55001 Lat: -16.06621 Elevation: -1.54 m Colony: coral

Status: Sent Dating technique: AMS

Sample ID: B14-325 Long: 123.55001 Lat: -16.06621 Elevation: -3.25 m Colony: coral Status: Sent Dating technique: AMS

234 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Sample ID: B14-554 Long: 123.55001 Lat: -16.06621 Elevation: -5.54 m Colony: Porites Cylindrica Status: Sent Dating technique: AMS

Kimberley Marine Research Program | Project 1.3.1 235

Reef Growth and Maintenance

Appendix VIII

Australian Journal of Maritime and Ocean Affairs paper

236 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Kimberley Marine Research Program | Project 1.3.1 237

Reef Growth and Maintenance

238 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Kimberley Marine Research Program | Project 1.3.1 239

Reef Growth and Maintenance

240 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Kimberley Marine Research Program | Project 1.3.1 241

Reef Growth and Maintenance

242 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Kimberley Marine Research Program | Project 1.3.1 243

Reef Growth and Maintenance

244 Kimberley Marine Research Program | Project 1.3.1

Reef Growth and Maintenance

Kimberley Marine Research Program | Project 1.3.1 245

Reef Growth and Maintenance

246 Kimberley Marine Research Program | Project 1.3.1