This page intentionally left blank
THE GEOMORPHOLOGY OF THE GREAT BARRIER REEF
Over the last 25 years considerable i...
347 downloads
842 Views
12MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
This page intentionally left blank
THE GEOMORPHOLOGY OF THE GREAT BARRIER REEF
Over the last 25 years considerable information on the geomorphological evolution of the world’s largest coral reef system, the Great Barrier Reef, has become available. This book reviews the history of geomorphological studies of the Great Barrier Reef and assesses the influences of sea-level change and oceanographic processes on the development of reefs over the last 10 000 years. It presents analyses of recently attained data from the Great Barrier Reef and reconstructions of the sequence of events that have led to its current geomorphology. The authors emphasize the importance of the geomorphological time span and its relevance for present management applications. This is a valuable reference for academic researchers in geomorphology and oceanography, and will also appeal to graduate students in related fields. D A V I D H O P L E Y is Adjunct Professor in the School of Earth and Environmental Sciences (formerly the School of Tropical Environment Studies and Geography) at James Cook University, Queensland, Australia. He has spent over 40 years working on the Great Barrier Reef and has been a consultant in coastal and coral reef management since 1997. S C O T T S M I T H E R S is Senior Lecturer in the School of Earth and Environmental Sciences at James Cook University. He has worked on the Great Barrier Reef and in the Pacific and Indian Oceans. His broad research interests are in the Quaternary evolution of coastal environments, especially coral reefs and tropical coasts. K E V I N P A R N E L L is Associate Professor in the School of Earth and Environmental Sciences at James Cook University. After completing a Ph.D. at JCU, he worked on temperate beach systems at the University of Auckland before returning to JCU in 2003, undertaking research on reef and tropical beach systems.
THE GEOMORPHOLOGY OF THE GREAT BARRIER REEF Development, Diversity, and Change DAVID HOPLEY SCOTT G. SMITHERS KEVIN E. PARNELL James Cook University
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521853026 © D. Hopley, S. Smithers and K. Parnell 2007 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2007 eBook (EBL) ISBN-13 978-0-511-28524-0 ISBN-10 0-511-28524-8 eBook (EBL) hardback ISBN-13 978-0-521-85302-6 hardback ISBN-10 0-521-85302-8
Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.
Contents
Preface Acknowledgements 1 Geomorphology and the Great Barrier Reef 1.1 Introduction 1.2 The role of geomorphology in the understanding of coral reefs 1.3 A chronicle of geomorphology and reef research 1.4 The history of geomorphological study of the Great Barrier Reef to 1982 1.5 Outline of the following chapters 2 Foundations of the Great Barrier Reef 2.1 Introduction 2.2 Geological and geomorphological development of the coast 2.3 Evolution of the Coral Sea 2.4 The continental shelf of north-east Australia 2.5 The late establishment of the Great Barrier Reef 2.6 The Pleistocene Reef 2.7 Conclusion 3 Sea level: a primary control of long-term reef growth and geomorphological development 3.1 Introduction 3.2 Quaternary sea-level change 3.3 Late Pleistocene sea level 3.4 Postglacial sea level 3.5 Historical sea-level change on the Great Barrier Reef 3.6 Conclusion v
page ix xii 1 1 3 5 7 16 18 18 19 27 30 34 37 40 42 42 44 49 58 87 90
vi
Contents
4 Oceanography, hydrodynamics, climate, and water quality as influences on reef geomorphological processes 4.1 Introduction 4.2 The climate of the Great Barrier Reef region 4.3 Oceanography and hydrodynamics 4.4 High-frequency waves 4.5 High-intensity events 4.6 Mainland influences 4.7 Oceanographic and climatological stressors 5 Spatial analysis of the morphology of the reefs and islands of the Great Barrier Reef 5.1 Introduction 5.2 Remote sensing and the Great Barrier Reef 5.3 The history of spatial data collection and analysis 5.4 Great Barrier Reef lagoon areas and volumes 5.5 Reef and reef island statistics and classification 5.6 Reef types and reef management 5.7 Conclusion 6 The non-reefal areas of the continental shelf 6.1 Introduction 6.2 Surficial sediments 6.3 Subsurface sediments and the Pleistocene surface 6.4 Low sea-level drainage patterns 6.5 The age of shelf sediments 6.6 The Halimeda bioherms 6.7 Conclusion 7 Fringing and nearshore coral reefs 7.1 Introduction 7.2 Distribution and settings 7.3 Fringing reef structure 7.4 Holocene reef growth 7.5 Fringing reef morphology and processes 7.6 Conclusion 8 The mid-shelf reefs of the Great Barrier Reef 8.1 Introduction 8.2 The data base 8.3 Criteria used to classify the selected reefs 8.4 Evidence of morphological evolution from the internal structure of reefs
92 92 94 99 111 118 125 135 138 138 138 139 146 147 162 165 166 166 167 171 175 180 183 190 191 191 192 202 207 222 231 233 233 235 247 255
Contents
9
10
11
12
8.5 Evolution of windward reef fronts 8.6 Rates of geomorphological development: discussion and conclusions The coral reefs of the outer shelf of the Great Barrier Reef 9.1 Introduction: shelf-edge morphology 9.2 Modes of shelf marginal reef growth and major influences on the growth morphology 9.3 Detailed structure and evolution of the shelf-edge reefs 9.4 Conclusion Islands of the Great Barrier Reef 10.1 Introduction 10.2 Classification and geomorphology of reef islands 10.3 Island distribution 10.4 Reef island formation 10.5 Reef island dynamics 10.6 Discussion: reef island prospects and potentials 10.7 Conclusion The accumulation of the Holocene veneer to the Great Barrier Reef 11.1 Introduction 11.2 The depth to the antecedent surface 11.3 The fabric of the Pleistocene foundation 11.4 Date of recolonization during the Holocene transgression 11.5 Rates of growth and accretion 11.6 The timing of reefs reaching modern sea level 11.7 Reef growth relative to sea-level rise 11.8 Holocene reef structure and facies development 11.9 Comparisons with reefs elsewhere 11.10 How does the Great Barrier Reef compare? 11.11 Conclusion The Holocene evolution of the Great Barrier Reef province 12.1 Introduction 12.2 The glacial maximum low sea level – 20 000 years BP 12.3 The early transgression to 12 000 years BP 12.4 The start of the Holocene – 10 000 years BP 12.5 The final 2000 years of the transgression – 7000 years BP 12.6 The mid to late Holocene 12.7 Conclusion
vii
260 265 271 271 272 276 309 311 311 315 343 346 353 360 365 367 367 368 370 372 375 380 383 386 391 403 408 411 411 412 416 422 424 426 429
viii
Contents
13 Geomorphology’s contribution to the understanding and resolution of environmental problems of the Great Barrier Reef 13.1 Introduction 13.2 Sediments and reefs 13.3 Nutrient excess and the Great Barrier Reef 13.4 Geomorphological assessment for conservation 13.5 Management of reef islands 13.6 Global climate change, geomorphology, and coral reefs 13.7 Conclusion References Geographic index Subject index
431 431 432 442 447 450 459 467 469 519 526
Preface
In the preface to The Geomorphology of the Great Barrier Reef: Quaternary Development of Coral Reefs published by one of the present authors in 1982, the opportunity for a synthesis of ideas on the geomorphology of coral reefs was identified. Almost 25 years later and with a wealth of new research and publications, there is again the need for a holistic view of the evolution of the present geomorphological features of the world’s largest coral reef system, which it is hoped this book will provide. However, it is very different from the 1982 publication which attempted to fill a wide area of coral reef science, using the Great Barrier Reef as an example. This volume is much more focused on the Great Barrier Reef (GBR) region and the way its features have evolved especially during the Holocene period of the last 10 000 years. Much of the data for this period has come from programs of drilling into the reef to depths up to 25 m during the 1980s and 1990s, some of it for specific engineering or non-geomorphological purposes. By far the largest programs, however, were those headed by Professor Peter Davies (now Sydney University) of what was then the Bureau of Mineral Resources, Canberra, and one of the present authors (D. H.) and his postgraduate students. These and other drilling programs have created a data bank which could only be imagined in 1982 but it is not the only area in which the geosciences have added to the understanding of the development and processes which sustain the Reef. Studies of sedimentation patterns, hydrodynamics, and other geomorphological processes are integral areas of coastal geomorphology but over the last ten years in particular on the GBR such studies have often been undertaken by non-geoscientists. Whilst the quality of the data collected is unquestionable its use and interpretation has sometimes suffered from a lack of understanding of geomorphological processes, a theme that is taken up in the latter part of this book. The geomorphological timescale is also a feature of the present work. In the past 15 years there has often been a division between geologists who see reefs as ix
x
Preface
robust systems, surviving major climate and sea-level change over millions of years, and ecologists with a contrary view, monitoring the decline in reef systems over the last 50 years or more and interpreting them as fragile. The timescales used by each discipline are critical to the contradictory interpretations both of which are correct within their own dimensions. The boundary between the two is not sharp and is covered by the period considered basic to geomorphological understanding. Even since sea level reached its present position 6500 years ago, the GBR has changed enormously. It will be shown that maximum growth rates and maximum number of habitats occurred in early to mid-Holocene times. According to parameters by which ecologists may evaluate the health of a reef system, the GBR is already in a state of natural decline without any consideration of human impact. This needs to be acknowledged by management agencies that may only recognize the dynamic nature of the reef system at an ecological scale, for example, the importance of natural disturbances in creating biological diversity. However, these disturbances are superficial and changes, for example, to reef morphology and natural sediment build-up are measured at the geomorphological timescale and this provides the background trend upon which ecological periodicity is superimposed. Thus the usefulness of geomorphology for reef management provides the theme for the final chapter in this book, drawing on the information provided earlier. The book moves from long- and short-term processes (sea-level change and oceanography) through an analysis of the GBR on a basic spatial division (inter-reefal areas, fringing reefs, mid-shelf reefs, outer shelf reefs, and reef islands). The final chapters provide a more holistic view of the data, describing the processes and rates of GBR evolution during the Holocene, and the way in which the Reef has changed dramatically over a relatively short period of 10 000 years, changes that were witnessed by the original Australians. An enormous amount of new information has become available over the last 25 years and we have attempted the task of summarizing this and incorporating it into our ideas of how the GBR has evolved. Even as the manuscript was being written it was clear that the data flow is if anything increasing and it is our conclusion that, as in other disciplines, compilations that build on the foundations laid by earlier workers but incorporating the new data will be needed more frequently than the approximately 25 years since the publication of The Geomorphology of the Great Barrier Reef in 1982. Geomorphology is essential for the understanding of coral reefs and it is through compilations like this that professional geomorphologists can communicate their thoughts, ideas, and data to other disciplines.
Preface
xi
Please note that SI units are used throughout this work except where taken from other works. Abbreviations include time units: millions of years (Ma) and thousands of years (ka). Many radiocarbon dates and other radiometric dates are as recorded in the quoted literature. Unpublished dates from the present authors are all reported as conventional radiocarbon years, without environmental correction.
Acknowledgements
Each of the three authors has been supported in this work by a large list of individuals but collectively we are especially grateful to our cartographic team headed by Adella Edwards of the School of Earth and Environmental Sciences, James Cook University and trainee cartographers under her supervision Klaudia Jochim and Mario Bretschneider. Maps and diagrams are an essential part of this book and their efforts and skills are very much appreciated. A large part of the work incorporated in this book comes from postgraduate students supervised by the authors. Their theses are referenced in the text but included in the long list of individuals are Roger Barnes, Trevor Graham, David Hoyal, Nick Harvey, Peter Isdale, Kay Johnston, Joanie Kleypas, Frazer Muir, Bruce Partain, Cecily Rasmussen, Alison Slocombe, Ann Smith, Andy Steven, Thon Thamrongnawasawat, and Rob van Woesik. The results of a cooperative drilling program focused on the central Great Barrier Reef between James Cook University led by David Hopley and a Bureau of Mineral Resources team led by Peter Davies, together with John Marshall, have provided a major input into the interpretation of the evolution of the Reef. The help, support, and friendship of Professor Peter Davies, now of Sydney University, is gratefully acknowledged. A number of individuals and organizations have provided photographs or have given permission to use diagrams, and these are acknowledged with the appropriate figures. Many of the figures produced for this book incorporate data which is # Commonwealth of Australia (Geoscience Australia) 2001–2004, and/or data provided courtesy of the Great Barrier Reef Marine Park Authority. The assistance of Paul Tudman of the Great Barrier Reef Marine Park Authority in accessing spatial data is gratefully acknowledged. We have endeavored to acknowledge the source of all diagrams used in the book that are not our own. We apologize for any inadvertent omissions. xii
Acknowledgements
xiii
John Dawson, Diane Walker, and Debora de Freitas assisted with reference compilation. Phillippa, Samantha, and Riley Smithers are thanked for their encouragement and tolerance, especially as the book neared completion, but also for whenever dad is away – on a reef, in a boat, or just away without them. Finally, the input of Patricia Hopley in compilation and editing of the manuscript and her organizational support to all three authors has helped to keep the manuscript moving forward and we gratefully thank her for this. David Hopley Scott G. Smithers Kevin E. Parnell
1 Geomorphology and the Great Barrier Reef
1.1 Introduction The Great Barrier Reef (GBR) is the largest coral reef system in the world. It extends from 248 300 S in the south to 98 300 S in the north, a distance of about 2300 km along the north-east shelf of Australia (Fig. 1.1). Accurate estimates of dimensions and other geographical data are available only for the Great Barrier Reef Marine Park (345 500 km2) or the Great Barrier Reef World Heritage Area (348 000 km2) which also includes islands excluded from the Park. Within this area are 2900 reefs occupying over 20 000 km2 or 9% of the 224 000 km2 shelf area (Hopley et al., 1989). However, this administrative area does not include the contiguous shelf of Torres Strait, data for which are more scant. The Strait is 150 km wide and east of the line of high islands, which link Australia to Papua New Guinea, the shelf has a width of over 200 km. Estimated total shelf area here is about 37 000 km2 and, relying on comparative data from the adjacent Great Barrier Reef Marine Park (which ends at 108 420 S) there may be a further 750 reefs and shoals with a total area of about 6000 km2. The GBR is also one of the best studied in the world. Although first described during James Cook’s voyage of exploration in 1770, because of science’s preoccupation with atolls, it did not become a major focus until after the establishment of the Great Barrier Reef Committee in 1922 and the ground-breaking year-long Royal Society Expedition to Low Isles near Cairns in 1928–29 (see below and Bowen and Bowen, 2002). Hopley (1982) summarized the geomorphological knowledge of the Reef as it stood at about 1980. Since then the amount of research has increased exponentially and this book is written with the intention of synthesizing this recent work to produce a new holistic picture of the evolution of the GBR. There is much that can be learnt from the GBR which is also applicable to other reef systems of the world. Its size, extent, and variety of morphology 1
2
Geomorphology and the Great Barrier Reef
Figure 1.1 The Great Barrier Reef and major locations mentioned in Chapters 1 and 2.
1.2 The role of geomorphology in the understanding of coral reefs
3
together with its location close to the center of marine biodiversity (Briggs, 1992, 1999; Wallace, 2002) give it a range of reef morphology that cannot be matched elsewhere. It may not contain atolls but almost every other form of reef is found here. This reflects the latitudinal extent of 158 but even more important are the cross-shelf gradients with distances from mainland coastline to the edge of the continental shelf of up to 300 km (Hopley, 1989a). Thus, whilst the experience of Australia’s largest reef may be most applicable to other shelf barriers such as those found in Papua New Guinea, Indonesia, Madagascar, New Caledonia, and Belize, it shares features with most reef systems elsewhere in the world. With increasing global concern for coral reefs (Wilkinson, 2004), the GBR has importance from two other aspects. First, because of its size, distance offshore, and the absence of a subsistence economy dependent on reef resources living on its adjacent shoreline, there remain many parts of the Reef that may be regarded as pristine and against which other reefs may be compared. This condition has been aided by a large part of the Reef being under the management of the Great Barrier Reef Marine Park Authority (GBRMPA) for almost 30 years. In 2004 the area under complete no-take protection was increased from 4.5% to 33.3%. However, not every part of the GBR is unaffected by anthropogenic activities. The effects of mainland runoff are of major concern especially from the high-rainfall (>3000 mm) region south of Cairns where the GBR comes within 30 km of the coastline. Shipping movements, commercial fishing activities, and a marine tourism industry worth over A$2 billion annually also have impacts on the Reef and large areas of the Reef have been affected by coral bleaching especially in 1997–98 which was the hottest year on record. Thus strategies to tackle these problems including those associated with global warming may also be shared with most other coral reefs. 1.2 The role of geomorphology in the understanding of coral reefs Coral reefs attract a wide range of disciplines as they are built and destroyed by living organisms, are subjected to many physical and chemical processes, and produce landforms which on a geoscientific scale are rapidly changing. Most of these disciplines have contributed to over 25 years of careful biophysical monitoring of the GBR and this has helped to identify natural variability in reef systems and in the environmental parameters affecting them (Done, 1992a). Disturbances have been identified as playing a major role in shaping the community structure of coral reefs but the synergistic effects of natural and new anthropogenic stresses on reef systems are considered as pushing reefs into disturbance regimes from which they cannot recover, a situation termed ‘‘turn-off’’ by Buddemeier and Hopley (1988).
4
Geomorphology and the Great Barrier Reef
To ecologists identifying decline in reef communities during the period of their monitoring programs, reefs have been interpreted as ‘‘fragile’’ ecosystems. In contrast, geologists, observing reefs surviving and evolving within ever-changing environments, perceived reefs as ‘‘robust.’’ The debate polarized the two disciplines in the 1990s (e.g. Davies, 1988; Done, 1991, 1992b; Grigg, 1992, 1994a, b; Kinzie and Buddemeier, 1996). However, Done (1992b) also recognized that the paradox was largely a matter of scale and, quoting Buddemeier and Hopley (1988), pointed out ‘‘the importance of understanding ecological change over annual-to-decadal time scales in bridging the gap between geological and ecological perspectives’’ (Done, 1992b, p. 655). More recently, Grigg (2002) revisited the debate and concluded that both sides were correct, ‘‘depending on the scale of inquiry in space and time.’’ Whilst there is no demarcation line between areas of knowledge, there is clearly a space between ecology and geology which from the point of view of coral reefs may be filled by geomorphology which provides the continuum between the other two disciplines. An analogy may be made with atmospheric study. Ecology represents the day-to-day weather, monitoring of which can put together annual seasonal cycles. Geomorphology represents climate based on records which, for coral reefs, may go back beyond the period of instrumental monitoring. Widening the analogy, observations of tropical cyclones can provide sufficient data to provide risk assessment but the record may be far less than 100 years long. Geomorphological interpretation of storm deposits in beach ridges may allow a longer-term assessment (e.g., Chappell et al., 1983; Nott, 2006) which can give greater confidence to the instrumental record. The climatic analogue may be extended further by relating geological investigation to major climatic changes in the past. Spatially, geomorphology also bridges the gap. At one end of the scale, the study may be of single coral colonies, for example interpreting small-scale sea level changes from undulations in the surface of a microatoll (Smithers and Woodroffe, 2000). At the largest scale, it may provide global-scale comparisons, as for example for the effects of different relative sea level histories in the Holocene on reef development (Hopley, 1982, ch. 13). At and beyond this scale, geomorphology merges into geology. In Australia and elsewhere, geomorphology has developed as part of geography, the essential spatial discipline. Spatial analysis is thus a fundamental part of geomorphology though more recently with the development of computer-based geographical information systems (GIS), other disciplines have encroached upon this area of study. The integrity of geomorphology, however, depends on other elements including study of both modern-day processes and historic evolution. When other non-related disciplines attempt what is
1.3 A chronicle of geomorphology and reef research
5
essentially geoscientific research the results may be seriously misleading. For example, Pastorok and Bilyard (1985) published a table which estimated the degree of impact of sediment on coral reefs, with levels of more than 50 mg cm2 d1considered as being severe to catastrophic. Their figures were widely quoted and suggestions made that they should be used as controls for assessing impacts on GBR waters. However, when the first measurements of sedimentation rates in inshore areas of the Reef were made (Mapstone et al., 1989; Hopley et al., 1990) sedimentation rates more than twice those quoted by Pastorok and Bilyard were found to be everyday occurrences. The cause of this misinterpretation was a lack of appreciation of the local adaptability of corals and even more so the geomorphological processes in the areas in which they obtained their data. These included the largely limestone islands of Barbados and Guam which have little surface runoff and thus naturally low sedimentation rates to which the local corals are adapted. The relationship between reefs and sedimentation rates is discussed in Chapter 13. As Risk (1992) noted: ‘‘a ‘monitoring’ program that does not include sedimentologists (geomorphologists?), chemists and oceanographers as well as biologists is in danger of being useless; without an integrated approach, biological monitoring is a sterile exercise incapable of identifying causes. Ecology is not, and should not be, the sole preserve of biologists.’’ A further example which indicates the degree of specialization that geomorphology brings to reef research is the impact of greenhouse induced sea level rise on coral reef islands. Without an understanding of the processes involved in island formation and erosion far too many commentators, including some scientists, merely raised the waterline against the atolls and cays, predicting that many may disappear altogether in the not-too-distant future (e.g., Falk and Brownlow, 1989; Wells and Edwards, 1989). However, where geomorphologists have taken into account the changes in sediment production on adjacent reef flats and more efficient delivery to the islands, results were very different with the possibility of some islands actually expanding (e.g., McLean, 1989; Parnell, 1989; Hopley, 1993, 1997a; Kench and Cowell, 2002). Other environmental changes may make the atolls uninhabitable but it is misleading to suggest that this ecological niche will not survive. This theme is also taken up in Chapter 13. Geomorphology can make important contributions to other environmental disciplines. It is also an essential ingredient to many management decisions. 1.3 A chronicle of geomorphology and reef research Geomorphological observations of coral reefs are almost as old as the first modern scientific studies which accompanied the voyages of the early European
6
Geomorphology and the Great Barrier Reef
explorers into tropical waters. The major objective of the early scientists or ‘‘naturalists’’ as they were then called was to observe and make collections of the botanical and zoological species that were so new to European eyes. However, they could not avoid seeing and commenting on the proliferation of coral reefs in the areas which they surveyed, especially in the Pacific Ocean. Thus, without the benefit of any underwater observation, many of the first accounts of coral reefs were of their shape, extent, and distribution, essential components of modern geomorphology, defined by Bloom (1978) as ‘‘the systematic description and analysis of landscapes and the processes that change them.’’ It was from the geomorphological observations of naturalists such as Banks, von Chamisso, Quoy, and Gaimard and of navigators such as Cook, Freycinet, and Beechey that the first great coral reef ‘‘problem’’ was identified (for greater discussion see Hopley, 1982, ch. 1; Bowen and Bowen, 2002). Extensive geomorphological data on the apparent simplicity and recurring pattern of the Pacific Ocean atolls was drawn together by Charles Lyell (1797–1875) in the second volume of his Principles of Geology (1832) which devoted the entire final chapter to a summary of all that was known of coral reefs, giving strong support to the idea that atolls had grown on the rims of submerged volcanic craters. The theory was further exemplified by Charles Darwin (1838) who highlighted the apparently anomalous thickness of reefs in relation to the depth at which reefbuilding organisms seemed to flourish (about 100 m). He reasoned that three main types of reef which had been identified by the early explorers and subsequently by scientific voyages such as that of the Beagle – fringing, barrier, and atoll – were genetically related and controlled by slow subsidence. This geomorphological ‘‘problem’’ was to dominate coral reef research for the next 100 years with alternative hypotheses involving antecedent platforms cut by waves, rising depositional banks and sea level change postulated (for discussion see Hopley, 1982; Woodroffe, 2002a). Only deep drilling of an atoll could resolve the problem and in 1896–98 the Royal Society organized the Funafuti Coral Reef Boring Expedition under the leadership of T. Edgeworth David of Sydney University in Australia. Although extending down to 340 m with the upper 194 m in coral limestone, overlying dolomite, this drilling did not conclusively answer the questions regarding the origin of coral reefs as the lower section was interpreted by some as fore reef talus. Only the deep drilling associated with nuclear weapon testing on Bikini, Enewetak, and Mururoa atolls in the 1950s and later finally resolved the problem. Over 1000 m of shallow-water reef limestone was recovered, overlying basaltic (volcanic) foundations. Numerous unconformities marking periods of subaerial exposure also pointed to the major part played by sea level fluctuations in the evolution of modern reef morphology.
1.4 The history of geomorphological study of the Great Barrier Reef to 1982
7
Whilst the concept of geomorphology was well established during the nineteenth century it was only during the first part of the twentieth century that it developed as a clearly defined and identified discipline. In the intervening period the evolving geomorphologist was a physiographer, physical geographer, or physical geologist, terms that endured until the mid twentieth century. W. M. Davis, regarded as the father of geomorphology, developed not a geomorphological cycle of landscape evolution but a geographical cycle (Davis, 1899). When Alfred Steers (the first British coastal geomorphologist) and his colleague Michael Spender mounted a geomorphological expedition to the GBR in 1928–29 as a companion program to the Royal Society’s larger program on Low Isles (see, for example, Bowen and Bowen, 2002), it and its successor in 1936 were termed ‘‘Geographical Expeditions.’’ Even later, one of the first holistic geomorphological texts was called Principles of Physical Geology (Holmes, 1944). Not surprisingly, the part played by geomorphology in the scientific study of the GBR has been obscure, to the extent that in a recent review of the history of science on the Reef (Bowen and Bowen, 2002) geomorphology does not rate a mention. At least in part it is the aim of this work to indicate that not only does geomorphology have a pivotal role to play in the modern understanding of coral reefs and the GBR in particular, but there is a lineage that can be traced back to the early voyages of exploration. 1.4 The history of geomorphological study of the Great Barrier Reef to 1982 1.4.1 The nineteenth century The GBR contains no atolls and for this reason did not play a major role in the nineteenth-century debates on coral reefs. The Beagle sailed around the southern shores of Australia and Darwin never had the opportunity to view the GBR. His 1842 book makes only a brief mention of it with Darwin claiming that it supported the concept of subsidence. More than 70 years later in 1914 W. M. Davis, the leading physical geographer of the time, spent two weeks sailing up the Queensland coast but his interest in the detailed morphology of the Reef was very limited. Through some observation of coastal landforms but largely by deductive argument Davis tried to show that the Queensland coast and adjoining GBR had evolved through repeated patterns of continental uplift and shelf subsidence (Davis, 1917, 1928). He spent only one night actually on the Reef, at Green Island near Cairns which he found ‘‘was an entertaining experience but as might have been expected, entirely fruitless as far as the origin of the reef is concerned’’ (Davis, 1928, p. 347).
8
Geomorphology and the Great Barrier Reef
Until there was some concept of the magnitude and diversity of the coral reefs of the region, large-scale geomorphological interpretation remained speculative. Navigation through many parts of the Reef was limited and even as late as 1960 the best navigational charts showed huge tracts of reef completely blank and still with acknowledgements to the surveys of Flinders, King, Blackwood, Stanley, Yule, and Denham undertaken between 1802 and 1860. Holistic appreciation of the GBR came only when a complete aerial survey combined with the first satellite imagery became available in the 1960s and 1970s. Nonetheless, from the mid nineteenth century onwards most researchers tried to link their field observations to one or other aspects of the ‘‘coral reef problem.’’ Some (e.g., Jukes, 1847; MacGillivray, 1852; Rattray, 1869; Penck, 1896; Davis, 1917, 1928) including the early members of the Great Barrier Reef Committee (see below) supported Darwinian-style subsidence. Others such as Agassiz (1898), Gardiner (1898), and Andrews (1902) fitted their observations into various antecedent platform hypotheses (for fuller discussion, see Hopley, 1982, ch. 1). Finally, in the twentieth century Daly’s (1915) glacial control theory involving sea level change affected observations and interpretations of workers such as Marshall et al. (1925) and Steers (1929, 1937). However, retrospectively the greatest value of much of this early work relates to observation and description of individual features and conclusions relating to the more recent evolution of the GBR. It was these observations that were to be the focus of significant research in the second half of the twentieth century when radiocarbon dating provided a timescale for interpretation. Jukes (1847) for example was one of the first to note features along the Queensland coast which he attributed to ‘‘apparently recent elevation of the land.’’ Scientific staff of other survey vessels did little to advance the ideas of Darwin, whom they supported, but they did describe many new features of the islands and mainland such as shingle ridges and cemented deposits. One of the most observant of the early workers was the Harvard zoologist Alexander Agassiz (1898). In 1896, on a specially chartered vessel he spent two months on a reconnaissance survey of the GBR as far north as Lizard Island (148 400 S). His hypotheses on the origin of the Reef as a thin veneer over a wavecut platform may be seen as extreme, and some of his interpretations such as storm-deposited reef blocks being the last remnants of a much higher reef are now completely untenable. However, his descriptions of many islands and the shapes of reefs, reef flat zonation (including the distribution of soft corals), and beach rock and conglomerate are highly accurate. He was the first to note the terrigenous sediment just behind the outer reef and Breaksea Spit as a northward encroachment of siliceous sand limiting the southern extent of the GBR.
1.4 The history of geomorphological study of the Great Barrier Reef to 1982
9
1.4.2 The first part of the twentieth century, 1900–50 Building upon the work largely carried out on surveying voyages of the nineteenth century, the next boost for GBR geomorphological research came from the first deep drilling on a coral reef planned to endorse Darwin’s subsidence, carried out not on the GBR, but on Funafuti Atoll. The drilling accomplished there between 1896 and 1898 was organized by the Royal Society but was led by Professor T. Edgeworth David of Sydney University and had other Australian interests. A tradition of coral reef research was established at Sydney University. E. C. Andrews, a student of David and a member of the Funafuti Expedition, formed a wide interest in coral reefs summarized most succinctly in his presidential address to the Royal Society of New South Wales many years later (Andrews, 1922). Charles Hedley was also a member of the Funafuti Expedition and in 1922 became the first Scientific Director of the Great Barrier Reef Committee. As the GBR is located in Queensland, further impetus to geoscientific research was given with the appointment of H. C. Richards to the Foundation Chair of Geology and Mineralogy at Queensland University in 1919. In 1922 Richards presented an address to the Queensland branch of the Royal Geographical Society of Australasia on ‘‘The problems of the Great Barrier Reef’’ (Richards, 1922). Subsequently, the Governor of Queensland, Sir Matthew Nathan, supported an appeal to a wide array of scientific societies and educational institutions to nominate representatives on a Great Barrier Reef Committee of the Society. The Committee was set up in 1922 with members from 34 institutions. The initial chairman was Nathan, but Richards took over shortly afterwards, with Charles Hedley appointed Scientific Director. Hedley traveled widely along the Queensland coast using the steamer which serviced the lighthouses. Also, three Sydney University graduates were given scholarships to work on specific projects. Results of all this work, much of which was geomorphological in nature, were published in 1925 as the first volume of the Transactions of the Royal Geographical Society (Queensland). However, shortly afterwards there was a major rift between the Committee and its parent Society. Bizarrely, the Great Barrier Reef Committee became a separate body without a parent institution. However, the Committee did provide the stimulus for research and publication on the GBR, including the first drilling on Michaelmas Cay near Cairns to 183 m in 1926, one of the last projects of Charles Hedley. Eleven years later a second hole was sunk to 223 m on Heron Island at the southern end of the Reef. Both were intended to clarify the subsidence controversy and did provide valuable information on the development of the GBR.
10
Geomorphology and the Great Barrier Reef
Figure 1.2 Low Isles, a low wooded island near Cairns and site of the 1928–29 Royal Society Expedition.
Most importantly the Committee held talks with the British Association for the Advancement of Science the result of which was an expedition funded by the Commonwealth Government, the Great Barrier Reef Committee, and the Royal Society. The base was Low Isles (Fig. 1.2) near Cairns, with 23 scientists led by C. M. Yonge spending a year on the island between 1924 and 1929. The expedition is well covered by Bowen and Bowen (2002) except for the geomorphological work. Most of this was carried out by Alfred Steers from Cambridge University, the first true geomorphologist to spend time on the Reef (Fig. 1.3). He was accompanied by Michael Spender and E. C. Marchant, the party working for six weeks with the main expedition on Low Isles. As well as producing the first detailed map of a low wooded island the group also explored other parts of the GBR, mainly the islands, highlighting the usefulness of the islands in deciphering much of the recent geomorphological history of the Reef. Far from being the fiasco claimed by Bowen and Bowen (2002) the Steers-led expedition was the stimulus for much subsequent work, leading to the establishment of a strong continuing interest in coral reefs in the Geography Department of Cambridge University (as stated by Steers in talks with one of the authors (D. H.) in Townsville in 1967). Publications of this purely geomorphological work and a second expedition to the GBR in 1936 refocused geoscientific research away from armchair-based
1.4 The history of geomorphological study of the Great Barrier Reef to 1982
11
Figure 1.3 Alfred Steers (left) a member of the 1928–29 Royal Society Expedition talking to David Stoddart (back to camera), leader of the 1973 Expedition and Richard Orme (second from left) a member of the Expedition and other participants of the 1973 Second International Coral Reef Symposium, during a field trip on the shingle ramparts of Low Isles during the 1973 Symposium.
hypotheses and towards field-based studies (Steers, 1929, 1937, 1938; Spender, 1930), as outlined by Hopley (1982, ch. 1). The carefully surveyed maps of the islands were of such quality as to allow quantitative comparison with surveys carried out in 1973 on the Royal Society–Universities of Queensland Expedition (see below). Whilst the work of the main party on Low Isles in 1928–29 was largely biological they did carry out research which could only be described as geomorphological and which also heralded the new era of careful field measurements. Most prominent was the experiment carried out by Sheina Marshall and A. P. Orr who deployed jars (whose dimensions were carefully given) at five positions on the Low Isles reef flat between December and June, with sediments being collected weekly (Marshall and Orr, 1931). Such an
12
Geomorphology and the Great Barrier Reef
experiment was far beyond its time and was also carried out so carefully that it was possible to replicate it exactly some 63 years later (Johnston, 1996). The results are discussed in Chapter 13. Of great help to the 1928–29 expedition was vertical aerial photography carried out by the Royal Australian Air Force. This was not the first aerial photography of the Reef as from 1924 there had been experiments using seaplanes to trial aerial survey of areas such as shipping passages (Bowen and Bowen, 2002, p. 303). Alfred Steers recognized the importance of this tool for geomorphological research (Steers, 1945) and during the Second World War further photographs of parts of the GBR and the adjacent coastline were taken. Partly instrumental in this were two geologists, Rhodes W. Fairbridge and Kurt Teichert, who persuaded the Royal Australian Air Force to undertake a project to improve the accuracy of the photo-interpretation of coral reefs (see Fairbridge and Teichert, 1948), Not surprisingly they concentrated on the well-studied Low Isles and produced a comparative analysis of this reef (Fairbridge and Teichert, 1947, 1948; Teichert and Fairbridge, 1950). Traditional hydrographic survey of coral reefs can give but a generalized outline of these most complex of landforms (as seen on the naval hydrographic charts predating the 1920s). Fairbridge was possibly the first person with a geoscientific background to view the variety of reefs, the planimetric details of which had not been previously seen. Such a view gives new insight into the role of waves and currents in the formation of reefs whilst the appreciation of the range of morphology led for the first time to an evolutionary type of classification of modern reefs (Fairbridge, 1950). 1.4.3 New tools for research, 1950–82 From 1964 onwards the whole of the GBR was systematically photographed by the Australian Commonwealth Government at scales of between 1: 50 000 and 1: 80 000. Geomorphological research on the Reef after the Second World War had been almost non-existent, though with the establishment of a permanent research station by the Great Barrier Reef Committee on Heron Island in 1951 more systematic research at least on the southern end of the Reef had commenced. Most significant was the work of Graham Maxwell of the Geology Department, Sydney University who, with his students, set about mapping the surface sediments of the whole of the Great Barrier Reef Province (e.g., Maxwell et al., 1961, 1964). He was therefore in an excellent position in the mid 1960s to combine his wide field knowledge with the new perspectives of almost the entire Reef from the aerial photography. The result was his magnificent Atlas published in 1968 which not only showed the distribution of surface
1.4 The history of geomorphological study of the Great Barrier Reef to 1982
13
sediments but illustrated the wide variations in reef size and morphology within the context of the latest bathymetric survey and ecological data. Like Fairbridge he envisaged an evolutionary classification of reefs through growth then dissolution, a process he termed ‘‘resorbtion.’’ Whilst this scheme is no longer viable (see Hopley, 1982 for discussion) the work was at the forefront of a new wave of geomorphological research. Other geomorphological work of Maxwell remains highly relevant today (e.g., Maxwell, 1970, 1973a). More or less contemporary with the publication of Maxwell’s Atlas were a number of important events for further research on the GBR. Satellite imagery was available from 1972 and both the quality of the imagery and the techniques for processing them have improved greatly since then. Radiocarbon dating (and, later, other radiometric dating techniques) was coming into wide use in geomorphological study, providing an absolute chronology for evolutionary and process studies especially in the Holocene. Maxwell was himself one of the first to obtain radiocarbon dates for reefal sediments (Maxwell, 1969). Hopley also obtained many dates from a wide range of features which had been previously described for example by Steers, including emerged corals, cemented deposits, and other depositional materials, first from the high islands of North Queensland (e.g., Hopley, 1968, 1971) and later from the outer reefs (e.g., Hopley, 1977). A higher mid-Holocene relative sea level for the inner shelf of North Queensland was established from these studies. Seismic reflection survey of inter-reefal areas was also being introduced into the GBR in 1973 (Orme et al., 1978a, b) and shortly afterwards seismic refraction techniques allowed some insight into the internal structure of the reefs themselves (Harvey, 1977a, b; Harvey et al., 1979). Also providing information on internal structure was a new wave of reef drilling, initially on the 1973 Royal Society–Universities of Queensland Expedition (see below) using land-based drills (Thom et al., 1978), but shortly afterwards using hand-operated drills (Hopley, 1977) and more adaptable hydraulic rigs which could be deployed in a variety of positions on the reefs (Davies et al., 1979) (Fig. 1.4). All these tools were important in testing what was at the time a resurrected idea for the explanation of complex reef morphology. In 1974 E. G. Purdy published a paper which suggested that large- and small-scale reef forms were inherited from karst relief formed by subaerial weathering of earlier reefs during periods of glacially lowered sea level (Purdy, 1974). The idea was not entirely new having first been suggested by Japanese workers Yabe (1942), Asano (1942), and Tayama (1952) and exemplified by MacNeil (1954). Drilling into mid-Pacific atolls such as Enewetak, Bikini, Midway, and Mururoa in association with atomic weapon testing in the 20 years after the Second World War had
14
Geomorphology and the Great Barrier Reef
Figure 1.4 The James Cook University hydraulic drilling rig on Gable Reef typical of the equipment used during the 1970s onwards to obtain cores through the reef to about 30 m.
not only confirmed subsidence of mid-oceanic atolls but had also identified ‘‘solution unconformities’’ (Schlanger, 1963), buried surfaces which had been subjected to subaerial erosion. Reefs were shown to have a ‘‘layer cake’’ structure consisting of units approximately 20 m in thickness laid down over the previously exposed reef during high sea level periods. The last of the layers has been added during the Holocene. Purdy (1974) focused attention on the depth and relief of the unconformities especially that separating Holocene and Pleistocene reef. In Australia, Davies (1974) reinterpreted the core from the 1934 Heron Island drilling and established a depth of 20 m for the unconformity. In 1973 after several years of planning with input from both Sir Maurice Yonge and Alfred Steers of the 1928–29 Royal Society Expedition, a second expedition led by David Stoddart of Cambridge University was mounted to research the northern GBR (Fig. 1.3). This Royal Society–Universities of Queensland Expedition had available the new tools of radiometric dating, drilling, and seismic survey and although much of the Expedition’s work was focused on the reef islands, members were aware of both Purdy’s and Davies’ yet to be published work. Determination of the depth to Pleistocene below both reefs and inter-reefal areas was an important part of the work carried out. Results of the Expedition were published as both a monograph and in two volumes of the Philosophical Transactions of the Royal Society (Series A, vol. 291, pp. 1–194 and Series B, vol. 284, pp. 1–162).
1.4 The history of geomorphological study of the Great Barrier Reef to 1982
15
1.4.4 The modern era, post-1982 At the end of the 1970s new innovative tools were opening up a whole new range of geomorphological research directions on the GBR, and first results from this work were starting to accumulate. Thus it seemed to one of the current authors an appropriate time for a review of coral reef processes and specifically to the formation of the GBR (Hopley, 1982). That monograph was much more comprehensive in its coverage than the current work as few monographs devoted entirely to coral reefs in general were available, and it was attempting to fill a perceived gap in the literature. Subsequently there have been many volumes devoted to many aspects of coral reefs including ecology (e.g., Dubinsky, 1990), geology (e.g., Birkeland, 1997; Wood, 1999), biogeography (Veron, 1995), and taxonomy (Veron, 2000), but, with the exception of the small book devoted mainly to the description and classification of coral reefs by Guilcher (1988), review of geomorphological work has been limited to single chapters in less specialized texts (e.g., Nunn, 1994; Woodroffe, 2002a). On the GBR the amount of research has increased exponentially over the last 25 years. Probably the best known is the biological and ecological work which has been widely published but the increase of knowledge has expanded in all fields. For example in 1982 almost all of what was known about the oceanography of the GBR was contained in one 135-page monograph (Pickard et al., 1977). In the last ten years three complete texts on oceanography have been published (Wolanski, 1994, 2001; Furnas, 2003). The increasing interest in coral reefs worldwide is illustrated by the attendance and the number of papers presented at the congresses of the International Society for Reef Studies held every four years (Salvat, 2002). In 1985 in Tahiti there were fewer than 600 participants presenting 424 papers. At the last two, in Bali (2000) and Okinawa (2004), numbers of both participants and papers exceeded 1500. Perhaps not surprisingly, the geoscientific research carried out over this same period has been somewhat buried in the avalanche of biological, ecological, and management studies. Nonetheless, within this field too, there has been a similar increase in knowledge in evolutionary and process studies. They include: * *
* *
increasing knowledge of the physiography and bathymetry of the GBR extensive data banks from shallow (25 m) drilling and radiocarbon dating of cores from all areas of the reef (Fig. 1.4) an Ocean Drilling Program (ODP Leg 133) carried out just off the GBR in 1990 drilling through the full extent of the northern GBR in 1995
16 * *
Geomorphology and the Great Barrier Reef new paleoenvironmental analysis from coral cores measurement of processes operating on the Reef including rates of coral and reef growth, sediment movement, and rates of sedimentation.
These and many more studies carried out since 1982 form the foundation for the present work. 1.5 Outline of the following chapters The organization of coasts into a series of overlapping temporal and spatial scales has been superbly exemplified by Woodroffe (2002a, ch. 1). This synthesis of recent work on the GBR is approached in a similar manner to explain how the present morphology has evolved. The basic premise is that most modern reefs, possibly excluding many fringing reefs, have grown over older Pleistocene reefal foundations which were drowned during the latter part of the Holocene transgression. How these Pleistocene reefs evolved and the earliest foundation of the GBR within the geological setting of north-east Australia is the subject of Chapter 2. Once it was established, the major large-scale influence on the evolution of the reef has been the oscillation of sea level within amplitudes of 100 m or more associated with the Pleistocene glaciation. Chapter 3 analyses the sea level history as it has affected the GBR, especially over the last 10 000 years. At the present time the major short-term driving forces are those associated with climate, oceanography, and water quality (Chapter 4). Spatial patterns determine the organization of the next six chapters. Chapter 5 is an overall survey of GBR geography and geomorphology providing an analysis of the basic dimensions of the Reef. Chapter 6 is an introduction to the continental shelf of north-east Australia, describing the sediments and features of the inter-reefal areas. The next three chapters provide a cross-shelf analysis of the reefs: fringing and inner shelf reefs (Chapter 7), mid-shelf reefs (Chapter 8), and outer shelf reefs (Chapter 9). However, each has its own particular theme. Fringing reefs are examined in terms of the surface and deposits on which they have been built and their far from continuous growth history in the Holocene. Mid-shelf reefs are related to the evolutionary classification of Hopley (1982) and, with the information now available on rates of calcification and reef growth, tentative time-lines for evolution from one reef type to another are given. In Chapter 9, the regional differences in the reefs of the outer shelf are described and explained in terms of shelf edge morphology and possible tectonics. Capping, and therefore younger than all these reefs, including the fringing reefs, are more than 400 reef islands (300 within the Marine Park, plus over 100 in Torres Strait). Chapter 10 goes beyond a simple classification to examine details of evolution and dynamics.
1.5 Outline of the following chapters
17
The final chapters provide a more holistic analysis of the data. Chapter 11 provides an overview of the data from all the reefs, especially from the shallow drilling programs, and provides some insight on rates of geomorphological processes through the Holocene. It also makes comparisons between the GBR and other reefs of the world. In Chapter 12 the rapidly changing paleogeography of the GBR region during the postglacial transgression is described, summarizing the evolutionary themes of earlier chapters. The last chapter returns to the essence of geomorphology and its usefulness within the GBR to management and conservation processes. This is achieved through five themes: reefs and sedimentation, reefs and eutrophication, geomorphological input into conservation issues, reef islands, and finally global climate change and coral reefs. Given the speed with which new knowledge on the geomorphology of the GBR has accumulated over the last 20 years and as is illustrated by the following chapters, it is concluded that a synthesis of the data may be required more frequently than the 25-year interval since the last major review, Hopley (1982).
2 Foundations of the Great Barrier Reef
2.1 Introduction For much of its length even the innermost reefs of the Great Barrier Reef (GBR) are beyond sight of the mainland. The main reef tract occupies the outer 30% to 50% of the shallow (<60 m) waters of the continental shelf which varies from about 250 km in width in the south central GBR to less than 40 km near Cape Weymouth. Whilst fringing reefs occur around many continental islands, some of which are located up to 70 km from the mainland shore, reefs attached to the mainland are rare, especially south of Cairns. Nonetheless, no study of the evolution of the GBR could ignore the geology and geomorphology of the adjacent mainland of north-east Australia. The continental shelf upon which it rests has formed from terrigenous sediments eroded from the Eastern Highlands (Symonds et al., 1983), and as Lloyd (1977) noted, the geology of the shelf is closely related to the onshore geology of Queensland. For a large part of their history the GBR foundations have been part of the mainland as, during major glaciations of the Quaternary, sea level dropped to expose the shelf edge. Today, the direct influence of the mainland (including Papua New Guinea) in the form of runoff, sediment, and nutrients may be restricted to the Reef north of 188 S but along the entire coastline south to Fraser Island the sedimentary and geomorphological record of the last 8000 years or so greatly augments our understanding of the final phases in the evolution of the GBR which are the focus of this book. However, the fundamentals of the geological structure and physiological features of north-eastern Australia date back to the start of the Cainozoic about 65 Ma ago. Australia at this time was moving northwards at a rate of about 7 cm a year having separated from Antarctica and the other Gondwana continents 30 Ma earlier (Johnson, 2004). Eastern Australia had already commenced its own period of rifting with the Tasman Sea opening up over a 30 million year period between 18
2.2 Geological and geomorphological development of the coast
19
84 and 54 Ma ago. The Coral Sea opened up much later, between 58 and 48 Ma ago, with uplift of the continental margins forming the Eastern Highlands and independent subsidence of a number of continental blocks producing the distinctive morphology of the Coral Sea including the Queensland, Marion, and Eastern Plateaux (Fig. 2.1). These events would eventually produce the conditions suitable for building the continental shelf upon which the GBR rests. 2.2 Geological and geomorphological development of the coast As the Coral Sea opened up, uplift of eastern Queensland produced a drainage divide close to the east coast. Coastal streams cut back rapidly and the major rivers in particular were able to exploit the north-north-west–south-south-east structures of the older Tasman geosyncline and younger structures such as the Triassic Bowen Basin. The proto-Burdekin and Fitzroy Rivers were especially effective and moved the main drainage divide as much as 400 km inland. Working behind the coastal ranges these rivers were able to isolate large areas of westward sloping but eastward draining tablelands. Nearer to the coast the most resistant rocks (mainly granites) of what had been the initial and highest coastal divide remained as the highest peaks in Queensland overlooking a narrow coastal plain (Fig. 2.2). These include Mounts Bartle Fre`re (1611 m), Bellenden Ker (1593 m), Elliot (1234 m), and Dalrymple (1259 m). These mountains may not be especially high, but given their proximity to the coast and the coincident high annual rainfall totals for these areas (>3000 mm), they have the potential for very high sediment yields to the coastline. The Fitzroy (142 537 km2) and Burdekin (130 126 km2) Basins, though draining much drier inland areas, also have potential for high sediment yield into GBR waters (see Chapters 4 and 13). Contemporaneous with the uplift of eastern Australia were a series of volcanic eruptions the products of which include older, eroded central volcanoes and younger shield volcanoes with long lava flows (Fig. 2.1). Central volcanoes are generally older in the north, believed to be the result of plate movement over a stationary mantle plume (Johnson, 2004). Behind the Queensland coast these range in age from 33 Ma at Cape Hillsborough to 24 Ma in the Glasshouse Mountains. However, even more extensive are the broad shield volcanoes located close to the Main Divide, dating from about 8 Ma to as young as 13 ka years. These include the Atherton Tableland (1800 km2) behind Cairns where basalt flows between 3.9 and 1.6 Ma ago infilled the longitudinal valley between the coastal ranges and the Main Divide, with some flows extending down to the coast. The McBride (5500 km2), Chudleigh (2000 km2), Sturgeon (5200 km2), and Nulla (7500 km2) basalt provinces are of similar age with some lava flows
20
Foundations of the Great Barrier Reef
Figure 2.1 Major structural features of north-east Australia and the Coral Sea.
2.2 Geological and geomorphological development of the coast
Figure 2.2 River basins and mountain peaks adjacent to the Great Barrier Reef.
21
22
Foundations of the Great Barrier Reef
Figure 2.3 The crater rim and surrounding coral reef of Waier Island in the Murray Islands, Torres Strait.
extending more than 150 km from their source craters (Stephenson et al., 1980, 1998; Johnson, 2004). Volcanism on a more subdued scale has continued in eastern Queensland until the recent past. Numerous scoria cones remain from eruptions after 200 ka with some believed to be younger than 20 ka. Included in these are the pyroclastic cones and minor basalt flows of the Murray Islands blasted through the early GBR on the outer shelf in the far north (Fig. 2.3). These are almost certainly of Pleistocene age (Stephenson et al., 1980). Uplift and denudation of the Eastern Highlands over the last 50 million years has provided much of the material which has constructed the continental shelf upon which the GBR is established (Section 2.4). The present Holocene high sea level stand leaves visible only the uppermost section of this depositional province. Descriptions of the coastal plain have been made by Coventry et al. (1980) with the greatest detail available for the Townsville coastal plain (Hopley and Murtha, 1975). The sediments are mainly alluvial and colluvial including numerous abandoned river courses now marked by sandy rises crossing the coastal plain which is dominated by weathered clay soils. The unconsolidated deposits overlie bedrock typically by 30–150 m though numerous outcrops occur as coastal hills on the coastal plain, as headlands along the coast, and as islands offshore. Greatest depth of sediments on the coastal plain is associated with the lower reaches of the major streams. The Fitzroy River, for example, has an estuarine infill 3–18 km wide and at least 45 m thick. The deltaic deposits of the Pioneer River at Mackay are up to 30 m thick, the
2.2 Geological and geomorphological development of the coast
23
Herbert River >93 m, and the Barron River at Cairns >40 m. Greatest detail is available for the Burdekin River delta (Hopley, 1970) where deltaic deposits are an average 70 m in thickness and reach a maximum of 150 m at the apex of the delta. As much as 38 m of Holocene sediments overlie a distinct weathered Pleistocene surface recognizable across the delta. The distributaries of the delta are highly dynamic and continue to prograde (e.g., Pringle, 2000). Both landforms and the monitoring of contemporary sediment loads in streams are indicative of large sediment yields to the mainland coastline inside the GBR (e.g., Pringle, 1991; Nott et al., 2001; Thomas et al., 2001), figures which are estimated to have increased by up to 900% since European settlement and land clearing (Great Barrier Reef Marine Park Authority, 2001). Flood plumes which historically may have limited inshore coral reef development are also considered to have proportionally increased their area of impact over the last 150 years extending to fringing reefs of the inshore islands and even mid-shelf reefs following major floods (Devlin et al., 2001). A geomorphological assessment of the impact of increased sediment yield is made in Section 13.2.2. The mainland coastline (most recently described by Hopley and Smithers, 2003) is mainly depositional, large sediment compartments separated by prominent headlands or by short stretches of hard rock coastline as for example near Cape Clinton (north of Rockhampton), adjacent to the Whitsunday Passage, between Innisfail and Cape Grafton (south of Cairns) and around Cape Tribulation. Fringing reefs are associated with only a few of these rocky coastlines, including Cape Tribulation and Hydeaway Bay south of Bowen (Fig. 2.4).
Figure 2.4 Emerged mid-Holocene fringing reef at Cape Tribulation.
24
Foundations of the Great Barrier Reef
Elsewhere the sediment compartments clearly reflect the south-to-north movement of sediments under the influence of the prevailing south-easterly trade winds. Typically, the coastline is made up of ten or more beach ridges up to 500 m wide but close to major rivers the sequence may widen to over 5 km (Fig. 2.5a). The majority of the ridges are Holocene in age, but a fragmentary Pleistocene series is found in some areas indicating at most only a marginally higher sea level than has been experienced in the Holocene (Section 3.3.1). However, there are large accumulations of dune materials of Pleistocene age in discrete locations along the coast. Dune cappings may raise Holocene beach ridges to a maximum of 10 m height at the exposed northern end of beaches but in a number of locations there are Pleistocene dunes rising to over 30 m height. Opposite the southernmost GBR this may not be surprising as the world’s largest (and one of the highest) sand islands, Fraser Island, has dunes up to 240 m high. Formed of sediments swept by longshore drifts from the south, Fraser Island has accumulated over several phases of the Pleistocene and its continuation across the continental shelf as Breaksea Spit is a possible contributor to the termination of the GBR in this location (see Chapter 1). Elsewhere along the coast the major dune sand masses are also considered to be Pleistocene in age, displaying spectacular colored variations as the result of weathering, and massive parabolic landforms (Fig. 2.5b). They include (from south to north): large dunes near Cape Capricorn on Curtis Island, and south of Cape Clinton; a large white silica sand complex on Whitsunday Island; at the northern end of Hinchinbrook Island where the Pleistocene sequence may extend at least 30 m below modern sea level (Pye, 1982a; Grindrod and Rhodes, 1984); the 700 km2 silica sand complex of Cape Flattery and Cape Bedford and the 400 km2 complex near Shelburne Bay in the far north. The dune fields appear to have been derived from high sediment yielding sedimentary or granitic rocks and to have formed at least in part during low sea levels via temporary storages of sand exposed on the continental shelf (Pye, 1982b; Pye and Bowman, 1984). Within the lee of headlands, in all estuaries, and in channels between islands and the mainland, mangrove swamps predominate (Fig. 2.6). The most extensive system is found in the Hinchinbrook Channel and adjacent Missionary Bay where 31 species have been identified. On the wetter coastlines in favorable locations, mangroves extend right across the tidal flats but in drier areas and especially where the tidal range exceeds 4 m the mangroves are limited to the seaward margin and to drainage channels. Wide hypersaline mud flats dominate the coastline in these areas. They are particularly wide (<25 km) around the delta of the Fitzroy River and behind Princess Charlotte Bay (Chappell and Grindrod, 1984). Isolated chenier ridges may occur at irregular intervals across the flats.
2.2 Geological and geomorphological development of the coast
Figure 2.5 Depositional coastlines. (a) Wide beach ridge sequence north of the Burdekin River. Sand originated from offshore sources and brought ashore during the latter part of the transgression. (b) Large dune field south of Cape Flattery.
25
26
Foundations of the Great Barrier Reef
Figure 2.6 Mangroves at the northern end of Hinchinbrook Channel.
The geology of the mainland is extended seawards in the many islands which occur along the Queensland coastline. Hopley et al. (1989) estimated that there were 768 high islands in the Great Barrier Reef Marine Park with some kind of coral reef attached, but there are also others with no reef which were not incorporated into the calculation. There are also the many high islands of Torres Strait, which continue the geological structure and lithology of Cape York Peninsula. A number close to 1000 high islands may be a reasonable estimate for the whole GBR. By far the largest number occur between 208 and 228 S in the Whitsunday, Cumberland, and Northumberland Groups. Most islands occur on the inner 30% of the continental shelf though as Hopley (1982) noted on the narrower shelf of the northern GBR in particular there may be significant outcrops of continental rocks within some of the midshelf reefs including the Sir Charles Hardy and Forbes Islands, Quoin Island, and Howick Island. Lizard and North and South Direction Islands have a similar mid-shelf location surrounded by reefs. In the south, where the shelf is much wider only Tern and Redbill Reefs have small granitic outcrops the probable raison d’eˆtre for their location in an otherwise reefless area. There is a strong possibility that other reefs have rocky foundations close to the surface. Pandora Reef 70 km north of Townsville and inside the Palm Islands has on its surface a large number of boulders of porphyritic granite similar to
2.3 Evolution of the Coral Sea
27
the rocks of adjacent islands. They are concentrated in two areas and have suggested (Hopley, 1982, figs. 10.13 and 11.2) that bedrock may be buried at very shallow depth between these two locations. The overall influence of basement rocks within the continental shelf on the GBR and its evolution can only be surmised but is likely to have been significant. 2.3 Evolution of the Coral Sea The importance of the Coral Sea to the establishment and maintenance of the GBR cannot be overstated. Coral reefs existed here millions of years before the foundations of the GBR were laid and provided a gene pool for the GBR both in its initial establishment and subsequently after each major lowering of sea level when reef growth off the north-eastern Australia mainland was more or less annihilated (Davies et al., 1989; McKenzie and Davies, 1993). Extensive seismic studies of the Coral Sea and especially the Queensland and Marion Plateaux in the 1980s (Davies et al., 1989) have been supplemented with the new stratigraphic data which has come from Holes 811 to 826 drilled on Leg 133 of the Ocean Drilling Program in 1990 (McKenzie et al., 1993). The GBR is but the youngest of a series of carbonate platforms which have existed in the region for more than 25 Ma. When the Coral Sea had finished opening up about 50 Ma ago, it had produced the Coral Sea Basin >4600 m deep and underlain by oceanic crust, and a series of marginal plateaux underlain by modified continental crust (Davies et al., 1989) (Fig. 2.1). The northernmost of these plateaux, adjacent to the GBR, is the Eastern Plateau, a complex fault-bounded block with an average depth of 1500 m and an area of 31 000 km2. It is separated from the GBR by the Pandora and Bligh Troughs and separated from the Queensland Plateau to the south by the Osprey Embayment. The Queensland Plateau is one of the largest features of this type in the world, having an area of about 165 000 km2 (Davies et al., 1989). About half of the area is less than 1000 m deep, with living reefs found along its southern margin. The Queensland Plateau is separated from the GBR by the Queensland and Townsville Troughs. The Marion Plateau, 77 000 km2, forms a deep-water extension of the Australian continental shelf. It is bounded on the east by the Cato Trough. Within the Coral Sea it was on the plateaux that reef growth was first developed. However, for the first 25 Ma of their existence, Australia’s position was far too far south for reef growth, but as the continent’s position moved steadily northwards at about 7 cm yr 1 these shallow-water banks were moved from temperate to subtropical and tropical zones (Fig. 2.7). The studies carried out by the Ocean Drilling Program, Leg 133, have determined a complex
28
Foundations of the Great Barrier Reef
Figure 2.7 Australia’s northward drift into (a) tropical waters and (b) combined with the superimposed surface water temperature envelope for north-east Australia and the Miocene ‘‘phosphate spike’’ which may have inhibited reef growth (after Davies, 1988).
2.3 Evolution of the Coral Sea
29
subsequent history culminating in the development of the GBR (McKenzie et al., 1993). Concentrating on the Queensland and Marion Plateaux these studies have indicated changing environmental conditions controlled by fluctuations in eustatic sea levels, pulses of subsidence of the plateaux not necessarily simultaneous (Katz and Miller, 1993), and changes in oceanographic conditions large enough to switch on and off the tropical reef systems which occupied the plateaux (Isern et al., 1993). The northernmost part of Australia first moved into the tropics about 24 Ma ago and bryozoan-rich sediments of temperate zones were replaced by tropical reef systems in suitably shallow waters (Fig. 2.7). However, the change was too rapid to be related to continental drift alone and the onset of tropical oceanographic convergence off north-eastern Australia is considered to have played a major part. Both the Queensland and Marion Plateaux remained within the photic zone (<100 m suitable for coral growth) for a considerable period of time. However, towards the late middle Miocene falls in eustatic sea level became more pronounced culminating in a major low about 10.4 Ma ago. The Marion Plateau, which was stable at this time, became emerged and remained so until about 5 Ma ago. In contrast the Queensland Plateau experienced subsidence about 13.7 Ma ago and was able to maintain carbonate productivity throughout the late middle Miocene. A second subsidence pulse affected both plateaux between 7 and 6 Ma, drowning previously emerged reefs on the Marion Plateau. However, more or less simultaneously there was a change in oceanographic conditions with cooler waters not allowing reef growth on either platform (Isern et al., 1993). The return of tropical conditions with warming of surface waters did not take place until 3.5 Ma ago (Isern et al., 1993) unfortunately coinciding with the most recent period of subsidence 3–2 Ma ago which more than compensated for eustatic sea level lowering at about the same time. The Queensland Plateau sank beneath the photic zone apart from a few isolated areas, mainly on its southern margin where reefs still exist today (Holmes, Tregrosse, Flinders, Lihou, Willis Reefs, etc.). Whilst the Queensland Plateau became drowned, the previously emergent Marion Plateau did not renew reef growth during the sea level rise of the late Miocene as cool surface waters inhibited reef growth initially and, by late Pliocene/Pleistocene the platform, being attached to the Australian continent, had become the site of significant siliciclastic deposition, preventing coral growth in all but a few areas (e.g., Marion Reef) when surface temperature rose again. Thus before the GBR was even initiated the previously extensive coral reef growth on the Coral Sea plateaux had been very much reduced though still remaining significant for the establishment of the GBR.
30
Foundations of the Great Barrier Reef
2.4 The continental shelf of north-east Australia Subsequent to the Coral Sea Basin opening up, the continental shelf of northeast Australia commenced its development (Mutter and Karner, 1980; Symonds et al., 1983). The sequence developed by Symonds et al. (1983) remains a useful model though related specifically to the central GBR (Fig. 2.8). With some modification to its timing in view of more recent studies, five phases can be recognized: (1) Post 50 Ma – opening of the Coral Sea Basin 50 Ma ago produced a fractured continental margin which had been the planated Tasman geosyncline. To seawards were the major rift basins which were to form the Pandora, Bligh, Queensland, and Townsville Troughs and, whilst what is now the inner part of the continental shelf remained a stable block beneath the coastal escarpment, the outer shelf was also formed from a series of rift basins which were filled initially with fluvial and colluvial sediment. (2) 24–15 Ma – the major marine transgression of this time, together with marginal subsidence east of a hinge line identified on the inner shelf (see below) produced a transgressive onlap phase of sedimentation in the form of marginal marine fan deltas, through restricted shallow marine to open marine oozes and turbidites. (3) 15–5 Ma – this was a period of lower sea levels with a major regression at about 10.4 Ma. It produced a dominantly offlap phase of fluvial and deltaic sedimentation with thin interbedded shallow-water sediments from short periods of high sea level. Sediment supply outstripped any subsidence and major progradation of the shelf took place, by approximately 10 km off Cairns and 40 km off Townsville. (4) 5–<1 Ma – this phase coincided with the subsidence pulse of 3–2 Ma ago. This was balanced by massive shelf aggradation of more than 200 m formed of low sea level alluvial deposits in conjunction with thin high sea level clastic shallow-water onlapping sediments. Symonds et al. (1983) recognized wave and fluvial dominated deltaic deposits near the top of the sequence. (5) The final reef-building phase – by 2.6 Ma ago the continental shelf was in place in appropriate neritic water depth for reef building (McKenzie and Davies, 1993; Watts et al., 1993). Widespread reef building, however, does not appear to have commenced until after 700 ka for a number of reasons discussed below. Once initiated, the proto-reefs became preferential sites for future growth, especially after each glacial low sea level stand. Symonds et al. (1983) suggest that carbonate sediments from the reefs now became incorporated into the progradational facies of the shelf margin and that shelf edge reef growth channeled major low sea level sedimentation onto the slope, changing the style of sedimentation on the slope from mainly progradational to mounded submarine fans. The continental shelf has continued to develop since the first appearance of the GBR.
2.4 The continental shelf of north-east Australia
Figure 2.8 Evolution of the continental shelf of the central Great Barrier Reef (from Symonds et al., 1983).
31
32
Foundations of the Great Barrier Reef
This continued development probably involves further tectonic activity. As noted by Lloyd (1977) the GBR shelf is related to the onshore structural features (Fig. 2.1) some of which may have continued activity well into the Cainozoic. Symonds et al. (1983) recognized a middle to inner shelf hinge line between Cairns and Townsville, paralleling the coastline and determining the western boundary of shelf subsidence. The southern extension of the hinge line can be extrapolated to the onshore Millaroo Fault Zone which forms the western boundary of the major Bowen Basin. Elsewhere the general shape of the coastline can be closely linked to major structural features. Its general alignment is that of the ancient Tasman geosyncline whilst details such as the northern area of Princess Charlotte Bay are defined by older structures. The bay itself is the northern expression of the landward Laura Basin which is defined on its western side by the Palmerville Fault, the extension of which determines the alignment of northern Cape York. The Laura Basin is but one of a series of Tertiary basins (Benbow, 1980) which straddle the coastline or are imposed on the continental shelf including the Hillsborough Basin adjacent to the south central GBR, the Capricorn Basin of the southern GBR, which is defined on its western side by the Bunker Ridge and to the east by the Swain High, and the Halifax Basin impinging on the outermost shelf opposite Townsville (Grimes, 1980; Mutter and Karner, 1980). Volcanism appears to have been associated with the early foundation of several of the basins and subsidence has continued until at least the late Pliocene. Whilst this was prior to the foundation of the GBR, Grimes (1980) believes that some earth movements continued until the Quaternary. The Ocean Drilling Program showed that subsidence pulses on and adjacent to the GBR shelf were taking place until at least the late Pliocene (Katz and Miller, 1993). Of particular importance is the Halifax Basin which underlies 800 km of the GBR opposite Townsville (Mutter and Karner, 1980, fig. 11). It contains up to 3000 m of mainly terrigenous sediments dating back to the opening up of the Coral Sea in the Cretaceous. However, more recent subsidence as late as the Pleistocene has been suggested (e.g., Benbow, 1980) and later in this work (Chapter 9) a case will be made for even more recent subsidence of the Halifax Basin affecting reef development on the outer shelf of the central GBR. Whilst the GBR shelf is tectonically stable it does experience some earthquake activity (University of Queensland, 2005). Within the Coral Sea and GBR region locations have included (from north to south): in eastern Torres Strait near the Murray Islands, offshore from Cairns, offshore between Townsville and Mackay, and in the Capricorn Channel and Capricorn–Bunker Group of reefs in the southern GBR (Fig. 2.9). In November 1978 Heron Island
2.4 The continental shelf of north-east Australia
Figure 2.9 Seismic events recorded along the Queensland coast between 1866 and 2000 (from University of Queensland, 2005).
33
34
Foundations of the Great Barrier Reef
experienced a 5.2 magnitude earthquake located to the east of the island and in 1998 a 4.7 magnitude event centered to the west. In 2003 an earthquake 40 km east of Ingham was felt on the mainland between Cairns and Townsville and the Queensland record contains 493 tectonic events in the Townsville region, the largest having a magnitude of 5.5. Many of these have been out to sea. Also affecting the shelf has been volcanic activity sometime during the Pleistocene and subsequent to reef development. The area affected was the outer shelf of Torres Strait where six islands (Maer, Dauer, Waier, Darnley, and Stephens Islands, Black Rock and Bramble Cay) consisting of pyroclastic cones with some basaltic flows in varying degrees of erosion occur. The pyroclastic materials include reefal limestone indicating eruption through the previously established reef (Stephenson et al., 1980) (Fig. 2.3). 2.5 The late establishment of the Great Barrier Reef Whilst mid-oceanic atolls such as Enewetak, Midway, and Mururoa may have over 1000 m of reefal limestones representing almost continuous reef growth since the early Cainozoic, interrupted only by low sea level periods, the GBR is one of the youngest reef systems on earth. Only in the last 20 years has this young age been established. Previously from the four deep holes drilled on the reef, an age as great as the Miocene had been suggested for its foundation (Marshall, 1983a). Drilling had commenced on the GBR in 1926 with the scientific borehole in Michaelmas Cay put down by the Great Barrier Reef Committee (Chapter 1). A further hole was drilled in 1937 in Heron Island and subsequently oil companies drilled exploratory wells in the Gulf of Papua (Anchor Cay, Borabi, and Pasca) and in the Capricorn Basin (Aquarius, Capricorn, and Wreck Island). Unfortunately only three of these were actually located on a reef, Michaelmas, Heron and Wreck (Anchor Cay well was located just off the reef and did not penetrate reef framework above 225 m and even below this depth carbonate sediments are dominantly coralline algal rhodoliths: Marshall, 1983b). As Davies et al. (1989) noted, the boreholes on Michaelmas, Heron, and Wreck Islands had a remarkably similar stratigraphy with 100–150 m of what was presumed to be Plio–Pleistocene reef limestones overlying a thinner sequence of foraminiferal limestone and siliciclastics or mixed siliciclastics and carbonates. The Miocene or Pliocene age for the initiation of the GBR was based on poor recovery and extrapolation of evidence which was recognized as ambivalent even 40 years ago (Lloyd, 1973; Hill, 1974).
2.5 The late establishment of the Great Barrier Reef
35
Figure 2.10 Ribbon 5 Reef site of deep drilling in 1995 and submersible reconnaissance of the outside of the reef in 1984 (photograph: Great Barrier Reef Marine Park Authority).
A much later age for all except the northernmost reefs in the Gulf of Papua was recognized once the relatively thin section of the reef was established from seismic interpretation. An age no greater than the Pleistocene was suggested (Marshall, 1983a; Symonds et al., 1983). The 1990 Ocean Drilling Program confirmed this though the exact date of initiation was far from precise. Davies and McKenzie (1993) opted for an age of <500 ka, Feary et al. (1993) for as much as 1 Ma, and Montaggioni and Venec-Peyre´ (1993) for reef growth throughout the Pleistocene. The drilling in 1995 into Boulder and Ribbon 5 Reefs (Fig. 2.10) had hoped to provide a finite date, and has essentially done this (Fig. 2.11). The major reef-building turn-on event appears to have occurred between 452 and 365 ka ago though this was preceded by a transitional period of reef growth (International Consortium for Great Barrier Reef Drilling, 2001; Webster and Davies, 2003; Braithwaite et al., 2004). The question still remains as to why this turn-on was so late. At the largest chronological scale it required Australia to drift into tropical latitudes, which for the northern GBR was about 24 Ma ago, and for the southern Reef only in the last million years or so (Davies et al., 1987, 1989) (Fig. 2.7). The start of this period is thought to coincide with early Miocene increased oceanic phosphate levels which would have inhibited the growth of coral reefs (Kinsey and Davies, 1979a). Subsequently even in some of the optimal coral reef areas of the Coral Sea plateaux, it has been shown that growth may not have been
36
Foundations of the Great Barrier Reef
Figure 2.11 Core logs for the holes drilled through Boulder and Ribbon 5 Reefs (after Webster and Davies, 2003; Braga and Aguirre, 2004).
2.6 The Pleistocene Reef
37
continuous. Reef growth declined on the Queensland Plateau in the late Miocene as surface water temperature dropped below 208 C, the result of reorganization of oceanic circulation in the Coral Sea (Isern et al., 1993). Renewed warming took place only in the late Pliocene and early Pleistocene but apparently still before the GBR had been initiated. The massive lowering of sea level centered around 10 Ma ago, which aided the terrigenous progradation of the Queensland shelf, would also have prevented any reef build-up. An examination of how the continental shelf off north-eastern Australia has been constructed clearly indicates two further features that would have provided constraints on reef development. The first was the massive amount of sediment which was being shed from the continental land mass since its uplift 50 Ma ago, providing what may have been very marginal conditions for reef turn-on (Perry and Larcombe, 2003). It may have been necessary for the shelf to have built out far enough from the continental land mass for the outer edge to have been sufficiently removed from extensive fluvial input during shelf flooding events. This would have been especially important for the area of the wet tropics centered on Cairns and opposite the two major rivers, the Fitzroy and the Burdekin. Nonetheless, these conditions would have been achieved well before the commencement of the Pleistocene. The turn-on event (or events) would have required the Queensland shelf to have been seeded from other reefal areas with reefs of the Coral Sea identified as such nursery grounds (McKenzie and Davies, 1993). Also necessary would have been appropriate circulation patterns in the Western Coral Sea (Isern et al., 1993), which seem to coincide with high sea level interglacials. Also suggested as a turn-on mechanism has been an up to 48 C increase in water temperature at about 400 ka ago. Webster and Davies (2003) note that Stage 11 at about 410 ka ago was the warmest interglacial in the last 450 ka and may have been the final trigger for extensive development. The event may also have been somewhat arbitrary once the general environment for reef growth was established. It was dependent on the coincidence of suitable oceanic temperature and circulation and an appropriate depth of water (between 30 and 80 m) on the outer shelf produced by a specific eustatic sea level high (see Section 3.2.2). 2.6 The Pleistocene Reef In the far north of the GBR, especially within the Gulf of Papua, reef growth may have started soon after northern Australia entered tropical waters. Subsurface reefs of Miocene and Pliocene age have been identified in a number of places including the western margins of the Pandora Trough where such a
38
Foundations of the Great Barrier Reef
reef 1.5 km thick occurs beneath Portlock and Boot Reefs. Buried reefs of similar age are found along the adjacent far northern GBR shelf (Davies et al., 1989). Elsewhere a middle to late Pleistocene age is identified for the Reef and whilst the direct evidence comes only from Ribbon 5 and Boulder Reefs (International Consortium for Great Barrier Reef Drilling, 2001; Webster and Davies, 2003; Braithwaite et al., 2004) similar relatively thin sequences of reefal framework seen in the boreholes on Michaelmas, Heron, and Wreck Reefs, and in seismic sections appear to corroborate this age for the Reef along its entire length. The hole drilled to 210 m on Ribbon 5 Reef (Figs. 2.10 and 2.11) contains the more complete record. The reef is located on the edge of the continental slope (see Chapter 9) and especially in the lower section there is evidence for much downslope transport of material (International Consortium for Great Barrier Reef Drilling, 2001). Details of the core have been published by the International Consortium (2001), Webster and Davies (2003), Braithwaite et al. (2004), and Braga and Aguirre (2004). The lowest section from 210 to 158 m is composed of non-reefal grainstones and packstones interpreted as debris flows and turbidites. Some coral lithoclasts were recovered (Seriatopora, Tubipora, Millepora) but were considered to have originated from higher up the slope. Braithwaite et al. (2004) suggest that these may indicate older carbonates in the region than were recovered at Ribbon 5. Between 158 and 96 m is a rhodolith-dominated section which was interpreted by Braithwaite et al. (2004) as representing gradual warming and shallowing of the environment. It also contains two in situ coral framework reef units, the oldest being between 138 and 130 m and containing robust, branching species of Stylophora pistillata, Acropora humilis, and Pocillopora sp. A sharp irregular surface with a change in d18O record separates this unit from the next coral framework section between 130 and 117 m. Basal grainstones are overlain by a 5 m in situ framework unit (Webster and Davies, 2003). The upper section of the Ribbon 5 core is almost entirely reefal. It is composed predominantly of in situ coral framework in which six separate units were recognized by Webster and Davies (2003): 96–94 m, 94–85 m, 85–64 m, 64–25 m, 25–16 m, and 16–0 m. Similar assemblages of corals are found throughout including a robust branching community representing the shallow reef edge, a community dominated by massive Porites and faviids typical of lower-energy reef front or reef slope, and a similar assemblage which lacks encrusting forms and is interpreted as representing the leeward reef flat (Webster and Davies, 2003). The coralline algal sequences are in good agreement with these units and in this upper 96 m occur mainly as crusts on corals, also suggesting shallow reef settings (Braga and Aguirre, 2004). Each reef
2.6 The Pleistocene Reef
39
section appears to represent a sea level cycle with early flooding of a foundation in cool lower-energy conditions, progressing to shallow high-energy conditions later in the cycle. The thickest framework unit between 64 and 25 m is very similar to some of the Holocene sections described in Chapters 8 and 9 with sections of up to 7 m of rubble. Halimeda, bryozoans, and encrusting foraminifera increase towards the top. The uppermost Pleistocene unit, between 25 and 16 m, is composed of a Halimeda-rich grainstone facies with minimal in situ coral. This may be similar to the uppermost Pleistocene in other reefs and is discussed further in Chapters 9 and 11. The reef units are separated by distinct d18O changes and in at least one instance by an apparent marine erosion surface at 130 m (International Consortium for Great Barrier Reef Drilling, 2001). Clear solution unconformities as are demonstrated for example in cores from Pacific atolls (e.g., Szabo et al., 1985) are recognized only in the uppermost part of the core, at 16 m and 25 m (Webster and Davies, 2003) or 28 m according to International Consortium for Great Barrier Reef Drilling (2001) and Braithwaite et al. (2004). These latter authors also suggest a further ‘‘karst’’ surface with paleosols at 36 m. In a reinterpretation of the Heron Island bore, Davies (1974) identified solution unconformities at 20, 35, 75, 95, and 140 m, the upper two of which may equate with those seen in Ribbon 5. Only Webster and Davies (2003) provide details of the 86 m core from inner shelf Boulder Reef (Fig. 2.11). The upper 34 m are carbonate dominated, consisting of four reef units between 0 and 11 m (Holocene), 11 and 20 m, 20 and 28 m, and 28 and 34 m. Poor recovery typified much of this part of the core and at the base of the 11–20 m section there was much rubble. The lower two sections contain or are set within siliciclastic muds. Muds, clay, and minor sands also dominate the core section between 34 and 86 m with only thin coralbearing horizons at 47–56 m and 60–64 m. Whilst sea level oscillation has been the major driving force in the deposition of the typical ‘‘layer cake’’ model of reef accretion, Webster and Davies believe that the changes in the coral assemblages within each unit also reflect the amount, or in some cases, the lack of accommodation space, wave energy, and sediment input. However, the same group of coral assemblages appear to be present throughout the construction of the GBR, tempting Webster and Davies to reiterate the geological robustness of reef systems (see Section 1.2). The evidence of the Ribbon 5 and Boulder Reef cores suggests that the GBR was established after an early period of ephemeral reef development as represented in the thin reef layers below 96 m on Ribbon 5 and below 34 m on Boulder Reef. The latter in particular may equate to the turbid environment
40
Foundations of the Great Barrier Reef
inshore reefs of the present GBR (Smithers and Larcombe, 2003) as discussed in Chapter 7. Ephemeral reefs such as these may have ‘‘turned-on’’ several times before becoming permanent enough to provide a significant foundation for subsequent reef growth. Aging of the sequences has proved problematical, with some contradictions in the ages obtained, problems of contamination, and lack of agreement between the level of samples and known world sea levels at the time (Braithwaite et al., 2004). Much of the material, in the lower cores at least, is not in situ and, as Braithwaite et al. note, the differences in thickness of the reef units suggest a lack of accommodation space at time of deposition and/or subsequent erosion meaning that the sequence is far from complete. Paleomagnetic analysis indicates a normal polarity throughout the Ribbon 5 core implying that the rocks were deposited entirely within the Brunhes polarity interval, post 770 ka. Strontium isotope ratios provide an apparent age of 770 280 ka for material from 206–210 m, in agreement with deposition close to the Brunhes–Matuyama boundary. Three dates for material from between 122 and 88 m all gave dates of 600 280 ka. However incompatible uranium series ages for materials at 56 m, 95 m, and 118 m were 322 119 ka, 564 78 ka, and 616 51 ka. A further date of 414.92 19 ka for a sample from 185 m is also considered to be contaminated (Braithwaite et al., 2004). A uranium series date of 125.7 0.6 ka from the top of the Pleistocene does seem to indicate a last interglacial age for this unit. Radiocarbon ages less than 8000 years from the upper 16 m confirm the Holocene age of the uppermost layer. Similar Holocene dates come from the top of Boulder Reef core with an electron spin resonance (ESR) date of 210 40 ka from the base at 86 m implying that inner shelf growth commenced later than on the outer shelf. 2.7 Conclusion The data from Ribbon 5 suggest a foundation age for the GBR of between 452 and 365 ka for the base of the main reef section (Webster and Davies, 2003). The base of the rhodolith section, which contains two reefal units, is estimated at between 606 and 520 ka. Coral lithoclasts occur in the base of the hole and, as Braithwaite et al. (2004) note they indicate that ‘‘there were remnants of an older structure nearby and therefore that the oldest in situ rocks recovered do not reflect the initiation of carbonate deposition in the area.’’ Further drilling of the GBR may push back the age of initiation quite significantly. An area of special interest may be the Pompey Complex of the south central GBR (see Chapter 9). Not only are these reefs the largest in the whole province (many larger than 100 km2), but several high sea level phases
2.7 Conclusion
41
may be demonstrated in their morphological development, as argued in Chapter 9. Further, these reefs are the only ones that demonstrate significant karst features in the form of blue holes up to 100 m deep (Backshall et al., 1979). Hopley (1982) argued for several periods of subaerial exposure to produce such features, which would also have required mature recrystallized reefal limestones for their formation. The Pompey Reefs are also almost 200 km from the terrigenous influence of the mainland coast. They are closer to the Marion Plateau reefs which, though struggling, did manage to re-establish themselves by the late Pliocene. After 50 Ma of preparation a reef was in place off the north-eastern coast of Australia during the last interglacial high sea level period at 125 ka, which may not have been very different from the present GBR. However after the last interglacial, sea level fell away once more exposing the reefs to subaerial processes (Section 3.3.1). The early part of the last glacial cycle produced interstadial sea levels, exposing all reefs except those of the shoulder of the continental shelf (see Chapter 9). Interstadial levels were initially (to 80 ka) high enough to drown the shelf margin (Chappell et al., 1996). Subsequently, until the rapid rise post 18 ka, most of the continental shelf remained dry with a coastline oscillating between the current –40 m and –120 m isobaths, i.e. the shelf shoulder and the upper continental slope (see Chapter 3). The present characteristics of the GBR, which are the topic of the remainder of this book, were acquired during the 100 000-year period of exposure and the rapid Holocene drowning event.
3 Sea level: a primary control of long-term reef growth and geomorphological development
3.1 Introduction Coral reefs such as the Great Barrier Reef (GBR) are the net result of the productive efforts of marine animals and plants that build structures able to withstand waves and currents. These structures are predominantly composed of skeletal material produced by a myriad of simple organisms that can achieve high productivity and nutrient-cycling efficiency in seas that are generally nutrient poor. Scleractinian corals and coralline algae are the major contributors to reef growth on most coral reefs. Both utilize solar energy via photosynthesis to realize calcification rates that are significantly higher than those possible if they were solely heterotrophic (Chalker and Dunlap, 1983; Barnes and Chalker, 1990). Calcification rates, and thus the accumulation of calcium carbonate products, or reef growth, proceed most rapidly in the euphotic zone, where photosynthetically active radiation (PAR) can be accessed and utilized. In tropical seas, where the majority of corals live, this may extend to depths exceeding 100 m under exceptional circumstances; for example, corals are known to grow at depths below 140 m in the Red Sea, and to reach similar depths on outer reefs of the GBR such as Myrmidon Reef off Townsville (Hopley, 1994; see Fig. 9.12). Generally, however, PAR rapidly diminishes at depths below 30 m, or even shallower in turbid settings (Anthony and Fabricius, 2000). As a result, reefs grow best within 30 m of the sea surface, with reef accretion rates usually highest above 20 m (Hopley, 1982; Carter and Johnson, 1986). A range of environmental controls such as water temperature, salinity, substrate stability, and aragonite saturation state influence where coral reefs flourish (Hubbard (1997), Kleypas et al. (2001), and Guinotte et al. (2003) provide good summaries) but light, and thus depth, is acknowledged as a primary control of photo-enhanced calcification and coral reef growth (Falkowski et al., 1990). Although individual corals may survive in deeper waters, they do 42
3.1 Introduction
43
not accumulate calcium carbonate faster than it is removed, and thus true coral reefs do not develop. The confinement of reef growth near the sea surface makes them excellent indicators of gross sea-level position; the upper 30 m in which most coral reefs actively grow represents less than 1% of the mean 3729 m depth of the world’s oceans. Radiometrically dated sequences of fossil reefs at various elevations have been used to reconstruct past sea levels, most successfully at sites where reefs are tectonically uplifted at relatively constant rates. The spectacular uplifted coral terraces of Huon Peninsula on Papua New Guinea’s north coast are an iconic location for sea-level studies (e.g., Chappell, 1974; Aharon et al., 1980; Aharon and Chappell, 1986). At Huon, flights of up to 19 uplifted reefs rise more than 700 m above the current shoreline, pushed upwards due to their location at the point of collision between the Pacific and Indo-Australian plates (see Fig. 3.6a). Reef terrace flats preserve fossil microatolls (Section 3.4.2) that identify them as reef flats that were constrained by sea level at the time of growth. Radiometric dating of these terraces can determine when over the last 300 000 or so years (the age of the oldest dated terrace) relative sea levels have been stable enough for reef flats to form. Sea level at any location is a function of both the absolute elevation of the sea surface and vertical movements of the land, the balance producing a relative sea level. Comparable rates of tectonic uplift and sea-level rise at Huon produce relative sea-level stability and thus reef flats form. The same sea-level change on stable or subsiding coasts would produce a relative sea-level rise, which if too rapid may prevent reef flat formation. Where long-term uplift or subsidence rates are known and reef terraces can be dated, the elevation and thus sea level at which a reef flat originally formed can be back-calculated. The co-ordinated analysis of sea-level records derived from uplifted coral reefs such as those at Huon, Barbados (Mesolella et al., 1970; Edwards et al., 1987; Bard et al., 1990), Indonesia (Chappell and Veeh, 1978; Jouannic et al., 1988), and Vanuatu (Cabioch and Ayliffe, 2001; Neef et al., 2003), together with those established from emergent reefs on stable coasts (e.g., Chen et al., 1991; Stirling et al., 1998; Toscano and Lundberg, 1998), and those recovered from previously inaccessible drowned reefs in stable or subsiding areas (e.g., Moore and Fornari, 1984; Webster et al., 2004) has contributed enormously to our knowledge of the amplitude and timing of Quaternary sea-level changes. For reasons outlined later in this chapter data from the GBR provide relatively little information on sea level throughout most of the Quaternary, but Quaternary sea-level changes have significantly affected the development of the Reef. Geological evidence suggesting that ‘‘true tropical reef’’ growth on the GBR began as late as 452–365 ka (Webster and Davies, 2003; Braithwaite et al., 2004) (Fig. 2.7) was discussed in Chapter 2. The number of quality deep
44
Sea level: a primary control of long-term reef growth
cores is limited, but they confirm the late Quaternary heritage of much of the GBR and document the influence of the rapid and frequent sea-level changes characteristic of the late Quaternary as turn-on and turn-off events for reef growth, as earlier envisaged by Daly (e.g., Daly 1915, 1919). The details of Quaternary sea-level changes follow in this chapter, but were sufficient to vertically move the euphotic window in which reefs thrive by more than 100 m, and to cause large lateral shifts in the location and extent of reef growth on the continental shelf (Chapter 12). For example, during sea-level lowstands the entire continental shelf was emergent and subject to terrestrial processes, restricting reef growth to the shelf edge. As discussed in later chapters, the modification of older reef surfaces during lowstands is an important control of modern reef growth and geomorphic development during highstands. As sea levels rose following the last ice age, shorelines over low gradient parts of the shelf retreated at rates exceeding 10 m a1 for centuries to millennia. The growth and development of the GBR in the late Quaternary, when sea level and associated environmental conditions fluctuated significantly and repeatedly, provides an important context for assessments of future reef-growth trajectories. As concern about anthropogenically forced environmental change heightens, a knowledge of the rates and magnitudes of past changes and of the GBR’s response to them is essential to understanding the significance of present and future trends, and for mitigating or managing them effectively. The structure of this chapter is as follows. The general nature of Quaternary sea-level change is first assessed, focusing on the Late Pleistocene. Sea-level change through the last glacial–interglacial cycle, when the present geomorphology of the GBR has largely developed, is then examined. Sea level since the last glacial maximum is covered in most detail as the modern GBR developed during this period. The evidence used to establish sea-level histories over the GBR is discussed, and hydro-isostatic and tectonic influences on relative sea-level histories during the mid to late Holocene are addressed. The chapter ends with an account of recent sea-level changes on the GBR determined from instrumental records. 3.2 Quaternary sea-level change Quaternary sea-level change is principally forced by climate changes, with sea level during the Late Quaternary fluctuating by more than 120 m at various times due mostly to the expansion and contraction of high-latitude continental ice sheets and subsequent changes in ocean volume. Between 17 and perhaps as many as 27 glacial and sea-level cycles occurred during the Quaternary, depending on exactly when it began. Each cycle involves a change from interglacial
3.2 Quaternary sea-level change
45
(sea-level highstand) to glacial (sea-level lowstand) conditions with a periodicity of 100–120 ka. Oxygen isotope records derived from foraminiferans extracted from deepsea cores (collectively referred to as the marine isotope record) provide the longest records of these cycles and indicate that they have been a strong feature of global climate for the last 800 ka (e.g., Shackleton and Opdyke, 1973; Imbrie et al., 1984; Shackleton et al., 1990). Time-series of carbon dioxide (CO2) concentrations in air bubbles trapped in ice cores extend back more than 400 ka (Petit et al., 1999) and accord reasonably with marine isotope records (Mudelsee, 2001). The influence of Quaternary climatic cycles on global sea level has been estimated by comparing marine isotope records with sea level reconstructed from fossil reefs, but these only go back 340 ka (Chappell, 1994), and detailed data cover just the last 140 ka (Chappell et al., 1996; Yokoyama et al., 2001a; Chappell, 2002). Early Quaternary sea level is thus inferred from the marine oxygen isotope record of ice-volume fluctuations preserved in deepsea cores (Chappell and Shackleton, 1986), which assumes that deep-sea temperatures and salinities have remained constant (Nunn, 1994; Pillans et al., 1998). Figure 3.1 shows the relationships between the marine oxygen isotope record, the Vostok ice core CO2 record of temperature, and the eustatic sea-level record derived from dated coral terraces. Milankovitch’s (1941) astronomical theory is widely accepted as an explanation for Quaternary climate cycles, although why they were poorly developed before 800 ka is unclear. Recent research using improved sampling (especially of lowstand corals) and dating techniques, and based on marine isotope records more precisely reconciled with water depth than earlier studies, also questions the neat synchrony of orbital and Quaternary climate and sea-level cycles (Cutler et al., 2003; Siddall et al., 2003, 2004; Potter et al., 2004; Thompson and Goldstein, 2005). These studies confirm the broader 100–120 ka cycles established with other proxies for the last 400 ka, but also identify shorter (3–9 ka) oscillations that are incompatible with orbital frequencies (minimum periodicity of 21 ka). 3.2.1 Quaternary sea-level change: eustasy and isostasy Global sea-level changes resulting from changed ocean volume are often called eustatic sea-level changes, following the terminology of Suess (1885). Traditionally eustatic sea-level changes were defined as ‘‘uniform, worldwide and simultaneous,’’ so that two eustatic sea levels were considered parallel. This definition is now recognized as oversimplistic (Sloss, 1991), with geoid deformation and the redistribution of mass over the earth’s surface introducing
46
Sea level: a primary control of long-term reef growth
Figure 3.1 Proxy records of Quaternary climate and sea-level change. (a) Marine oxygen isotope record derived from benthic foraminifera of deep-sea core ODP 677. Oxygen isotope stages marked in italics (after Shackleton et al., 1990). (b) (i) Carbon dioxide (CO2) variations and (ii) deduced temperature fluctuations for the last 420 ka derived from the Vostok ice core (after Petit et al., 1999; Masselink and Hughes, 2003). (c) (i) Reconstruction of sea level derived from radiocarbon-dated uplifted coral reef terraces, Huon Peninsula, Papua New Guinea; (ii) oxygen isotope record from marine isotope record derived from benthic foraminifera of deep-sea core V19–30; (iii) adjusted sealevel record based on these records for the last 240 ka (after Woodroffe, 2002a).
3.2 Quaternary sea-level change
47
a horizontal component to eustatic adjustments, the magnitude of which can vary geographically. Morner’s (1976) concept of a regional rather than global ¨ eustasy is now more widely accepted, and geophysicists continue to develop sophisticated models to explain regional relative sea level and shoreline responses to regional eustatic sea-level changes in various parts of the world (e.g., Peltier, 2002; Mitrovica and Milne, 2003; Lambeck 2004). Data from farfield sites distant from the direct impacts of ice sheets such as the GBR are important for constraining these eustatic sea-level models and for tuning models of earth rheology (Milne et al., 1999). Tooley (1993) provides an excellent review of the concept of eustasy. Glacial eustasy refers to sea-level changes produced by direct mass exchange between ice sheets and oceans, which dominate the eustatic component of Quaternary sea-level fluctuations (Daly, 1934). As discussed, geoidal factors introduce regional variation in the magnitude and timing of eustatic sea-level changes, and local and regional responses to shifts in water volume load on land as ice during glacials, or in the ocean as sea water during interglacials, also occur. These shifts cause adjustments in the isostatic balance between the earth’s crust and the underlying asthenosphere on which it ‘‘buoyantly floats.’’ Areas under increased load are depressed, and ‘‘unloaded’’ areas rise. Glacioisostasy refers to isostatic deformations caused by changes in ice mass load; these interactions can be spatially and temporally variable in near-field areas due to complex forebulge dynamics as ice sheets grow or ablate (e.g., Mitrovica, 2003). Reconstructing sea-level histories from close to previously glaciated areas can be very difficult. North-eastern Australia was not glaciated in the Quaternary and therefore was not directly affected by glacio-isostasy. However, as eustatic sea levels oscillated during the Quaternary higher sea levels flooded the continental shelf, producing a water load and subsequent hydro-isostatic effect. On wide continental shelves water load varies with depth, so that deeper shelf edge areas subside under greater hydro-isostatic load. Continental shelves are relatively rigid, so outer shelf subsidence can flex shallower inshore areas upward to produce a relative sea-level fall. Spatial variations in hydro-isostatic effects have greatly influenced relative sea-level histories and reef growth patterns on the GBR during the Holocene, and presumably also during previous highstands (Section 3.4.3). 3.2.2 Late Quaternary sea-level cycles Based on the last 400 ka that is known reasonably well, during at least the last four interglacials sea level was probably within 5–10 m of its present height,
48
Sea level: a primary control of long-term reef growth
and each interglacial lasted around 10 ka (50% of a precession cycle: Broecker, 1998). A quasi-continuous record is available only for the last glacial cycle (Section 3.3), but fragmentary evidence of earlier cycles suggests that they also involved progressive but irregular sea-level fall, with oscillations of up to 30–50 m amplitude (Rohling et al., 2004). These oscillations reflect shorter duration (2–3 ka) shifts between warmer (interstadial) and colder (stadial) periods. Approximately 20 ka before each cycle ends, climate warms and sea level rises rapidly around 100–130 m to near present level. Evidence of sea levels between 400 and 800 ka ago is patchy, and beyond 800 ka ago becomes increasingly scant. Coastal deposits such as the uplifted barrier dunes near Naracoote, South Australia document a history of sea-level highstands extending back 700 ka (Cook et al., 1977; Idnurm and Cook, 1980; Murray-Wallace et al., 2001), but preservation of direct evidence of sea level of great antiquity is rare. However, the marine isotope record does suggest that the prominent climate and sea-level cycles of the late Quaternary were not developed before 800 ka, where higher frequency (40 ka) cycles of lower amplitude (d18O mostly within 3.2–4.2 range) occurred (Shackleton and Opdyke, 1973) (Fig. 3.1). Lambeck et al. (2002) comprehensively review global climate–sea level interactions over the last 3 million years. Deep drilling and seismic investigations provide evidence of the influence of Quaternary sea-level changes on the GBR. Although earlier deep cores from Heron Island and Michaelmas Cay were incorrectly interpreted as entirely younger than 20 ka (Richards and Hill, 1942), later analysis confirmed that both developed via the successive superposition of reefs formed during episodic higher sea levels, with accommodation space provided by erosion during emergence and general shelf subsidence (Davies, 1983; Muller et al., 2000). ¨ Major solutional unconformities and erosion surfaces were identified between reef growth units, which together with diagenetically altered carbonate indicative of subaerial exposure provided some of the earliest evidence of the influence of Quaternary sea-level oscillations on the growth and formation of the GBR. Fossil inclusions at the base of the Heron Island core suggested that the reef was perhaps more than 2 million years old, but poor recovery and chronological control limited confident interpretation. The deep cores drilled through the northern GBR to redress these shortcomings were described in detail in Section 2.5. The cores contain distinct reef units with systematic and repeated internal changes in taxa formed during successive highstand reef growth episodes (Webster and Davies, 2003); taxa changes develop as each reef unit grows nearer to sea level. A precise chronology for pre-Holocene reef units is difficult to achieve, but paleosols (some with plant residuals), stained horizons, and isotopic shifts point to prolonged periods of subaerial
3.3 Late Pleistocene sea level
49
weathering and erosion during intervening lowstands (Webster and Davies, 2003; Braga and Aguirre, 2004). Most seismic work on the GBR has used low-powered technology and focused on identifying the last interglacial surface and overlying Holocene facies units and has provided limited detail on earlier reef growth episodes (Harvey and Searle, 1983; Symonds et al., 1983; Johnson and Searle, 1984). Seismic reflectors that ‘‘probably represent the eroded remnants of former reefs and indicate that reefs have developed by the incremental addition of reefal limestone during high sea-level phases’’ (Searle, 1983, p. 270) were, however, occasionally recognized. Facies interpreted from seismic profiling were directly correlated with subsurface cores at Michaelmas Cay, and suggested that the last interglacial surface was 8–12 m below datum (Searle and Harvey, 1982). Based on this, Searle and Harvey speculated that the 27.5 m unconformity recognized at Heron Island (Davies, 1974) possibly represented an earlier emergence, but did not verify it. Because most of the continental shelf is shallow (see Chapter 5), and the continental slope is steep, reef growth over most of the GBR can only proceed during sea-level highstands, or on the shelf edge (at above 60 to 70 m) during higher interstadials (Chapter 8). Critically, this represents only around 10–15% of the last 450 ka since ‘‘true tropical reef growth’’ began on the GBR (Webster and Davies, 2003), suggesting that the GBR developed in as few as 40–50 000 cumulative years comprising numerous shorter growth episodes. Potential explanations for reef turn-on at around 450 ka remain uncertain (Section 2.5). It is nonetheless clear that the GBR’s development was dominated by long periods of subaerial exposure and associated destructive processes punctuated by shorter episodes of relatively rapid reef construction. Importantly, during successive phases of reef growth and emergence driven by major climatic and sea-level changes, coral reefs of similar taxonomic composition and physical character have successfully re-established themselves many times on the GBR (Webster and Davies, 2003). 3.3 Late Pleistocene sea level Sea level during the last 140 ka is known in greatest detail, and has most significantly influenced the geomorphic development of the modern GBR. This interval extends from the penultimate glacial maximum (MIS-6) to present, and includes slightly more than one full glacial cycle (Fig. 3.2). Much of the detail of this record has been resolved over the last decade, and ongoing research and analytical improvements will continue to advance knowledge of late Pleistocene sea levels. Relatively few pre-Holocene data from the GBR are
50
Sea level: a primary control of long-term reef growth
Figure 3.2 The relative sea level through the last glacial cycle from Huon Peninsula, supplemented with data from Joseph Bonaparte Gulf and with GBR data from the Late Pleistocene and early Holocene added. The 30 m and 60 m sea levels are marked. See Fig. 3.3 for shelf configuration at these sea levels (after Woodroffe and Horton, 2005).
included in these reconstructions as most evidence of sea level for this period lies in difficult-to-access deeper water on the continental shelf shoulder and slope. Consequently, late Pleistocene sea level in the GBR is largely inferred from evidence from other locations. 3.3.1 Late Pleistocene sea-level change Efforts to constrain when the penultimate glacial maximum occurred and sea level at that time derived from marine isotope records generally suggest an age around 133 ka and a sea level around 125 5 m (McCulloch and Esat, 2000). This sea level has not been directly established, but a depth similar to the last glacial maximum is inferred from marine oxygen isotope records
3.3 Late Pleistocene sea level
51
(e.g., Chappell and Shackleton, 1986) and the elevation of lowstand deposits of shallow-water mollusks (Ferland et al., 1995). Uranium series dates of 131 2 ka from corals from 80 m in Aladdin’s Cave, Huon Peninsula, are the first direct data to constrain sea level following the penultimate glacial maximum (McCulloch and Esat, 2000). They are believed to have grown during a cooler period similar to the Younger Dryas, when continental ice expansion slowed sea-level rise (McCulloch and Esat, 2000). Most significantly for reef growth, the last interglacial sea-level maximum was achieved by 129–128 ka (Stirling et al., 1998; Esat et al., 1999), suggesting that sea level rose very rapidly into the last interglacial. Based on the Huon coral record, the last 70–80 m of the penultimate transgression occurred in just 1–2 ka, a rate triple that inferred from marine isotope records (e.g., Broecker and Henderson, 1998; Muhs et al., 2002), but not exceeding that experienced during melt-water pulse 1A (Section 3.4.1). Speculation that the penultimate transgression was preceded by a brief oscillation that rose to within 15 m of current sea level 135 ka before falling below 80 m also exists (Esat et al., 1999), but this awaits confirmation (Broecker, 1998; McCulloch and Esat, 2000). Analyses of data carefully screened for quality from several locations support this hypothesis, albeit with slightly different timing and amplitude (Gallup et al., 2002; Cutler et al., 2003; Thompson and Goldstein, 2005). If such rapid high-amplitude sea-level oscillations occurred, they would have significant implications for reef growth on the GBR and globally. The last interglacial period was the last time prior to the Holocene that sea level was near to present. It is when the Pleistocene reefal foundations that underlie many reefs on the GBR formed, and is therefore an important period in the Reef’s history. The last interglacial is generally defined by MIS-5e in deep-sea cores, which extends from 129–128 ka to 116–114 ka. The exact duration of the last interglacial highstand is contentious, with credible onset and termination ages as divergent as 136 ka (Stein et al., 1993; Rubin et al., 2001) and 110 ka respectively (Muhs et al., 1994; Szabo et al., 1994). The few direct dates available for Pleistocene reefs on the GBR are mostly within this range (Chapter 11), although Marshall and Davies’ (1984) study of the southern GBR included dates between 138 ka and 172 ka, the latter conceded to be of questionable accuracy. Last interglacial reef near sea level is rare on the GBR, confirmed only at Digby Island in the southern GBR (Kleypas, 1991). Small outcrops of last interglacial reef may exist in Torres Strait (Barham, 1983), but are not confirmed. Sea level during the last interglacial is generally believed to have peaked several meters higher than present (e.g., Stirling et al., 1995; Lambeck et al., 2002); a eustatic level of 5 m is widely used where local evidence is lacking.
52
Sea level: a primary control of long-term reef growth
Sea level underwent several broad oscillations after the last interglacial, represented by MIS-5d to MIS-5a, in which stages 5c and 5a correspond with two major interstadials (Fig. 3.2). The sea level fall from MIS-5e to MIS-5d after 115 ka was relatively abrupt and reached 60 m below present (Lambeck and Chappell, 2001), although this depth is debated (Cutler et al., 2003). Sea level then rose to at least 15 m below present to peak in MIS-5c around 103–98 ka (Toscano and Lundberg, 1999; Cutler et al., 2003; Potter et al., 2004), before falling to around 60 m during MIS-5b by 90 ka (Cutler et al., 2003). Sea level during MIS-5a rose again to within 15–25 m of present level about 85 ka (Ludwig et al., 1996; Toscano and Lundberg, 1999; Lambeck et al., 2002). An additional highstand at 77 ka has been suggested based on high-precision uranium series dating of Barbados corals, and tentatively linked with one of the five Huon reef terraces developed during MIS-5e–a (Potter et al., 2004). Sea level was 65 m below present when MIS-5a ended around 75 ka (Lambeck et al., 2002). Post MIS-5 sea level fell irregularly to the last glacial maximum. During MIS-4 and MIS-3 sea level rapidly fluctuated, with cycles of 15 m amplitude occurring every 6–7 ka (Chappell, 2002; Lambeck et al., 2002). After rising near to 40 m midway through MIS-4, sea level then fell to around 70 m by 65 ka before again rising near to 50 m by the start of MIS-3 (Lambeck et al., 2002). Chappell (2002) presented a high-resolution sea-level record for this period based on high-precision uranium series dated corals and detailed topographic and stratigraphic analysis. This analysis identified successive peaks in sea level through MIS-3 at 60–58 ka (50 m), 52 ka (46 m), 44.5 ka (56 m), 38 ka (71 m), and 33 ka (72 m), with associated lowstands at 62 ka (76 m), 54 ka (62 m), 46 ka (70 m), 40 ka (80 m), and 34 ka (81 m). With the exception of the most recent highstand (33 ka), these sea-level peaks coincide remarkably with North Atlantic Heinrich events – depositional horizons produced by ice breakouts from glaciated areas – and benthic marine isotope cycles (Shackleton et al., 2000). This coincidence suggests that rapid eustatic sea-level fluctuations driven by northern hemisphere ice-sheet instability affected the Huon reefs, and thus probably also the GBR. Recent research suggests persistent sea-level variability of 3–9 ka period and 6–30 m amplitude driven by ice-sheet dynamics extends from at least MIS-7 (Thompson and Goldstein, 2005). The lowest sea levels of the last glacial cycle occurred during the last glacial maximum at 30–19 ka (Lambeck et al., 2002), coinciding with MIS-2 in deepsea cores (Fig. 3.2). Shallow-water sediment facies and coral data from Huon Peninsula indicate that the sea-level fall to the last glacial maximum began 32 ka, rapidly dropping 30–50 m in 1–2 ka (Lambeck et al., 2002). The details
3.3 Late Pleistocene sea level
53
of sea-level height and stability during the glacial maximum are contentious (see Lambeck et al., 2002; Peltier, 2002), but 125 5 m is an accepted ‘‘best estimate’’ based on ice-volume models and the Barbados coral record (Nakada and Lambeck, 1988; Fairbanks, 1989; Bard et al., 1990; Fleming et al., 1998). Yokoyama et al. (2001b) believe paleodepth-constrained micropaleontological indicators in cores from Joseph Bonaparte Gulf, northern Australia record a slightly deeper lowstand (130 to 135 m) between 22 ka and 19 ka, but others argue that sea level remained stable around 120 m (Lambeck et al., 2002) (Fig. 3.2). Veeh and Veevers (1970) attempted to establish directly the time and depth of the last glacial maximum lowstand adjacent to the central GBR using radiometrically dated corals retrieved by submersible from the outer shelf slope. They concluded that sea level fell to 175 m sometime between 17 ka and 13.6 ka, but overestimated the lowstand depth because the paleodepths of the dated corals were poorly constrained (Yokoyama et al., 2001b). Undated paleoshorelines detected on continuous seismic profiles on the shelf shoulder between 114 and 133 m depth have also been associated with the last glacial maximum lowstand because of their depth (Carter and Johnson, 1986), as has a zone of maximum cave/notch development (suggestive of active erosion) around 100 to 140 m observed on submersible surveys of the shelf slope east of Ribbon 5 and Myrmidon Reefs (Hopley, 1994, 2006; Chapter 9). Evidence of glacial lowstand sea levels on the GBR are summarized in Table 3.1. Following the termination of the glacial maximum at 19 ka sea level rose rapidly to be close to present by around 6.5 ka ago (Section 3.4). 3.3.2 Late Pleistocene sea-level changes on the Great Barrier Reef Quaternary sea-level fluctuations have significantly influenced the growth and geomorphological development of the GBR, with the formation of stacked reefs on the continental shelf during interglacial highstands (at least since 450 ka), and limited growth during glacial maxima due to the steepness of the shelf slope. Between these extremes sea levels associated with stadial and interstadials have located reef growth on the shelf shoulder, where submerged shelf-edge reefs with a variety of morphologies have grown in the last glacial cycle, and presumably also in earlier ones (Hopley, 2006; Chapter 9). Although discrete reef terraces have developed in locations such as east of the Pompeys (Hopley, 2006) and Mustard Patches (Carter and Johnson, 1986), elsewhere along the shelf edge reef units merge and do not form distinctive structures easily associated with particular stillstand events (Harris and Davies, 1989).
Evidence
17 1 13.6 1
25 1.05 15.6 0.5 20.2 0.6 Estimated last interglacial
>38 Estimated last interglacial
Age (ka)
Veeh and Veevers (1970)
Hopley (1970) Hopley (1970) Hopley (1970) Hopley (1971)
Hopley (1970)
Barham (1983)
Kleypas (1991)
Reference
a
LAT, Lowest Astronomical Tide; MHWS, mean high water springs.
The evidence listed below must be considered ambiguous, with firm ages, origin, depth at formation, and sea-level significance questionable (see Harris and Davies (1989) for discussion) Ribbon 5, Myrmidon Notches and caves ‘‘maximum cave and Zone from 100 to Maxwell (1968), Reef pitting development’’ 140 m 140 m Hopley (1994) Bowl Reef Submerged shorelines/reefs This book, Chapter 9 (see Fig. 9.14b) Submerged shorelines/reefs 114 m 18 Carter and Johnson (1986)
Evidence of last glacial maximum sea level Central Queensland Shelf Dated coral (Southern GBR) Beachrock 175 m 150 m
Weathered and oxidized dune and beach 1.2–4.5 m above ridge sands. Discontinuous occurrence present levels between Bowen and Rollingstone Dune calcarenite Dune calcarenite Elevated beachrock 5 m above MHWS Boulder spits
Mainland coast, e.g. Elliot River, Cungulla
Mt. Inkerman Cockermouth Island Camp Island High Islands
Pleistocene reef exposure
Reef flat (1.4 m above LAT) Reef flat
Elevationa
Torres Strait
Evidence of last interglacial sea level Digby Island Pleistocene reef exposure
Location
Table 3.1. Reported last interglacial and last glacial maximum sea-level elevation and ages from the GBR region
3.3 Late Pleistocene sea level
55
Hopley (1982) discussed the consequences of late Pleistocene sea-level fluctuations on the GBR, and although some details of sea-level history have been revised, the broad trends identified then remain valid. Shelf bathymetry was clearly an important control of when and where reef growth occurred during the last glacial cycle. Figure 3.3 depicts the emergence of the continental shelf at present, 30, 60, and 120 m sea-level lowstands. As discussed further in Chapter 12, when sea level was 30 m below present, most existing reefs and their last interglacial substrates were emergent as limestone islands (Fig. 3.4), some possibly with surrounding shoals. The mainland coast was further east, most markedly opposite broad coastal plains and less so near major headlands. The impacts of the 30 m regression differed geographically over the GBR, with perhaps the greatest effects occurring north of Cairns. At 30 m the coastal plain between Cape Melville and Cairns extended to the mid-shelf, and inshore reefs were completely exposed. A narrow channel existed behind the Ribbon Reefs, but it was shallow and poorly connected to the open sea. Further north, Torres Strait was closed, and the coast located further east in the Gulf of Papua. South of Cairns to the southern GBR the mainland shore was also further east than at present, but greater shelf depth, a less continuous outer barrier, and flooding from behind via the Capricorn Channel would have allowed coral growth over more of the continental shelf than occurred further north. At lower stillstands reef growth was further restricted to the shelf edge, with only a narrow fringing reef on the outer slope and the detached reefs of the far north persisting at the last glacial maximum (Fig. 3.3d). The paleogeography of the GBR at various sea-level stages is discussed in Chapter 12. As previously noted, recent studies achieving higher chronological and elevation precision than previously possible reveal an emerging picture of significant high-frequency sea-level oscillations through the late Pleistocene. This research indicates that the structure of stadial and interstadial phases is more complex than formerly understood (e.g., Chappell, 2002; Potter et al., 2004; Thompson and Goldstein, 2005), with higher-frequency oscillations of significant amplitude (15 m amplitude during interstadials and 35 m amplitude during glacials: Siddall et al., 2003) occurring over these broader cycles. Rates of sea-level rise inferred from this recent research are extremely rapid; for example, a rate of 50 mm a1 sustained for several thousand years is required during the penultimate deglaciation (McCulloch and Esat, 2000), and up to 16 mm a1 between 62 ka and 60 ka if Chappell’s (2002) sea-level positions are sound. However, as noted earlier and discussed in Section 3.4.1, these rates are not unimaginable, with rates during melt-water pulses of the most recent postglacial trangression widely accepted to exceed 40 mm a1 for 500 years (Fairbanks, 1989; Bard et al., 1990; Edwards et al., 1993). If
Figure 3.3 Continental shelf in Great Barrier Reef region at (a) present sea level; (b) 30 m; (c) 60 m; and (d) 125 m (last glacial maximum).
3.3 Late Pleistocene sea level
57
Figure 3.4 Holocene reef core (left) and last interglacial reef material (right), diagenetically altered during subaerial exposure, from Yam Island, Torres Strait.
sustained, depth increases associated with swift sea-level rise may lift the euphotic zone above the reef surface and inhibit reef growth, producing ‘‘give-up’’ reefs (Neumann and Macintyre, 1985; Chapter 11). The possibility that mid and outer shelf reefs on the GBR could be drowned in situ by rapidly rising seas was raised by Carter and Johnson (1986), but many submerged reefs they identified are within the photic zone, making it unlikely that rapid sealevel rise alone caused their demise (Hopley, 2006; Chapter 9). ‘‘Keep-up’’ reefs that have continuously grown upward for the last 14 ka are described from Tahiti which demonstrate reefs can survive rapidly rising seas (Montaggioni et al., 1997). However, contrary to earlier views based on traditional sediment deposition models (e.g., James and Kendall, 1992), data now show that maximum offshelf sediment flux occurs as transgressive events flood the outer shelf (Dunbar and Dickens, 2003; Page et al., 2003; Page and Dickens, 2005). Seas rising to this level may thus have initiated offshelf sediment transport and suppressed reef growth by increasing turbidity and compressing the euphotic zone as rising sea-levels increased depth. Of course, rapid sealevel falls also occurred through the late Pleistocene, and emergence presents obvious challenges for reef growth. Cutler et al. (2003) estimate that the mean rate of sea-level fall from MIS-5a to MIS-4 was possibly above 10 mm a1, and
58
Sea level: a primary control of long-term reef growth
a rate triple that is probable as sea level fell from MIS-3 to the last glacial maximum (Lambeck et al., 2002). For several hundred thousand years the GBR has been repeatedly affected by significant fluctuations in one of its primary environmental controls – sea level. The conclusion drawn from the previous glacial cycle must be that changing sea levels are more normal than stable ones. Ironically, as discussed below, it is during the postglacial period, when modern reefs have flourished, that one of the most rapid and sustained episodes of eustatic sea-level rise occurred, to be followed by a prolonged period of eustatic sea-level stability. 3.4 Postglacial sea level The modern GBR developed as the postglacial marine transgression flooded the continental shelf. Not surprisingly, advancing knowledge of postglacial sea level has been a long-standing research focus, to provide a fundamental context in which to interpret the GBR’s development (e.g., Hopley, 1983a; Larcombe et al., 1995a), but also for other reasons, such as resolving isostatic contributions to relative sea level in different locations (e.g., Chappell et al., 1982; Lambeck and Nakada, 1990). This research continues; detailed sea-level histories remain unavailable for large areas of the GBR. Where sea-level histories do exist, four issues remain the focus of debate: (i) was postglacial sea-level rise smooth or pulsed; (ii) when was modern sea level first reached; (iii) has sea level been higher than present during the Holocene; and (iv) where applicable, when and how did relative sea level fall following this highstand? 3.4.1 Postglacial sea-level rise At the end of the last glacial maximum 19 ka, eustatic sea level rose quickly from around 125 5 m to be near to present by approximately 6.5 ka (Lambeck et al., 2002). Sea-level curves from far- and intermediate-field sites are broadly consistent over most of this period (Fig. 3.5), although the detail is still debated. Early disagreement whether sea level rose smoothly or irregularly from the glacial maximum (e.g., Fairbridge, 1961; Jelgersma, 1961) is now largely attributed to difficulties in decoupling eustatic and isostatic contributions to relative sea-level change (Masselink and Hughes, 2003). The eustatic component of postglacial sea-level rise was dominated by melt-water discharge until around 6.5 ka, after which ocean volume has changed relatively little (Lambeck and Chappell, 2001). Since 6.5 ka local variations in sea level mainly reflect isostatic adjustments and tectonic histories, the influences of which on the GBR are discussed in Section 3.4.3.
3.4 Postglacial sea level
59
Figure 3.5 Postglacial sea-level history reconstructed from the long reef cores from Barbados, Tahiti, and Huon Peninsula (after Woodroffe and Horton, 2005), with earliest Great Barrier Reef Holocene coral (Hayman Island (Hopley et al., 1978); Boulder Reef (Davies and Hopley, 1983)) and microatoll (Yam, Hammond Island (Woodroffe et al., 2000), Fisher Island, Low Wooded Island (Polach et al., 1978)) data included.
The consensus is now that sea level during the last deglaciation did not rise smoothly and continuously, but included two periods of rapid and sustained sea-level rise separated by a short period of stability during the Younger Dryas (12.5–11.5 ka), when ice melt briefly ceased (Fairbanks, 1989; Chappell and Polach, 1991; Lambeck et al., 2002). Episodes of extremely fast sea-level rise related to rapid ice-sheet decay and melt-water discharge are also recognized. Two such episodes are recognized in the Barbados coral record, and are widely referred to as melt-water pulse 1A (MWP-1A) and melt-water pulse 1B (MWP-1B) (Fairbanks, 1989; Bard et al., 1990). The precise timing, duration, amplitude, and rates of sea-level rise associated with these events, particularly MWP-1B, remains contentious. MWP-1A is generally accepted as ‘‘a real feature of the postglacial eustatic sea-level history’’ (Webster et al., 2004,
60
Sea level: a primary control of long-term reef growth
p. 249), beginning at 14.5 ka, after which sea level rose by 40–55 mm a1 for around 500 years (Fairbanks, 1989; Edwards et al., 1993). In contrast, the details of MWP-1B are less certain (Bard et al., 1996; Shennan, 1999; Bassett et al., 2005), but it is speculated that MWP-1B began 11.5 ka, lasted for around 200 years, and produced rates of sea-level rise comparable to those of MWP-1A (Fairbanks, 1989). Additional melt-water pulses beginning at 19 ka (Yokoyama et al., 2001a; Clark et al., 2004) and 7.6 ka (Blanchon and Shaw, 1995) are reported, but neither is evident in the Tahiti record (Bard et al., 1996), questioning their global significance (Peltier, 2002; Bassett et al., 2005; Woodroffe and Horton, 2005). Few data from the GBR exist to contribute to our knowledge of late glacial to early Holocene sea level. We know of only two radiocarbon ages that show postglacial reef growth earlier than 9 ka, one from Boulder Reef (9.6 ka from 16 m: Davies and Hopley, 1983) and the other from near the base of the fringing reef at Hayman Island (9.3 ka from 15 to 20 m: Hopley et al., 1978). Early postglacial to Holocene sea-level data are rare because the pre-Holocene substrate beneath drilled reefs is generally quite shallow and thus was not inundated until late in the transgression. Pleistocene foundations at 28 m at Myrmidon Reef are the deepest drilled on the GBR (Marshall, 1985) (although not necessarily the maximum depth of Pleistocene reef); drill and seismic investigations suggest Pleistocene reef is commonly 15–25 m below the Holocene veneer (Chapter 11). Assuming an average depth to Pleistocene of 20 m for most mid to outer shelf reefs, and accepting Thom and Roy’s (1983) sea-level envelope approximates sea level on the GBR at this time (Section 3.4.3), little Holocene reef growth could have begun before 10–9 ka. Fringing reefs probably grew at lower levels around the base of some Pleistocene platforms (see Fig. 12.7: Bowl Reef), during this and previous transgressions (Hopley, 1982; Johnson et al., 1996), but from the last glacial maximum to the early Holocene, reef growth was largely restricted to the upper edge of the continental shelf (Chapter 12), where depths now make sampling difficult, and from where few useful sea-level data have been gained. The eustatic sea-level histories for this period derived from coral records from Barbados, Tahiti, and Huon Peninsula are presented in Fig. 3.5, with the earliest dates from the GBR also included. Coral reefs responded variably to rapid postglacial sea-level change. Some reefs drowned (Blanchon et al., 2002; Webster et al., 2004), but others kept up and remained close to sea level (Montaggioni et al., 1997). On shallower gradient shelves backstepping upslope occurred, with some deeper reefs eventually catching up to sea level once it stabilized in the mid-Holocene (see Davies and Montaggioni, 1985). For reasons outlined earlier, direct evidence
3.4 Postglacial sea level
61
of how the GBR responded to early postglacial sea-level rise is unavailable. Several researchers have proposed that submerged reefs on the deeper continental shelf shoulder ‘‘backstepped’’ onto the shallower shelf during the transgression, but they may also be the result of earlier transgressions (Davies and Montaggioni, 1985; Carter and Johnson, 1986; Harris and Davies, 1989; Hopley et al., 1997). Hopley (2006) produced a model linking the development of submerged shelf-edge reefs to sea level and slope morphology but without a chronology the history of these features remains speculative. Submerged reefs are, however, the focus of planned research made possible with the use of improved submersible and remotely operated vehicle (ROV) technologies, and they offer great promise as archives of early postglacial sea-level data for the GBR region hitherto unavailable. Most reefs that have been cored on the GBR have earliest but not necessarily basal radiocarbon ages younger than 8 ka, and seem typically to have adopted a catch-up strategy as they grew toward a stabilizing higher midHolocene sea level (see Hopley (1982) and Davies and Hopley (1983) for examples, and Chapter 11 for discussion). Detailed discussions of the Holocene growth histories of different types of reef on the GBR can be found in Chapters 7, 8, and 9. Much of the data used to reconstruct sea level during the Holocene on the GBR is derived from these growth histories. The indicators from which these data are derived are examined further below.
3.4.2 Sea-level indicators Sea-level histories for the GBR, as elsewhere, are largely reconstructed with data derived from a variety of sea-level indicators of varying precision (Davies and Montaggioni (1985), van der Plassche (1986), and Pirazzoli (1991) provide excellent reviews of sea-level indicators). The key attributes of useful sea-level indicators are: (i) (ii) (iii) (iv)
a narrow vertical depth range; a reproducible elevation; good geological preservation; and can be accurately dated.
Finite sea-level indicators are usually limited at a particular intertidal, shallow subtidal, or possibly peritidal level and allow quantitative assessment of the relationship of the sea surface to the indicator. This relationship is sometimes referred to as the ‘‘indicative meaning’’ which for a particular indicator in its local environment may have an elevational range quantified and reduced to a datum, or may be expressed relative to a reference tide level (e.g., mean high
62
Sea level: a primary control of long-term reef growth
water springs (MHWS)) with an associated error range (van de Plassche, 1986). They are preferred for high-resolution sea-level studies, but rarely form during periods of rapid transgression. As coastal configuration and tide curves changed during the postglacial transgression the vertical distances between tidal datums such as mean sea level (MSL) and the MHWS tide (or of biological, chemical or physical interactions with them) may also have varied. For example, the elevation separating MSL and MHWS is larger in a mesotidal setting than in a microtidal setting. Thus where tidal traits have changed interactions between sea-level rise, changed tidal dynamics, and the indicator’s indicative meaning, may complicate sea-level reconstructions. The main sea-level indicators used to establish sea-level histories on the GBR are described below. (a) Corals and coral reefs Coral and coral reef cores provide many sea-level data on the GBR. With a few exceptions, these indicators must be considered relatively imprecise directional sea-level indicators (that simply indicate if sea level was above or below the indicator), as the depth at which a coral grew can rarely be precisely determined, and thus the indicative range may be 30 m or more. However, corals do have a good preservation potential and can be accurately dated using radiocarbon and uranium series methods. As a result, coral-derived sea-level data generally has relatively large elevation but small age-error terms. At the broadest level, Holocene reef growth over transgressive surfaces recognized by the occurrence of paleosols, diagenetic alterations, and calcrete stringers (e.g., Hopley, 1982; Webster and Davies, 2003) represents a firm directional sea-level indicator where in situ coral (Fig. 3.6b) immediately overlies the unconformity, although verifying in situ condition and determining the lag between inundation and coral growth can be challenging. Four main reef facies are represented in cores from a wide area of the GBR: a coralline encrusting facies; a branching coral facies; a head coral facies; and a sand and coral rubble detrital facies. Facies trends have been used to diagnostically indicate sea levels on other reefs (e.g., Montaggioni et al., 1997) but not successfully on the GBR (Davies and Montaggioni, 1985). Sand aprons prograde leeward of many GBR reef flats, with subtidal sands fining upward but intertidal sands coarsening upward (Davies, 1983). If preserved, these sequences may function as a directional sea-level indicator, or possibly a coarse but finite sea-level indicator if reef flat corals are also identified (Davies and Montaggioni, 1985). On the GBR there is no equivalent to the Caribbean coral Acropora palmata, which only grows within 5 m of the surface and can be used to identify keep-up
3.4 Postglacial sea level
63
Figure 3.6 Sea-level indicators. (a) Uplifted coral terraces at Huon Peninsula (photograph: A. Bloom). (b) In situ branching Acropora exposed below reef flat surficial sediments over higher mid-Holocene back reef, Pioneer Bay, Orpheus Island (see Fig. 7.8). (c) Fossil microatoll, Iris Point, Orpheus Island (see Figs. 7.4 and 7.5.) with a small backpack for scale. (d) Oysters, Balding Bay, Magnetic Island. (e) Beachrock, Low Isles. (f) Mangrove peat exposed on the beach near Bowen.
reefs in which vertical accretion tracked sea-level rise (Lighty et al., 1982; Toscano and Macintyre, 2003). Species assemblages argued to be depth definitive on other reefs (Montaggioni et al., 1997) are also not recognized. Corals are, however, zonally distributed across the GBR along energy gradients (Done, 1983), and various coral morphotypes are stratified over environmental gradients, including depth and wave energy (Chappell, 1980). Further analysis may reveal relationships between species, morphotypic assemblages, and depth that may indicate sea level, but the internal structure of many GBR
64
Sea level: a primary control of long-term reef growth
reefs is dominated by detrital sediments (e.g., Davies and Hopley, 1983; see Chapters 7–9), and thus reconstructing sea levels by comparing extant taxa and/or morphotypic depth distributions with those recovered in cores will be challenging. Coral reef growth on the GBR typically extends upward to 0.25 m of mean low water springs (MLWS) in open-water habitats, but can grow higher where water ponds above this level (Hopley, 1983a; Davies and Montaggioni, 1985). On the GBR water may be ponded behind shingle ramparts or algal rims to near MSL, which can be 2 m above the open-water low tide (Hopley, 1983a, 1986a, 1987). Where in situ reef flat material can be identified an indicative range can be resolved with greater confidence, although differences in tidal range, wave exposure, and species assemblages introduce some variability. Algal ridges are conspicuous and distinctive reef crest features on many midPacific atolls and some higher-energy Atlantic reefs (e.g., Nunn, 1993), and a coralline algal ridge is indicative of reef growth to near the sea surface (near MLWS, but higher in higher-energy settings (e.g., Stoddart, 1969; Adey, 1978; Woodroffe et al., 2004)). Algal ridges are a less useful sea-level indicator on the GBR. On the outer ribbon reefs they form a wide (200 m) but thin (0.1–0.15 m) crust over coral rubble, with a low preservation potential probable in this extreme energy environment (Davies and Montaggioni, 1985). On the inner shelf reefs of the northern GBR they are narrower (<20 m wide), elevated between mean low water neaps (MLWN) and MLWS, but also often very thin and easily destroyed (Scoffin, 1977). The use of coral or coral reef material as sea-level indicators in high-energy or storm-prone settings can be problematic as material can be transported above growth position and sea level overestimated without careful verification of in situ status. Timetransgressive variations in storm activity or wave exposure may add a further complication; for example, during the Holocene ‘‘high-energy window’’ (Hopley, 1984) larger waves may have deposited material higher onshore than now possible since the outer barrier reached sea level and excludes large oceanic storm swells from the reef lagoon. (b) Microatolls Microatolls are disk-shaped corals with living outer rims and relatively planar dead upper surfaces that are common on many reef flats (Fig. 3.6c). They may grow to several meters in diameter and vary from centimeters to meters in thickness. Microatolls form when upward coral growth is constrained by prolonged exposure near to low water, and lateral growth predominates (Scoffin and Stoddart, 1978; Stoddart and Scoffin, 1979). Both massive and branching corals form microatolls, but those developed by massive corals are
3.4 Postglacial sea level
65
most common and are better preserved. Forty-three species from 23 genera are known to form microatolls on the GBR (Rosen, 1978), and fossil microatolls approaching 6000 years old are reported from several locations (Hopley, 1982; Chappell et al., 1983). In open-water habitats on the GBR, where tidal fluctuations are unrestricted, the upper surfaces of Porites microatolls are usually constrained close to the MLWS tide (Scoffin and Stoddart, 1978). In moated habitats, where the ebbing tide is held behind features like algal rims or rubble ramparts, microatoll upper surfaces form close to the elevation of the ponded water, which may be as high as the mean high water neap (MHWN) tide (Scoffin and Stoddart, 1978). Tolerances to emergence and exposure vary with species; Goniastrea and Platygyra sp. typically grow higher than Porites when they occur on the same reef flat. The exact relationship of the upper limit to coral growth to tidal datums and desiccation can vary with exposure to wavelet wetting, diurnal and seasonal tidal patterns, and a variety of other factors (Scoffin and Stoddart, 1978), but on the central GBR the tops of microatolls within the same field are usually within 10 cm (Chappell et al., 1983). Even greater precision is possible where tidal range is low and growth habitat can be determined (Smithers and Woodroffe, 2000). Microtopographic undulations over the upper surfaces of well-preserved microatolls can be used to track interannual variations in the elevation of the confining water level, within the limits of coral growth (Woodroffe and McLean, 1990; Smithers and Woodroffe, 2001). The relatively precise relationship between microatoll elevation and a specific tidal level means that they are one of the most precise finite sea-level indicators available, and they have been used extensively in sea-level studies on the GBR and elsewhere (Section 3.4.3). In many locations fossil microatolls can be readily identified and compared directly with nearby living counterparts that can be surveyed to a tidal datum. Where the tidal curve has not significantly changed through time, the relative elevations of microatoll tops can be used to reconstruct former constraining water levels. Microatolls can be radiometrically dated and thus histories of water-level change can be established. Where it can be verified that the microatolls were not moated, these records are very good proxies for sea-level change. Chappell et al. (1983) used this approach to determine the nature of late Holocene sea-level change on the central GBR, and others have used fossil microatoll sequences to establish sealevel histories in other areas (e.g., Woodroffe et al., 2000). Microatolls are excellent sea-level indicators, but several issues need consideration when they are used. Confirmation is required that they: (i) are microatolls and not ‘‘pseudomorphs’’ with planar tops caused by erosion
66
Sea level: a primary control of long-term reef growth
(verified by examining a colony’s internal structure); (ii) are undisturbed by storms or other agencies; (iii) grow in an open-water habitat (this applies to most but not all investigations – see Davies and Montagionni (1985) for diagnostic criteria for open-water microatolls); and finally (iv) when subsampling microatolls for dating, especially large long-lived specimens, subsample location should be systematically selected and replicated where inter-colony and between site comparisons will be made. For example, ad hoc sampling across a large microatoll several meters in diameter (e.g., randomly between perimeter and center) may introduce an age uncertainty exceeding 100 years. (c) Oysters and encrusters Encrusting organisms including oysters, barnacles, and calcareous tubeworms occupy well-defined vertical zones along the coast and have been used as sealevel indicators, including on the GBR. The oysters Saccostrea cucculata and Crassostrea amasa form conspicuous encrustations from MSL up to MHWN, although surveys from island and mainland locations reveal that the upper and lower extent of C. amasa (Endean et al., 1956), and the precise elevation and range of S. cucculata encrustations, may vary considerably within this range. In northern Queensland S. cucculata forms distinctive visor-like structures 0.3–0.6 m thick (Fig. 3.6d) between MHWN and MSL (Beaman et al., 1994), with the exact elevation varying with exposure. On open coasts visors are usually poorly preserved as intense weathering and wave erosion destabilizes the rocks to which they attach (Scoffin, 1977). At Balding Bay on Magnetic Island a fossil encrustation of S. cucculata occurs in a protected sea cave 1.65 m above the modern oyster visor, and has been argued as evidence of mid-Holocene sea levels 1.65 m above present persisting until around 4 ka (Beaman et al., 1994). This interpretation differs from that established with microatolls elsewhere on Magnetic Island, which suggests a shorter midHolocene highstand of around 1 m at 6 ka and decline to present soon after (Chappell et al., 1983). Although the fossil and modern oyster encrustations at Balding Bay are separated by 1.65 m, it is arguable whether the fossil oyster’s protected environment is directly comparable to the modern oyster’s exposed habitat, raising concerns that the fossil oysters may have grown higher than possible on an exposed shore. Barnacles and serpulid tubeworm agglomerations have been used as sealevel indicators elsewhere in Australia (reviewed by Baker et al., 2001a), but have had limited application on the GBR. Endean et al. (1956) documented in detail the vertical distributions of a range of intertidal organisms on islands of the Queensland coast, and many of these organisms have potential as sealevel indicators. The barnacles Octomeris brunnea and Tetraclitella sp. form
3.4 Postglacial sea level
67
significant crusts that are preserved in sheltered locations, such as the caves at Balding Bay, where they grow in close association with the oyster S. cucculata. A calibrated radiocarbon age of 5310 years exists for the Balding Bay fossil barnacles (Beaman et al., 1994), but their sea-level significance remains ambiguous. Just south of the GBR at Valla Beach, northern New South Wales, a late Holocene sea level 1 m above present was inferred from the relative heights of modern and dated fossil calcareous tubeworms (Galeolaria caepitosa) and barnacles (Flood and Frankel, 1989). (d) Beachrock and cemented deposits Beachrock is literally lithified beach sediment, formed when beach sediments are cemented together, usually by calcium carbonate cements. Lithification occurs on relatively stable beaches beneath a cover of loose sediments that are later eroded to expose the beachrock outcrop. Cemented clasts can vary from fine sands to boulders and be terrigenous or biogenic. They are generally indistinguishable from uncemented sediments in equivalent settings on the same beach. Sedimentary structures such as bedding planes are preserved in many beachrocks composed of finer textured sediments, but are less common in coarser outcrops (Fig. 3.6e). Some beachrocks are weakly cemented, with only the outer crust firmly indurated, but highly indurated massive beachrocks cemented through thicknesses exceeding 1 m also occur. Beachrock forms on reef islands (cays and high islands, Chapter 10) and on mainland beaches. Debates concerning the form of beachrock cements and beachrock development are summarized elsewhere (e.g., Hopley, 1982; Scoffin and Stoddart, 1983). Recognition that modern beachrock typically forms in the intertidal led to its use as an indicator of intertidal elevation; the vertical ranges of beachrock outcrops vary with tidal range. The exact upper limit of formation is, however, often poorly constrained, particularly where the tidal range (Hopley, 1986b) and/or wave energy (Hopley, 1971) are high. On the GBR the upper limit is commonly around MHWS, but the precise upper limit to cementation can be difficult to define as the degree of cementation may grade up the beach. This probably in part reflects the speed with which beachrock can form; outcrops on Magnetic Island developed in six months (Hopley, 1986b), and similarly rapid formation is reported from other reefs (Emery et al., 1954; Russell, 1962; Scoffin, 1970). It is possible with such rapid cementation that beachrocks may form at tidal levels achieved relatively infrequently where beaches are sufficiently stable. Thus precisely defining an upper limit to cementation and consistently relating this to a tidal level is problematic. On the northern GBR the upper limits of modern beachrocks are between 1.6 and 2.9 m above tidal datum (Davies and Montaggioni, 1985), where MHWS is around 2.3 m.
68
Sea level: a primary control of long-term reef growth
After considering these limitations, Hopley (1986b) suggested that the upper limit of beachrock formation on the GBR should conservatively be considered to be as high as the Highest Astronomical Tide (HAT). The base of beachrock outcrops on the GBR is commonly around MLWS but should not be used for sea-level reconstructions as cementation may have occurred as sea-level rose above this level at an earlier stage of lithification. Constituent biogenic clasts may be dated to derive a beachrock’s maximum age, but the cement must be dated to determine a minimum age and this is technically demanding. An adequate volume of uncontaminated cement is difficult to harvest and the possibility of multigenerational cementation and rejuvenation of the apparent age presents problems. An example from Herald Island illustrates the difficulty in precisely dating beachrock (Hopley, 1971). A Trochus shell from near the top of the outcrop at 4 m above HAT was radiocarbon dated at 4280 100 years BP whilst a bulk sample of fine-grained beachrock comprising approximately 50% cement at the base of the same outcrop was radiocarbon dated at only 3540 90 years BP. McLean et al. (1978) provide another example from the northern GBR, where neighboring shingle clasts on a cay beach differed in age by up to 3 ka. They also described a rampart rock, firmly indurated by a fine-grained cement in many sections, deposited over reef flat microatolls which consistently date around 3.5 ka. Replicate dates from the cement suggest an age around 2 ka, but interstices filled with weakly consolidated muds remain within the outcrop. Other cemented outcrops (cay sandstone, dune calcarenite, coffee rock, and silcrete) that superficially resemble beachrock but have no consistent relationship to sea level also occur near the coast (Hopley, 1986b). Conglomerate platforms, emerged reefs, and rubble ramparts can also be similar in appearance and elevation to beachrock, but each should be distinguished for improved interpretation. Finally, although many beachrocks document former shoreline positions and have remained in situ, large beachrock slabs can be dislodged during storms, and are often recemented intact. Only in situ beachrock should be used as a sea-level indicator. (e) Mangrove deposits Mangroves grow in the intertidal zone, with most species confined between MHWS and MLWS tide levels. Individual mangrove species have narrower vertical ranges, and where in situ stumps or roots can be identified it has been suggested that sea level may be estimated to within 1 m (Hopley, 1982). However, detailed survey in different north Queensland estuaries shows marked variation in the elevation and intertidal range occupied by particular mangrove species; up to 0.5 m elevation difference was determined for particular
69
3.4 Postglacial sea level
Table 3.2 Published elevations of Rhizophora stylosa Tidal rangea (HAT–LAT)
Elevation range relative to LAT at coast for R. stylosa (vertical range)b
Reference
3.40 m
0.39–4.30 m LAT (3.91 m)
Bunt et al. (1985)
3.30 m
2.39–3.26 m LAT (0.87 m)
Bunt et al. (1985)
3.30 m 3.60 m 3.89 m
1.48–3.96 m LAT (2.48 m) 2.76–4.47 m LAT (1.71 m) 1.45–2.90 m LAT (1.45 m)
4.01 m
1.46–3.26 m LAT (1.80 m)
4.01 m 4.01 m
1.40–2.90 m LAT (1.5 m) 2.01–2.46 m LAT (0.45 m)
Sandfly Creek
4.01 m
1.56–1.86 m LAT (0.30 m)
Cocoa Creek
4.01 m
1.86–2.64 m LAT (0.78 m)
Cocoa Creek
4.01 m
1.76–2.96 m LAT (1.20 m)
Bunt et al. (1985) Bunt et al. (1985) Grindrod and Rhodes (1984) Woodroffe S. A. et al. (2005) Spenceley (1982) Woodroffe S. A et al. (2005) Woodroffe S. A et al. (2005) Woodroffe S. A et al. (2005) Horton et al. (2003)
Location Normamby River Endeavour River Morgan River Murray River Missionary Bay Saunders Beach Cockle Bay Sandfly Creek
a b
Predicted range at coast as per National Tide Tables (2005). Note that upper limit may exceed HAT at the coast as the elevation of tidal planes are modified up through an estuary.
mangrove species in communities less than 20 km apart and was linked to local differences in tidal conditions (Spenceley, 1982; Bunt et al., 1985) (Table 3.2). Such variability clearly introduces problems for sea-level reconstruction, especially where coastal and tidal configurations have changed through the Holocene. Larcombe et al. (1995a) suggest an error of 1.5 m for locations where the modern range of mangrove deposits is unknown. Mangrove deposits are described as peats or muds. Mangrove peats are highly organic units dominated by mangrove-derived macrofossils. On the GBR, mangrove peats exceeding 0.15 m thickness appear restricted to offshore islands, or exposed shorelines where finer material is winnowed away. Mangrove peats in cores from the mainland are usually thin (<0.15 m) compared to offshore units, and we know of nowhere on the GBR where mangrove peats are presently forming. Unpublished radiocarbon dates indicate a phase of mangrove peat accumulation between 6 and 5 ka, coincident with the ‘‘big swamp phase’’ in Northern Australia when mangrove expansion peaked near the end of the transgression (Woodroffe et al., 1985). Remnant mangroves
70
Sea level: a primary control of long-term reef growth
dated as 8740–8550 years old but complete with intact bark and leaves have recently been recovered in cores from the northern GBR beneath 70 cm of marine sediment, and it has been speculated that they were quickly buried by a pulse of rapid sea-level rise (Hull, 2005). Mangrove muds are dominated by fine terrigenous sediments (mostly fine silts and clays) but contain mangrove pollen and sometimes remnants of mangroves and mangrove community associates such as oysters and other molluskan fragments (Grindrod and Rhodes, 1984). They accumulate beneath mangrove communities and are typically grey or greyish olive and buttery in texture. Mangrove muds are widely described, especially from the mainland coast and inner shelf where they commonly overly very fine beach and subtidal flat sands and are interpreted as a transgressive deposit (Grindrod and Rhodes, 1984; Carter et al., 1993; Larcombe and Carter, 1998). They also occur in paleotidal channels across the shelf and in paleoestuaries nearer the shelf edge (Grindrod et al., 1999; Moss and Kershaw, 2000). Local variations in the indicative meaning of mangrove deposits and of deposit taphonomy must be recognized when using them for sea-level reconstructions. As discussed, mangrove elevation may vary spatially as tidal curves differ, and mangrove deposit elevation can change over short distances. The surface elevation of mangrove muds in Mutchero Inlet near Innisfail, for example, falls by 0.9 m over 4 km in the upstream direction, possibly due to freshwater influx from upstream (Gagan et al., 1994). Especially on exposed coasts, mangrove deposits are vulnerable to reworking during storms, and some, but not all (Beaman et al., 1994), mangrove deposits may have been reworked during the postglacial transgression (Larcombe et al., 1995a). Bunt et al. (1985) estimated that disturbances including erosion and accretion produced elevation errors of 1 m. Compaction and/or organic degradation can also lower a deposit, resulting in sea levels being underestimated. The impact of post-depositional modifications is minimized where peats are thin and overlie less compactible strata such as bedrock or beach sands. Compaction of 10–20% is estimated for Holocene mangrove sequences overlying stiff Pleistocene clays on the central GBR (Larcombe et al., 1995a). Well-preserved mangrove wood can be radiocarbon dated, and where it is clearly in situ, can be located on the plant (e.g., a buttress root), and is close to modern counterparts, both relative elevation and age can be confidently established. Establishing precise chronologies from mangrove muds is more challenging, as bulk samples are required to yield sufficient carbon and samples may integrate carbon with a wide range of true ages. Accelerator mass spectrometric (AMS) dating requires much smaller samples and can reduce the potential size of these errors. However, conspicuous paleochannels visible
3.4 Postglacial sea level
71
on tidal flats along the GBR coast clearly document the lateral mobility of mangrove-lined creeks, and the continual reworking of organic deposits. (f) Microfossils Microfossils such as pollen and foraminiferan tests have been used as sea-level indicators on the GBR. For example, palynological examinations of continental shelf and deep-sea cores collected east of the GBR reveal cycles of mangrove pollen abundance as transgressions flooded the continental shelf shoulder, and declines coinciding with sea-level regressions and periods of stability (Grindrod et al., 1999; Moss and Kershaw, 2000). Foraminiferan tests preserved in coastal sedimentary deposits are another microfossil used as a sea-level indicator on the GBR. Many modern foraminiferans have narrowly defined environmental niches (Murray, 1991), and intertidal species are often stratified according to tidal levels (e.g., Scott and Medioli, 1978; Gehrels, 1994; Hayward et al., 1999). Where they are well preserved in situ they are useful indicators of sea level. Assemblages associated with broader tidal or environmental zones have been identified within the GBR (Haslett, 2001) and elsewhere in Australia and overseas (Scott et al., 1996; Wang and Chappell, 2001; Cann et al., 2002; Hill et al., 2003), but elevational precision is often poor. The foraminiferan ‘‘transfer function’’ approach (see Horton et al. (1999) for details) has been applied overseas to develop precise sea-level records from foraminiferan data (e.g., Edwards, 2001; Gehrels et al., 2001; Debenay et al., 2002; Sawai et al., 2004; Edwards and Horton, 2005) and has been used recently on the GBR with impressive results (Horton et al., 2003; Woodroffe S. A. et al., 2005). A precision of 0.07 m was achieved for mesotidal mangrove coasts near Townsville (Horton et al., 2003). This technique has great promise for reconstructing sea-level histories on other sedimentary coastlines of the GBR where high-precision sea-level indicators such as microatolls and encrusting oysters are poorly represented, and research to refine this methodology continues (Woodroffe S. A. et al., 2005). This method can also resolve paleodepth and may thus be applied to improve the precision of directional indicators such as mangrove muds. Foraminiferans are small and light; age is usually determined by AMS dating a bulk sample of foraminiferan tests picked from sediment under a microscope. (g) Other indicators Several geomorphic or sedimentological features with more limited application have also been associated with sea level on the GBR. These include chenier ridges (Chappell et al., 1983), intertidal facies (Harvey et al., 2001), freshwater swamp, fluvial, and other terrestrial facies or materials (Hopley,
72
Sea level: a primary control of long-term reef growth
1970; Crowley and Gagan, 1995; Smithers et al., 2006), shore platforms (Stanley, 1928; Driscoll and Hopley, 1967), reef islands, boulder tracts, and ramparts (Scoffin, 1977; Hopley, 1982; Davies and Montaggioni, 1985), and bioerosion and cementation signatures (Scoffin, 1977). 3.4.3 Holocene sea levels on the Great Barrier Reef Sea-level change during the Holocene, particularly the length of time that sea level has been near to present and whether or not it has been higher, is an important influence on coral reef growth and has relevance to many other environmental studies, including coastal management. As a far-field site not directly affected by ice load where many different sea-level indicators occur, the GBR has contributed significantly to our understanding of Holocene sealevel change globally, and of the mechanisms (such as upper earth rheology) that produce regional differences in relative sea-level change. As prefaced in Section 3.4.1, discussions on Holocene sea level on the GBR usually focus on four main issues, which are examined below. (a) Was postglacial sea-level rise smooth or pulsed? Broad agreement now exists that postglacial sea-level rise was neither smooth nor continuous, but included phases of rapid rise associated with melt-water pulse events, and brief periods of relatively slow rise or possibly sea-level stability during cooler episodes like the Younger Dryas (Section 3.4.1). As Harris (1999, p. 235) stated, ‘‘the debate has long since moved beyond this simple view’’ that postglacial sea-level rise must be either smooth or episodic, ‘‘and the more recent literature quite clearly demonstrates that there were indeed episodes of relatively rapid rise interspersed with times of relatively slower rise.’’ However despite agreement that decaying continental ice sheets pulsed meltwater into the oceans and produced erratic postglacial eustatic sea-level rise, research suggesting episodic sea-level rise on the GBR remains contentious. The postglacial sea-level envelope of Thom and Chappell (1975) and the revision by Thom and Roy (1983) are based upon data from south-eastern Australia, but have been widely adopted in discussions of the GBR (e.g., Hopley, 1982, 1983a, b; Davies and Hopley, 1983). Thom and Roy’s (1983) envelope begins at 13–12 ka ago and does not cover the approximately 6000 years immediately following the late glacial, including MWP-1A or other pulse events (Section 3.4.1). A sea-level envelope was proposed rather than a single curve, recognising the limited age and paleodepth precision of many of the data (Thom and Roy, 1983). The sea-level envelope shows an overall continuous rising trend from 13–12 ka ago to near present sea level by 6.5 ka
3.4 Postglacial sea level
73
Figure 3.7 Holocene sea-level curves or envelopes that have been applied in studies of the Great Barrier Reef: (a) Thom and Roy (1983); (b) Fairbridge (1961); (c) Larcombe et al. (1995a).
ago (Fig. 3.7a), but as Thom and Roy note, sea-level rise may have varied within their envelope, and may even have fallen between 9 and 8 ka ago. Despite these caveats, most authors adopting their envelope for interpretations of reef growth have assumed that sea level rose continuously (e.g., Hopley, 1982; Davies and Hopley, 1983). An alternative view that postglacial sea-level rise was episodic was proposed by Carter and Johnson (1986) and Carter et al. (1986), based largely on bathymetric, seismic, and core evidence from the continental shelf shoulder on the central GBR where a series of drowned shorelines were interpreted and linked to episodic sea-level stillstands or minor regressions. Many of these submerged shorelines correlate well with episodic stalls in the postglacial transgression recognized in New Zealand (Gibb, 1986). The episodic postglacial sea-level rise model has, however, been questioned because some of the submarine features used as evidence are not laterally continuous and are thus ambiguous sea-level indicators (Harris and Davies, 1989), and others interpreted as notches have been observed on ROV surveys to be depositional
74
Sea level: a primary control of long-term reef growth
features with limited sea-level constraints. Furthermore, without ages it is impossible to establish whether these features formed during episodic postglacial stillstands or during previous interstadials (Larcombe et al., 1995a). Carter and others have published additional stratigraphic evidence in support of episodic Holocene sea-level rise (e.g., Gagan et al., 1994; Larcombe and Carter, 1998), but this has also been challenged and debate continues (Harris, 1999; Larcombe and Carter, 1999). Larcombe et al. (1995a) published a single Holocene sea-level curve for the central GBR based on 364 radiocarbon dates from a variety of marine and coastal sediments from across the region which they argued presented new evidence for episodic postglacial sea-level rise (Fig. 3.7c). They noted (p. 37) that ‘‘It is noteworthy that even this large and detailed dataset cannot unequivocally define the precise nature of postglacial sea-level rise, particularly at times of rapid sea-level change’’ but nevertheless inferred a complex sea-level curve and envelope that depicts sea level at 11 m at 8.5 ka regressing to 17 m at 8.2 ka before rising rapidly to –5 m by 7.8 ka. A correlation between temporally well-constrained oxygen isotope excursions in ODP core 820 A (see Peerderman and Davies, 1993) and inferred ice-volume-forced sea-level stillstands and regressions was presented in support (but not proof (Johnson et al., 1996)) of episodic sea-level change, but the proposed oscillation between 8.5 and 7.8 ka is unrepresented in the isotope record. Harris (1999) suggested that difficulty in climatically forcing the ice-volume changes needed to produce the proposed sea-level fluctuation in the 700 years available questions the reality of this oscillation, especially as isotope evidence from the Antarctic and Greenland ice-core records does not show coincident ice-volume changes of necessary size. However as discussed earlier (Section 3.3), studies concluding that rapid sea-level fluctuations of comparable amplitude have occurred persistently during the late Pleistocene are increasingly reported in quality journals (e.g., Siddall et al., 2003; Potter et al., 2004; Thompson and Goldstein, 2005). Rapid sea-level oscillations since the mid-Holocene have also been inferred in the Pacific (Nunn, 1998) and Indian Oceans (Morner et al., 2004), ¨ but these are somewhat contentious (Kench et al., 2005a; Woodroffe, 2005). Larcombe et al. (1995a) have made a valuable contribution to progressing sea-level investigations on the GBR by collating the regional dataset, but their interpretation of the data and presentation of a single curve for the region can be criticized on several grounds. They provide a detailed account of the precision with which the sea-level indicators included in their dataset represent a sea-level position, and note that data from coral cores consistently lie below coastal and inner shelf sediments as coral reef growth may lag behind sea level. However, we believe that the 5 m precision inferred for some of the coral core
3.4 Postglacial sea level
75
data used to conclude depths exceeding 5 m occurred within the period 8.5–7.5 ka cannot be easily supported, as the paleodepth of the sample at the time of deposition often cannot be reliably established (see Section 3.4.2a). Pooling data collected by various researchers over several decades from different settings over a broad geographical area (48 of latitude) is also problematic, especially within a region where hydro-isostasy may have vertically displaced the 6 ka shoreline by up to 4 m (Section 3.4.2b) (Chappell et al., 1983; Lambeck and Nakada, 1990). Larcombe et al. (1995a) discuss the effects of hydro-isostasy, but appear to underestimate the resultant difficulty in correlating age and depth across the shelf and between sections affected differently by isostatic adjustments, especially when complexities introduced by the large variation in tidal range through the central GBR are also considered. Finally, the inclusion of several additional radiocarbon dates to the dataset, some from the central GBR and others from outside the region, further questions several of the reported episodic oscillations. For example, the regression from 11 m to 17 m between 8.2 and 7.9 ka is challenged by several dates within this interval from depths between 9.9 and 12.9 m below the reef flat at Boulder, Cockermouth, and Redbill Reefs; and the 2 m stillstand at 6 ka is at odds with abundant, mostly microatoll evidence that modern reef flats had begun to form by then (e.g., Iris Point, Cape Tribulation; Cockermouth, Low Wooded Island) (McLean et al., 1978; Hopley and Barnes, 1985; Partain and Hopley, 1989; Kleypas and Hopley, 1993). Further research to better define Holocene sea level on the GBR from the last glacial maximum to the present is clearly needed. No useful data presently exist to reconstruct sea-level behavior in the GBR region for most of the first 10 000 years of the postglacial transgression, and too few quality data exist to compile a high-resolution record of Holocene sea-level change for most of the region in which high-frequency and/or subtle oscillations can be confidently identified, confirmed, or discounted. (b) When was modern sea level reached and has sea level been higher on the Great Barrier Reef during the Holocene? Two of the main issues regarding Holocene sea-level change on the GBR – when was present level reached and did it go higher? – are inextricably linked and are thus addressed together here. Several earlier reviews of the evidence of when sea level was reached on different parts of the GBR exist (e.g., Flood, 1983a; Hopley 1983a, b; Larcombe et al., 1995a) and should be consulted for detail. As indicated previously in this chapter, the precision with which sea-level indicators can resolve when modern sea level was reached can be limiting, but broad trends are nevertheless discernable. Figure 3.8 shows the
76
Sea level: a primary control of long-term reef growth
Figure 3.8 Oldest radiocarbon ages within 1.5 m of the surface on the Great Barrier Reef. Ages in italics are from a maximum of 1.5 m below the reef surface; ages in plain text are within 1 m of the surface. A square symbol denotes reef core material; a triangle denotes a microatoll which can be confidently interpreted as sea-level constrained. Note the younger ages through the outer central Great Barrier Reef. See Table 3.3 for details.
3.4 Postglacial sea level
77
oldest published radiocarbon ages within 1 m of the reef surface, conservatively representing the time at which Holocene sea level was reached, and the oldest sea-level features dated on the mainland. The data are also presented by latitude in Table 3.3 along with the source, type of evidence, and location on the shelf (inner, mid, outer). The original conclusions drawn from Hopley’s (1982, 1983b) earlier analyses of regional patterns of Holocene relative sea level on the GBR are not altered by the addition of recently published data, with the general trend being that older reef-top ages of between 6.8–5.5 ka dominate most of the inner shelf whereas dated sea-level indicators on the outer reef tend to mostly be several thousand years younger. However, significant differences in the density of data occur. The most data are available for the central GBR between Bowen and Cairns, but few sea-level data exist for large areas of the GBR, including north of Cape Melville, and over substantial areas of the southern GBR such as the Swain Reefs and Pompeys. The data at a regional level are briefly summarized below. Torres Strait to Cairns The evidence from this region, where the shelf is relatively narrow, suggests a mid-Holocene highstand 1 m above present was reached before 5.5 ka on the inner shelf (Bird, 1970; Maxwell, 1973a; Hopley, 1978a; Chappell et al., 1982; Hopley and Partain, 1987; Woodroffe et al., 2000), although McLean et al. (1978) believe that dated shallow core materials from Bewick Island and dated microatolls from Fisher Island (6310 90 years BP) and Low Wooded Island (6080 90 years BP) suggest it was achieved earlier. No emergence is known on the outer half of the shelf, but radiocarbon dates suggest that the outer reefs did not lag far behind those closer inshore. Many radiocarbon dates between 5 and 4 ka exist for this area, and ages between 5190 120 and 5940 190 years BP within 1 m of the reef flat are reported from the shelf edge at Yonge, Carter, and Ribbon 5 Reefs (Hopley, 1977, 1994; Davies et al., 1985). Cairns to the Palm Islands Between Cairns and the Palm Islands the shelf widens but the story is very similar to that of the northern GBR, with dated microatoll sequences across fringing reef flats on inshore high islands suggesting maximum Holocene sea level was reached on the inner shelf between 6.2 and 5.5 ka and reefs formed at this level have emerged by around 1–1.5 m since (Chappell et al., 1982; Graham, 1993). Earlier estimates of greater emergence based on beachrock and boulder shorelines (Hopley, 1971) have been revised following confirmation that cementation should conservatively be considered to occur up to HAT level and not MHWS, and these are now compatible with the microatoll evidence (Hopley, 1983a). The sea-level envelope of Grindrod
5770 70 6340 80 4740 130 6310 90 4980 80 5110 105 5660 205 5800 87 5730 96 5850 170 5800 110 5800 130 5220 78 4870 70 4960 80 4750 69 3700 90 4740 120 5910 110 6080 90 3750 110 4520 110
108 130 1428 480 <1
<1 <1 <1 <1
148 100 1448 150 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1
148 300 148 310 148 310 148 330 148 330 148 330 148 340 148 360 148 390 148 390 158 030 158 050 158 060
158 070 1458 250 <1 158 070 1458 250 <1
Warraber Is.
Hammond Is. 108 350 1428 140 <1
118 360 128 160 138 550 148 060
Yam Is.
Raine Is. Fisher Rf. Stainer Rf. King Is.
Flinders Is.
Noble Is. Houghton Is. Houghton Is. Carter Rf. Leggatt Is. Leggatt Is. Hampton Is. Yonge Rf. Nymph Is. Nymph Is. Long Rf. Lark Pass Rf. Low Wooded Is. Three Is.-3 Three Is.-3
1448 360 1448 580 1448 580 1458 360 1448 400 1448 400 1448 520 1458 370 1458 150 1458 150 1458 340 1458 440 1458 230
1448 030 1438 120 1438 500 1448 200
6340 80
1428 450 <1
98 500
Location
ANU-1380 GaK-7667
ANU-9189 ANU-9210 ANU-1287 GaK-6480 ANU-1286 ANU-9183 ANU-1207 Beta 40806 ANU-8959 ANU-1285 GaK-6483 GaK-6683 ANU-1604
ANU-2319
ANU-6623 ANU-1640 ANU-1639 ANU-2323
Wk-7769
Wk-7774
Wk-7760
Age error Laboratory (yr) code
Depth below Lat. S Long. E MLWSa Reef location
M C
M M M C M M M C M M C C M
M
C M M M
M
M
M
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
Polach et al. (1978) Hopley (1982)
Woodroffe et al. (2000) Woodroffe et al. (2000) Woodroffe et al. (2000) Hopley (1994) Polach et al. (1978) Polach et al. (1978) Chappell et al. (1983) Chappell et al. (1983) Zwartz (1995) Zwartz (1995) Polach et al. (1978) Hopley (1977) Polach et al. (1978) Zwartz (1995) Polach et al. (1978) Hopley (1994) Zwartz (1995) Polach et al. (1978) Hopley (1977) Hopley (1982) Polach et al. (1978)
Materialb Inner Mid-shelf Outer Reference
Table 3.3 Published radiocarbon ages from the GBR within 1.5 m of the reef surface
4140 130 4130 110 4475 125 5100 130 2680 120 1930 110 1460 110 3800 110 3530 60 5130 140 3760 100 5295 100 3993 45 3230 140 5020 160 5855 145
168 130 1458 530 <1.5 168 340 1458 310 <1 168 340 1458 310
168 350 1468 020 <1
168 480 1468 120 <1
168 560 178 140 178 420 178 430 178 500 178 530 178 580
178 580 1468 080 <1
178 580 1468 360 <1 188 030 1468 540 1
188 110 1468 100 <1
Michaelmas Rf.-1 Thetford Rf.-1 Channel Rf.-1 Hedley Rf.-1 Potter Rf. Ellison Rf. Taylor Rf. 17065 Rf. Dunk Is.
Lugger Bay
Moss Rf. Barnett Shoals Goold Is.
1468 270 1468 280 1468 320 1468 240 1468 330 1468 440 1468 100 <1 <1.5 <1 <1 <1 <1 <1
4110 130 2570 100 5390 60 6860 60 5690 96 6150 70
<1 <1 <1 <1.5 <1 <1
158 390 158 450 168 010 168 020 168 050 168 050
Cairns Rf. East Hope Is. Emmagen Rf. Rykers Rf. Myall Rf. South Myall Rf. Opal Rf.-3 Yule Point Yule Point
1458 330 1458 260 1458 290 1458 290 1458 300 1458 300
5910 190
158 250 1458 260 <1
Williamson Rf. Boulder Rf.
6460 110
5940 80
158 220 1458 350 <1.5
Ribbon No 5 158 220 1458 470 <1
EWG-1
GaK-8934 GaK-8933
Wk-16745
GaK-7683 GaK-7686 ARL-235 Wk-1396 GaK-11082 Wk-1393 EWD-1
GaK-7681
GaK-7675
GaK-7674 GaK-2729 EWY-3
GaK-6686 GaK-6687 2567 2570 ANU-9214 2564
CSIRO 234
CSIRO 227
ANU-3516
M
C C
C
C C C C C C M
C
C
C C M
C C C C M C
C
C
C
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
Chappell et al. (1983)
Hopley (1982) Hopley (1982) Graham (1993) Graham (1993) Graham (1993) Graham (1993) Chappell et al. (1983) Perry and Smithers (2006) Hopley (1982) Hopley (1982)
Hopley (1982)
Hopley (1982) Bird (1971a) Chappell et al. (1983) Hopley (1982)
Davies and Hopley (1983) Davies and Hopley (1983) Davies and Hopley (1983) Hopley (1982) Hopley (1982) Partain (1988) Partain (1988) Zwartz (1995) Partain (1988)
5910 110 5490 100 1920 80 4550 90 3610 60 4280 100 5530 130 1200 45
1468 290 1468 290 1478 260 1468 310
188 360 188 360 188 380 188 400
188 410 1468 310 <1.5
188 420 1468 360 <1
188 450 1478 160 <1 188 480 1478 310 <1.5 188 500 1488 090 <1
198 020 1468 380 <1 198 020 1468 300 <1
198 050 1468 340 <1
Fantome Is.
Great Palm Is. Keeper Rf. Wheeler Rf. Viper Rf.-2
Herald Is. Rattlesnake Is. Paluma Shoals
<1 <1 <1 <1
5970 100 5830 87 4100 80 5520 100
5170 69 6260 120 5390 78 5340 78
1468 290 1468 290 1468 290 1468 290
188 360 188 360 188 360 188 360
Hazard Bay Iris Point Iris Point Little Pioneer Bay Pioneer Bay Pioneer Bay Grub Rf. Fantome Is. <1 <1 <1 <1
3640 60
188 280 1478 330 <1
Myrmidon Rf.-1 Bowl Rf.-3
3890 50
5255 65
Wk-10800
GaK-2014 GaK-7688
GaK-7272 SUA-1528 Wk-438
GP/3
Beta 5709
SUA-1679 ANU_7901 SUA-1907 ANU-2464
ANU-8298 ARL-201 ANU-6953 ANU-8003
Wk-591
Wk-427
Beta 1055
Age error Laboratory (yr) code
188 150 1478 230 <1
Britomart Rf. 188 140 1468 440 <1.5
Location
Depth below Lat. S Long. E MLWSa
Table 3.3 (cont.)
C
BR/C M
C C C
M
C
C M C M
M C M M
C
C
C
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
Smithers and Larcombe (2003)
Partain (1988) Zwartz (1995) Hopley (1983a) Chappell et al. (1983) Johnson and Risk (1987) Chappell et al. (1983) Hopley (1982) Hopley (1983a) P. Davies pers. comm. Hopley (1971) Hopley (1982)
Johnson et al. (1984) P. Davies pers. comm. P. Davies pers. comm. Zwartz (1995) Barnes (1984) Zwartz (1995) Zwartz (1995)
Materialb Inner Mid-shelf Outer Reference
Reef location
5990 80 4810 45 1310 100 4670 80 6160 120 5760 80 3040 80 4570 150 6400 80 970 80 5930 100 5730 70 5390 180 7480 130 6560 70
208 030 1488 530 <1.5 208 040 1488 290 <1
208 380 1508 480 <1 208 450 1518 000 <1 208 470 1498 230 <1
208 470 208 520 208 580 218 010 218 42 218 570 238 200
238 300 1528 040 <1
Hayman Is. Hydeaway Bay Molar Rf. Cockatoo Rf. Cockermouth Is. Gable Rf. Scawfell Is. Redbill Is. Penrith Is. Percy Is. High Peak Is. Wreck Rf. A
One Tree Is.
Fitzroy Rf.-A 238 380 1528 090 <1
Fairfax Rf.-A 238 520 1528 220 <1.5
1508 300 1498 360 1508 050 1498 530 1508 160 1508 410 1518 580
Wk-303
Wk-331
Wk-253
SUA-1774 Beta-33213 GaK-7820 Beta-33202 Beta-33208 Beta-41067 Wk-369
GaK-7268 SUA-2114 ANU-6631
Beta-83809 Wk-16738
GaK-4894 EWS/1
GaK-7025 GaK-8931 GaK-4736 EWC/1
EWM/4
C
C
C
C C C C C C C
C C C
C C
M M
C C C M
M
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
Kleypas (1991) Kleypas (1991) Hopley (1984) Kleypas (1991) Kleypas (1991) Kleypas (1991) P. Davies pers. comm. P. Davies pers. comm. P. Davies pers. comm. P. Davies pers. comm.
Chappell et al. (1983) Hopley (1982) Hopley (1982) Hopley (1975) Chappell et al. (1983) Hopley (1975) Chappell et al. (1983) Kan et al. (1997) S. Smithers (unpub. data) Hopley (1982) Kleypas (1991) Kleypas (1991)
It was assumed from observations of contemporary reefs that most reef flat surfaces are close to MLWS, and so this depth class is estimated relative to this datum. b M, microatoll where explicitly stated; C, coral material, not specifically identified as microatoll; BR, beachrock.
a
5290 120 5925 115
198 590 1488 220 <1 208 020 1488 160 <1
Middle Is. Stone Is.
<1 <1.5 <1.5 <1.5 <1 <1 <1
6070 130 6420 160 5980 120 5820 115
<1 <1 <1 <1
198 120 198 160 198 440 198 520
Darley Rf. Stanley Rf. Holbourne Is. Camp Is.
1488 150 1488 060 1488 220 1478 530
5325 80
198 090 1468 520 <1
Magnetic Is.
82
Sea level: a primary control of long-term reef growth
and Rhodes (1984), established from Hinchinbrook Island mangrove deposits, suggests that sea level on the inner shelf in this region was achieved between 7 and 6 ka but provides no detail on the nature of a subsequent highstand. Further offshore, near-surface dates older than 5 ka are recorded from Britomart and Taylor Reefs (Johnson et al., 1986), but no evidence of Holocene emergence occurs on the outer reef (Hopley, 1983b). Graham (1993) reported young (3 ka) reef-top ages on inner, mid, and outer shelf reefs off Innisfail, but concluded that the younger inner dates were derived from younger sections of reef that had prograded out from older reef sections that reached sea level thousands of years earlier. Palm Islands to the Whitsundays Sea level in this region was the focus of vigorous debate through the 1970s and 1980s (Gill and Hopley, 1972; Belperio, 1979a; Hopley, 1980), but it is now generally agreed that it was reached by 6 ka. Microatolls across fringing reef flats indicate maximum sea level was reached at 6–5 ka on the inner shelf and was followed by emergence of around 1–1.5 m, a conclusion also supported by the level of intertidal cementation on offshore islands such as Rattlesnake, Herald, Holbourne, and Middle Islands (Hopley, 1975, 1982). Comparison of modern and fossil oysters on Magnetic Island suggest a mid-Holocene highstand up to 1.65 m above present was reached by around 5.6 ka and persisted for 1500 years (Beaman et al., 1994), although the oyster evidence does not discount the possibility that the highstand was reached earlier. A relatively large number of sea-level data are available for this region, but questions remain regarding the exact timing and amplitude of the mid-Holocene highstand due to uncertainties about the completeness of the record covering the early part of this period and the precision of some of the data (Harvey et al., 2001). On the outer reef in this region there are no emergent features, and reef-top ages are generally less than 3 ka (Hopley, 1982, 1983b). Whitsundays to Bundaberg Inshore data from this region are scarce, but a few dates are available from reef flat corals and shallow reef cores that suggest that relative sea level was reached prior to 6 ka and was followed by emergence. At the northern end of the region a date of 5990 80 from <1 m depth on Hayman Island suggests sea level was attained by this time, and similar ages were reported from within 1 m of the reef surface on various reefs drilled in the Cumberland Islands by Kleypas and Hopley (1993) (Fig. 3.8, Table 3.3). Although no evidence of higher sea level has been reported from the Whitsunday Islands, elevated mangrove deposits in the Torilla Plains area on the eastern edge of Broad Sound suggest emergence of 4–6 m since the
3.4 Postglacial sea level
83
mid-Holocene (Cook and Polach, 1973). These raised mangrove deposits have been linked to tectonic uplift to the east of the Broad Sound fault (Cook and Mayo, 1977); chenier plain sediments indicate that sea level on the western side of Broad Sound was reached by 5 ka and has varied by <1 m since (Cook and Polach, 1973). Interestingly, corals growing within 1 m of present MLWS at Redbill Reef, more than 80 km off the mainland coast, are dated at 6550 150 years BP, and ages around 4 ka from beneath the algal rim on the same reef are reported and interpreted as possibly indicating 1.2–1.6 m of emergence. The reef flat at nearby Penrith Island also appears to have reached sea level quite early, by around 5.6 ka (Kleypas and Hopley, 1993). As far as we are aware, the oldest reef-top age on the Pompeys, further to the east, is 4670 80 years BP. Kleypas and Hopley (1993) suggested that tectonic movement along structural lineaments in the southern GBR may have influenced the time at which reefs reached sea level, and the distribution of reefs that preserve evidence of higher mid-Holocene sea level in this area. As indicated above, the differential emergence of coastal deposits on either side of Broad Sound have also been ascribed to tectonics. In the Capricorn Bunker Group, at the southern end of the GBR, reef-top dates between 6.4 and 5.4 ka from Wreck, One Tree, Fitzroy, and Fairfax Reefs suggest sea level there was reached between 7 and 6 ka. Modern sea level on the Great Barrier Reef: a discussion As summarized in Hopley (1983a), significant progress in resolving debates about the reality and possible reasons for geographic differences in Holocene sea-level records was achieved in the late 1960s through the early 1980s when various workers revisited Daly’s earlier ideas on isostasy and further refined the concept of hydro-isostatic adjustment on shallow continental shelves. Early modeling by Walcott (1972) depicted how ocean basins deformed as ocean volumes changed through glacial cycles, with later modeling of global viscoelastic response to changing ice loads and ocean volumes accounting for the major variations in sea-level records determined for different locations (Clarke et al., 1978). This modeling predicted that shorelines on the GBR would emerge by around 1.5–2 m soon after melt-water discharge ceased, and possibly rise even higher due to ‘‘tilting’’ of the flooded shelf. Widespread evidence of emergence close to the mainland and its absence offshore was noted early on by various researchers working on the northern GBR (e.g., Hedley, 1925a; Spender, 1930; Fairbridge, 1950), a pattern Thom and Chappell (1975) argued supported claims of hydro-isostatic adjustment of the shelf to water load following the postglacial transgression. Chappell et al. (1982, 1983) later extended this work to develop a hydro-isostatic model to account for regional differences in the
84
Sea level: a primary control of long-term reef growth
Figure 3.9 Hydro-isostasy and relative sea level on the Great Barrier Reef. (a) Predicted sea level (m) at 6 ka; (b) time at which sea level reached present represented as zones; (c) characteristic sea-level curves for the zones represented in (b). Note that actual relative sea-level variations vary latitudinally as shelf morphology varies (after Nakada and Lambeck, 1989).
timing and elevation of late Holocene sea levels over the northern and central GBR. This model exploited the fact that the physiography of the continental shelf and coastline are spatially variable (Fig. 3.3a), and thus loading by the rising postglacial sea and any hydro-isostatic response would also vary regionally. A theoretical isobase map of hydro-isostatic warping for north Queensland 5.5 ka (Chappell et al.’s estimate of when highest sea level was achieved) was produced using coral and sedimentary field data from a transect extending from Britomart Reef to the Gulf of Carpentaria (Chappell et al., 1982) (Fig. 3.9). Geographic variations in reef-top age were examined by Hopley (1982) to assess the impacts of shelf warping on relative sea level over the GBR, and reveal very similar results to the hydro-isostatic models that demonstrate variations in sea-level history along and across the GBR (see also Hopley, 1983b). The outputs of Chappell et al. and Hopley vary slightly in
3.4 Postglacial sea level
85
some areas, most particularly on the outer shelf off Townsville where regional tectonics may have augmented isostatic downwarping, but both show emergence on the inner shelf and stability or slight submergence on the outer shelf, with both trends more pronounced where the shelf is broad (Fig. 3.9). Hydro-isostatic adjustment influences the pattern of relative sea-level change experienced at a given location, with uplifted areas achieving modern sea level earlier than downwarped areas. Geophysical modeling by Nakada and Lambeck (1989) suggests four different sea-level curves for offshore zones roughly parallel to the mainland coast (Fig. 3.9). Using the area offshore of Townsville as an example, the inner zone (I), which extends to around 65 km offshore, reached present sea level prior to 6 ka before continuing higher. Zone II, which extends from around 65–140 km offshore, reached present sea level around 5.5 ka, zone III which extends from 140 to 200 km offshore of Townsville, reached present sea level around 4 ka, and zone IV, the outer zone off the contintental shelf reached present sea level recently. These model outputs are well supported by reef-top ages (Fig. 3.8, Table 3.3), but it should be noted that they cannot yet precisely constrain sea-level history due to uncertainties related to the sea-level indicators from which the data are derived and ambiguity regarding the details of deglaciation. For example, models that assume melt-water discharge from northern hemisphere ice sheets ceased by 6.5 ka generate an inshore sea-level curve for north Queensland that exhibits a pronounced highstand after 6.8 ka followed by rapid fall to present levels (Nakada and Lambeck, 1989; Lambeck and Nakada, 1990) but those that include Antarctic melting until 4 ka (Goodwin, 1998) produce a sea-level curve with a more rounded and longer highstand after 6 ka (Peltier, 2002). Furthermore, it is important to note that reef-top ages clearly indicate that most midshelf and outer shelf reefs reached sea level after any highstand, and therefore they provide no evidence of its occurrence. Acceptance of hydro-isostasy was a critical advance for sea-level researchers as it provided a geophysical framework that accounted for geographical variations in sea level along and across the GBR, and confirmed that different parts of the GBR have developed under the influence of regionally different sea-level histories. Modern sea level was reached at different times at different places, and where emergence has occurred, the amount of emergence may differ due to hydro-isostatic influence. However, as foreshadowed above, although hydro-isostatic models yield sensible results in areas where data are available, most data used to verify these trends are relatively imprecise, coverage of the GBR is incomplete, and the accuracy of modeled outputs could be greatly improved. The collection of more precise sea-level data from a much broader geographical range is required, and is presently being pursued. Even
86
Sea level: a primary control of long-term reef growth
in the central GBR near Townsville, where the density of sea-level data is relatively high, fundamental questions such as when modern sea level was reached and the elevation of any highstand are still not adequately resolved. Indeed, although the influence of hydro-isostasy is widely accepted, the nature of its influence remains controversial, with north–south latitudinal variations in deformation predicted by models (Nakada and Lambeck, 1989) argued to be ‘‘limited or non-existent’’ (Haworth et al., 2002, p. 581). As indicated by these authors, the continental shelf south of the GBR is relatively narrow (mostly <30 km) and thus as others have pointed out it is unlikely that hydroisostatic warping would be as significant as observed in areas where the shelf is wider (Chappell et al., 1982, 1983; Hopley, 1982, 1983b). Data best constrain the hydro-isostatic models on the central GBR and north to around Cape Melville, with fewer data available for the southern end of the reef where the shelf significantly broadens. Flood (1983a) reviewed the data available at that time and the dataset has not been significantly augmented since. Of greatest significance to the Holocene sea-level history for the area are data from Redbill, Penrith, and Lady Elliot Island Reefs, all of which are more than 80 km offshore and suggest a mid-Holocene highstand up to 1.6 m above present (Flood, 1983a; Hopley, 1983b). If the interpretation of these data is correct, then it is likely that the highstand nearer the mainland would have been even higher, but no published data exist to test this hypothesis. (c) When and how did sea level fall from the highstand to present? The timing and pattern of post mid-Holocene highstand sea-level fall is equivocal, with spatial variations occurring over the GBR due to isostatic and possibly other factors. In an early analysis of evidence from the inner shelf between Townsville and Princess Charlotte Bay, Chappell (1983) suggested that sequences of fossil microatolls across reef flats indicated that a highstand around 1 m above present was reached around 6 ka and was followed by a linear or ‘‘smooth’’ regression to present sea level. Others have suggested that the highstand peaked soon after 6 ka but then persisted for longer. Beaman et al. (1994) argued that radiocarbon dates from fossil oysters elevated 1.65 m above modern oyster beds provided evidence that the mid-Holocene highstand continued from at least 5.2 ka until at least 3.7 ka, the end date in good accord with another dated encrustation from northern New South Wales (Flood and Frankel, 1989), although there the possibility has been raised that the highstand may have been maintained until less than 1.8 ka. A recent re-examination of the dataset used by Chappell (1983) to support a smooth fall of sea level following the mid-Holocene highstand further questions both the timing of the mid-Holocene peak and the nature of the
3.5 Historical sea-level change on the Great Barrier Reef
87
regression. Although they focused mainly on the New South Wales coast, Baker and Haworth (2000a, b) demonstrated that a stepped or rapidly oscillating regressive sea-level pattern has equal statistical validity to the smoothly falling model, and noted that their polynomial analysis identifies a peak around 4 ka. Baker and Haworth linked late-Holocene sea-level oscillations to sea surface temperature fluctuations, evidence for both of which they argued existed elsewhere in the southern hemisphere (Baker et al., 2001b). If late Holocene sea level did oscillate synchronously over a wide geographic area as Baker and coworkers assert, an eustatic rather than isostatic explanation would seem more likely, and thus the orthodoxy that late Holocene sea levels are dominated by isostatic influence may need to be re-evaluated. The detail of late Holocene sea level on the GBR is at present insufficiently known to determine with confidence whether late Holocene sea-level fall has been smooth or oscillatory. This is largely because data from appropriate locations are scarce, and because many of the existing data have insufficient depth and chronological precisions to resolve meaningfully any high-resolution oscillations. 3.5 Historical sea-level change on the Great Barrier Reef Four tide gauges – at Cairns, Townsville, Mackay, and Bundaberg – have operated on the mainland coast inside the GBR for 25 years or more. The sealevel trends at each gauge vary, with Townsville increasing at þ1.12 mm a1 and Mackay at þ1.24 mm a1, but slight falls recorded at Cairns (0.02 mm a1) and Bundaberg (0.03 mm a1) (Mitchell et al., 2000) (Fig. 3.10). The trends calculated for Townsville and Mackay differ markedly from those calculated using a shorter dataset covering 1950–59, which indicated falls averaging 0.49 mm a1 and 0.16 mm a1 respectively (Colquhoun, 1979). Shorter records collected by the Australian National Tidal Facility at Cape Ferguson near Townsville (beginning September 1991) and Rosslyn Bay near Yeppoon (beginning June 1992) depict sea-level trends of þ3.1 mm a1 and þ2.5 mm a1 (ABSLMP October 2005), which also diverge from the trends established from the longer records. The differences in trend established for each location when using records of different length reflect the influence of factors that may affect relative sea level, such as natural climate variations (El Nino–Southern Oscillation (ENSO), ˜ Pacific Decadal Oscillation (PDO)), possibly longer-term climate change (enhanced greenhouse effect), and as discussed earlier in this chapter, vertical land movements. As an example, ENSO events have an average return period of four to seven years, and usually last 12–18 months. During neutral conditions (neither El Nino prevailing easterly trade winds blow across ˜ or La Nina), ˜
88
Sea level: a primary control of long-term reef growth
Figure 3.10 (a) Map of Great Barrier Reef region showing sea-level trend for 1993–2003 detected by TOPEX/Poseidon and Jason-1 satellites: zone 1, rising trend of 4–6 mm; zone 2, 4–2 mm; zone 3, 0 mm (Colorado Center for Astrodynamics Research, 2006). (b) Sea-level trends at the four tide gauges on the Great Barrier Reef coast with records longer than 25 years (Mitchell et al., 2000).
3.5 Historical sea-level change on the Great Barrier Reef
89
the South Pacific and raise the surface of the Western Pacific about 0.5 m above that in the Eastern Pacific near Ecuador, and a strong sea-level signal associated with ENSO is recognized across the Pacific (Church et al., 2004). The PDO is a fluctuation in the Pacific Ocean that manifests similarly to ENSO cycles, but has a longer frequency of 20–30 years (Trenberth and Hurrell, 1994). During negative PDO phases, such as the one active since about 2004, sea surface temperature and height in the western Pacific show an increasing trend. To derive a meaningful sea-level trend from gauge data, the influence of these factors must be understood and accounted for, and this is best achieved as sea-level records become longer. Douglas (2001) concluded that at least 70 years’ data are required to recognize meaningful sea-level trends; no records of this length exist for the GBR region. Knowledge of historical sea-level change on the GBR is limited by the lack of long-term tidal records and offshore data. Satellite altimetry data are available from 1993 and document recent sea-level behavior in this region. The TOPEX/Poseidon and later Jason-1 satellites identified regional sea-level trends over the GBR for the period 1993–2003, with variation observed from north to south. From Torres Strait to just south of Mackay a rising trend of 4–6 mm a1 was detected, with lower rates of rise grading to stability and possibly even a slight sea-level fall detected for the southern GBR (Colorado Center for Astrodynamics Research, 2006). The sea-level trends determined from north to south along the GBR extend across the shelf to include outer reef areas (Fig. 3.10). Sea-level trends for the Pacific determined from satellite data closely match tide-gauge records, and satellite altimeter and tide-gauge data for the GBR region also accord reasonably for the common period. Several studies recently concluded that the best estimate of global averaged eustatic sea-level rise between 1950 and 2000 is 1.8 0.3 mm a1 (Church, 2001; Church et al., 2004) but higher rates close to 3 mm a1 are estimated from many coastal tide gauges and the satellite altimetry data for the common period 1993–2000 (Church et al., 2004; White et al., 2005). As discussed, the higher rates of sea-level rise since 1993 detected at many coastal gauges probably reflect decadal variability driven by factors including ENSO and PDO cycles. The effect of volcanic eruptions on ocean heat and sea-level rise has also been demonstrated recently (Church et al., 2005), and may have contributed to the apparent acceleration of sea-level rise since the early 1990s. Church et al. (2004) argued that accelerating rates of global sea-level rise could not be substantiated once natural variability on interannual to decadal timescales was filtered; however, recent work based on global sealevel reconstructions extending back to 1870 suggests a significant acceleration over the twentieth century (Church and White, 2006).
90
Sea level: a primary control of long-term reef growth
The reasons for spatial differences in relative sea-level signal established using both gauge and satellite data over the GBR remain uncertain. Although geographic differences in the tidal data derived trends may reflect variable vertical movement of land on which a gauge is located due to compaction, subsidence, or other processes, satellite altimetry measures the absolute elevation of the sea surface and thus trends derived from these data should be independent of changing land elevation. Mechanisms to explain spatial differences in the rate of change in the sea surface are many and complex but possibly include the interaction of currents and regional variations in thermosteric and halosteric effects; sea surface temperature and sea surface height distributions derived from satellite data are closely correlated (e.g., Levermann et al., 2005). 3.6 Conclusion Sea level and sea-level change through the late Quaternary have significantly influenced the development of the GBR. Since ‘‘true reef growth’’ began on the GBR just 450 ka ago, reef growth over most of the GBR has occurred episodically when higher sea levels inundated the continental shelf. However, the length of time that the entire shelf has been submerged is very limited, totaling perhaps 40–50 ka. For long periods the continental shelf was largely exposed, and reef growth was restricted to the shelf edge and slope (see Chapter 9). Significantly, coral reefs of similar taxonomic composition and physical character have re-established during successive highstands, usually over older reefal foundations. Sea level during the last glacial cycle is known best and demonstrates the relatively fleeting persistence of interglacial highs, and the dominance of lower sea levels that restrict reef growth to the shelf edge. During these relatively rare and brief highstands geographic differences in the shelf’s isostatic response to water load has produced spatially varying relative sea-level histories. As discussed in later chapters, these differences have important implications for the pattern and nature of postglacial reef growth over the GBR. Over much of the inner shelf, emergence since 6 ka has produced reefs of very different geomorphological character to those where emergence has not occurred, such as on the outer central GBR. These differences are explored in detail in Chapters 7–9. Significant and repeated sea-level change clearly dominates the period over which the GBR has formed. The modern GBR has grown through periods of rapid sea-level rise and environmental change during the early postglacial transgression, and many parts of the inner GBR have experienced relative sealevel fall over the last few thousand years. Reefs with different Holocene relative
3.6 Conclusion
91
sea-level histories have different geomorphologies, but detailed Holocene sealevel histories are yet to be established for large areas of the GBR. Instrumental records of sea-level change are too short to reveal meaningful long-term trends, but also clearly show that recent interannual sea-level trends vary spatially within the Reef, and that the maximum rate of recent sea-level rise is around 20 times slower than rapid rates experienced during phases of the postglacial transgression. However, sea-level behavior is but one of a whole range of modern oceanographic processes which affect the Great Barrier Reef and which are discussed in the next chapter.
4 Oceanography, hydrodynamics, climate, and water quality as influences on reef geomorphological processes
4.1 Introduction Reef morphology is the product of contemporary physical, biological, and chemical processes, acting on inherited surfaces and frameworks, but morphodynamic feedback exists whereby simultaneously, reef shape and the configuration of reef complexes significantly affect the way the basic driving forces such as waves, tides, long period oscillations, and freshwater plumes operate. Although descriptions of the hydrodynamics and oceanography of the Great Barrier Reef (GBR) were well established in the literature in the 1970s and early 1980s, particularly with the work of Wolanski and Pickard (Pickard et al., 1977; Wolanski, 1983; Wolanski and Bennett, 1983; Wolanski and Pickard, 1985), there had been little consideration of the physical processes affecting reef geomorphology of the GBR (Parnell, 1988b), except at the individual reef scale (e.g., Davies and West, 1981; Frith, 1983a). In the last two decades there has been considerable research building on the understanding of basic driving forces and processes (excellently summarized by Wolanski (1994), in the book Physical Oceanographic Processes of the Great Barrier Reef), particularly with respect to the links between physical and biological processes (Wolanski, 2001). Another significant theme, driven to a large extent by management agencies, has been the effect of terrestrial runoff on reefs (Furnas, 2003), although most consideration has been given to coral assemblages rather than to reefs over larger time and space scales. However, biological researchers have now recognized that small-scale experiments provide limited insights into phenomena that operate at larger space and longer timescales (Hughes et al., 1999). Increasingly, attention has been on the understanding of the effects of climate change on the GBR. With respect to oceanographic processes, Wolanski (1994) divides the reef into four regions (Fig. 4.1). The regions are: (a) Torres Strait, a largely shallow marine environment, mostly less than 15 m deep. The northern extension of 92
4.1 Introduction
Figure 4.1 Oceanographic regions of the GBR. Climate data are presented for locations labeled (*) in Table 4.1.
93
94
Oceanography, hydrodynamics, climate, and water quality
the GBR, mainly outside the Great Barrier Reef Marine Park, is bordered by the Warrior Reefs to the west and an area of high reef density to the east, separated by the Great North East Channel which opens to the Gulf of Papua. Torres Strait is well known for strong tidal currents flowing in relatively narrow channels. (b) The Northern region extends south to 168 S, with water depths generally less than 30 m. Reefs and shoals occur right across the narrow shelf, with elongate reefs separated by narrow channels on the shelf edge. Waters to the north and to the south of Barrow Point (148 220 S) maintain little connection due to reefs being densely packed across the shelf, allowing little water transfer. (c) The Central region, extending from 168 S to about 208 S, has a generally low density of reefs, with very few reefs on the inner shelf. Although generally less than 40 m water depth, in places the shelf slopes gently to about 100 m at the shelf break. This region is most affected by inflow from large rivers. (d) The Southern region, south of 208 S, is generally deeper with water depths reaching 140 m. On the extensive middle to outer shelf, reef density is generally high, with the inner shelf being largely reef free. Many studies (Maxwell, 1968; Orme and Flood, 1980; Belperio, 1983a, b; Johnson, 1996) have identified three shore parallel belts of sediment across the GBR shelf: (a) the inner shelf, to about 20 m water depth, characterized by mainly terrestrial sediments (<30% carbonate); (b) the mid-shelf, with water depths 20 m to 40 m with mixed sediments (30–80% carbonate); and (c) the outer shelf with >80% carbonate sediments in water depths greater than 40 m. This chapter gives a brief overview of the climate and oceanography of the GBR region, emphasizing geomorphic relationships. Cyclones and tsunami that have significant impacts on reef and shoreline geomorphology are discussed. The geomorphic effects of terrestrial runoff are considered. 4.2 The climate of the Great Barrier Reef region The most noticeable climatic influences in the GBR region are the persistent south-easterly trade winds and the marked seasonality of climate. The winter dry season is associated with anticyclones crossing Australia at about 308 S, causing a ridge of high pressure along the coast to the north, and resulting in south-easterly winds that rarely exceed 30 knots (15 m s1). In summer, lowpressure systems normally dominate (Fig. 4.2a), resulting in generally northerly quarter winds up to about 15 knots (7.5 m s1), although weak ridges causing periods of south-easterly winds may persist. Sea breezes occur at the coast and across the inner shelf causing a 458 anticlockwise rotation in wind direction during the middle of the day (Wolanski, 1994). From December to
4.2 The climate of the Great Barrier Reef region
Figure 4.2 (a) Monthly mean sea-level atmospheric pressure and (b) monthly mean sea surface temperature 1950–97 by latitude; (c) Monthly Southern Oscillation Index 1950–2005. ((a) and (b) Copyright # 2001. From Oceanographic Processes of Coral Reefs: Physical and Biological Links in the Great Barrier Reef by Lough, J. (Wolanski, E., ed.). Reproduced by permission of Routledge/Taylor and Francis Group, LLC.)
95
96
Oceanography, hydrodynamics, climate, and water quality
April, moist air masses move over northern Australia, influenced by the southward shift of the Inter Tropical Convergence Zone (ITCZ). Rainfall, temperature, and evaporation data for a range of stations shown in Fig. 4.1 are presented in Table 4.1. Cloudiness and rainfall is significantly affected by orographic effects, with the highest rainfall on and adjacent to the coastal ranges between Daintree and Ingham. Coastal Cape York receives mean annual rainfall totals between 1100 and 2100 mm. South of Ingham, rainfall decreases, with coastal annual totals generally in the order of 1000 mm to 1200 mm, although there are pockets of higher rainfall (e.g., around Mackay). Rainfall is both highly seasonal (with most rainfall occurring between December and March) and highly variable. Rainfall is greatest when the summer monsoon circulation is strong, with activity usually occurring in bursts associated with 30–60 day Madden–Julian circulation (Hendon et al., 1989; Hendon and Liebmann, 1990; Lough, 2001). Averages are not good predictors of rainfall. Townsville, for example, has an annual mean rainfall of 1100 mm, but with an annual rainfall 90th percentile of 1785 mm and a 10th percentile of 587 mm. Intense rainfall events contribute significantly to high totals. Table 4.1 shows 24-hour rainfall totals at coastal locations up to 800 mm (at Port Douglas), and these events have the potential to stress coral reefs, particularly if they coincide with low spring tides. Rainfall only exceeds potential evaporation in the wettest parts of the region, between Daintree and Ingham. Related to extremes in many of the climatic variables is the passage of cyclones. Tropical cyclones may occur during the months November to May in areas south of about 108 S. The GBR experienced a mean of 2.8 cyclones per year, or a total of 80, over the period 1969–97 (Puotinen et al., 1997), most of which were category 1 and 2 (winds 17–33 m s1 and central pressure 970–1000 hPa), with only about 10% being category 3 (winds >33 m s1 and central pressure less than 970 hPa) and above. In the period 1997–2005, there have been 12 cyclones that have impacted the GBR (average of 1.5 per year). Of these, four were well to the east of the GBR and did not pass through the GBR lagoon or cross the coast, with two (Joint Tropical Cyclone Warning Centre) or three (Australian Bureau of Meteorology) being category 3 or above (Fig. 4.3). The highest concentration of cyclones has occurred in the central GBR region (Puotinen et al., 1997). The majority of cyclones originate in the Coral Sea, tracking in a generally southerly direction. Many do not cross the coast, ending as rain depressions in the Coral Sea. A smaller number originate in the Gulf of Carpentaria, crossing Cape York and regenerating in the Coral Sea. For those cyclones forming in the Coral Sea that do cross the coast, the strongest winds are in the forward left quadrant. The impact of cyclones on reef geomorphology is discussed in Section 4.5.
1888 1950 1887 1956 1887 1942 1921 1887 1884 1941 1962 1881 1920 1980 1968 1940 1886 1987 1870 1934 1959 1870 1934 1956 1899 1872 1957 1939 1885 1942
2004 1993 1955 2004 2004 2004 2004 2004 2004 2004 2000 2004 2004 2004 2004 2004 2004 2004 1987 2004 2004 1987 1987 2004 1987 1958 2004 2004 1990 2004
Source: Bureau of Meteorology.
Thursday Is. (a) Thursday Is. (b) Cape York Lockhart River Coen Post Cooktown Willis Island Low Isles Port Douglas Cairns Fitzroy Island Innisfail S. Johnstone Lucinda Ingham Townsville Ayr Bowen (a) Bowen (b) Hayman Island Mackay (a) Mackay (b) Pine Islet Heron Island Cape Capricorn Gladstone (a) Gladstone (b) Lady Elliot Island Bundaberg (a) Bundaberg (b)
Station 30.8 29.9 29.8 31.6 31.6 32.0 30.7 32.2 30.3 31.4 30.6 30.8 31.1 30.2 32.3 31.4 32.1 31.7 31.5 30.5 30.0 30.4 30.3 29.7 27.7 29.9 31.1 29.2 30.3 29.9
Mean maximum temperature First Last record record Jan (year) (year) (8C) 25.1 24.8 24.0 23.6 22.6 24.1 25.7 25.6 23.7 23.6 24.5 22.7 22.4 25.4 23.1 24.2 22.3 23.8 24.0 24.8 23.4 23.2 24.3 24.1 36.1 22.2 22.4 23.9 21.3 21.3
Mean minimum temperature Jan (8C) 28.3 27.7 27.6 27.0 27.5 26.3 26.0 25.6 24.6 25.7 23.5 24.1 23.8 30.2 32.3 25.0 25.2 24.6 24.3 23.1 21.2 22.0 21.3 21.4 19.6 22.2 22.6 21.0 22.0 21.9
Mean maximum temperature July (8C)
Table 4.1. Climate data from coastal and island weather stations
22.7 22.5 21.5 19.3 16.6 17.9 21.9 20.4 16.8 17.0 19.0 15.1 15.0 25.4 23.1 13.5 11.5 13.6 14.2 16.6 12.8 11.5 17.3 16.5 25.0 11.4 13.2 16.6 9.9 10.0
Mean minimum temperature July (8C) 1717 1746 1749 2140 1174 1665 1117 2088 2013 2003 2680 3559 3307 959 2047 1122 1058 865 1010 1338 1567 1665 873 1047 798 1020 918 1141 1141 1011
2124 2213 2313 3133 1664 2385 1593 2992 2752 2792 3627 4777 4659 1327 3227 1785 1689 1721 1672 2058 2451 2415 1144 1429 1093 1395 1231 1572 1688 1500
1230 1280 1218 1398 793 1117 581 1367 1268 1266 1774 2510 2161 334 1133 587 493 380 461 762 947 1075 696 771 457 619 539 741 692 633
Mean annual Decile 9 Decile 1 rainfall rainfall rainfall (mm) (mm) (mm) 238 453 279 462 393 326 336 470 801 403 517 539 441 174 489 549 478 327 392 402 389 627 282 282 258 478 229 454 430 258
2057 1486 1305 2396
6.2 4.6 4.4 7.2
1573 1572
4.8 5.3
2034
2021 1506 4.7
Highest Mean daily Annual daily evaporation evaporation rain (mm) (mm) (mm)
98
Oceanography, hydrodynamics, climate, and water quality
Figure 4.3 Tropical cyclones in the vicinity of the GBR from 1997–98 season to 2005–06 season. The tracks of cyclones of category 3 and above are drawn in a heavier line. Data are sourced from the Australian Bureau of Meteorology and the Joint Tropical Cyclone Warning Centre. Maps and analysis of cyclone tracks from 1969 to 1997 can be found in Puotinen et al. (1997).
Monthly mean sea surface temperatures (SSTs) range from a summer maximum of more than 29 8C north of 148 S to about 21 8C at 248 S during winter (Fig. 4.2b). In summer SSTs greater than 28 8C extend south to 228 S, with minimum temperatures north of 108 S being about 25 8C. Anomalies in SSTs
4.3 Oceanography and hydrodynamics
99
with respect to long-term monthly means range 2 8C, with anomalies persisting over significant periods of time due to the thermal inertia of the sea (Lough, 1994). From data collected at Myrmidon Reef, Lough (2001) showed the annual range in SST to be 4.8 8C, but the difference between warmest daily and coolest daily temperatures to be 9.5 8C, with a diurnal range of about 1 8C. Reef organisms therefore experience considerably higher temperature variability than is indicated by monthly average variation. The climate of the GBR region is strongly influenced by El Nino–Southern ˜ Oscillation events (Lough, 1994), as recorded in the Southern Oscillation Index (SOI) (Fig. 4.2c). ENSO (El Nino) events occur when the eastern ˜ equatorial Pacific is unusually warm, with anti-ENSO (La Nina) ˜ events occurring when it is unusually cold. Events evolve over a period of 12–18 months, and once evolved tend to follow similar patterns. A strong El Nino ˜ event occurred in 1997–98, with SSTs 1–2 8C above average in the late summer on the GBR, and was associated with a coral bleaching event. Two years later, at the same time of year, SSTs were 1–2 8C below average. ENSO events are associated with weak monsoon circulation, fewer tropical cyclones, and lower rainfall. Anti-ENSO events are associated with increased monsoonal activity, and higher rainfall. Although rainfall (and consequently runoff) is the most strongly associated ENSO-linked variable, Lough (1994) found correlation with SST and wind fields. Burrage et al. (1995) found correlation between long-term sea-level data from Townsville and ENSO cycles. A number of studies have investigated the relationship between El Nino ˜ events and cyclone behavior, suggesting that cyclones form and remain further east during El Nino ˜ conditions, with more cyclonic activity near the coast during anti-ENSO years (Revell and Goulter, 1986; Dong, 1988; Nicholls, 1992; Lough, 1994). Puotinen et al. (1997) found a correlation between a declining SOI index and fewer cyclones tracking near the Queensland coast, and between an increasing SOI index and more cyclones tracking near coast.
4.3 Oceanography and hydrodynamics 4.3.1 Tides, low-frequency motions, and water circulation Based on the tidal form factor F (Foreman, 1977) the tides of the GBR region are classified as mixed with diurnal and semidiurnal components except in the vicinity of Broad Sound, where they are semidiurnal (Fig. 4.4a). The M2 (lunar semidiurnal) and K1 (lunisolar diurnal) constituents dominate (Wolanski, 1994) (Fig. 4.4b, c). There is significant latitudinal and cross-shelf variation. Tides are controlled by two amphidromic nodes, near New Zealand and in the
100
Oceanography, hydrodynamics, climate, and water quality
Figure 4.4 Tidal characteristics of the GBR region by latitude. (a) Form factor (Foreman, 1977) F < 0.25, semidiurnal; F > 3.0, diurnal; 0.25 < F < 3.0, mixed; (b) M2 and (c) K1 tidal constituents; (d) tidal range. (Copyright # 1994. From Physical Oceanographic Processes of the Great Barrier Reef by Wolanski, E. Reproduced by permission of Routledge/Taylor and Francis Group, LLC.)
Coral Sea, with water propagating through the Capricorn Channel and through the area of low reef density in the central section. The tidal waves are restricted by dense offshore reefs (in the northern section and the Swains) (Middleton et al., 1984) and are amplified as they cross the shelf, converging near Broad Sound, where tidal ranges exceed 8 m. North of Townsville, tidal range is typically about 2.5–3 m, but increases significantly north of 128 S (Figs. 4.4d and 4.5) from 3 m to more than 3.6 m in Torres Strait. The spring tidal range at the shelf edge is typically about 3 m. A significant driving force is the East Australian Current (EAC) that generally flows south in the Coral Sea adjacent to the GBR, but which flows north (as the Hiri Current) from a seasonally varying bifurcation point somewhere between 148 and 208 S (Church, 1987; Brinkman et al., 2001). To the south of
4.3 Oceanography and hydrodynamics
101
Figure 4.5 Tidal range (Indian Springs range calculated as two times the sum of the amplitudes for constituents M2, S2, K1, and O1) from tidal model data supplied by Luciano Mason (Australian Maritime College, Tasmania).
the bifurcation point, currents strengthen (Dight et al., 1990a). In the northern section, the near-continuous reef edge inhibits intrusion of EAC into the GBR lagoon. The importance of this current is noted by Brinkman et al. (2001) indicating that it can assist in GBR lagoon flushing, spread larvae,
102
Oceanography, hydrodynamics, climate, and water quality
assist nutrient-rich upwelled water to penetrate the GBR, and inhibit river plumes spreading onto the outer shelf. Also operating at the shelf edge are large internal tides. Wolanski and Pickard (1983), in studies at Myrmidon Reef, showed vertical excursions of the thermocline as great as 110 m. Internal tides at the shelf edge have the ability to move cold, nutrient-rich water onto the shelf and into the GBR lagoon (Furnas and Mitchell, 1996). There have been few attempts to measure mean circulation directly, although satellite imagery has made the identification of some large-scale features possible using remote-sensing techniques (Burrage et al., 1996), particularly when combined with other long-term field measurements. A comprehensive drift card experiment involving 16 releases at 13 locations along the length of the GBR (Collins and Walker, 1985) found highest correlation with wind speed and direction (although only 4.9% of cards were found). Indirect evidence was used to infer a deep southerly directed current that would dominate in the absence of wind. This conclusion is supported by evidence presented by Wolanski and Pickard (1985), showing a long-term 0.2 m s1 current near Green Island, probably due to the presence of the southwardflowing EAC. Numerical modeling by King and Wolanski (1992) indicated dominance of the southward-moving current on the mid to outer shelf, but with a wind-driven northward-moving current dominant on the inner shelf and mainland shoreline. Time-series data of currents from a range of studies involving deployed instruments (Andrews, 1983; Wolanski and Bennett, 1983; Middleton et al., 1984; Wolanski and Pickard, 1985; Griffin et al., 1987; Wolanski et al., 1989; Wolanski and King, 1990) show significant correlation with wind speed and direction, the result of wind-driven low-frequency barotropic shelf-trapped waves, propagating north. These low-frequency motions, with periods of days to many months (but with most energy in the 13–40-day period), are associated with sea-level oscillations up to 0.35 m (Wolanski, 1994). A long-term dataset from Green Island showed speeds of up to about 50 cm s1 associated with low-frequency oscillations (Wolanski and Pickard, 1985). Measured currents are correlated with tidal range and the intensity of lowfrequency circulation, but are significantly affected by topography. The tidal component of current is generally cross-shelf and small (Church et al., 1985), except in the vicinity of Broad Sound where currents are larger with a more dominant longshore component. It is only in areas devoid of reefs that low frequency currents (encompassing tidal currents and circulation derived from low-frequency shelf waves) can be accurately resolved, and with respect to the geomorphology, generalization is of little value. Most studies have
4.3 Oceanography and hydrodynamics
103
found significant variability over small distances (e.g., Wolanski et al., 1989), sometimes quite at variance with tidal models (Andrews and Bode, 1988), with King and Wolanski (1996) showing that the variation is partly due to tidally induced residual currents that result from the interaction of the tide and the complex reef topography. These residual currents can be as significant as other non-tidal currents in some areas. Consequently, most studies of the relationship between water movement and the movement of sediments and biological dispersal have necessarily been at the small spatial scale. Close to the mainland coast, primarily affecting fringing reefs on the mainland and offshore high islands, consistent reversing tidal streams dominate (Hamner and Hauri, 1977; Parnell, 1987, 1988a, b), indicating dominance over low-frequency wind-driven circulation. 4.3.2 Topographically induced circulation Modification of regional currents by topographic features such as reefs, channels, headlands, bedforms, and pronounced depth changes (such as the shelf edge and reef front) is significant and geomorphologically important. Such flows can be very complex, and are most significant with respect to sedimentation and the movement of small organisms. In the case of the huge Halimeda banks of the northern GBR (Section 6.6), tidal jets delivering nutrients from deep water control their location. Reef density effects Reef density varies significantly over the GBR. In areas of low reef density, the reefs are essentially invisible to regional currents. In areas of high reef density, however, the intrusion of the EAC and influence of low-frequency shelf wave currents is significantly lessened, particularly on spring tides when water movement is more significantly disrupted at low tide. On spring tides, tidal and low-frequency currents are steered away from the reefs, but this occurs to a lesser extent on neap tides. This has biological implications, with trapping of plankton in areas of high reef density (particularly on spring tides), whereas in areas of lower reef density, regional circulation carries them away. Counterintuitively, in such waters periods of high regional currents can result in larvae being in contact with reefs for longer periods. Wolanski and Spagnol (2000) termed these conditions ‘‘sticky waters.’’ Tidal currents in areas of high reef density, however, can be extreme. In Torres Strait, speeds greater than 3 m s1 may occur, and in the very narrow channels between reefs of the Southern region, speeds greater than 4 m s1 have been recorded (Hopley, 1982; Chapter 9).
104
Oceanography, hydrodynamics, climate, and water quality
Tidal jets, oceanic nutrients, and Halimeda banks Restriction of flow through narrow channels causes high water velocities. The shelf-edge ribbon reefs of the northern region are a very good example of this, with significant geomorphic effects (Fig. 4.6). Strong currents in the narrow (1–2 km wide) channels between the long (2–8 km) and narrow, shelf-aligned reefs, induce upwelling of deep ocean water from beyond the shelf edge (brought towards the surface by internal tides), by Bernoulli-type processes operating at tidal frequency over periods of spring tides. The upwelled water is then carried through the passage (tidal jet), with little water flowing across the reefs themselves (Young et al., 1994). The cold nutrient-rich water remains near the bottom, to be dispersed in vortices generated by the tidal jet. Water
Figure 4.6 Schematic diagram of tidal jet and the eddying of nutrient-rich waters as found as found in the GBR lagoon adjacent to the ribbon reefs in the northern region, and the relationship of eddies to the location of Halimeda banks
4.3 Oceanography and hydrodynamics
105
can be held in the vortices, extending up to 10 km across the shelf (Young et al., 1994), for a considerable period of time, providing areas of nutrient-rich water, near the lagoon floor. Eddies also form on the seaward side of the reefs on the reversing tide, but are considerably weaker than those on the landward side (Young et al., 1994). Secondary circulation caused by friction in the shallow (30–40 m deep channel) results in a slick of planktonic matter in the center of the channel in the otherwise nutrient-depleted water (Wolanski and Hamner, 1988; Wolanski et al., 1988). The highly enriched waters are conducive to the growth of the calcareous alga Halimeda and over time large Halimeda banks have grown. These are significant inter-reefal features of the GBR, the geomorphology of which is explored fully in Chapter 6. Drew (2001) noted that not all entrances are associated with Halimeda banks, and speculated that for Bernoulli processes to transport nutrient-rich water effectively, channels need to be narrow (<1 km) and neither too deep or too shallow (40–45 m deep at the outer sill being ideal), associated with reefs at least 4 km long. Drew also observed that there are areas with apparently ideal conditions not associated with Halimeda banks (such as around Princess Charlotte Bay), but speculates that this may be the result of factors unrelated to nutrient availability such as water turbidity. Circulation around reefs and islands The modification of regional circulation around reefs has long been recognized. Circulation around reefs involves highly complex, three-dimensional water movements. Reefs present a barrier to water flow, at times forcing all water to move around them, and when not exposed, allowing some water to flow over the top. In the lee of individual reefs complex three-dimensional eddies form, and these can result in local upwelling. The hydraulics of eddying in the wake of regular shapes is well understood (usefully summarized in the context of reef environments by Wolanski (1994)), but becomes complex for irregular reefs and islands, and particularly for shallow submerged features where water moves both around and over the feature. Early studies in the GBR (Hamner and Hauri, 1977, 1981; Wolanski and Thomson, 1984) demonstrated the presence of eddy circulation, and this was quantified and modeled in later studies around Rattray Island in the northern Whitsunday Islands (Wolanski et al., 1984; Falconer and Mardapitta-Hadjipandeli, 1986; Falconer et al., 1986; Wolanski, 1986a), Myrmidon Reef (Wolanski, 1986b), and in various fringing reef situations (Parnell, 1987, 1988a, b). Around Rattray Island (Fig. 4.7a, b), where tidal streams dominate circulation, eddy characteristics were similar for all tidal phases (Falconer et al., 1986), and eddy evolution and decay occurred quickly following reversals of the tidal stream (Wolanski et al.,
106
Oceanography, hydrodynamics, climate, and water quality
Figure 4.7 Eddy circulation in the lee of an island or reef: (a) plan view and (b) side view, barotropic case (Wolanski, 1986a); (a) plan view and (c) side view, baroclinic case (Wolanski, 1986b); (d) plan view and (e) side view, bayhead fringing reef (Parnell, 1987). After Parnell (1988b).
1984). The downstream extent of the eddy was approximately the same as the island width, and there was considerable upwelling in the eddy (Wolanski, 1986a). Eddies were linked to the creation of banks and shoals, and the sorting of sediments with all movable particles advected towards the center of the eddy where coarser particles remained, with finer particles being moved to the eddy extremities. The baroclinic case was modeled in studies on Myrmidon Reef (Fig. 4.7c) by Wolanski (1986b), where the downstream influence extended to more than twice the reef width. Eddies were found to dominate circulation on embayed fringing reefs (Fig. 4.7d, e) (Parnell, 1987, 1988a). The radius of the eddy was dependent on the incident angle between the headland shedding the eddy and the dominant current, the degree of indentation of the bay, and the morphology of the reef front. The radius of the eddy was greatest where the incident angle was highest, and was strongest in situations where the bay was enclosed by headlands at both ends, with water being deflected against the opposing shore. Eddy circulation was established quickly following the establishment of a tidal
4.3 Oceanography and hydrodynamics
107
stream and velocities within the eddy were independent of tidal stream velocity. Wolanski (1993) investigated the characteristics of a narrow free shear layer (an area inhibiting movement of water across it) comprising numerous small eddies, in a similar situation at Bench Point in the Whitsunday Islands (Fig. 4.7e). Drogues released on the landward side of the shear layer remained almost stationary. Current measurements indicated that the shear layer extended to the seabed. Such lines are often visible downstream of headlands. Circulation around non-fringing reefs of the GBR is similar to that established for places like Rattray Island, with the presence of eddies and shear layers. However, in areas of medium to high reef density, the reefs will not be hydrodynamically isolated. Wolanski (1994) points out the difficultly this poses for numerical modeling where spatially uniform boundary conditions are generally assumed. The importance of eddies to the dispersion of planktonic larvae has been noted by Hamner and Hauri (1981) and Oliver and Willis (1987), with fronts and slicks being associated with eddies behind reef structures (Kingsford et al., 1991). In cases where eddies are driven by low-frequency currents rather than reversing tidal currents, particles can be held within them for considerable periods of time. Free shear layers can concentrate particulate organic matter, sufficient to make the water within them more turbid (Wolanski, 1993). Most of the studies of circulation around and between coral reefs have used numerical modeling in the context of larval dispersal (Black et al., 1990, 1991; Dight et al., 1990b; Wolanski and King, 1990; Black and Moran, 1991; Oliver et al., 1992; Scandol and James, 1992). Wolanski et al. (1989) found that in calm weather larvae aggregated along topographically controlled fronts and stayed in the lagoon or in the boundary layer around reefs, but under higherenergy conditions the bulk of the larvae were advected away from the home reef. Modeling studies by Black and Moran (1991) noted that currents moving larvae off their originating reef take them around, and not to, adjacent reefs, a finding significant in relation to recruitment and the importance of selfseeding. However, Wolanski (1993) warned of problems with the techniques employed by Black et al. (1990) and Black and Moran (1991), and noted that the findings had not been validated by extensive hydrodynamic data. This concern was reinforced by Oliver et al. (1992), who found ‘‘virtually no correlation’’ between modeled and observed larval concentrations and recruitment around Bowden Reef. They found the most obvious shortcoming was the failure of the models to predict extreme patchiness, due to fine-scale threedimensional flows being ignored in depth-averaged models. The Helix Reef experiment (Sammarco and Andrews, 1988, 1989) showed that reefs were primarily self-seeding and that small-scale circulation leading to localized
108
Oceanography, hydrodynamics, climate, and water quality
high residence times and accumulations of larvae was most important, but that there was sufficient far-field dispersal to ensure gene flow between reefs. In an interesting extension to the recruitment debate, Sammarco et al. (1991) argued that reefs are asymmetric with respect to prevailing south-easterly trade winds, waves, and currents, and that reef shape is determined by seasonal currents at the time of coral spawning, with preferential recruitment (and extension) occurring on the low-energy parts of the reef at those times. They argue that these concepts can be used in the interpretation of paleocurrents. Although an appealing concept, the importance of the morphology of older foundations needs to be recognized. The foundations of modern reefs are shaped in many ways over long periods of time, including subaerial (in particular karst) processes. 4.3.3 Circulation on and within reefs Circulation in reef lagoons was first studied in the large open atoll lagoons of the Pacific (Munk et al., 1949; von Arx, 1954; Atkinson et al., 1981), with wind-driven surface circulation dominating and return flow at depth. Average residence times were in the order of one month. Reef lagoons in the GBR are smaller, and have a larger tidal range, with lagoons ranging from being completely enclosed to being open except on the windward edge. In enclosed lagoons, ponding may occur, reducing the tidal range in comparison to the open water. There have been numerous studies of lagoon circulation on the GBR, including on Boulder Reef (Davies and Hughes, 1983), Davies Reef (Frith, 1983b; Andrews et al., 1984; Pickard, 1986) Britomart Reef (Wolanski and Jones, 1980; Wolanski and King, 1990) and One Tree Reef (Ludington, 1979; Davies and West, 1981; Frith, 1982, 1983a; Wilson, 1985; Frith and Mason, 1986). Lagoonal currents were generally less than 0.2 m s1 except in openings in the reef crest. Flushing times were between 12 hours and six days. In One Tree Reef lagoon, wind- and wave-generated currents dominated tide. When ponding occurs, an overturning wind-driven circulation dominates during the extended low water. In the much more open Davies Reef, wind-driven circulation and tide-driven circulation dominated at different times depending on the tidal range and wind strength. On Britomart Reef early studies indicated that tidal flows dominated, but Symonds et al. (1995) found wave-generated currents to be significant. Black et al. (1990) modeled the residence time of neutrally buoyant material such as larvae on reefs, concluding that on a ‘‘typical’’ reef 45% of particles will remain on the reef ten days after release.
4.3 Oceanography and hydrodynamics
109
On large reefs higher percentages are expected, and on smaller reefs most will be lost in the first one to three days. On bayhead-enclosed fringing reefs circulation is dominated by the tidal stream and its interaction with the local morphology (Parnell, 1987). Velocities over the reef flats are generally less than 0.2 m s1. Fringing reef bays with welldeveloped eddy circulation are well flushed. In Pioneer Bay, Orpheus Island, for example, Parnell (1987) found the exchange ratio (the water replaced over a tidal cycle) to be up to 90%. Under normal energy conditions, water velocities in reef lagoons and on lee fringing reefs are generally insufficient to entrain sand size sediment, with major transport and reworking of sediment occurring during cyclone and other high-energy events (Davies and Hughes, 1983; Frith, 1983a; Parnell, 1988a). The distribution of new sediment on the lagoon floor following an extreme energy event within a lagoon can promote some sediment transport under moderate conditions for a period of time (Frith, 1983a). However, it is clear that some sediment does move under generally low-energy conditions. Bioturbation, particularly by Callianassa shrimps, initiating transport by ejecting sediment into the water column, is very important in moving sediment through reef systems (Tudhope and Scoffin, 1984; Parnell, 1987). De Vaugelas (1985) found that in a lagoon environment Callianassa could cause the complete turnover of the top 30 cm of sediment in eight months. Tudhope and Scoffin (1984) emphasized the importance of Callianassa in the sediment sorting process, leading to a 5–60 cm thick layer of gravel-free sediment overlying a poorly sorted gravel-rich sediment, in reef lagoons. At a smaller scale, patch reefs within a lagoon modify the current field in their vicinity, causing areas of locally reduced and increased velocities, promoting deposition or erosion (Hamner and Hauri, 1981). Wolanski and Jones (1980) reported eddy formation, edge effects, convergences, and divergences around individual coral heads, leading to periods of water stagnation of up to 30 minutes. Small-scale fluid mechanics are known to be important in the ecology of coral heads (Wolanski and Jones, 1980), and Nakamura and van Woesik (2001) speculate that significant water flow past a coral head can reduce the chance of coral bleaching compared to corals in stagnant water or low flow. Coral reefs are normally considered as static barriers with water moving over and around them. Studies of water movements within the reef framework have normally been considered with respect to the management of groundwater in the Ghyben–Herzberg lens beneath cays. The dynamics of groundwater flow in the Ghyben–Herzberg lens is generally well understood
110
Oceanography, hydrodynamics, climate, and water quality
Figure 4.8 The Ghyben–Herzberg lens of fresh water is truncated at the Pleistocene unconformity due to the highly karstified nature of the Pleistocene sediments.
(Wheatcraft and Buddemeier, 1981; Falkland, 1993), but as demonstrated by Herman et al. (1986), Buddemeier and Oberdorfer (1990), and Oberdorfer et al. (1990), where the Ghyben–Herzberg lens intersects the Pleistocene unconformity, it is truncated due to significant flows of salt water through the highly karstified framework (Fig. 4.8). Despite the potential significance of flow through reef framework on nutrients, contaminants, and reef diagenesis, there has been little research on the GBR. Oberdorfer and Buddemeier (1986) on Davies Reef and Parnell (1987) on a fringing reef flat on Orpheus Island addressed the issue in two quite different environments. On Davies Reef, horizontal velocities ranged from 0.2 to 400 m d1, with calculated Darcian pore velocity of 3.2 m d1. Vertical movement was suggested as being upward on the rising tide and downward on the falling tide. On the fringing reef flat, horizontal velocities of 40 m d1 were calculated based on first arrival time of fluorescent tracer between boreholes, probably indicating flow through higher-permeability zones rather than bulk flow. Downward flow was significantly greater than upward flow, and there was little communication between surface waters and subsurface waters. Diffusive spreading of water within the framework occurred when the reef flat was covered, but a small hydraulic gradient encouraged seaward advection when the reef flat was exposed. In summary, water flow through substrate is fast, but is largely confined to high-permeability zones, such as unconsolidated gravels, cracks, and voids.
4.4 High-frequency waves
111
4.4 High-frequency waves 4.4.1 Wave forcing There are a number of wave types that impact the GBR with frequencies higher than those of the tides. Forcing mechanisms are swell waves impacting the GBR from the Coral Sea and wind waves generated under non-cyclonic conditions within the GBR lagoon, cyclone-generated waves, and very infrequent tsunami. Cyclones and tsunami are discussed in Section 4.5. The Queensland Environmental Protection Agency (EPA) maintains a network of wave recorders along the Queensland Coast. Data are available on the Internet, and since 2000 an annual summary, covering all sites, and including details of significant meteorological events, has been prepared. Summary data from Cairns, Townsville, Mackay, the Capricorn Coast, and North Stradbroke Island (near Brisbane) are presented in Fig. 4.9. There are no recording sites north of Cairns. Although wave models (e.g., WAM and Wavewatch III) and data from satellite observations (TOPEX/Poseidon, Jason, ERS-2) are now available, there has been no systematic analysis of waves reaching the GBR from the Coral Sea. None of the EPA-operated gauges are located on the eastern side of the GBR although gauges located on the coast to the south (Mooloolaba and on the seaward side of North Stradbroke Island) have application particularly to the southern region of the GBR. Wolanski (1986c) measured waves in 100 m water depth off Myrmidon Reef for three months during the dominant trade wind season in 1980. There was considerable variation in wave heights, peaking at 4.2 m. Following shifts in wind direction, both short- and long-period waves responded simultaneously, indicating waves were generated within the Coral Sea. Spectral analysis indicated a peak in wave period at about 10 s, but under lighter wind conditions a secondary peak at about 7 s was observed. Wave data for the 2003–04 year collected from offshore North Stradbroke Island (278 300 S) showed maximum Hsig of about 7 m, and a 90th percentile Hsig of 2 m. Waves from the south-east dominated the record. Peak spectral periods (Tp) had a modal range between 7 and 9 seconds. These data are consistent with the complete dataset from this site (1976–2004). The GBR provides an effective barrier to waves entering from the Coral Sea. Wave records at coastal stations are dominated by locally generated waves with a period of 3–4 s, but many records have small secondary peaks at 8–9 s, evidence of leakage of some longer-period energy through the GBR matrix. For some wind directions, particularly the south-east, the GBR lagoon offers
112
Oceanography, hydrodynamics, climate, and water quality
Figure 4.9 Wave data from five wave recording stations operated by the Queensland Environmental Protection Agency. ( a) Significant wave height; (b) peak spectral period (s) 1 November 2003–31 October 2004 (histogram), and all data (line); (c) wave direction and height, 1 November 2003–31 October 2004 (Cairns recorder is non-directional). The wave buoys on the Capricorn Coast, at Mackay, and at Townsville are fetch-limited by reefs or mainland coast in some directions from which significant waves may come.
4.4 High-frequency waves
113
long fetches, and many of the coastal gauges show energy in the 5–7 s range generated in an alongshore direction within the GBR lagoon. Although there are no data, in the Northern region, with high reef density and a narrow continental shelf, normal wave periods are unlikely to exceed 4 s. Wave heights within the lagoon under normal conditions are generally low. For wave recording sites between Cairns and the Capricorn Coast, 90th percentile Hsig heights range from 1 to 2 m with the highest heights being for sites with significant fetch. Although there is a reasonable understanding of wave characteristics for sites outside the GBR and near the mainland shore, data inside but near the reef matrix are almost non-existent. Hardy et al. (2000) argue that understanding is best achieved with numerical modeling and have adapted the WAM wave model for use in this environment, particularly with respect to cyclonic waves. This model is further considered in Section 4.5.1. 4.4.2 Wave interaction with reefs The influence of waves on reef morphology has long been recognized, but as is the case with beaches, reef morphology affects processes – a morphodynamic feedback loop. Significant differences in reef shape (including profile) over relatively short distances exist, often dependent on whether the reef is subjected to generally high or low wave energy. For example, high-energy reef fronts frequently have spur-and-groove systems which are generally aligned in the direction of dominant waves (Hopley, 1982). The precise effect of spur-and-groove systems on wave hydraulics and breaking has not been studied, but Munk and Sargent (1954) and Gourlay (1996a) suggest they can be expected to have significant effects on the wave-breaking process, by increasing wave attenuation efficiency. On most reefs the windward edge is normally immediately obvious, with characteristic zonation (Hopley, 1982; see also Fig. 9.5). GBR shelf reefs characteristically have very steep windward edges, with waves having little significant interaction with the seabed only a few hundred meters from the reef front. For example, a 10 s period wave will start interacting with the seabed in water depths of about 75 m, with orbital velocities at the seabed increasing to about 0.3 m s1 (sufficient to entrain some sand-sized sediment) in water depths of 30 m. Shoaling is rapid and the surf zone is narrow, with most energy expended over a relatively short distance at either the reef edge or at the reef crest (Fig. 4.10), the zone of dissipation being characteristically a hard coralline algal pavement. Most energy (60–95%) is lost during the wave shoaling, breaking, and reforming process (Roberts et al., 1975; Lee
114
Oceanography, hydrodynamics, climate, and water quality
Figure 4.10 Wave setup on a coral reef (after Gourlay, 1996a).
and Black, 1979; Roberts, 1981; Nelson and Lesleighter, 1985; Young, 1989; Hardy et al., 1991a; Gourlay, 1994) but unlike beaches with low slope, most of the energy is dissipated by wave breaking rather than bottom friction (Massel, 1992). The importance of energy dissipation to nutrient uptake by reef organisms is discussed by Hearn et al. (2001), suggesting it is a paramount reason why coral reefs can maintain high productivity in low-nutrient tropical waters. A relatively small proportion of the wave energy reaches the back reef or lagoon. However, tidal range is an important controlling factor (Hardy et al., 1991a, b). At times of low tide, little or no wave energy crosses the reef flat. At mid tide, most waves break, energy is dissipated, and some waves re-form as bores, passing over the reef flat as solitary waves, eventually re-forming in the deeper lagoon. At high tide, waves may break and re-form or, under lowenergy conditions or very high tidal range, may pass over the reef flat into the back reef or lagoon. Kench and Brander (2006) reported 74% attenuation of wave energy across an exposed reef flat at Lady Elliot Island at high tide. On a lee reef flat, where the smaller waves pass over the reef without breaking, no energy was lost. A parameter used to determine wave breaking is the wave-limiting parameter g ¼ H/h, where H is wave height and h is still water depth. At the reef edge g ranges from 0.7 to 1.1, with the value depending on the nature of wave breaking, with larger waves breaking by plunging and smaller waves by spilling (Gourlay, 1994). However, Gourlay (1994) and Nelson (1994) showed that g never exceeds 0.55 for re-formed waves on the reef flat. Where H0/h < 0.4, waves pass over the reef without breaking (where H0 is deep-water wave height). There is a broadening of the wave energy spectrum following
4.4 High-frequency waves
115
wave breaking (Young, 1989; Hardy and Young, 1991) with energy being distributed across a range of frequencies. Wave setup, an increase in water level over the reef crest, is caused by radiation stress (the excess flux of momentum) (Longuet-Higgins and Stewart, 1964) which results from the wave-breaking process. It is of prime importance in driving water across the reef crest and reef flat, allowing wave energy to enter the back reef area and contributing to back reef and lagoon water circulation. Gourlay (1996a, b) usefully summarizes the effects of wave setup on reefs: (a) wave setup increases with increasing wave height and period (up to a point after which it remains constant); (b) wave setup is larger on lower tide levels than at high tide; (c) in low tidal range locations setup-generated water levels at low tide may exceed normal high tide water levels; (d) wave setup does not occur if the water depth is sufficient for the waves to pass over the reef without breaking; (e) wave setup is maximized in closed lagoon situations; (f) wave-generated net flow also increases as wave height and period increase, but unlike setup, it is higher at high tidal levels; (g) the shape of the reef profile affects the amount of energy dissipated at the reef rim, and therefore the amount of wave setup; (h) offshore wave groupiness (surf beat) significantly affects setup with the amount of setup for regular waves of a certain height being greater than for irregular waves with the same average height; (i) setup is unaffected by reef width. In locations subject to relatively consistent moderate to high wave energy, setup can result in the reef surface on the windward edge being elevated over that which would exist under low wave energy conditions. Secondary setup may occur at cay shorelines or on fringing reef beaches, and this has recently been recorded in experiments on Lady Elliot Island (P. Kench, pers. comm.). The flows associated with wave setup have also been extensively studied. Gourlay (1993) demonstrated the importance of reef morphology (basically roughness) on reef flat current. Hearn and Parker (1988) and Symonds et al. (1995) showed reef flat and lagoonal currents were proportional to off-reef wave height (after tidal effects had been removed). With respect to wavegenerated flow velocity across the reef, Gourlay and Colleter (2005) summarize as follows. The relative submergence of the reef top is defined as S ¼ d/H0 where H0 is the off-reef wave height and d is the water depth over the reef top. Wave setup is maximum when S ¼ 0 (a situation analogous to a beach), and decreases to zero when S ¼ 2.5, as the waves cease to break on the reef. The wave-generated flow across the reef increases from zero when S ¼ 0 to a maximum when S 1.75, decreasing to zero when S ¼ 2.75. There are two wave-generated flow regimes: (a) reef top control where the pressure gradient across the reef,
116
Oceanography, hydrodynamics, climate, and water quality
generated by the wave setup, and bottom friction determine velocity (high values of S), and (b) reef rim control, where critical flow at the downstream reef rim (that is where the flow re-enters the back reef) determines reef top flow velocities at low values of S. Maximum velocities occur at the transition between these regimes. Although modeling studies of wave setup on reefs now abound, field verification remains limited. Gourlay and Colleter (2005) stress that all the work that has been done to date on modeling setup-generated flows are limited by the assumptions made with respect to reef morphology and surface roughness (Massel and Brinkman, 2001). The most significant of these has been the assumption of a horizontal reef surface. Some advances have been made, for example by allowing for variation in offshore reef profile in the models (Massel and Gourlay, 2000); however, considerable work remains to be done. Most research into wave processes on coral reefs has taken the ocean-tolagoon transect approach. The effects of longshore components of radiation stress have not been researched, despite the fact that longshore currents have been shown to exist (Symonds et al., 1995). Brander et al. (2004) demonstrated that transect approaches do not adequately describe wave transformations that take place over the reef flat. Refraction and diffraction at the reef crest, on the reef margins, and across the reef platform are significant processes in the determination of sediment distribution, and in the formation and location of sand cays and other reef top sediment deposits. Simple models developed by Gourlay (1988) showed that stable lee cay formation is associated with oval or elliptical reefs with the long axis aligned with the dominant wind, and that windward shingle cays were the result of a zone of wave interference (Fig. 4.11). Gourlay (1988) suggested that cays on more circular reefs are likely to be much more unstable, due to the change in the wave focal zone with relatively small shifts in wave direction. On wide reef flats, high-frequency wind waves can be generated. The frequency and direction can be quite different from waves that have traveled from the reef edge. Brander et al. (2004) and Kench and Brander (2006) show wave energy occurring in a range of spatially and temporally varying frequency bands across reef flats, with internally generated wind waves dominating at lower tide levels. An understanding of all wave frequencies present is important with respect to sediment transport. Wave-generated water velocity at the bed is a function of wave height and period, and water depth, and significant oscillatory motion can be generated with small high-frequency waves in shallow water. Wave-generated oscillatory motion generates high bed shear stress and is very efficient at mobilizing sediment, to be transported by other unidirectional currents that exist at the time. Kench (1998) demonstrated that
4.4 High-frequency waves
117
Figure 4.11 Wave transformations and the development of sand and shingle cays (after Gourlay, 1988).
currents induced by waves generated on the reef are able to mobilize carbonate sediments under mean wave energy conditions. For most reef flats maximum sediment mobilization occurs within a narrow depth window, where the water is deep enough to allow high and long waves to exist, but shallow enough to ensure significant bed velocities. Wave-induced circulation on fringing reefs has received little attention on the GBR. Roberts et al. (1975), Roberts (1981), and Lugo-Fernandez et al. (1998, 2004) reported on Caribbean cases, but these fringing reefs with significant lagoons are quite different from the majority of GBR fringing reefs on the mainland coast and on the windward side of islands. On fringing reefs, subjected to significant wave energy, channels on the reef flat may develop to allow water carried shoreward by wave processes to escape, a process analogous to rip formation on beaches (Roberts, 1981). Longshore currents are likely to be significant, and there is likely to be wave setup generated at the reef front and again on the beach.
118
Oceanography, hydrodynamics, climate, and water quality
4.4.3 Waves and wave-generated currents on the coast In the coastal boundary layer created by predominant south-easterly winds, sediment transport has been shown to be northward (Belperio, 1983a; Orpin and Ridd, 1996; Woolfe et al., 1998a). This finding is supported by an examination of sediments along the 10 m isobath (Lambeck and Woolfe, 2000) showing increased sediment maturity in a northward direction between Bowen and Cape York, with pockets of ‘‘immature sediment’’ in north-facing bays, reflecting their isolation from the dominant coastal processes. Increased variability in the Northern region is attributed to local sediment trapping, resulting from limited fetch and greater accommodation space, and supports the hypothesis of a hydrodynamic discontinuity at Princess Charlotte Bay (Wolanski, 1994).
4.5 High-intensity events 4.5.1 Cyclones Tropical cyclones have long been recognized (Moorehouse, 1936) as very significant for the geomorphology of reefs and reef islands over a range of timescales (Scoffin, 1993), primarily due to extreme waves, but with associated storm surge and runoff effects (Section 4.6) being significant near the coast. Depending on its track, the effects of a tropical cyclone can extend over very large areas, or over relatively small areas of reef, but always with significant spatial variability depending upon track (with reefs to the left and to the right of the track experiencing different levels of wave energy, from different directions), reef density, and sheltering effects. High wave energy from an uncharacteristic direction can have major geomorphic consequences. Cyclones in the Coral Sea can result in very large waves on the shelf-edge reefs, but have little influence in the GBR lagoon due to wave dissipation. Cyclones crossing the GBR lagoon can result in catastrophic seas locally. Larcombe and Carter (2004) argue the importance of cyclones in maintaining the sedimentary characteristics of the GBR continental shelf. Based on a dataset from TC Joy, from near Cairns in 1990, and from analogies with shelf currents elsewhere, they describe northward along-shelf currents of 100–300 cm s1 associated with cyclones, which cause the erosion of the middle shelf seabed, northward transport of bedload and shoreward transport of suspended sediment load. They argue that this process of ‘‘cyclone pumping’’ creates and maintains the three shelf parallel sediment facies (an inner shelf terrigenous shore-connected zone (0–22 m depth), a zone of sediment starvation, erosion and northward
4.5 High-intensity events
119
transport (22–40 m depth), and a zone of reef connected carbonate sediment in water depths >40 m) (see Chapter 6). Only a small proportion of reefs are likely to be impacted by cyclones in any given year, but it is likely that virtually all reefs will have experienced at least some cyclonic effects in recent time. Puotinen (2004) suggests proximity to path is a simple but reasonable predictor of reef disturbance. Puotinen et al. (1997) found that every location within the GBR region had been within 100 km of the track of at least one cyclone between 1969 and 1997. In a study of the effects of TC Ivor (March 1990), van Woesik et al. (1991) and Done (1992c) found that damage was severe but patchily distributed within 50 km of the cyclone track, but uniformly low at distances greater than 50 km from the eye. The nature of cyclone damage to corals is variable, with damage to the reef framework, physical damage and breakage of corals, dislodgement of coral heads, stripping of soft corals and other organisms, and damage due to sediment movement all being likely. On the windward edge, corals and sediments down to 20 m depth can be eroded (Scoffin, 1993), with debris accumulating on the reef slope, at the foot of the slope, in crevices, on the reef flat, and in the lagoon or back reef. Reef blocks with diameters up to 4 m tossed onto reef flats attest to the power of cyclonic waves. The vulnerability of corals to physical damage is very dependent on growth form, with arborescent and tabulate forms being most vulnerable, and massive corals being highly resistant to cyclonic waves even if only a small percentage of their base is firmly attached (Massel and Done, 1993). Vulnerability increases with age, with new recruits presenting little drag, and therefore less likely to be broken or dislodged. Done (1992c) describes the process of cyclone wave attack as one of ‘‘attrition,’’ where ‘‘cyclone waves exfoliate reefs, chunk by chunk, over the period of storm waves,’’ with the reefs sustaining the most damage being ecologically ‘‘reset.’’ The importance of cyclones in processes of sediment transport in inter-reefal areas of the GBR lagoon was demonstrated by Gagan et al. (1988, 1990), following the passage of TC Winifred across the shelf near Innisfail in 1986. The entire seabed out to the mid-shelf reefs was mobilized and reworked. Some sediment derived from mainland sources was deposited on the inner shelf to about 20 m water depth, but in deeper water transport was landward. Mobilization depths of the carbonate-dominated sediments on the mid shelf averaged 6.9 cm with a maximum of 14.4 cm. One year after the cyclone, the inner shelf terrigenous-dominated sediments remained reasonably intact (perhaps due to continued deposition of terrigenous sediment), but were completely bioturbated further offshore. Storm surge (low atmospheric pressure and wind setup causing an elevated water level, particularly in the left forward quadrant of the cyclone path)
120
Oceanography, hydrodynamics, climate, and water quality
affects the mainland shore and nearby islands and reefs. Storm surges combine with shoreline wave setup and extreme wave run-up to inundate low-lying areas. Numerous storm surges greater than 3 m have been recorded in GBR coastal waters (Hopley, 1982). While significant storm surge does not affect the shelf reefs to the same extent, elevated water levels due to low pressure, combined with elevation in water levels due to wave setup and increased wave height combine with increased sediment supply to increase sediment transport significantly during cyclonic events. The paucity of quality cyclone wave data, and to a lesser extent, storm surge data, has led to a reliance on modeling waves. Early models by Young and Sobey (1980) and Young (1988) were for open water situations, predicting, for example, waves of up to 13 m for a 950 hPa cyclone traveling at 30 km h1. Limited testing of this model against wave gauges indicated significant overestimation of significant wave heights by 20–30%, although peak wave periods (varying between 10 and 13 s) were reasonably well predicted. Hardy et al. (2000, 2003, 2004), James and Mason (2005), and a series of reports produced by the Queensland Government titled Queensland Climate Change and Community Vulnerability to Tropical Cyclones describe a methodology for modeling wind and pressure fields, waves, and storm surges associated with tropical cyclones in the GBR region. The output from the models takes a recurrence interval approach. The basis of the approach is the generation of a synthetic set of tropical cyclones, based on the record of cyclones since 1969, since the time reliable records were kept. An autoregressive approach was used to generate a 3000 year or 9911 cyclone record, to enable the calculation of return periods to at least 1000 years. The method is described in James and Mason (2005). Wave modeling was undertaken using a modification of the WAM model (Hardy et al., 2003), and when combined with the synthetic tropical cyclone database, resulted in wave statistics at 150 000 points with a 1500 m resolution over the GBR region for return periods between 20 and 1000 years. Significant wave height recurrence intervals are shown for four transects across the GBR in Fig. 4.12. Outside the reef significant wave periods are high (20-year return periods ranging from 6 to 10 m), and 500-year return periods from 12 to 16 m. Shelf reefs tend to experience smaller cyclonic waves than inner shelf reefs, where significant fetches within the GBR lagoon can result in significant waves developing. This is particularly noticeable for the Swain Reefs, where considerable fetch length exists except to the east and north whereas in the Northern region (represented by Cape Grenville in Fig. 4.12), the density of reefs across the shelf limit fetch inside the outer shelf reefs. A model of wind fields during cyclones (McConochie et al., 2004) is used in conjunction with a storm surge model and tidal data to calculate combined
4.5 High-intensity events
Figure 4.12 Modeled wave return periods on four transects across the GBR. Data sourced from the Atlas of Physical Processes in the Great Barrier Reef World Heritage Area (Marine Modelling Unit, 2006).
121
1.82 1.92 1.87 1.77 2.04 2.54 2.50 2.15 3.77 4.77 2.94 2.82 2.61 1.22
2.16 2.54 2.54 2.01 2.89 3.66 3.43 2.71 4.38 5.57 3.79 3.80 3.63 1.42
1000-year RP (AHD) (m) 0.34 0.62 0.67 0.24 0.85 1.12 0.93 0.56 0.61 0.80 0.85 0.98 1.02 0.20
1000 year RP 100 year RP (m)
b
Return periods incorporate tide height data. AHD, Australian Height Datum, which approximates Mean Sea Level. c HAT, Highest Astronomical Tide. Source: Data are calculated based on Hardy et al. (2004).
a
Lockhart River Bathurst Bay Cooktown Cape Tribulation Cairns Cardwell Townsville Bowen Mackay Clairview (Broad Sound) Yeppoon Gladstone Moore Park (Bundaberg) Surfers Paradise
100-year RP (AHD b) (m) 1.79 2.01 1.65 1.63 1.78 2.20 2.15 1.95 3.47 4.42 2.57 2.42 1.89 1.11
HAT c AHD (m) 0.03 0.09 0.22 0.14 0.26 0.34 0.35 0.20 0.30 0.35 0.37 0.40 0.72 0.11
100-year RP (HAT) (m)
Table 4.2. Modeled return periods a for water levels associated with storm surge at various coastal locations
0.37 0.53 0.89 0.38 1.11 1.46 1.28 0.76 0.91 1.15 1.22 1.38 1.74 0.31
1000-year RP (HAT) (m)
4.5 High-intensity events
123
storm surge and tide inundation return periods along the Queensland coast (Hardy et al., 2004). Model output from a range of sites is reproduced in Table 4.2. Typically, north of the Whitsunday Islands, 1000-year return periods range from 2–3.5 m above AHD (approximately Mean Sea Level). South of the Whitsunday Islands they range from 4–5.5 m, reaching a maximum in Broad Sound, the area with the highest tidal range. The results of the model show that 1000-year return period storm surge is less than 1.5 m above HAT (Highest Astronomical Tide) (and for most places less than 1 m), the difference between 100-year and 1000-year return periods is small (<1 m), and the 100-year return period is <0.5 m above HAT. Intuitively, the storm surge heights for long return periods seem small, and are clearly heavily influenced by the incorporation of tidal probabilities into the analysis. For example, the 2.9 m (Hopley, 1974a, b) storm surge associated with Cyclone Althea that affected Townsville in 1971 occurred close to low tide. The recorded water level of 2.4 m AHD made this approximately a 1-in-100-year event. Had this event occurred at or close to a period of HAT (2.15 m AHD), then the combined water level would have been about 4.5 m AHD, well over the 1000-year return period event of 3.43 m AHD. Similar statistical methods to enable the extension of the cyclone record have been used by McInnes et al. (2000), who reported a 1000-year return period surge for Cairns of 2 m above HAT. From a geomorphic perspective, the stochastic methodology has a number of problems, despite the fact that the statistical methods and numerical methods are well founded. They exclude from consideration overland flood effects and wave setup, run-up, and overtopping (Hardy et al., 2004), all of which can have a significant bearing on inundation and sedimentary processes during a cyclone event. Perhaps of even more concern, for both surge and wave analysis, is the use of the 1969–2001 cyclone record for the stochastic development of an extensive cyclone database. Nott and Hayne (2001) provide geomorphic evidence based on beach ridges and terraces from a number of sites, concluding that category 5 cyclones occur every 200–300 years within all regions of the GBR between latitudes 138 S and 248 S, rather than the previously assumed once every several millennia. Nott and Hayne (2001) and Nott (2003a) noted that there were no category 5 cyclones recorded in the GBR region in the twentieth century, unlike the previous 100 years which Nott (2003a) describes as a period of severe cyclogenesis. These findings call into question the reliance on the 1969–2001 cyclone record for the database for the models referenced above. By back-calculating from inundation heights, Nott and Hayne (2001) calculate the storm surge component of inundation to be between 10% and 70%
124
Oceanography, hydrodynamics, climate, and water quality
of total inundation, the remainder being the result of wave run-up and setup, suggesting that surge heights alone do not provide reasonable estimates of inundation or physical processes that operate on shorelines during cyclonic events. Nott and Hayne (2000) found deposits (surge plus run-up) likely to be the result of TC Mahina that crossed the coast at Bathurst Bay on 22 March 1899 at 3–5 m above HAT (despite anecdotal reports of a 13-m surge 30 km to the south). This is considerably in excess of the 1000-year return period event of 2.5 m AHD predicted by Hardy et al. (2004). Nott (2003a, 2004a, b) argues the necessity to incorporate pre-historic data in the analysis of cyclone effects to account for the lack of stationarity through time. For example, in the historical record, the largest storm tide (surge plus wave setup) in Cairns has been 0.7 m above HAT, but in the prehistoric record there are records of three events with storm tides between 2.5 and 4.5 m in the period 1800–70. Tropical Cyclone Ingrid crossed the coast 57 km south-east of Lockhart River on 10 March 2005 (Bureau of Meteorology, 2005). Although it had been category 5 previously, it was a compact category 4 cyclone at the time. Tides over the period of the cyclone were approaching maximum range, although TC Ingrid crossed the coast at about mid tide. Maximum observed inundation levels were about 2.2 m HAT. A peak storm surge of 0.28 m was recorded at Cooktown (310 km south of where the cyclone crossed the coast) on the morning of 9 March 2005. Because this occurred at the time of high tide near HAT, the water level reached 0.24 m above HAT, corresponding to a 1-in100-year event based on the storm surge models of Hardy et al. (2004), despite the cyclone being a considerable distance away. Counter-intuitively, a negative surge of 0.33 m was recorded at Cooktown at the time the cyclone crossed the coast, Cooktown being on the southern side of the cyclone path. (Environmental Protection Agency, 2005). 4.5.2 Tsunami Evidence of tsunami impacting the GBR region is provided by Nott (1997, 2000, 2003b) and Bryant and Nott (2001). Boulders, too large to be moved by cyclonic wind waves (some weighing over 100 tonnes, 8–10 m above sea level: Bryant and Nott, 2001), are found at a number of mainland (between Cairns and Cooktown) and high continental island (Orpheus Island) sites opposite gaps in the GBR (such as Trinity Opening and Grafton Passage). There have been no sizable tsunami impacting the GBR in historical time, and the timing of deposition of boulder deposits is unclear, although Nott (1997) shows clustering of 14C dates of corals pinned between boulders around 400–500 years
4.6 Mainland influences
125
BP and 700–900 years BP. The origins of tsunami impacting the GBR are presently unknown, but are almost certainly some distance from Australia. Evidence of tsunami on the shelf reefs is limited. Nott (1997) infers that tsunami may be responsible for the placement of reef blocks on reef platforms adjacent to openings in the reef. Using equations of Massel and Done (1993), 16 m waves would have been necessary to move these blocks, and although not excluding the possibility of cyclone waves being responsible, Nott (1997) questions why large reef blocks are found near reef gaps and not elsewhere. The limited evidence of pre-historic tsunami on the shelf reefs is not surprising, given the recent experience of the 26 December 2004 tsunami in the Indian Ocean. In the Maldives, tsunami waves resulted in little change that is likely to be preserved in the geological record (Kench et al., 2006), although displacement of large boulders was observed in channels between islands on the atoll rim in a number of locations, with one case of the excavation of large (2 2 1.5 m) slabs of beachrock associated with an island breach being recorded (Australian Government Mission and the Maldives Marine Research Centre, 2005).
4.6 Mainland influences The influence of mainland-derived water, sediment, nutrients, and other pollutants on the health of the GBR has been a major concern of management agencies and scientists in the last 25 years, and remains the topic of considerable discussion and debate. Early emphasis was on the quantities of the various inputs and flood plume dynamics, with emphasis later moving towards consideration of water and sediment quality and the recycling of nutrients and sediments within the system. Mainland-derived materials are an important environmental influence on inter-reefal areas and nearshore reefs (see Chapters 6, 7, and 13). This chapter describes the nature of those inputs, and processes affecting their movement in the GBR lagoon.
4.6.1 Water, sediment, and nutrient inputs into the Great Barrier Reef lagoon Fresh water from river runoff Runoff from rivers and ephemeral coastal streams on the east coast of Queensland between Cape York and Fraser Island enters the GBR continental shelf waters. The Queensland Department of Natural Resources and Mines (QNRM) identify 35 catchments (Fig. 4.13), with a combined catchment area
126
Oceanography, hydrodynamics, climate, and water quality
Figure 4.13 Catchments of the GBR, with average annual runoff (km3) adjusted for ungauged parts of the catchment for the period 1968–94 (Furnas, 2003).
of 423 000 km2. Seasonally ephemeral streams on high islands with a land surface area of 1070 km2 (Furnas, 2003) also have local effect. From data collected between 1968 and 1994, adjusted for streams that are not gauged, average annual discharge from GBR catchments is estimated to be 66 km3 yr1, ranging between 20 and 180 km3 yr1 (Furnas and Mitchell, 2001). The large,
4.6 Mainland influences
127
essentially inland catchments of the Burdekin (130 000 km2) and the Fitzroy (143 000 km2) dominate land area and total runoff. Highest runoff is from the Burdekin River with annual discharges (at Clare) ranging from 54 km3 (1973–74) to 0.3 km3 (1968–69) with an average of 11.3 km3 (data 1950–2004). Flows have been regulated since the construction of the Burdekin Dam in the early 1980s. All other GBR catchments with the exception of the Normanby (24 000 km3) and Burnett (33 000 km3) have areas less than 10 000 km3. Some rivers (particularly those in the wet tropics) have multiple major flows each year, others have one major flow each year, while the Burdekin and the Fitzroy tend to have major flows less frequently (Devlin and Brodie, 2005). Discharge from all rivers is highly seasonal and highly variable within seasons. The Burdekin River, for example, had a maximum daily discharge of 2.4 km3 in 1974. Furnas (2003) provides a detailed summary of the hydrology of GBR catchments. Most of the fresh water carried into the GBR lagoon comes from flood events, the largest of which change the character of estuaries, essentially emptying them of salt water. Fresh water enters the sea as a buoyant plume that is frequently clearly separated from surrounding water by a discrete front (Fig. 4.14). An estuarine salt wedge exists, but in high flow events it will occur outside the estuary. Wind, currents, and density effects eventually cause
Figure 4.14 Plume from the Hull River near Tully subsequent to Cyclone Winifred in 1986.
128
Oceanography, hydrodynamics, climate, and water quality
mixing of the plume with surrounding waters. In the absence of winds, Coriolis effects are sufficient to cause the plume to move north, adjacent to the coast, and the combined discharges from many rivers can result in a band of lowsalinity water extending for considerable distances. Wind stress is the dominant factor in the movement of river plumes. Winds from the south-east push fresh water north and towards the coast. Under these conditions, fresh water is typically held to within 20 km of the coast (Devlin et al., 2001). When winds blow from the northerly quarter fresh surface waters are directed southward and offshore (Devlin et al., 2001). Reefs on the mid shelf are only directly impacted by very-low-salinity water in extreme rainfall events (Davies and Hughes, 1983) and extended periods of northerly winds (Devlin et al., 2001). These findings have been generally confirmed by various studies of terrestrial tracers (e.g., Sammarco et al., 1999), although Wolanski and Spagnol (2000) report fine terrestrial sediment 30 km offshore from Cairns. Dispersion of flood plumes ultimately occurs over distances of many hundreds of kilometers, with pockets of reduced-salinity water breaking off and dispersing reaching some mid-shelf reefs, particularly to the north of Cairns where the shelf is narrow and topographic features can deflect water offshore. Indeed, lowsalinity water was recorded on Low Isles during the 1928–29 GBR Expedition. There have been numerous measurements (Wolanski and Jones, 1981; Wolanski and van Senden, 1983) and modeling studies, reviewed in King et al. (2001), of river plumes since the early 1980s, as well as observations of terrigenous sedimentation on reefs (Davies and Hughes, 1983). Some mortality on high island fringing reefs has been recorded (van Woesik et al., 1995). Burrage et al. (2002) used an airborne sea surface salinity imaging device and groundtruthing to examine the Herbert River plume in 1999/2000. These and other studies using sedimentary and other evidence (e.g. Gagan et al. 1987) indicate that significantly lowered salinity water rarely reaches mid-shelf reefs, generally being confined to within 20 km of the coast. However, some aerial observations during Cyclone Sadie in 1994, showed discolored water attributed to the flood rather than bottom resuspension reaching mid-shelf reefs between Cairns and Cardwell (Brodie et al., 1997; Devlin, 1997). Wolanski and Ruddick (1981) showed that, at times, water from the Fly River (Papua New Guinea) reaches the GBR lagoon. Inshore reefs, however, particularly those adjacent to catchments of the wet tropics, are frequently affected by low-salinity water. Submarine groundwater discharge Although total volumes are likely to be small relative to runoff, locally, discharge from groundwater may be significant, particularly with respect to
4.6 Mainland influences
129
the high concentrations of nutrients that can be delivered. Discharge from groundwater is immediately obvious at low tide along many mainland sand beaches, with water discharging close to a significant break in slope that is characteristic of North Queensland beaches. Submarine groundwater discharge is also a component of the freshwater budget on coral islands (Falkland, 1993). Stieglitz (2005) identifies discharge from unconfined coastal aquifers and discharge from large dune systems as potentially significant sources of groundwater. Discharge from confined coastal aquifers or ‘‘wonky holes’’ (Stieglitz and Ridd, 2000) which exist in incised river paleochannels, occurs at numerous sites over the inner continental shelf during the wet season. 4.6.2 Sediment and nutrient concentrations The supply of sediments to the GBR lagoon is highly correlated with river discharge, with respect to total quantity, seasonality, and variation over longer time periods. Suspended sediment concentrations in rivers increase rapidly when water levels start to rise, reaching a peak at or slightly before peak discharge. However, suspended sediment concentrations tend to reduce quickly as discharge decreases. High concentrations are frequently associated with the first significant rainfall event following a dry period, as weathered materials enter rivers. Maximum suspended sediment concentrations in wet catchments are substantially less than in dry catchments, primarily due to different land cover. In large floods suspended sediment concentrations in the Burdekin River reach between 1 and 3 g l1, or 1–3 million tonnes of sediment per km3 of discharge (Furnas, 2003). Low flow concentrations range from 0.05 to 0.2 g l1, contributing almost nothing to total sediment discharge. While data for suspended sediments is quite good, if spatially variable, little is known about sediment carried as bedload. Belperio (1979b) estimates bedload sediment transport in the Burdekin to be about 10% of the suspended load, all of it occurring during floods. Rivers transport nutrients as free dissolved ions, as part of dissolved organic compounds, and as suspended particles. Nutrient concentrations in rivers depend on a range of factors including soil character, physical and biological processes, and land cover and use, and vary considerably through time and between rivers. Furnas (2003) shows a high correlation between particulate nitrogen and particulate phosphorus concentrations, and suspended sediment concentrations. The relationship between discharge and concentration for dissolved nutrients is more complex, and is described for a number of rivers by Furnas (2003).
130
Oceanography, hydrodynamics, climate, and water quality
4.6.3 Sediment and nutrient volumes Estimates of the volumes of sediments and nutrients that reach the GBR lagoon are unreliable. Estimates of sediment discharge based on accumulation within the GBR lagoon range from 15 to 28 million tonnes per year (Belperio, 1978, 1983b; Moss et al. 1993 as cited in Furnas, 2003; Neil and Yu, 1996; Prosser et al., 2001; McKergow et al. 2005). Furnas (2003), using a method based on volume-specific sediment export coefficients which are established for rivers with known characteristics and applied to other similar rivers, estimated the total input of fine sediments to the GBR to be 14.4 million tonnes per year, with the Burdekin and Fitzroy contributing 42% of the total with only 22% of the runoff. McKergow et al. (2005) using the SED NET model (Prosser et al., 2001) provided an estimate of 16 million tonnes per year, with this model predicting higher sediment inputs from Cape York and coastal catchments. By comparison, estimates of pre-1850 sediment exports to the GBR are less than 5 million tonnes per year (Furnas, 2003; McKergow et al., 2005). McCulloch et al. (2003b), based on Ba/Ca ratios in long-lived Porites corals, estimated a five- to ten-fold increase in sediment delivery after about 1870, with highest delivery being associated with drought-breaking floods. Using similar methods based on volume-specific export coefficients, Furnas (2003) has estimated nutrient exports (Table 4.3) from runoff and other sources, speculating that inputs of nitrogen have doubled and phosphorus Table 4.3. Annual inputs (tonnes 1000) of nitrogen and phosphorus into the Great Barrier Reef lagoon Nitrogen
Phosphorus
Source
Range
Average
Range
Average
Land runoff (total) Land runoff (soluble) Pre-1850 land runoff (total) Pre-1850 land runoff (soluble) Coral Sea upwelling Rainfall Sewage discharge (upper estimate) Aquaculture (upper estimate)
10–120 6–60 4–66 2.9–38 4.4–40 14–44 2.3
43.0 20.0 23.0 12.0
1.3–22 0.3–4.7 0.36–7 0.16–1.8 0.63–6.3 0.84–2.6 0.6
7.0 1.5 2.4 0.6
Source: Furnas (2003)
0.2
0.0
4.6 Mainland influences
131
have tripled since European settlement. Some nutrient inputs from other sources are also recognized in Table 4.3, but contributions from other biological sources, such as cyanobacteria and benthic microalgae, and groundwater are largely unknown (Alonghi and McKinnon, 2005). Total inputs from upwelling and rainfall, particularly of the generally immediately bioavailable soluble forms of nitrogen and phosphorus, are comparable to land-derived sources. 4.6.4 Sedimentary processes associated with mainland-derived sediments Almost all of the sediment carried as bedload in rivers that reaches the marine environment is deposited in the estuary systems and in delta deposits near the mouths of the river estuaries. Bryce et al. (1996, 1998) have shown that estuary systems may be sites of net accumulation of both fluvial-derived and marinederived sediments. A substantial proportion of the finer sediments and particulate nutrients transported as suspended load in the rivers flocculate as saline water is encountered (prior to salinity increasing to 10 ppt). These near-source deposits, however, do not accumulate these fine particles, indicating that fine sediments are resuspended by waves and moved away. Dissolved nutrients move and are dispersed in the freshwater plume. Devlin and Brodie (2005) reported 10–100 times non-flood ambient concentrations of dissolved nutrients in plumes up to 200 km from source, with an inference that there is little biological uptake of nutrients in the initial stages of the plume (possibly due to light limitation). However, Alonghi and McKinnon (2005) concluded that most terrestrial-derived nutrients are, in the long term, restricted to the coastal zone. 4.6.5 Sedimentary processes and resuspension Wolanski (1994) notes a similarity between areas of lowered salinity during a Burdekin River flood and the distribution of terrigenous mud on the sea floor. Substantial quantities of terrigenous fine sediment are deposited and stored in large intertidal and subtidal deposits, forming an inner-shelf wedge less than 5 m thick that typically extends to about 20 m water depth (Larcombe and Woolfe, 1999a; Neil et al., 2002; Section 6.2). The largest deposits are in northward-facing bays that have some protection from the south-easterly winds, and in these locations muddy deposits extend to the shoreline, normally with associated mangrove communities (Woolfe et al., 1998a). In more exposed locations, a sandy zone inshore of the muddy deposits, the result of finer
132
Oceanography, hydrodynamics, climate, and water quality
material being removed by wave resuspension, provides substrate suitable for coral colonization (Larcombe et al., 2001; Smithers and Larcombe, 2003). The muddy sediment deposited in the sediment wedge is resuspended in shallow waters (for example, Cleveland Bay, Townsville) under even low to moderate energy conditions, and in deeper water during cyclone events (Gagan et al., 1987, 1988; Wolanski and Ridd, 1990; Neil et al., 2002). Turbidity on the inner shelf is not limited by sediment availability, but can be energy limited across most of the inner shelf (Larcombe et al., 1995b; Woolfe and Larcombe, 1998; Larcombe and Woolfe, 1999a, b). Only near the inner shelf break, where sediment characteristics change, and at sites a significant distance from river sources is sediment supply a limiting factor to turbidity. Larcombe and Woolfe (1999a) note that the inner shelf sediment wedge is prograding ( up to 1.5 m yr1), and its thickness increasing, but at rates too low to be reasonably addressed as a management issue. Mid-shelf reefs are unlikely to be engulfed. Tidal currents initiate sediment resuspension only in spatially restricted areas like Broad Sound (Kleypas, 1996); however, tidal currents exert some control on suspended sediment concentrations by promoting mixing and water exchange, particularly on spring tides. Higher near-bed shear stress under waves is the dominant mechanism causing sediments to be resuspended. In very shallow water, even small, locally generated wind waves (<3 s period) are able to resuspend sediment, but more generally, the longer 5–9 s period waves characteristic of the GBR lagoon are able to resuspend sediment across much of the inner shelf to at least 10 m water depth, even for waves of small height (Larcombe et al., 1995b). Orpin et al. (1999) showed that muddy sediment at 20 m is only resuspended a few days a year, but in water depths <10 m resuspension occurs frequently (>110 days a year in their study area). Under waves, turbulence is able to distribute the sediment through the water column over relatively short periods of time. Significant transport of sediment results when waves resuspend sediment in the presence of unidirectional currents. In a study of sedimentation and turbidity around a small high island near Cairns, Wolanski et al. (2005) found sediment resuspension during a highenergy south-easterly event down to water depths of 12 m on the windward side, and 5 m on the leeward side of the island, with sediment accumulation below those depths. In calm weather, accumulation of sediment may occur for extended periods, suggesting that exposure to regular wave energy is an important factor in determining sediment stress to corals. Wolanski and Spagnol (2000) recorded suspended sediment concentrations up to 1000 mg l1 in 2 m water depth. High concentrations of suspended
4.6 Mainland influences
133
sediments were found on and around reefs of Cleveland Bay, Townsville, by Larcombe et al. (1995b). Near-bed concentrations of >200 mg l1 were recorded within the bay, with reefs experiencing suspended sediment concentrations of up to 50 mg l1, sometimes exceeding 20 mg l1 for periods of over 24 hours. They conclude that similar concentrations characterize many of the fringing reefs of the GBR, and that concentrations are likely to have been similar to modern levels for the last 6000 years. Further examples are in Section 13.2. Wave-driven sediment resuspension from the seabed may be an important source of nutrients, releasing and transforming the nitrogen species, and stimulating the growth of phytoplankton (Chongprasith, 1992; Wolanski, 1994). Alonghi and McKinnon (2005) suggested intense recycling of nutrients in the microbial loop is enhanced by wave resuspension and mixing. Seagrass beds play an important part in the ecosystem of the GBR, particularly with respect to their role as fish habitat, and as a food source for dugong and turtle. They also transform and store sediments and organic matter, in quantities disproportionate to their size (Alonghi and McKinnon, 2005), although these can be released during periods of dieback. Areas of seagrass are associated with accumulation of sediments (see Section 13.3.2), although seagrass beds are particularly subject to disturbance by highenergy events (and therefore probably are not sites of long-term accumulation), and recovery can be slow (Furnas, 2003). The interactions between seagrass growth and terrestrial runoff effects are poorly understood (Waycott et al., 2005). 4.6.6 Effects of runoff, sediments, and nutrients on corals The effects of terrestrial input to coral reefs have been the subject of considerable debate. Although sediment quantities have undoubtedly increased, there is more contention over whether that is causing adverse impacts through increased turbidity and sedimentation (concepts that are frequently not distinguished). Few coral reefs exist near the mouths of significant rivers, most probably due to sedimentation and sediment mobility, but there are numerous examples of reefs in water with characteristically high turbidity (Smithers and Larcombe, 2003; Smithers et al., 2006). In these situations, sediment may be deposited on corals for short periods before being advected away, possibly providing a transient stress about which little is known (Wolanski, 1994). Reefs in high-turbidity situations are fundamentally similar to other reef systems, but do have differences, with Perry and Larcombe (2003) terming
134
Oceanography, hydrodynamics, climate, and water quality
them ‘‘different, not disturbed.’’ Around Broad Sound, an area of characteristically turbid water of up to 8 m tidal range, van Woesik (1994) and Kleypas (1996) found a decrease in colony size of hard corals, a decrease in diversity, a shift in coral growth morphology, and a lack of major framework-building corals. Kleypas (1996) suggests a 6–7 m maximum tidal range as a limiting factor for growth in this area. The effects of high nutrient levels have also been debated. Van Woesik et al. (1999) and Fabricius et al. (2003) correlated increased nutrients and/or sediments with reduced coral cover, species richness and abundance, increased recruit mortality, and changes in the relative abundance of corals and algae. Szmant (2002) concluded that over-enrichment can cause localized coral reef degradation, but the case for widespread effects is not substantiated, and in the GBR there is no evidence that sediments are becoming enriched over time or that there is an increase in water column chlorophyll. McCook (1999) and McCook et al. (2001) found no evidence that current high levels of nutrient inputs were leading to widespread algal growth over otherwise healthy established coral reefs. McLaughlin et al. (2003) summarized the adverse effects of sediment on corals and reefs but noted that reduced salinity, nutrient loading, other toxic substances associated with anthropogenic influences (such as biocides), and sediments are all interrelated, and use this as a justification for using runoff as a proxy indicator of the adverse effects of terrestrial inputs to reef systems. Their analysis indicated that, at a global scale, annual runoff of more than 1010 m3 yr1, in a 0.58 (latitude – longitude) grid cell is associated with strongly reduced occurrence of coral communities. Macdonald et al. (2005), however, responded that analysis at this scale is unreliable, arguing that their data sources (Reefbase and Reef Check) are biased towards clear water reefs, and there is poor understanding of what constitutes a coral reef as opposed to a coral community, or a degraded reef as opposed to a healthy coral reef in marginal (particularly sediment-affected) environments. This discussion reflects the considerable debate that exists on the effects of terrestrial runoff on reefs and reef environments, and is further discussed in Chapters 7 and 13. In an attempt to summarize the extensive literature on the effects of terrestrial inputs Fabricius (2005) ranked pollution (including sedimentation and eutrophication) as a threat to reefs similar in severity to coral bleaching and overfishing effects, and noted four fundamental processes that need to be assessed when considering terrestrial runoff effects: (a) dissolved inorganic nutrients reducing coral calcification and fertilization, and the role they play in
4.7 Oceanographic and climatological stressors
135
curbing organic enrichment of benthos, sediments, and suspended particulate organic matter; (b) enrichment with particulate organic matter enhancing feeding rates and growth in some corals and, to an even greater extent, other heterotrophic filter-feeders (such as the larvae of Acanthaster planci), and thereby changing community structures; (c) light limitation due to turbidity in deep waters affecting certain species and reducing coral recruitment; (d) sedimentation, affecting recruitment and survival in the early stages of life in a wide range of coral species, by smothering by sediment or sediment-trapping macroalgae, and affecting other species in a variety of ways. Fabricius (2005) noted that the type and severity of response at a particular location is co-dependent on these factors and other aspects of the physical and biological environment. Fabricius (2005) argues that reef systems simplify with increasing exposure to terrestrial runoff. However, most of the evidence presented is based on controlled experiments where corals are subjected to a range of treatments, either in the laboratory or in situ (Koop et al., 2001). Changes in community structure along water-quality gradients (Fabricius et al., 2005) not surprisingly have also been reported. There is no doubt that extreme cases of changes to sediment and nutrient loadings can have significant adverse effects on coral reefs (Kaneohe Bay, Hawaii, being an excellent example: Smith et al., 1981), but there remains little direct evidence that the changes that have occurred in the GBR have had any significant effect on reefs. There is a substantial body of evidence that turbidity conditions have probably not been significantly different for the last 6000 years, and that the reefs that grow on the inner shelf are different not disturbed. 4.7 Oceanographic and climatological stressors This chapter has concentrated on the description of present-day climatological and hydrodynamic processes that affect the geomorphology of the GBR. Some of the discussion has been within the context of change to coral communities and reefs, and much of the literature in this field has been written with future change (particularly climate change) in mind. The value of proxy records found in corals (Isdale, 1984; Lough and Barnes, 1997, 2000; Lough, 2004) and sediments is well recognized. The GBR has evolved and exists in a background of high variability of the natural environment associated with climatic variables, and at least over the last few hundred years, there is good evidence to suggest that climatic cycles, particularly ENSO events, are important factors contributing to variability
136
Oceanography, hydrodynamics, climate, and water quality
(Lough, 1994; Hendy et al., 2003). Hughes and Connell (1999) recognize that coral reefs are subject to frequent disturbances, and acknowledge the difficultly of basing understanding on investigations at one or two sites over short periods of time. They recognize the importance of considering multiple stressors, and the temporal relationship between disturbance events, even over long periods of time. However, the urgency given to the consideration of the effects of climate change on reef environments has meant that long temporal and large spatial perspectives on the effect of climate on reefs have tended to be overlooked. Changes are frequently attributed to climate change (Wilkinson, 1999), when it may be that the changes are within windows of long-term variability or associated with direct anthropogenic damage, such as has happened to many reefs in Asia, eastern Africa, the Caribbean, and Central and South America. Perhaps the most potentially damaging of the effects of climate change is coral bleaching, the understanding of the consequences of which were dramatically illustrated by the severe 1998 bleaching event. It has been suggested that mass bleaching events are likely to increase with the thermal tolerance of reefbuilding corals being exceeded more frequently (Hoegh-Guldberg, 1999), although it is known that other climatically related events, such as cold weather (Hoegh-Guldberg et al., 2005), can also cause bleaching. The consideration of climate change scenarios on coral reef habitat have concentrated on the effects of water temperature. Guinotte et al. (2003) concluded that climate change will cause many reef areas in the Pacific to become marginal due primarily to bleaching, with few new, currently low-temperature marginal areas becoming suitable areas for reef growth. The effects of other changes to environmental variables, such as changes to water circulation, wind speed and direction, cyclone intensity and frequency, and the effects of those changes on the GBR have yet to receive the same attention. Lough (2001) suggested that the effect that climate change has on ENSO is critical to the GBR given its influence on interannual climatic variability. If ENSO events become more frequent and/or intense, less frequent cyclones, less rainfall and runoff, and more dominant south-easterly winds would be the likely outcome. If anti-ENSO conditions become more frequent, increased disturbance through increased rainfall, floods, and cyclone activity may be expected. It may be possible that both sets of events intensify. The understanding of the climatology and the hydrodynamics of the GBR since the review (from a geomorphic perspective) by Hopley (1982) has advanced due to a large extent by improved instrumentation (including remote sensing) and improved computer modeling. Effort on some research themes
4.7 Oceanographic and climatological stressors
137
(e.g., regional-scale circulation) has reduced as understanding has improved, while other themes such as extreme events, sea surface temperatures, and terrestrial influences on the GBR have become more important, particularly in the context of environmental management and climate change. The contribution of geomorphology to some of the more significant environmental management themes is considered in Chapter 13.
5 Spatial analysis of the morphology of the reefs and islands of the Great Barrier Reef
5.1 Introduction The character and geomorphic diversity of the Great Barrier Reef (GBR) is still being discovered. Although described as a geomorphic system for well over a century, with small sections being studied in detail since the 1920s following the establishment of the Great Barrier Reef Committee and the Royal Society expedition in 1928–29 (see Chapter 1), it is only since the 1970s that LANDSAT imagery combined with extensive but incomplete aerial photography has enabled the extent and detail of the GBR to be described. Until a comprehensive bathymetric survey occurred in the 1980s, large areas of reef remained uncharted, particularly in the northern section where reefs are densest and away from shipping channels, and in the southern GBR, where the reef is complex and a considerable distance offshore. Since that time, the availability of spatial data derived using various remote sensing methodologies, and made accessible using Geographic Information System (GIS) technology, has increased and developed to the point where basic statistics and zoning information are accessible via Internet-based interactive GIS (Great Barrier Reef Marine Park Authority, 2004). However, a combination of supervised and unsupervised interpretation of satellite imagery has resulted in problems being introduced for morphological analysis with truthing either by examination of high-resolution imagery or ground survey not keeping pace with data acquisition.
5.2 Remote sensing and the Great Barrier Reef Although some aerial photograph images have been available since the 1920s (Bowen and Bowen, 2002), comprehensive coverage of the GBR was not available until satellite imagery became available. A LANDSAT receiving 138
5.3 The history of spatial data collection and analysis
139
station was established in Australia in 1979 (Bell, 1982), and work by Jupp et al. (1981a, b) and Kuchler (1984) researched the utility and methodologies of the new technology, and undertook considerable cross-checking with aerial photography and on ground surveys. The early explorations in remote sensing of the GBR are summarized by Jupp et al. (1985). Satellite-based remote sensing using hyperspectral techniques has continued to expand, but much of the published work has concentrated on experimental development and ground-truthing (Ahmad and Neil, 1994; Mumby et al., 1998, 1999, 2004a, b; Andrefouet et al., 2002; Hochberg et al. 2003; Kutser et al., 2003; Hedley et al., 2004; Joyce et al., 2004; Phinn et al., 2005) rather than on large-scale spatial analysis, where the particular benefit of remote sensing lies. While surprisingly little at the large scale, and little that could not be undertaken using other methods at the individual reef scale, has been achieved using satellite imagery, the advent of the new generation satellites such as IKONOS, providing 4 m resolution in the spectral bands (red, green, blue, and near-infrared) and 1 m resolution in the panchromatic band (Andrefouet et al., 2003; Elvidge et al., 2004) continues to provide hope that large-scale geomorphic interpretation using satellite imagery is possible. Satellite-derived data for other environmental variables (such as sea surface temperature, waves, cloud cover) is invaluable and well established (see Chapter 4). Other airborne remote sensing instruments and specialized photography have been used successfully. For example, Burrage et al. (2002) developed an airborne sea surface salinity mapping system useful in the mapping of flood plumes, while Hopley and Catt (1988) and Thamrongnawasawat (1996) demonstrated the use of digitized infrared photography for the monitoring of reef flat ecology. The technique was used for reef management applications in Thailand (Thamrongnawasawat and Hopley, 1995). 5.3 The history of spatial data collection and analysis The first full series of maps, with almost all reefs shown was produced by National Mapping, at a scale of 1: 250 000 in the early 1970s (the GBR Reconnaissance Series). These were based on LANDSAT scenes, with resolution of about 80 m, so that much reef detail was lost, and reef shape was crudely represented. However, some reefs were shown well out of position. In 1982, the Great Barrier Reef Marine Park Authority (GBRMPA) commissioned a series of maps, based on the LANDSAT derived images, but with reef outlines in much greater detail based on aerial photographs. From these maps, the first gazetteer was produced, which included reef names, location of the center of the reef, and the presence of islands, navigation aids, or other notable
140
Spatial analysis of the morphology of the reefs and islands
features. A reef numbering system based on latitude was developed, and this system is used to the present day. The production of the gazetteer provided the opportunity for calculation of reef area and geomorphic classification, based on the aerial photographs. An analysis of the data contained in the gazetteer was presented in Hopley et al. (1989), this becoming the source of the widely quoted counts of 758 fringing and 2146 non-fringing (totaling 2904) reefs in the Great Barrier Reef Marine Park. However, calculations only extended to the Marine Park boundaries, and do not include the reefs of Torres Strait north of 108 410 S, probably totaling around 750 (see Chapter 1). The use of GIS to manage spatial data began at the GBRMPA in 1992 (Hartcher and Shearin, 1996), the initial impetus being as a tool for zoning. The base data came from digitizing map sheets produced by the Australian Surveying and Land Information Group (AUSLIG) based on the original LANDSAT images. The advent of LANDSAT-7 in 1999 gave the opportunity to map the reef at a 30 m resolution. Considerable effort was put into obtaining good-quality ground control, obtained using helicopter surveys (Lewis et al., 2003a). Image interpretation was undertaken through a combination of supervised and unsupervised classification techniques. Resolution and turbidity issues precluded the remapping of most nearshore and fringing reefs, the data for which are currently being updated from other sources. At the present time the datasets consist of GIS accessible files showing indicative reef outline (which Lewis (2001) suggests equates to approximately 10.5 m below MSL), dry reef (but probably more appropriately called reef flat) and the coast and islands, including high islands and many cays (assumed to represent MHWS). The gazetteer and maps (Hopley et al., 1989) and the spatial data held in a GIS by the GBRMPA are largely independent, but can be linked through a unique reef identifier used in both datasets. Geomorphic reef type classification which formed an integral component of the gazetteer was largely lost in the implementation of the new GIS, although Lewis (2001) made some attempt to connect and analyze reef type by latitude within the GIS. Differences in interpretation are common, as illustrated in Fig. 5.1 and Table 5.1, where the interpretation of the reefs in the vicinity of Green Island is shown based on the two datasets, and it is often unclear which is more reliable. In general, the GIS data interprets the reefs as being larger than in the gazetteer, although knowledge of the bathymetry around Green Island, for example, indicates the GIS may overestimate area. Although mapping of the GBR has improved significantly as new technology has become available, there are significant problems in using the GBRMPA GIS data for the large-scale spatial analysis of reefs and island geomorphology. This is partly due to the purpose for which the database was
5.3 The history of spatial data collection and analysis
Figure 5.1 (a) GIS and (b) gazetteer maps of the area in the vicinity of Green Island.
141
142
Spatial analysis of the morphology of the reefs and islands
Table 5.1. Areas of reefs in the vicinity of Green Island (Fig. 5.1), as defined in the Great Barrier Reef Marine Park Authority GIS database and the gazetteer GIS Reef
Reef area (km2)
Dry reef area (km2) a
Island area (km2)
Gazetteer area (km2)
Green Vlasoff Upolu Oyster Thetford Arlington
35.41 9.69 14.69 17.71 12.84 127.75
3.48 (1) 0.87 (1) 2.11 (2) 2.56 (3) 2.29 (1) 42.04 (66)
0.20 0.04 0.07
7.10 6.25 12.10 14.50 7.90 99.50
a
The number of polygons defining the dry reef area is given in brackets.
developed, which was primarily the management of the zoning process. Reef types have only been included as a loose link from the gazetteer data and data verification has not been undertaken. There are 2977 named reef features included in the GIS, a number that is not significantly different from that reported by Hopley et al. (1989), plus approximately 726 new features that have been identified as separate reefs by the classification process. These features may or may not be accurately recorded, and there has been no attempt to classify them according to reef type. These features are of a number of different types. Some are new features that were not identified using early imagery, mostly submerged reefs such as some of the outer shelf submerged reefs (Hopley, 2006; Section 9.4), and including a small number of incipient fringing reefs. Some are likely to be spurious such as those seaward of the Ribbon Reefs in the northern section, probably chlorophyll-rich plumes seaward of the passages as recognized by Wolanski et al. (1988) (see also Section 6.6.4). Most, however, are splits of features that were recorded as single reefs in the gazetteer, probably the result of unsupervised classification recognizing areas of deep water in channels. This has occurred primarily in the area north of Shelburne Bay and within the mid and outer shelf reef matrix south of the Whitsunday Islands. In these cases the unique identifier has been applied to only one or a small number of the split features, affecting the calculation of aggregate statistics (such as reef area and reef type). Cockatoo Reef, a lagoonal reef at 208 450 S 1518 010 E, is a good example of the effect of splitting features. Figure 5.2a shows the interpretation of Cockatoo Reef from the gazetteer, with a recorded area of 90.60 km2. Figure 5.2b shows the GIS interpretation. Four polygons are identified as
5.3 The history of spatial data collection and analysis
Figure 5.2 Cockatoo Reef. (a) Cockatoo Reef as interpreted in the gazetteer (Hopley et al., 1989), with a total area of 90.60 km2; (b) The 24 reef features comprising Cockatoo Reef identified in the gazetteer, four of which (A, B, C, D) have associated attribute data and a total area of 69.59 km2.
143
144
Spatial analysis of the morphology of the reefs and islands
Figure 5.3 Bewick Reef environments and areas (a) as held in the GIS database and (b) as mapped in 2005.
comprising Cockatoo Reef with full attribute data, and a total area of 64.15 km2 (labelled A, B, C, D). There are however, 20 other polygons identified, with a total area of 5.44 km2. The GIS interpretation of Cockatoo Reef is therefore 24 separate reef features (with a total area of 69.59 km2). Unlike the gazetteer, the GIS database was not derived by geomorphologists, and many interpretations have been made that make little geomorphic sense. For example, areas of mangroves on reef flats are frequently classified as islands. An example of this problem is shown in Fig. 5.3 (Bewick Island, Howick Group; see also Fig. 8.3a). The reef area is mapped as two polygons, the inner one (1.47 km2) being coincident with the area mapped as island (Fig. 5.3a), and an outer doughnut shape of 0.78km2. The inner polygon does not contain easily identifiable linking attributes, and, although the data
5.3 The history of spatial data collection and analysis
145
are clear when examining it at the individual reef scale, it would be very easy in analyses of the aggregated data to record a reef area of 0.78 km2 for the Bewick Reef, rather than 2.25 km2. A different interpretation of the Bewick Reef is shown in Fig. 5.3b (mapped on a field visit in 2005). Bewick Reef has a sand cay (0.11 km2) at its north-western corner, the remainder of the ‘‘island’’ being mangrove forest covering most of the reef flat with some high shingle ridges around the outer edge of the forest. Some of the issues that arise due to poor classification are likely to be resolved with the implementation of new Geodatabase models in 2006–07. Other problems with the datasets have become apparent as knowledge of some environments has improved. For example, Halimeda banks in the northern section were interpreted as submerged reefs in the gazetteer, a situation that distorts the statistics relating to submerged reefs in some areas (further discussed in Chapter 6). Island data have not been retained in the reef GIS, but have been included as part of the ‘‘Coast’’ data layer. Although an island type attribute is included, it is aligned to management and does not clearly distinguish island geomorphology, and is not able to be used in the analysis in this chapter. Many island features identified in the gazetteer from aerial photography are not recorded at all in the GIS. The problem with the representation of islands is partly explained by the fact that many of the island data in the GIS were carried forward from the old AUSLIG maps. It is important to note, however, that the GIS database continues to be developed and many of the issues identified in the paragraphs above are likely to be resolved, particularly with the implementation of new Geodatabase models in the near future. Two significant, large-scale spatial products have been produced. Lewis (2001) produced a depth and elevation model of the GBR, including catchments that flow into the reef, based on Royal Australian Navy and AUSLIG hydrographic survey data, topographic data, the GBRMPA GIS datasets, and multibeam echo sounder and Laser Airborne Depth Sounder (LADS) data that exist for limited areas of coverage. An example of the model output is the reconstruction of shorelines at different sea-level stages (Chapter 3). The analysis based on the depth model includes consideration of reef slopes, areas, volumes of water by depth range, and detailed analysis of cross-shelf profiles. A recent rezoning of the Great Barrier Reef Marine Park, also known as the Representative Areas Programme (Fernandes et al., 2005), made extensive use of GIS datasets and techniques, including the production of material used in pre-implementation consultation, in the analysis of responses, and in the definition and publication of final results (Lewis et al., 2003b), and has resulted in several new layers of information in the spatial database.
146
Spatial analysis of the morphology of the reefs and islands
5.4 Great Barrier Reef lagoon areas and volumes The area of the GBR (constrained by the GBRMP boundaries to the north and south, and by the shelf edge, variously defined), has been estimated as 230 000 km2 (Pickard et al., 1977), 223 977 km2 (Hopley et al., 1989), and 238 700 km2 (Lewis, 2001). The total area of the GBRMP is estimated as Table 5.2. Surface areas and water volumes to the 150-m isobath, in the Great Barrier Reef Marine Park a
Depth (m)
Volume, horizontally defined (km3)
Area, vertically defined (km2)
Volume, vertically defined (km3)
Area, spatially constrained, vertically defined (km2)
Volume, spatially constrained, vertically defined (km3)
0–10 10–20 20–40 40–80 80–150 Total
2090 1915 2930 2215 400 9550
38200 20300 68300 94800 17100 238700
80 320 2060 5320 1770 9550
25600 15700 65000 114900 17500 238700
55 225 1760 5720 1790 9550
a
For an explanation of the calculation criteria, see Fig. 5.4 Source: Lewis (2001). (a)
0 –20 –40 –60 –80
(b)
0
Depth (m)
–20 –40 –60 –80 (c)
0 –20 –40 –60 –80
Figure 5.4 Volume calculations of the GBR lagoon: (a) horizontally defined; (b) vertically defined; (c) spatially constrained, vertically defined. Source: Lewis (2001). Data are presented in Table 5.2.
Transect length (km)
5.5 Reef and reef island statistics and classification 250
(a)
150
50
–10 Mean depth (m)
147
(b)
–25 –40 –55 –70
Volume (km3)
700
(c)
600 500 400 300 200 100 12
14
16
18
20
22
24
Latitude (° S)
Figure 5.5 (a) Shelf width (based on shore normal transects), (b) mean depth, and (c) water volumes, plotted as five-point running means, from 0.25 degree shore normal transects (see Lewis (2001) for an explanation of the methodology). Source: Lewis (2001).
341 300 km2 (Hopley et al., 1989) to 344 400 km2 (excludes State Islands, Queensland internal waters or seas and submerged lands, but includes Commonwealth Islands (195 km2)) (Great Barrier Reef Marine Park Authority, 2004). Based on a depth model, Lewis (2001) also estimated surface areas and volumes of water in depth ranges calculated in three ways (Fig. 5.4, Table 5.2). The shelf is narrowest and shallowest in the north, and deepest and widest in the south (Fig. 5.5). 5.5 Reef and reef island statistics and classification 5.5.1 Datasets and data quality Although reef numbers may be of interest and are frequently quoted, they are of only marginal value in geomorphic description due to the fact that
148
Spatial analysis of the morphology of the reefs and islands
significant numbers of reef features are not yet verified or classified, and there are large numbers north of the GBRMP boundary for which no data are available. Reef areas are of more significance, but total area is affected by interpretation (see the example of Green Island above). In the GIS, problems with reefs being split, and the inclusion of unverified and unclassified features, mean that its value for the calculation of descriptive statistics by reef type is limited. Another area of potential confusion is the fact that some areas have more than one distinct zone, and in such cases part of a single reef may be represented as, for example, ‘‘crescentic,’’ while another part is ‘‘reef patches.’’ In both databases, this situation is identified, but in this large-scale analysis, the features are treated as separate entities. In some cases, the boundary between two reef features can also be unclear. The analysis of reef classification by location is more valuable and distinct regional patterns can be determined despite the data problems. There remains a considerable amount of work to be done before an accurate representation of the reefs of the GBR is completed. Although it is likely that the gazetteer-generated data have underestimated the number of discrete reef features (2904) and the total area (20 055 km2) (giving a mean size of 6.91 km2), the gazetteer has benefited from supervised classification by geomorphologists. Interpretations of reef islands are likely to be quite accurate. The GIS, on the other hand, has probably more accurately determined discrete reef features (2977 named plus 726 new features, totaling 3703), if separation by deep (greater than approximately 10 m) water is used as the determining factor, but this has not generally been done based on morphology or common interpretation. The estimate of the total area of reef is also significantly higher using the GIS data (24 081 km2), but mean size lower (6.50 km2). Until such time as the GIS is re-examined with the view to reclassifying new and split features based on morphology and evolution, the gazetteer data remain the most useful for the examination of regional patterns. The reef classification used in the gazetteer (and carried forward into the GIS) is based on the evolutionary model proposed by Hopley (1982) (Fig. 5.6 and 5.7, Table 5.3). The development of this model, with examples, is discussed further in Chapter 8. A separate category is used for ribbon reefs. Fringing reefs (including incipient fringing reefs) are also incorporated in the analysis. The following analysis is based on data from the gazetteer, presented in Hopley et al. (1989). This is supplemented with data from the GIS as appropriate.
5.5 Reef and reef island statistics and classification
149
Figure 5.6 Evolutionary classification of coral reefs of the GBR (from Hopley, 1982).
5.5.2 Gross dimensions and numbers Of the 2904 reefs identified, 758 are fringing reefs including 213 incipient fringing (Table 5.4). The GIS identifies 754 fringing reefs (88 coastal fringing, 395 island fringing, 14 coastal incipient, and 257 island incipient). As discussed
150
Spatial analysis of the morphology of the reefs and islands
Figure 5.7 Examples of reefs in the context of the evolutionary classification. Juvenile stages: (a) reef patches just reaching sea level (Barnett Patches); (b) reef patches coalescing to form a crescentic reef (Yamacutta Reef). Mature stages; (c) fully formed crescentic reef (Lynches Reef); (d) partially open lagoonal reef (Fairey Reef); (e) lagoonal reef (Hoskyns Reef). Senile stages; (f) planar reef (Tryon Reef).
in Section 7.2, many nearshore reefs (close to but not directly attached to the mainland or high island shores) are almost certainly not recorded in either dataset. Most numerous of the non-fringing reefs are the submerged reefs, generally of small mean size, although many may be growing from common Pleistocene foundations. However, this category also includes some of the
5.5 Reef and reef island statistics and classification
151
Figure 5.7 (cont.)
largest identified features (up to 856 km2) on the mid shelf between 118 and 138 S, almost certainly a mix of reef and Halimeda banks. Of the shelf reefs with reef flat development, planar reefs are the most common, but crescentic and lagoonal reefs are the largest. The total reef area is approximately 8.4% (gazetteer) to 10% (GIS) of the total shelf area, based on the Lewis (2001) total shelf area estimate.
5.5.3 Analysis by reef type and latitude Table 5.5 contains a 18 latitudinal breakdown of reef area and shelf area (based on the gazetteer estimates). The shelf widens considerably to the south of 188 S.
152
Spatial analysis of the morphology of the reefs and islands
Table 5.3. Reef classification based on the model of Hopley (1982) (Figs. 5.6 and 5.7), with the addition of ribbon reef and fringing reef categories JUVENILE (enhancement of Pleistocene relief) (i) Unmodified antecedent platforms Pleistocene foundations without modern growth. As these cannot be differentiated from submerged reefs on aerial photographs, they are not separately included in the analysis (ii) Submerged reefs Reefs not at modern sea level but with some growth over the older foundations, usually most prolific on the highest parts of these Pleistocene foundations (iii) Irregular patch reefs Patchy reef flat development as the growth from the Pleistocene highs reaches modern sea level MATURE (horizontal extension of modern reef flats) (iv) Crescentic reefs Coalescence of patch reefs on the most productive windward margins to produce a crescent-shaped reef with open back reef area (v) Lagoonal reefs Extension of the reef flat around the margins of the foundations to enclose or partly enclose one or more lagoons SENILE (masking of the original relief) (vi) Planar reefs Infilling of lagoons by internal patch reef growth and sediment transport from windward margins to produce extensive reef flat, eventually with a widespread sediment blanket RIBBON (vii) Ribbon reefs Linear reefs growing from structurally or morphologically determined linear foundations FRINGING (viii) Incipient fringing reefs Corals growing over rocky foundations largely below low tide level, attached to mainland or continental island, but with no extensive reef flat (ix) Fringing reefs Identifiable reef flat development, attached to mainland or continental island
Table 5.4. Great Barrier Reef Marine Park reef numbers and areas, based on the classification presented in Table 5.3 Reef type
Number
Total area (km2)
Mean size (km2)
Submerged Patches Crescentic Lagoonal Planar Ribbon Incipient fringing Fringing Total a GIS total b
566 446 254 270 544 66 213 545 2 904 3 703
3 514 4 061 4 266 4 252 2 214 1 081 120 547 20 055 24 081
6.2 9.1 16.8 15.7 4.1 16.4 0.6 1.0 6.9 6.5
a b
Data from the gazetteer (Hopley et al., 1989). Data from the GBRMPA GIS datasets.
5.5 Reef and reef island statistics and classification
153
Table 5.5. Great Barrier Reef Marine Park shelf and reef areas and percentage of reef cover by latitude Latitude 8 S
Shelf area (km2)
Reef area (km2)
Reef cover (%)
108 410 –118 118–128 128–138 138–148 148–158 158–168 168–178 178–188 188–198 198–208 208–218 218–228 228–238 238–248 248–248 300 Total
5 329 13 738 7 042 7 324 11 300 6 186 7 518 8 136 17 015 27 566 26 878 38 775 26 651 14 163 6 361 223 977
835 2 883 2 582 1 280 1 606 716 825 756 1 119 2 265 2 259 2 202 289 415 23 20 055
15.7 21.0 36.7 17.5 14.2 11.6 11.0 9.3 6.6 8.2 8.4 5.7 1.1 2.9 0.4 9.0
Source: Hopley et al. (1989).
Total reef area has a bimodal distribution, being highest between 118 and 138 S and between 198 and 228 S. Highest percentage cover is between 118 and 128 S, where over one-third of the shelf is composed of reefs or Halimeda banks (Fig. 6.5). South of 178 S, reef cover is less than 10% of the shelf. However, this is strongly influenced by the wide and deep mid-shelf, almost devoid of reefs, with some areas (particularly the Pompey and Swain Reefs, 208 to 228 S), having high reef size and density. Total reef numbers (Table 5.6) are correlated with shelf area, and reach a maximum between 218 and 228 S. As previously discussed, consideration of mean size is strongly influenced by the interpretation of what constitutes an individual reef. The high mean areas between 118 and 138 S (Table 5.7) are strongly influenced by the large submerged features, now interpreted as predominately Halimeda banks. There is considerable variation in reef type by latitude (Hopley, 1982; Hopley et al., 1989). Figure 5.8a and Table 5.6 show numbers of reefs of different types by latitude, and Fig. 5.8b and Table 5.7 shows reef areas by type and latitude. Figure 5.9 shows the modal reef type in 0.58 latitude by longitude ‘‘squares’’. The distribution of reefs by type over the entire GBRMP is shown in Fig. 5.10.
0 0 1 0 4 0 0 3 7 16 19 80 76 0 7 213
108 410 –118 118–128 128–138 138–148 148–158 158–168 168–178 178–188 188–198 198–208 208–218 218–228 228–238 238–248 248–248 300 Total 28 22 17 8 38 13 24 15 35 12 206 52 2 73 0 545
Fringing 24 65 42 53 33 14 23 22 25 53 88 85 27 9 3 566
Submerged 22 81 35 36 17 20 20 15 30 68 31 55 16 0 0 446
Patches
The most abundant type for each degree of latitude is indicated in bold. Source: Hopley et al. (1989)
a
Incipient fringing
Latitude 8 S 11 14 6 2 15 11 11 22 27 39 16 64 16 0 0 254
Crescentic 1 5 9 7 1 1 0 0 4 37 61 129 8 7 0 270
Lagoonal
Table 5.6. Frequency of reef types by latitude in the Great Barrier Reef Marine Park a
43 69 34 40 63 32 16 0 7 28 44 132 20 15 1 544
Planar 0 9 14 5 12 19 0 1 0 1 2 2 1 0 0 66
Ribbon
129 265 158 151 183 110 94 78 135 254 467 599 166 104 11 2904
Total
0.0 0.0 0.1 0.0 0.4 0.0 0.0 0.1 3.6 5.6 2.3 17.2 41.4 47.1 1.9 119.7
14.6 45.8 46.8 31.4 95.8 17.5 40.7 53.3 40.1 16.4 107.6 32.6 4.4 0.0 0.0 547.0 341.7 881.7 1 527.0 223.5 11.0 5.1 14.5 21.3 17.7 134.9 86.0 148.3 28.5 53.7 19.5 3 514.4
199.9 1 126.7 479.0 224.2 43.9 165.1 98.6 215.1 371.2 685.4 187.0 204.6 61.4 0.0 0.0 4 062.1
The maximum area for each latitude is indicated in bold. Source: Hopley et al. (1989).
a
108 410 –118 118–128 128–138 138–148 148–158 158–168 168–178 178–188 188–198 198–208 208–218 218–228 228–238 238–248 248–248 300 Total 84.5 366.0 259.7 54.4 670.2 254.1 487.6 459.7 573.4 467.3 104.1 410.6 74.6 0.0 0.0 4 266.2
27.5 30.3 38.2 224.6 13.4 17.0 0.0 0.0 67.6 894.4 1 577.6 1 217.9 34.0 109.4 0.0 4 251.9
835.2 2 882.9 2 582.1 1 280.4 1 606.1 715.7 825.4 755.7 1 119.2 2 265.2 2 259.2 2 201.5 288.6 415.2 23.3 20 055.7
6.5 10.9 16.4 8.5 8.8 6.5 8.8 9.7 8.3 8.9 4.8 3.7 1.8 4.0 2.1 6.9
8.1 11.6 18.1 8.7 10.7 7.2 11.2 11.7 11.5 9.9 8.9 4.6 2.7 11.7 5.4 9.0
Mean area excluding Ribbon Total area Mean area fringing (km2) (km2) (km2) (km2)
167.0 0.0 209.3 223.1 116.0 115.3 421.0 101.3 433.1 338.3 72.3 184.6 184.0 0.0 0.0 6.2 25.6 20.0 39.3 21.9 147.7 46.9 160.9 9.4 29.9 14.4 205.0 0.0 1.9 0.0 2 213.0 1 081.4
Incipient fringing Fringing Submerged Patches Crescentic Lagoonal Planar Latitude 8 S (km2) (km2) (km2) (km2) (km2) (km2) (km2)
Table 5.7. Areas of different reef types and mean reef size in the Great Barrier Reef Marine Park by latitude a
156
Spatial analysis of the morphology of the reefs and islands
Figure 5.8 Reef distribution by latitude: (a) reef numbers and (b) reef area.
5.5 Reef and reef island statistics and classification
Figure 5.9 Modal reef types by 0.58 latitude by longitude squares (after Hopley et al., 1989).
157
158
Spatial analysis of the morphology of the reefs and islands
Figure 5.10 Distribution of reefs by reef type.
5.5 Reef and reef island statistics and classification
159
The distribution of reefs shows considerable morphological complexity, but regional patterns are clear. Incipient fringing reefs are concentrated inshore at the southern end of the GBR, close to the southern limit of coral growth (Chapter 7). Fringing reefs are widely distributed, but are most numerous towards the southern end of the reef province where offshore high islands are more numerous, as throughout the GBR fringing reefs attached to the mainland are relatively rare. The Whitsunday Islands are particularly important for fringing reef development. Only in the area from Cairns to Princess Charlotte Bay are there a significant number of mainland attached fringing reefs (Smithers et al., 2006; Chapter 7). Submerged reefs are frequently difficult to identify. The GIS database contains many newly recognized features almost certainly of this type. In the northern GBR, submerged shoals are the modal reef type, typically on the outer shelf, but many are likely to be Halimeda banks, probably overlying older reef foundations. Submerged reefs are also prominent towards the southern end of the GBR, where they commonly occur as shelf edge shoals from the south of Townsville (Hopley, 2006; Chapter 9). Reef patches are a juvenile reef form particularly common between 118 and 128 S and around 198 to 208 S, on the mid to outer shelf. They are low in number between 148 and 188 S and south of 228 S. Crescentic reefs are the dominant form over the central GBR, being the modal type by area from 148 to 198 S. They are a significant feature around Lizard Island, north of Cooktown, and examples are found in the northernmost GBRMP, and in the Swain Reefs. Lagoonal reefs have a distinctive and limited distribution, being restricted to the south of 198 S (particularly in the Pompey and inner Swain complexes), with a smaller cluster between 138 and 148 S. In area, they dominate reef types between 198 and 228 S. Planar reefs are found north of 168 S and south of 208 S. There are five areas of concentration: near Cape York, north of Princess Charlotte Bay, near Cairns, in the northern Swains, and in the Capricorn–Bunker group. The distribution is probably correlated with reef size as smaller reefs achieve this senile form very quickly (Hopley, 1982, 1983c; Chapter 8). Ribbon reefs are effectively limited to the shelf edge between 118 and 168 S. They occur where the shelf is narrow and where the shelf-edge water depths are less than about 50 m. They occupy the top of the drop-off at the shelf edge, with water depths within a few hundred meters of the reef edge frequently exceeding 1000 m. The few reefs classified (by shape) as ribbons to the south of 168 S are typically not shelf-edge, but are formed on the edge of older reefal foundations. Genetically, they are more similar to crescentic reefs.
160
Spatial analysis of the morphology of the reefs and islands
Hopley et al. (1989) undertook a cluster analysis of reefs to determine degrees of similarity between 0.58 squares as defined in Fig. 5.9. A number of methods and data presentations (using combinations of area, number, and reef type presence or absence) were attempted. At a six-cluster level, a high coefficient of similarity was maintained, with clusters being easily identified with fringing reefs, and the juvenile, mature, and senile reef types as outlined in Table 5.3. At higher cluster levels, regional fragmentation of the basic divisions occurred.
5.5.4 Analysis by island type and latitude Perhaps surprisingly, island data held within the GIS lack classification attributes, with the morphological data from the gazetteer not being carried forward, and with some cay features not being included at all. A Queensland Island Inventory is currently being developed by the Queensland Department of Natural Resources and Water, although at this stage, reef island classification is not incorporated, and cays are not well represented. The gazetteer assessment of reef islands, as reported in Hopley et al. (1989), remains the most complete assessment of reef islands in the GBRMP. There are 617 high islands (over 50% of which occur between 208 and 228 S, being the Cumberland and Northumberland Groups, including the Whitsunday Islands) with reefs. There are 300 reef islands (with a few reef platforms having more than one island (Table 5.8)). Four reef island types are identified (see Section 10.2 for a detailed description): *
*
* *
Unvegetated sand and shingle cays, continuously mobile and generally below the level of HAT (e.g., Figs. 8.4b and 13.10a). Vegetated sand cays. As mapping was carried out from aerial photography without ground-truthing, it is likely that many islands classified as unvegetated, particularly in the Swain Reefs, may have some vascular plants. This estimate is therefore likely to be low (e.g., Figs. 13.6, 13.8, 13.9a, and 13.10b). Vegetated shingle cays, usually on windward margins (e.g., Fig. 8.8b). Low wooded islands. These are complex island forms normally consisting of windward shingle ramparts, reef top mangroves, and a leeward sand cay (e.g., Fig. 8.3a).
The large majority of reef islands are found on planar reefs (the most senile form), although all types of reef may incorporate an island, with even fringing reefs occasionally supporting an unvegetated sand cay. Table 5.8 indicates two cases of fringing reefs with low wooded islands (Howick Reef and Clack Reef). These cases are planar reefs with small outcrops of continental rock.
161
5.5 Reef and reef island statistics and classification
Table 5.8. Numbers of reef islands by reef type in the Great Barrier Reef Marine Park Island type Reef type
Unvegetated cay a
Vegetated cay a
Low wooded island
Total
%
Reef patches Crescentic Lagoonal Planar Ribbon Fringing Total %
42 14 11 135 4 7 213 71.0
0 2 0 41 0 0 43 14.3
0 0 0 42 0 2 44 14.7
42 16 11 218 4 9 300 100.0
14.0 5.3 3.7 72.7 1.3 3.0 100.0
a
Approximately 3% of cays are classified as shingle cays. Source: Hopley et al. (1989).
Figure 5.11 Reef island distribution by latitude. Shingle cays are indicated in gray.
Planar reefs, and occasionally reef patches, have the accommodation space and the required morphology to produce a wave refraction pattern conducive to cay formation, able to concentrate sediment deposition in a particular part of the reef (Gourlay, 1988, 1994; Chapters 4 and 10). Such wave patterns are not normally found on lagoonal, crescentic, or ribbon reefs. The latitudinal distribution of islands (Fig. 5.11) largely reflects the distribution of planar reefs, with most north of 168 S and between 218 and 228 S (in the Swains
162
Spatial analysis of the morphology of the reefs and islands
complex). There is a cluster of islands all of which are vegetated in the Capricorn–Bunker group between 238 and 248 S. There are no low wooded islands south of 178 S. The central GBR (between 168 and 218 S) has very few cays and no vegetated cays, again reflecting the paucity of planar reefs. 5.6 Reef types and reef management The GBRMPA has recently implemented a new zoning plan which called extensively on the GIS in the planning process (Lewis et al., 2003b). A review of zoning in the early 1990s indicated that the amount of protected or ‘‘no-take’’ area was inadequate and the distribution was not optimal to ensure protection of the marine biodiversity (Fernandes et al., 2005). At the time of the review, only 4.5% of the GBRMP was in a no-take zone, and over 80% of these zones were around reefs, with other habitat types virtually unrepresented. A process commenced to define ‘‘bioregions.’’ A total of 30 reef and 40 non-reef (including eight deep-water off shelf) bioregions were identified, reflecting cross-shelf and latitudinal variability, based on a range of biophysical inputs, including the reef type classification of Hopley (1982). The establishment of no-take zones (known locally as ‘‘Green Zones’’) based on the bioregions (the Representative Areas Programme) was guided by a number of biophysical operational principles including (a) defining minimum dimensions (20 km long on the smallest dimension, except for along the coast), (b) establishing fewer large, rather than more small areas, (c) avoiding fragmentation of individual reefs into more than one zone, (d) having sufficient no-take areas in a bioregion to ensure against negative impacts in some part of the bioregion, (d) establishing minimum amounts of no-take zone for each reef and non-reef bioregion, (e) representing all habitats, (f) maintaining cross-shelf and latitudinal diversity, (f) providing for known connectivities (currents, migrations, etc.) (g) protecting special or unique places (Day et al., 2003; Fernandes et al., 2005). An extensive process of analysis and consultation followed, which resulted in the gazetting in 2003, and establishment on 1 July 2004, of the new zones. Table 5.9 shows the areas and proportions of each zone within the GBRMP, and their nearest equivalent IUCN categories. A detailed description of the activities permitted and restricted in the zones can be found in Great Barrier Reef Marine Park Authority (2004). Overall, approximately 34% of the GBR is in no-take zones (corresponding to IUCN categories IA and II), with a further 4% with lesser, but still a significant degree of protection (IUCN category IV). An analysis of reefs of different types in relation to no-take zones could only be undertaken for those features in the GIS database for which a reef type has
163
5.6 Reef types and reef management
Table 5.9. Great Barrier Reef Marine Park zones Zone Preservation Marine National Park Scientific Research Buffer Conservation Park Habitat Protection General Use Total
Nearest IUCN category IA II IA IV IV VI VI
Area (km2) 710 114 530 155 9 880 5 160 97 250 116 530 344 215
Percentage of the GBRMP <1 33 <1 3 1 28 34 100
Source: Great Barrier Reef Marine Park Authority (2004).
been recorded, therefore the total of 2912 is less than the total number of features with a reef identifier (2977) in the GIS. As discussed in Section 5.3, the analysis based on the GIS data excludes many split features and a small number of newly identified reefs. However, as the results are expressed as percentages, the results are likely to reasonably reflect the whole dataset. Table 5.10 shows the numbers and areas of reefs of the different types that are in no-take zones (defined as being Preservation (IA), Scientific Research (IA), or Marine National Park (II) zones), and in no-take and limited extraction zones (as above but including Conservation Park (IV) and Buffer (IV) zones, in which some types of fishing are permitted). The selection of features was based on the center of the reef being within the no-take zone. Other alternatives tried were, contained within (giving slightly smaller totals) or intersects with (giving a slightly larger estimate). The variation is not substantial, primarily because of the principles by which zones were established, which minimized splitting reefs between zones. A high proportion of fringing reefs are in protected zones. Incipient fringing reefs are less well represented. The juvenile reefs (submerged and reef patches) have close to the mean representation for the GBR. Planar and ribbon reefs have a higher than average proportion in no-take zones. A large number of these reefs have features such as islands (that may have significant conservation, historic, or visitor value) or are of types that may have been included due to their uniqueness or special attributes, the ribbon reefs being a good example. The inclusion of special or unique places was one of the operational principles used in the decision-making process.
86 391 14 255 564 450 256 583 541 72 2 912
Coastal fringing Fringing Coastal incipient Incipient fringing Submerged Reef patches Crescentic Lagoonal Planar Ribbon Total
Source: GBRMPA, GIS.
Number
Reef type 231 553 9 128 3 472 4 289 5 465 4 746 2 788 1 235 22 916
Area km2 23 23 29 17 24 40 29 29 40 43 30
Percentage in no-take zones (by number) 37 51 27 19 34 34 23 24 42 54 32
Percentage in no-take zones (by area) 40 57 71 25 27 43 32 31 43 44 38
Percentage in no-take and limited extraction zones (by number)
57 73 76 24 34 34 29 29 51 54 36
Percentage in no-take and limited extraction zones (by area)
Table 5.10. Percentage of classified reefs in the Great Barrier Reef Marine Park Authority GIS by number and area in no-take zones (Preservation, Scientific Research, and Marine National Park) and in no-take and limited extraction zones (Preservation, Scientific Research, Marine National Park, Buffer, and Conservation Park)
5.7 Conclusion
165
A long-term view of GBR management would suggest that concentration on the younger (juvenile and mature) reef types would be sensible. Senile reefs are sediment dominated, with only a small proportion of their area coral covered, with little accommodation space. Mature reef types have the greatest range of habitats, while the juvenile forms have high accommodation space and are less likely to be affected by water temperature induced coral bleaching. These ideas are further explored in Chapter 13. 5.7 Conclusion The spatial distribution of reefs shows considerable geomorphologic complexity, but it is clear that reef form is not random, with latitudinal and cross-shelf patterns. The reasons for these regional patterns are explored in the context of reef evolution in the following chapters, particularly Chapter 8 which examines the mid-shelf reefs. Obvious regional spatial patterns of reef distribution that were recognized in the 1970s and summarized in Hopley (1982) provided a basis for many of the research questions to investigate reef evolution that have been tackled in the past 30 years. The research has tested the descriptive model presented in Table 5.3 and Fig. 5.6 and it has proved to be robust as a descriptive tool, but it also provides a basis for the understanding of reef evolution, particularly the importance of the nature of the pre-Holocene foundations on reef morphology. An understanding of the spatial patterns of reef morphology in an evolutionary context is important for the long-term management of the GBR in a period of climate change, as present morphology will significantly influence future morphology and the characteristics of reef habitats. Huge advances in reef mapping and data storage and analysis capabilities have occurred in recent years that should make this task easier, but the full benefit of the technology has yet to be realized for the analysis of reef morphology, with much of the attribute information on reef morphology (and by implication, reef evolution) held in the gazetteer being only loosely transferred into the widely used GIS. The GIS has split many reef features into smaller units on the basis of deep-water channels, essentially establishing reefs as ‘‘surface’’ features, without considering reasonable interpretations of reef features that share common pre-Holocene foundations. The importance of the nature of the foundations and subsequent growth is discussed in detail with respect to different reef environments in the subsequent chapters and the importance of this understanding to reef management is revisited in Chapter 13.
6 The non-reefal areas of the continental shelf
6.1 Introduction In Chapter 2 the processes involved in the evolution of the continental shelf on which the Great Barrier Reef (GBR) is located were described. They involved alternating phases of offlapping terrigenous sedimentation and emplacement of onlapping marine facies. The period since the last interglacial is the latest of these episodes, the fall in sea level after 120 000 years BP exposing the shelf to varying degrees, with maximum exposure about 19 ka years ago (Chappell et al., 1996; Section 3.3.2) placing the shoreline on the shoulder of the shelf. Subaerial processes dominated most of the shelf for most of the period between the last interglacial and the rapid postglacial drowning after 19 ka years. However, it is during this last episode of drowning and subsequent Holocene stillstand (Chapter 3) that marine and nearshore processes have dominated and have produced the present pattern of superficial sediments. This chapter first describes the surface sediments of the shelf, then examines their three-dimensional properties and the nature of the late Pleistocene surface over which they have been deposited. Special attention is given to the shelf bioherms formed by calcarous algae, the Halimeda banks, the presence and extent of which has only become apparent over the last 25 years. Much of the information on surface sediments has been available since the publication of Maxwell’s classic Atlas in 1968. Subsequent studies have involved extensive seismic survey of the region, vibro-coring and visual survey of features such as the Halimeda banks by manned submersible and remotely operated vehicle (ROV), in addition to specific targeted sediment studies. The information gathered is complementary to that on the coral reefs which is discussed in Chapters 7, 8, and 9 and the interaction between the reefs and the adjacent shelf areas as discussed in Chapter 12. 166
6.2 Surficial sediments
167
6.2 Surficial sediments A major contribution of Maxwell’s (1968) Atlas of the Great Barrier Reef was the first description of the sediments of the reefs and surrounding continental shelf along the entire GBR. His classification (Maxwell, 1968, 1973b) was based on carbonate content and included: * * * * *
high carbonate facies (>80% carbonate) impure carbonate facies (60–80% carbonate) transitional facies (40–60% carbonate) terrigenous facies (20–40% carbonate) high terrigenous facies (<20% carbonate).
Further differentiation was made on the basis of the sand : mud ratio and organic and mineral components. So successful was this approach that it became the standard for many subsequent regional studies such as Maxwell and Swinchatt (1970), Frankel (1974) in Princess Charlotte Bay, Flood et al. (1978) in the area north of Lizard Island, Orme et al. (1978a, b) in the lee of the northern ribbon reefs, Orme and Flood (1980) in their reassessment of the sediments of the entire GBR, and the work of Harris (1988, 1991, 1995) on the far northern GBR adjacent to Papua New Guinea. From these studies important factors can be identified which determine the surficial sediment distributions on both regional and local scales. The general pattern is a cross-shelf one with two end members, a western terrigenous unit sourced from the mainland and an eastern shelf marginal carbonate unit sourced from the reefs. However, with a shelf width between 50 and almost 300 km the zone of influence of each of these end members is relatively small. Flood plumes related to even major events are generally retained within 20 km of the shoreline (see Devlin et al. (2001) for actual examples) and deposition of new sediments may be restricted to about 10 km from shore. For example, the floods associated with major Cyclone Winifred in 1986 produced a storm layer up to 30 km offshore in water depths of up to 43 m, but much of this was reworked material and newly deposited sediment extended only 12 km offshore (Gagan et al., 1988; Johnson, 1996). Distribution of sediments from the reefs may be even more restricted. Maxwell and Swinchatt (1970) suggested most reef productivity was incorporated into the reef structure and other studies have shown that reef-derived sediments rarely extend more than 2 km from a reef. For example, Flood et al. (1978) describe concentric elliptical patterns 2 km wide around reefs of the Howick Islands on the northern GBR. An exception to this pattern is found on the south central GBR, inside the Pompey Reefs where the impure carbonate zone reaches a maximum width
168
The non-reefal areas of the continental shelf
of 50 km and the high carbonate facies extends 25–40 km from the reefs (Maxwell, 1968). Whilst not all of this may be reef derived, Maxwell believed that the major influences were the presence of the largest area of reef within the entire GBR province (almost 5000 km2 of reef representing 14.1% of the total reef area within the GBR: Hopley et al., 1989) and the strong tidal currents reaching velocities of >4 m s1 as a result of a tidal range which can exceed 4 m (see Chapter 9). Maxwell (1970) describes the processes involved on the western side of the Pompey Reefs with scouring or deposition of isolated gravels close to the reefs and deposition of presumably reef-derived high carbonate muddy sand extending more than 20 km towards the west. Elsewhere the density of reefs on the outer shelf is insufficient to produce a high sediment cover and on the central GBR at least, the open shelf between reefs and more gently sloping shelf margin appears to favor the loss of sediment eastwards into the Queensland Trough (e.g., Dunbar and Dickens, 2003). Given the relative narrowness of the end member sediment units, the width of the shelf has a strong influence on local cross-shelf patterns (Fig. 6.1). Maxwell (1968, 1973b) describes the central shelf as an axial zone of impure carbonate and transitional facies. In the south this zone may be up to 50 km wide but in the north it may be almost non-existent due to the dispersal of terrigenous mud across the narrow shelf into zones elsewhere dominated by high carbonate sediments (Maxwell, 1973b). The inner shelf terrigenous sediment zone has been accumulating since modern sea level was achieved about 6.5 ka ago. It may contain up to 80% quartz and especially in the south is sand dominated. In the wet tropics, however, mud becomes much more dominant, a reflection of the more intensive weathering of the hinterland. In the far north, the inner shelf is composed largely of muddy sand. Distribution along the shelf allowed Maxwell (1973b) to identify the origin of the sediments to six groups of specific river catchments. Similar conclusions have been reached more recently by Lambeck and Woolf (2000). Examining the composition and textural variability of sediments along the 10-m isobath from Bowen to Cairns, they identified six zones related to sediment source. Each zone shows a northward trend of sediment maturity but with significant mixing south of the Tully River. To the north input from individual rivers is maintained close to source. North-facing bays were identified as having limited transport, and contain spikes of unmixed sediments. Significantly Lambeck and Woolf (2000) believed that ambient conditions, not cyclonic weather, were responsible for the patterns of sediment transport. An anomalous area around the Whitsunday Islands was recognized by Heap et al. (2002). The islands are surrounded by a thick sediment deposit including 1850 380 tonnes of Holocene siliciclastic sediments deposited in
6.2 Surficial sediments
Figure 6.1 Contrasting distributions of shelf sediments in the far north and southern Capricorn Channel areas of the Great Barrier Reef (after Maxwell, 1968; Frankel, 1974; Orme and Flood, 1980; Flood and Orme, 1988).
169
170
The non-reefal areas of the continental shelf
the last 6.5 ka. This is too great an amount to be accounted for by present discharge onto the inner shelf. It is suggested that the deposits, between 5 and 30 m thick, were deposited on the mid-shelf by early Holocene rivers and subsequently transported northwards into the vicinity of the islands, which have complex indented coastlines and have acted as major sediment traps. Heap et al. (2002) believe that there is significant accumulation of siliciclastic sediment around most continental islands of the middle shelf, providing a further zone of terrigenous sediment. At the northern end of the GBR is an input of fluvial sediment which exceeds that of all the rivers of Australia combined (Harris, 1988, 1991, 1995; Harris et al., 1993). The Fly River, emptying into the Gulf of Papua, has a discharge of 7000 m3 s1 and an approximate fluvial sediment export of 115 million tonnes yr1. Although about 50% of this is delivered through the delta to the adjacent shelf most travels north-eastwards along the Papua New Guinea coastline. A deltaic mud facies containing annual varves 2–5 cm thick progrades south-easterly from the delta at a rate of about 6 m yr1. It is within 10 km of Bramble Cay at the northern end of the GBR. Buried reefs occur north and east of Bramble Cay but Harris (1991) believes that these were buried by fluvial sediments at a low sea-level stage. The termination of the northern end of the GBR by the Fly River sediments, as has been speculated, is therefore indirect with Pleistocene foundations for Holocene reef growth being buried well before the Holocene transgression. With low-salinity waters trapped against the Papua New Guinea coast and very little terrigenous deposition south of Bramble Cay, the sea-floor sediments opposite Torres Strait show a similar pattern to elsewhere in the GBR (Harris, 1988, 1995). However, strong tidal currents play a greater role in the dispersal of sediments than anywhere else in the GBR except for the Pompey Reefs. Scouring close to channels in central Torres Strait has left bare limestone pavements of presumably Pleistocene age. Away from the channels sediments are graded from lag gravels to sands and gravels and, at greatest distance, fine silts and clay-size particles. In the lower-energy areas behind the reefs, and the shelf-edge deltaic reefs (see Chapter 9) is a muddy carbonate facies. Sediments are over 80% carbonate, derived from benthic foraminifera with bryozoans, mollusks, and Halimeda. Corals make up only a small proportion of the sediments. Terrigenous mud is largely derived from the reworking of the underlying Pleistocene clay deposits. Maxwell’s (1968, 1973b) axial transitional and impure carbonate zones make up by far the largest area of the continental shelf but without a major identifiable source of sediment, the rates of sedimentation are demonstrably low. In some areas, seismic profiles identify no Holocene sediments overlying
6.3 Subsurface sediments and the Pleistocene surface
171
the late Pleistocene surface (see below). The autochthonous production of biogenic sediments is the main source of sediments. Both Maxwell (1973b) and Orme and Flood (1980) indicate foraminifera as the most abundant of all organic sediments followed by mollusks. Bryozoans also make up a significant proportion of the sediments especially in the south where Orme and Flood (1980) believe they may represent relict deposits. At the time of Maxwell’s initial work the Halimeda bioherms, described later in this chapter, had not been discovered. Nonetheless, Maxwell (1968, 1973b) recognized areas in the Swain Reefs where Halimeda contributed over 60% of the sediment. Later work (e.g., Orme et al., 1978b; Orme and Flood, 1980; Flood and Orme, 1988) showed that the Halimeda content could be as high as 80% in areas behind the northern ribbon reefs and Scoffin and Tudhope (1985) delineated a high Halimeda (>40%) zone in water depths between 60 and 100 m in the central region of the GBR. The full extent of these Halimeda deposits is discussed later in this chapter. The central GBR Halimeda beds almost certainly represent a relict deposit. Given the depth of the GBR shelf and the lack of current velocities sufficient to entrain many of the sea-floor deposits, relict sediments are believed to make up a significant proportion of the shelf cover. Maxwell and Swinchatt (1970) mapped shallow water sands and gravels around Green Island off Cairns containing recrystallized carbonate bored by algae and sponges. Maxwell (1968) believed that many of the accumulations of quartz, bryozoans, and mollusks were influenced by ancient strandline controls. These include concentrations of quartz in the northern Pompey Reefs and around Redbill Reef in the south central GBR. Frankel (1974) identified linear concentrations of terrigenous and ironstained abraded carbonates and quartz sands in Princess Charlotte Bay originating as dune or beach deposits of a lower sea-level period. Ooids are rare in the GBR and in at least one of the two areas in which they have been found they are almost certainly relict. Diagenetically altered high-magnesium calcite ooids of early Holocene age have been dredged from depths of between 100 and 120 m in the Capricorn Channel (Marshall and Davies, 1975). Diagenetically unaltered ooids with quartz nuclei have also been described from channel sands lying between Lizard and Palfrey Islands in the northern GBR (Davies and Martin, 1976). 6.3 Subsurface sediments and the Pleistocene surface Over the past 25 years more than 6500 km of high-resolution seismic reflection profiling has been undertaken in GBR waters in some places complemented by vibro-coring and other sediment retrieval methods. The surveys cover the
172
The non-reefal areas of the continental shelf
entire GBR from the far north (Harris et al., 1993) to the Bunker–Capricorn Group (Davies et al., 1981). Individual legs have been taken longitudinally, parallel to the coast (e.g., Searle et al., 1980, 1982; Johnson and Searle, 1984) and as cross-shelf transects (e.g., Orme et al., 1978a; Graham, 1993). All environments of the GBR shelf have been covered including inshore (e.g., Johnson and Carter, 1987; Larcombe and Carter, 1998), Maxwell’s mid-shelf axial zone (Johnson and Searle, 1984; Heap et al., 2002; Fielding et al., 2003) and the outer reef and shelf-edge zones (e.g., Davies et al., 1981; Graham, 1993; Hopley, 2006; Chapter 9). This large data bank provides for a comprehensive assessment of the shelf area which was exposed during the last glaciation, the drainage patterns across this surface and the sediments deposited over it during the postglacial rise in sea level. 6.3.1 The Pleistocene surface and postglacial sediment cover of the mid-shelf The surface of the continental shelf exposed during the last glaciation closely corresponds to the modern-day shelf with only relatively thin veneers of sediment. Exceptions are around the coral reefs of the outer and mid-shelf regions, within river channels incised into the Pleistocene surface and close to the present coastline. The slope, away from the edge of the continental shelf, is modest, ranging from less than 0.5 m km1 to a maximum of about 4.3 m km1 (Searle and Hegarty, 1982; Carter et al., 1993). The steepest slope occurs where the shelf is narrowest. When penetrated the top of the Pleistocene is similar to the surface of the adjacent coastal plains, i.e., a stiff oxidized clay often containing calcareous nodules (e.g., Davies et al., 1983; Carter et al., 1993; Graham, 1993). This forms an ubiquitous and very distinct seismic reflection surface, often referred to as ‘‘Reflector A’’ following its first description by Orme et al. (1978a, b). In many places the surface has little or no sediment cover especially on the marginal shelf (e.g., Orme et al., 1978a) and across much of the mid-shelf axial zone (e.g., Johnson and Searle, 1984). In the latter case the very gentle shelf slope, especially in the central and southern GBR, combined with the shelf depth (25–45 m) are such that the postglacial transgression would have passed over this area extremely quickly, the shoreline position changing by as much as 5 km per century. Relict sediments dating back to the transgression are not large in volume and autochthonous production of modern biogenic sediments is very low. Elsewhere, away from the channel fills, typical postglacial sediment veneers are <2.5 m (Johnson and Searle, 1984; Heap et al., 2002) perhaps a little thicker on the narrower northern shelf. For example,
6.3 Subsurface sediments and the Pleistocene surface
173
Searle et al. (1982) quote maximum thicknesses of 20 m for the northern shelf, 14 m for the central GBR, and 8 m for the southern GBR, corresponding to shelf width, with a greater proportion of the shelf covered by nearshore terrigenous sediments and the biogenic sediments of the outer reef tract where the sediment cover is thickest. Where sediments are a little thicker they have been described as a transgressive veneer in the form of a sheet drape with occasional megaripples (Searle and Hegarty, 1982; Johnson and Searle, 1984; Graham, 1993). Vibro-cores have been obtained from the mid-shelf of the central GBR (Davies et al., 1983; Harris et al., 1990) and for Torres Strait (Harris et al., 1993). On the central GBR beneath a thin veneer of modern (?) bioclastic sand and gravel typically there is 1 m of light quartz sand, fining upwards, which in turn overlies >3.5 m of terrigenous mud or an olive-gray shelly gravel or up to 1.8 m of quartz sand. Within Torres Strait the sediments follow a similar sequence but appear to have an overall greater mud content. In both locations muddy sands and mangrove peats may patchily underlie the other transgressive sediments. 6.3.2 The outer reef tract Most research on the outer shelf has been carried out on the reefs themselves rather than the inter-reefal areas. Up to 30 m of Holocene reef has been added to Pleistocene foundations during the Holocene, the focus especially of Chapters 8, 9, and 12. However, whilst most of the area of Pleistocene reefal foundations may have been overgrown with a Holocene veneer, there are numerous examples where the Pleistocene reef has been larger than its Holocene equivalent. This was first noted for Pixie Reef near Cairns by Orme et al. (1978a), who also suggested that the shelf marginal reefs may have been more continuous in the past (Orme et al., 1978b) (see Fig. 12.3). Other examples are Helix Reef off Townsville (Searle et al., 1982) and Viper Reef a little further south where Davies et al. (1983) describe larger platforms now covered by Halimeda gravels at least 3 m thick. In the Bunker–Capricorn Group, Northwest Reef and Broomfield Reef were once a more extensive single reef joined together by Isbell Shoal (Davies et al., 1981). The Halimeda banks of the outer shelf, discussed later in this chapter, may also overlie Pleistocene reefal features. Around the modern reefs, biodetrital deposits are usually limited to the lee and lateral edges of the structures. Seismic reflection profiling has suggested maximum thickness of between 6 m (Searle and Hegarty, 1982) and 13 m (Searle, 1983) and extending no more than 1–2 km from the reefs. An exception as discussed in Chapter 9 is on either side of the Pompey Reefs where, because of high-velocity tidal currents, reefal sediments up to 10 m thick extend more
174
The non-reefal areas of the continental shelf
than 3 km from narrow reefal channels (Hopley, 2006). Cores collected from the outer shelf in close proximity to the reefs exhibit only one sediment facies, a high-carbonate, muddy sandy gravel (Harris et al., 1990). Where sampled, on the outer edge of the shelf off Townsville, the depth of sediment was 3.6 m, resting on an indurated Pleistocene limestone. Halimeda and benthonic foraminifera were the dominant components. Similar unconsolidated calcareous gravelly muddy sands up to 1.5 m in thickness also occur adjacent to the reefs of Torres Strait (Harris et al., 1993). Although morphological evidence of major fluvial channeling is lacking in the seismic surveys undertaken on the edge of the continental shelf (see below), Davies et al. (1983) describe up to 1.8 m of estuarine and fluvial sediments near Helix Reef and in the Flinders Passage area. Transgressive carbonate sediments typical of the reefal areas overlie up to 2 m of clay or fine to medium quartz sand of either fluviatile or deltaic origin, deposited when sea level was at its lowest during the last glaciation. 6.3.3 The nearshore zone In contrast to the mid-shelf area of the GBR which experienced a transgression by the sea in the late Pleistocene and Holocene times too rapid for significant sedimentation in any one location, the stillstand of the last 6500 years has resulted in the major accumulation of Late Holocene sediments from both fluvial and offshore sources. On the mainland these are seen as large sequences of beach ridges, dunes, and spits in higher-energy areas and mangrove swamps, salt-flats, and cheniers in lower-energy locations (Figs. 2.5 and 2.6). However there are also significant deposits immediately offshore. Along much of the coast the immediate sublittoral zone consists of high-energy, shallow-water sand bodies up to 12 m in thickness (Johnson and Searle, 1984) and extending a few hundreds of meters offshore (Graham, 1993). Beyond this is an even more extensive and continuous mud wedge. Larcombe and Carter (1998) for example show it as a continuous zone from Cape Upstart to the Herbert River delta on the central GBR, a distance of over 200 km. Its thickness is typically between 10 and 20 m (Johnson and Searle, 1984; Johnson and Carter, 1987), its distance offshore depending to a large extent on the proximity of a fluvial source. For example, near Cape Tribulation, Johnson and Carter (1987) show the wedge, about 10 m thick, extending about 3 km out to sea near the Cape, but to 11 km in the south close to the mouth of the Daintree River. Down to about 10 m of water depth the surface sediments appear to be actively moving (Lambeck and Woolf, 2000) but in places this sediment body extends into
6.4 Low sea-level drainage patterns
175
20–25 m depth (Graham, 1993; Larcombe and Carter, 1998) below wave base even in cyclonic weather. 6.4 Low sea-level drainage patterns There have long been speculations on the river patterns across the Queensland shelf at low sea-level stages. As Maxwell (1968) noted, most of these (e.g., Saville-Kent, 1893; Hedley, 1925d; Bryan, 1928; Steers, 1929) linked passages on the shelf edge to the larger streams on the adjacent coastline. The most elaborate of these reconstructions was that of Maxwell (1968) himself who recognized five major drainage groups based entirely on the modern bathymetry and reef distribution. However, from the very first seismic profiles across the GBR shelf undertaken in 1973 by Orme and colleagues, it was clear that such a simple relationship between modern bathymetry and ancient drainage did not exist (Orme et al., 1978a, b). Major incisions into the shelf representing old river channels were shown to be buried beneath more recent sediments and, in most cases, showed no evidence of their presence in the morphology of the seabed (Fig. 6.2). All subsequent investigations have come to the same conclusions (e.g. Searle, 1983; Johnson and Searle, 1984). Subsequently, there has been discussion as to
Figure 6.2 Infilled channel near Rib Reef off Townsville, probably the Pleistocene Herbert River. Seven meters of flat-bedded sediments overlie the channel, which is incised about 20 m into the Pleistocene surface and infilled with estuarine and fluvial sediments (after Searle et al., 1980; Hopley, 1982).
176
The non-reefal areas of the continental shelf
the style of fluvial morphology as it extends across the shelf during periods of major regression. Schumm (1993) has modeled river response to base level change and Graham (1993) has applied his conclusions to the observed features on the north central GBR shelf. Three main categories of channels are observed by Schumm: * * *
deeply incised master channels forming the conduits of major rivers smaller valleys that reflect distributaries or smaller streams crossing the shelf very small channels which form part of dendritic patterns wholly limited to the exposed shelf.
However, only in very rare situations will river profiles adjust their grade along their entire length with other factors combining with deep incision to adjust to the new sea level. Schumm (1993) believed that primary responses to changes to base level are adjustments in sinuosity, channel dimensions, and bed roughness. Only where lateral migration was restricted (e.g., by the solid limestone of older reefs on the outer shelf) would incision be the major response. Woolfe et al. (1998b, 2000) also suggested that incision into the shelf will be only local during low sea-level stands as the shelf gradient is too low except on the outer edge of the shelf. Seismic surveys have shown evidence of both fluvial incision and lateral movement of river courses without significant channeling during the period of shelf exposure. Whilst there are numerous cross-sections of incised channels more than 20 m in depth, these may not be continuous across the shelf, making it difficult to reconstruct complete paleo-drainage patterns. However, incision may be more common than suggested by Woolfe et al. (1998b, 2000). Large rivers such as the Burdekin may have sections incised 20 m into the general level of the seabed with a width of 260 m (Fielding et al., 2003). Elsewhere on the central GBR shelf, channels (probably the Herbert River) incised 17 m and up to 1000 m across have been identified (Searle et al., 1982; Johnson and Searle, 1984). In some locations, channel sequences up to 6 km wide have been described (e.g., Searle and Hegarty, 1982). As noted by Searle and Hegarty (1982) and Graham (1993) on the narrower and steeper shelf areas, such as the northern GBR and just south of Cairns, buried mid-shelf channels may be traced towards passes on the outer shelf where shelf-edge incision may be more prominent. Such relationships cannot be seen on the central and southern GBR. Where modern streams do connect with inner and mid-shelf paleochannels, the coarser fluvial fill may now act as a confined submarine aquifer carrying fresh water onto the continental shelf where it can reach the seabed via ‘‘wonky holes’’ (Stieglitz, 2005).
6.4 Low sea-level drainage patterns
177
In addition to being infilled with sediments, many of the identified channels are topped by positive relief in the form of broad mounds extending up to 2.5 km either side of the channel and up to 6.5 m above surrounding sea floor (Johnson et al., 1982) (Fig. 6.2). Seismic stratigraphy is similar wherever channel fill and overtopping occurs. Basal channel fill sediments are relatively seismically transparent and have been interpreted as well-bedded fluvial sediments, coarser towards the base. These are considered to have been laid down whilst sea level was falling and/or the coastline lay to seawards. They are unconformably overlain by a well-bedded finer sediment body representing estuarine deposition during the transgression. The overlying mound is weakly parallel bedded and has been interpreted as delta-front sediment deposited over the backfilled channels from a nearby river mouth. Not all buried channels have these positive relief features, and where present they tend to form disconnected ‘‘beads’’ over the sea floor. Johnson and Searle (1984) believe that this represents situations where mounds had to be sufficiently large to survive redistribution by the rising sea level (dependent on the rate of sediment delivery to the coast and the rate of sea level rise) or entrapment within coastal embayments. Almost all seismic surveys on the GBR shelf provide examples of paleodrainage patterns. In the far north Harris et al. (1993) indicate that rivers flowing from Papua New Guinea cut channels up to 60 m below the level of the Pleistocene surface. The drainage pattern in Princess Charlotte Bay described by Searle and Hegarty (1982) is very different from that described by Maxwell (1968) based only on bathymetry. In the west of the Bay the ancestral Stewart Rivers flowed between Grub and Corbett Reefs out to the main reef tract. The Kennedy and Normanby Systems, which coalesced, flowed between Corbett and Clack Reefs to Lowry Pass on the shelf edge, one of the few channels which has not been infilled. In the Cairns area the Daintree River has been traced out onto the shelf north of Low Isles (Johnson and Carter, 1987) and the Barron River, joined by other streams entered the sea via Trinity Opening (Orme et al., 1978a; Searle et al., 1982), in this instance confirming Maxwell’s (1968) interpretation. The relatively narrow shelf south of Cairns shows a typical series of buried channels on the inner and mid-shelf which tend to join into only two systems near the reef tract (Graham, 1993). Probably more so than elsewhere, channels in this area are ‘‘nested’’ in older channel complexes belonging to an earlier Pleistocene generation of erosion. Low sea-level river courses appear to have been re-established several times. Coastal streams north of Hinchinbrook Island (Tully and Hull Rivers, Liverpool and Maria Creeks) appear to have joined near Dunk Island before passing out to sea via Eddy Reef and between
178
The non-reefal areas of the continental shelf
Figure 6.3 Paleo-drainage channels of the Herbert and Burdekin Rivers.
Reefs 17-065 and 17-066. Further north other coastal rivers, such as the Russell and Mulgrave, and possibly the Johnstone River, may have made their way to the sea via Flora Pass (Searle et al., 1982) whilst the larger Geranium Passage was not a major low sea-level river course. The major interest on the wider shelf areas of the central and south central GBR is the low sea-level courses of Queensland’s larger river systems. The Herbert River, presently draining an area of 9843 km2 (Great Barrier Reef Marine Park Authority, 2001) is the fifth largest catchment draining into the GBR lagoon. Its ancestral course can be traced from a location south of its present mouth via an enigmatic gorge-like channel through the Palm Islands 35 km offshore and out to sea via the Palm Passage (Searle et al., 1982; Johnson and Searle, 1984) (Fig. 6.3). The longest river of all, the Fitzroy (drainage area 142 537 km2), presently passes out to sea immediately east of Rockhampton but the paleo-channels of this system have not been identified, although suggestions that the river once flowed northwards via Broad Sound as suggested by Jardine (1928) and Maxwell (1968) are not supported by any of the seismic evidence (Cook and Mayo, 1977). Only relatively small channels are seen incised into the pre-Holocene reflector. Maxwell (1968), however, did
6.4 Low sea-level drainage patterns
179
suggest that the most recent course of the Fitzroy was immediately out to sea where it would have been captured by the large embayment formed by the Capricorn Channel. Limited seismic survey within the Capricorn Channel has identified smaller channels throughout, but nothing that can be correlated with the large Fitzroy system (Marshall, 1977). In contrast, the second largest system, the Burdekin (130 126 km2), has been the focus of a number of studies. Hopley (1970) initially mapped the bedrock and late Pleistocene surface within the Burdekin delta showing that at all low sea-level stages the course was northwards into Bowling Green Bay (not eastwards into Upstart Bay as at present because of high bedrock outcrops). Based on bathymetric and widely spaced seismic data Belperio (1983b) and Johnson and Searle (1984) mapped a course across the shelf which passed west of Keeper and Faraday Reefs and off the shelf just south of Myrmidon Reef, a course which was restated by Harris et al. (1990). An unpublished dataset by Carr and Johnson referred to in Fielding et al. (2003) shows a slightly different course through the reef tract, the paleo-channel passing east of Grub, Yankee, and Bowl Reefs to the shelf edge. The most specific study, however, has been that of Fielding et al. (2003) who, via a series of short seismic reflection profiles, traced the ancestral Burdekin for 160 km across the shelf (Fig. 6.3). They show one major channel <1000 m wide with six minor tributaries <200 m wide, entrenched rather than incised into the shelf. From Bowling Green Bay, the course leads north to the 40-m isobath, then trends north-east. At –50 m just outside the reef tract the course was more difficult to recognize, possibly as it encountered a change in topography in front of karstified reefs. Between –70 and –80 m, within the reef tract the channel becomes more restricted and underfilled. No channel could be found on the outer 10 km of shelf where a topographic high at –60 to –70 m (probably the submerged reefs discussed in Chapter 9) was encountered. Fielding et al. (2003) believe the Burdekin either dispersed its load within this shelf marginal trough or, more likely, aggraded its bed to a stage where it could spill over onto a lower level but sediments were then subsequently removed by coastal erosion. Whatever the resolution, there is no channel graded to a –120 m sea level. However, the longitudinal profile of the paleo-channel consisting of flat and steeper sections does appear to equate to stillstands or small rises in sea level between 125 ka and 30 ka ago. Another type of shelf valley has been recently described by Harris et al. (2005). These authors show that some very deep valleys may be the result of tidal scour combined with the restricting influence of reef growth rather than fluvial action. They describe the over-deepened Darnley Valley on the northernmost GBR which is up to 220 m deep and attribute its formation to tidal scour which was strongest when sea levels were 30–50 m below present,
180
The non-reefal areas of the continental shelf
i.e., most recently between 9 and 7 ka ago but also in prolonged phases of sea level during oxygen isotope stages 3, 4, 5a, and 5c. A similar origin for the constricted passes between the Pompey Reefs in the south central GBR is also possible (see Chapter 9).
6.5 The age of shelf sediments Flooding of the GBR shelf commenced about 12 ka ago and, as the transgression progressed, triggered the formation of various sedimentary environments which have contributed to the Holocene sediment veneer. Organic material such as shells and mangrove fragments have been dated, producing finite ages for the various sediment bodies and rates of deposition. The changing environments of the transgression have been described by Harris et al. (1990), Hopley (1994), and Grindrod et al. (1999). These reconstructions will be discussed in greater detail in Chapter 12 but in summary consist of three phases: (i) drowning of the outer reef tract to form archipelagoes of limestone islands (ii) transgression across the mid-shelf area with depositional coastlines and estuaries dominated by mangroves (iii) achievement of modern sea level at about 6.5 ka ago at which time the modern coastal landforms were initiated and the nearshore terrigenous sediment wedge came into existence.
Sediments around the reefs would have first started to accumulate shortly after the drowning of the shelf margin, and accelerated once new coral growth commenced on the older Pleistocene foundations after 9 ka (see Chapter 12). The oldest dates from carbonate sediments near the reefs (Myrmidon, Bowl, and Keeper) are between 10 350 270 years BP and 10 100 150 years BP (Harris et al., 1990) with ages closer to the top of cores less than 1000 years BP. Close to reefs, rates of sedimentation are relatively rapid at 1.0–1.5 m ka1 but away from the protection of the reefs they do not exceed 0.2 m ka1 (Harris et al., 1990). As sea level rose further the middle shelf was drowned and, as Grindrod et al. (1999) have proposed, there was widespread mangrove development, even greater than found today. Increasingly, more fossil mangroves are being found beneath the middle shelf. Typical is a 70-cm deposit found 1–2 m below the surface and dated between 8740 and 8550 years BP (Australian Institute of Marine Science, 2005; Hull, 2005). Grindrod et al. (1999) report 13 dates between 10 050 190 years and 8700 610 years BP for fossil mangroves from the shelf between 168 400 S and 178 560 S (south of Cairns) from present water depths between 64.5 m and 28.5 m (Fig. 6.4). Some of the older dates
6.5 The age of shelf sediments
Figure 6.4 Fossil mangrove sites on the shelf south of Cairns, based on data from Grindrod et al. (1999) and Hull (2005).
181
182
The non-reefal areas of the continental shelf
from deeper sites appear to be associated with deep paleo-drainage channels within the reef tract such as that passing out to sea via Flora Pass whilst mid-shelf sites appear related to smaller less entrenched channels. An age : depth plot of these 13 samples shows only a broad relationship with the regional sea-level curve for all except three samples but some compaction and reworking of estuarine sediments could be expected (see Section 3.4.2). Harris et al. (1990) suggest estuarine channel fill rates of between 0.3 and 8.3 m ka1. Away from the channels the thin veneer of autochthonous sediment has accumulated very slowly. Dominated by foraminifera and mollusks, Harris et al. (1990) obtained ages between 8040 210 years and 6280 190 years BP for core material, whilst Maxwell (1969) dated grab samples from depths between 57 m and 36 m at between 4950 70 years and 2970 90 years BP. Slow rates of sedimentation are indicated, estimated at 0.2 m ka1 by Harris et al. (1990). The nearshore terrigenous sediments started to accumulate at least 1000 years prior to the peak of the transgression. Mangrove wood immediately underlying the large terrigenous sediment body which accumulated around the Whitsunday Islands (see above) was dated at 8580 230 years BP by Heap et al. (2002). Sedimentation rates for the terrigenous sediment were calculated at between 0.24 and 2.21 m ka1. A series of dates up to 8000 years BP has also been obtained from beneath or within the transgressive sediment wedge in Cleveland Bay, Townsville by Carter et al. (1993), though some are interpreted as belonging to a controversial oscillation of sea level at this time. Sedimentation rates of up to 0.68 m ka1 occurred within Cleveland Bay. Further north at Cowley Beach, Graham (1993) obtained dates of 7230 40 and 7130 110 years BP for mangrove deposits belonging to the last stage of the transgression and underlying the wide beach ridge sequence developed subsequently. Ages for beach ridges and cheniers in North Queensland can be more than 6 ka years (e.g., Bird, 1971b; Cook and Polach, 1973). The rate of deposition has depended to some extent on the proximity of a major river mouth, but much of the accumulation may also date from the earliest stabilization of sea level with sediment derived from the continental shelf (Fig. 2.5a). Coastal progradation in Cleveland Bay, for example, has been at a rate of 0.5–1.0 km ka1 since 6.5 ka ago (Carter et al., 1993) with vertical rates for the associated sand body of up to 1 m ka1. Although these figures for ages and rates of sedimentation are largely from the Townsville area and the adjacent GBR shelf, they appear to be characteristic of other areas of the GBR.
6.6 The Halimeda bioherms
183
6.6 The Halimeda bioherms 6.6.1 Distribution After coral reefs, the most significant structures built by living organisms on the Queensland continental shelf are the Halimeda bioherms. Although these calcareous algae have long been recognized as having major significance to coral reefs and coral reef sediments (since the Funafuti Boring Expedition in 1897 and the Siboga Expedition to Indonesia in 1899, according to HillisColinvaux and Orme, 1988) their ability to build significant structures was not appreciated until 1973 when, on the Royal Society–Universities of Queensland Expedition, Orme et al. (1978b) from seismic survey identified banks up to 18 m thick overlying the Pleistocene unconformity and corresponding in distribution to the extensive Halimeda meadows of the area surveyed, which was between Lizard Island and the outer ribbon reefs. Also on the 1973 expedition extensive Halimeda-covered surfaces with hummocky relief were noted behind Great Detached Reef (Hopley, 1978b, 1982). However, a major misinterpretation occurred regarding the extent of Halimeda banks north of Lizard Island. Black-and-white aerial photography of the areas was taken by the Commonwealth of Australia under remarkably clear conditions and an area behind the ribbon reefs displayed complex dark shapes on an otherwise light surface. Following arguments for antecedent karst control of reef morphology, Hopley (1978b, 1982) interpreted the features as a drowned karst surface (see Hopley, 1982, fig. 7.16). Only subsequently did ground survey show that the dark areas correspond closely with the tops of Halimeda mounds visible in depths of up to 25 m whilst the intervening unvegetated areas were light colored (Drew and Abel, 1988). Interpreted as ‘‘submerged reefs’’ many of the shoals were also mapped and classified as such on Great Barrier Reef Marine Park Authority plans (see Hopley et al., 1989). The full extent of the Halimeda bioherms or banks is probably still not known but they are most persistent on the northern GBR between 118 500 and 158 350 S, from Raine Island in the north almost to the southernmost limits of the ribbon reefs (Fig. 6.5). The banks may have an area of 2000 km2 behind the ribbon reefs (Drew and Abel, 1985) occupying up to 26% of the shelf between 168 270 and 158 020 S (Orme and Salama, 1988). The eastern limit of the banks is about 2–3 km in the lee of the ribbon reefs with an abrupt boundary from the algal to coralgal facies, whilst the western boundary is less abrupt (Orme and Salama, 1988). Less extensive Halimeda environments have been recognized elsewhere. On the central GBR Scoffin and Tudhope (1985) identify Halimeda meadows
184
The non-reefal areas of the continental shelf
Figure 6.5 Reported distribution of Halimeda bioherms on the northern Great Barrier Reef.
6.6 The Halimeda bioherms
185
perhaps several meters in thickness. They occur in water depths between 60 and 100 m and, whilst not major structures themselves, appear to occur in the same areas as the submerged shelf-edge reefs described in Chapter 9. In the far south Searle and Flood (1988) surveyed three banks in the Swain Reefs near Frigate Cay which appear identical to those of the northern GBR. Occurring in an open reef environment they have raised some questions as to the formation of these features. Even more extensive features occur on the edge of Australia’s north-west shelf around the Big Bank Shoals (Australian Institute of Marine Science, 2001). Elsewhere in the world Halimeda bioherms have not been widely recognized. Hillis-Colinvaux (1986) describes Halimeda meadows on the fore reef of Enewetak Atoll which may equate to the deeper sites on the central GBR. Hine et al. (1988) map the distribution of Halimeda bioherms off Nicaragua in the Caribbean but the most extensive area appears to be in the Makassar Strait, especially on the Kalukalukuang Bank (Phipps and Roberts, 1988; Roberts et al., 1988) and off the Mahakam Delta (Sydow, 1996). 6.6.2 The morphology of the Halimeda bioherms The morphology of the banks in the Cooktown area is well known from highresolution seismic and side-scan surveys, echo-sounding profiles, some scuba observation, and extensive area observation in a manned submersible. The banks occur in water depths of between 20 and 50 m tending to maintain the same relative relief of about 15 m over any particular area (Orme and Salama, 1988). They tend to be elongate up to 150 m long and though having a predominant north-to-south orientation they are sinuous and rarely parallel. Individual mounds are up to 100 m wide, flattish-topped with side slopes between 58 and 258 (Davies and Marshall, 1985; Phipps et al., 1985). The intervening valleys are 50–60 m wide with floors of fine, clean sand with Callianassa mounds. Breaking through the sand are occasional outcrops of probable Pleistocene reefal limestone in pinnacles up to 3 m high or occasional pavement surfaces several meters wide (Figs. 6.6 and 6.7). The vegetation cover of the banks is very dense and although only 30 cm high has been equated to a rainforest cover by Davies and Marshall (1985). An upper layer dominated by Caulerpa and lower storey of Halimeda both stabilizes and traps sediment. Thirteen species of Halimeda have been identified on the banks (Drew and Abel, 1988) with H. hederacea (48.4%) and H. copiosa (26.0%) the most prolific. Two species of Udotea (U. flabellum and U. argenteum) and Penicillus nodulosus are also common. Halimeda fragments, however, dominate the surface sediments. Occasional 4-cm corals are also found
186
The non-reefal areas of the continental shelf
Figure 6.6 Cross-section of Halimeda banks and intervening corridors based on observation from manned submersible.
Figure 6.7 (a) Vegetation cover and sediments on the Halimeda bioherms; (b) outcrop of Pleistocene limestone in the corridor between bioherms; (c) bare sandy floor of the intervening corridors.
6.6 The Halimeda bioherms
187
amongst the vegetation (Porites and Favia sp.), with an epifauna of mollusks, foraminifera, and bryozoans. Large holothurians are found on both banks and corridor floors. On the outcrops of the corridors there may be soft corals, sea whips, gorgonians, and some corals, especially Dendrophyllia. 6.6.3 Internal structure and age The first seismic reflection profiling across the Halimeda banks near Lizard Island by Orme et al. (1978b) revealed thick banks of bedded sediment extending down to a prominent reflector identified as the Holocene–Pleistocene unconformity. The composition of the banks was described as packstone or wackestone dominated by unbroken, gravel-size Halimeda plates in a mud matrix of carbonate and clay minerals. Subsequent investigations involving both seismic profiling and vibro-coring in the Lizard Island area (Orme, 1985; Orme and Salama, 1988), off Cooktown (Davies and Marshall, 1985; Marshall and Davies, 1988), and in the Swain Reefs (Searle and Flood, 1988) have confirmed very similar morphology and internal structure. The Pleistocene surface beneath the banks is everywhere planar or gently sloping, rising towards any adjacent reef structure. Searle and Flood (1988) indicate that unlike the Holocene reefs, the Halimeda banks do not appear to have required any antecedent surface for their growth, though Orme and Salama (1988) believe that in places at least the upper part of the Pleistocene consists of cemented Halimeda flakes. This has also been found in the Big Bank of the north-west shelf where up to 50 m of unconsolidated Halimeda sediments overlie a cemented Halimeda unit of Pleistocene age (Australian Institute of Marine Science, 2001). Where encountered in vibro-cores the top of the Pleistocene consists of a recrystallized limestone displaying evidence of vadose diagenesis with vugs partially filled with soil. The internal bedding closely follows the morphology of the surface of the banks (‘‘mounded with time lines,’’ according to Searle and Flood, 1988). However, in some places the bedding may be more complex with more than one generation of accumulation (Marshall and Davies, 1988). Constituents are everywhere Halimeda sand or gravel in an olive-gray muddy matrix. Some large foraminifera, mollusks, and bryozoans may also be present. The gravel content, >50% near the surface, diminishes with depth. The muddy matrix, at least on the northern GBR, is both carbonate and terrigenous, the latter being derived from major flooding with recurrence intervals of 5–10 years (Davies and Hughes, 1983). The Halimeda flakes, however, show little sign of movement (Orme and Salama, 1988) and production from a reef-derived source is not considered sufficient (Drew and Abel, 1983).
188
The non-reefal areas of the continental shelf
A minimum age for the banks is provided by a thin peat layer directly over the Pleistocene surface and overlain by carbonate sands and Halimeda gravels. The age of 10 070 180 years BP from near Petricola Shoal off Lizard Island (Orme, 1985) suggested to Marshall and Davies (1988) that the Halimeda banks may have commenced accumulation 1500 years before the Holocene reefs. These authors obtained dates of up to 4750 80 years BP for material down to 5.4 m providing an accumulation rate of 1.7 m ka1. In the Swain Reefs Searle and Flood (1988) obtained ages of 2200 80 years BP and 1310 120 years BP from 4.3 m and 2.0 m respectively giving accumulation rates of between 2.0 and 3.4 m ka1. They also projected a 5000 years BP startup date for the banks, very different from the northern GBR, perhaps reflecting the contrasting morphological and oceanographic conditions. 6.6.4 The environment of Halimeda bioherm construction Halimeda meadows are highly productive systems (Hillis, 1997). Biomass on the northern GBR has been measured as high as 4637 g dry wt m2 (Drew and Abel, 1985). Drew and Abel (1985) suggest average calcification rates of about 2 kg m2 yr1 producing vertical accumulation rates ranging from 3 up to as high as 14 m ka1, i.e., greater than suggested by the geological record (see also Drew, 2001). However, the reasons why what would have been originally Halimeda meadow converts into the three-dimensional Halimeda banks has been addressed by Hillis-Collinvaux (1988). Halimeda is held in place by an extensive holdfast system, the lower parts of the thalli buried in situ. This stabilizes several centimeters of sediment facilitating in situ accumulation of Halimeda plates. This is further encouraged by algal and cyanobacterial surface films. Without disturbance a sequence of partial burial followed by new growth from perennial portions is established and the bed is raised above the surrounding floor. Abel and Drew (1985) evaluated various environmental parameters which could favor the location of Halimeda meadows and subsequent bank formation. Depth is not a problem as even at 30–35 m the meadows of the northern GBR receive irradiance well above light saturation for most of the time. However, laboratory experiments indicated that the uptake of nutrients, especially nitrate, could have a major influence on rate of growth. Abel and Drew (1985) concluded that ‘‘the presence, intensity and frequency of nutrient intrusions may be a contributing factor in the geographical location of the Halimeda meadows in the GBR province.’’ Certainly some enhancement of nutrient levels is required as the thousands of square kilometers of Halimeda banks on the shelf require some
6.6 The Halimeda bioherms
189
3.3 tonnes km2 yr1, far beyond what is provided by the overlying water column (Drew, 2001). Outside the nearby ribbon reefs water depths of over 1000 m are found with the thermocline at about 80 m, below which the water is high in inorganic nutrients. The ribbon reefs with intervening passes well above the thermocline level were always considered too great a barrier for oceanic upwelling to be considered. However, Thomson and Wolanski (1984) noted cold water intrusion through the Raine Island Entrance on the northern GBR. Further research involving both modeling and field observation (Wolanski et al. 1988) concluded that the Bernoulli effect of fluid mechanics could be applied to the oceanography of the ribbon reefs and their passes. The theory is that strong currents in shallow water suck the underlying water upwards, bringing with it the inorganic nutrients. Working especially on the passage between Ribbon 3 and Ribbon 4, Wolanski et al. (1988) noted a strong jet through the pass, carrying with it a pair of vortices which continued to propagate onto the shelf even after the tide ebbed. Outgoing water came mainly from behind the ribbon reefs whilst the nutrient-rich vortices reach the Halimeda banks after about 12 hours (see Section 4.3.2 and Fig. 4.6). This nutrient-rich cold water is pumped into the GBR lagoon twice a day during peak spring tides. It is estimated that through each of the passes this water carries 60 tonnes of nutrients each year, more than enough for the growth of Halimeda meadows. Channel bathymetry especially width and shape have a strong influence on the formation of the tidal jets. Wider or shallow passes are less likely to be associated with the jets. Examining the whole of the northern GBR, Wolanski et al. (1988) concluded that the ideal was reefs at least 4 km long separated by channels less than 1 km wide and 40–45 m deep. These conditions are found along most of the northern GBR in close association with the Halimeda banks. They would also have commenced at the start of the Holocene when the shelf was first flooded and the passes were formed between the emerged Pleistocene reefs. A gap occurs in the distribution of the banks opposite Princess Charlotte Bay but this is thought to be caused by the turbid waters from nearby rivers. Wolanski et al. (1988) also believed that up to 25% of nutrients in the jets was taken up by phytoplankton before reaching the Halimeda. However, this is not totally lost as it is taken up by the zooplankton with nutrient-rich feces being remineralized. The outgoing jets on the falling tide also suck up cold water and chlorophyll-rich plumes have been observed seawards of the passes as phytoplankton take advantage of this nutrient-rich water. Satellite imagery indicates that this surface water also enters the GBR lagoon on the incoming tide and becomes located directly over the banks.
190
The non-reefal areas of the continental shelf
These explanations can be applied to the Halimeda banks of the northern GBR but this is the only area with the required reef morphology. Ribbon reefs and passes are not found in the Swain Reefs nor on the north-west shelf or overseas areas such as the Makassar Strait. However, upwelling occurs in many areas without involving the Bernoulli effect. For example, Roberts et al. (1988) suggested that strong upwelling of nutrient-rich Pacific intermediate water was encouraged along the steep western margin of the Kalukalukuang Bank. In addition large internal waves along the thermocline and reversals in the Pacific through-flow cause variabilities in the depth of the thermocline, perhaps resulting in pulses of cold, nutrient-rich water onto the margins of the bank. Shelf marginal upwelling elsewhere (including the Swain Reefs?) may similarly produce nutrient levels high enough for prolific Halimeda growth. 6.7 Conclusion Sedimentation on the tropical continental shelf of Queensland has been largely limited to the outer third of the shelf, dominated by carbonate productivity of the reefs and Halimeda banks, and to the inshore areas where terrigenous sediments have accumulated as coastal and subtidal deposits over 6 ka of relatively stable sea level. Changing environmental conditions are reflected in these distributions. Conditions in the early Holocene appear to have been more conducive to algal (Halimeda) growth than that of corals and substantial banks of Halimeda may have been present before a significant Holocene veneer began to accumulate on the reefs. Unlike the reefs these banks do not appear to have required antecedent features although cemented Halimeda deposits of Pleistocene age suggest that similar conditions for their accumulation may have existed in the past. These will be discussed further in Chapters 11 and 12. Apart from infilling the fluvial relief of the mid-shelf area, the gentle slope of this part of the shelf and the rapid rise of sea level at the time it was drowned have produced the largely featureless sea floor. Inshore, however, as the transgression ended and the stillstand began the environment changed rapidly and this too has affected the nearshore, especially fringing reefs (which will be discussed in Chapter 7). In many ways the shelf area provides a complementary history to that of the reefs helping to explain what have been some of the enigmatic questions of Holocene reef evolution.
7 Fringing and nearshore coral reefs
7.1 Introduction Fringing reefs and nearshore reefs occur close to mainland or high island shores. Fringing reefs are often shore-attached, with reefal sediments extending from the fore reef to shoreline, although back reef areas can be shallowly submerged to form flooded depressions up to several hundred meters wide. However, along most of the inner Great Barrier Reef (GBR) where the majority of fringing reefs occur, relative sea level has fallen by 1.5 m since the mid-Holocene highstand (Section 3.4.3), and it is more typical for back reef areas to be higher than the reef crest (discussed further in Sections 7.4 and 7.5). Nearshore reefs are very close but not directly attached to mainland or high island shores, and are separated from these shorelines by back reef areas dominated by non-reefal sediments that are shallower than 10 m (Larcombe et al., 2001). Following Darwin’s (1842) classification, fringing reefs are interpreted as the first stage in a long-term evolutionary sequence, in which reef growth is initiated over a substrate on the shoreline, but later becomes separated from the shore as the underlying substrate gradually subsides and a leeward lagoon develops. Fringing reefs are thus often considered as characteristic of stable or uplifting coasts, where the geomorphological transition from fringing to barrier reef is suppressed by a lack of subsidence. However, as discussed in this chapter, the influence of slow geological subsidence on modern fringing reef growth on the Great Barrier Reef and elsewhere is often relatively small compared to that of more rapid changes in relative sea level through the Holocene. Holocene sea-level changes and substrate topography together affect the accommodation space in which a fringing reef may grow, and have strongly influenced the geomorphological development of fringing and nearshore reefs (Smithers et al., 2006). Three decades ago, Steers and Stoddart (1977) noted that fringing reefs appeared morphologically simple and had been little studied to that point. 191
192
Fringing and nearshore coral reefs
In the years since, there have been many geomorphological studies of fringing reefs, including morphostratigraphic and geochronological investigations of around 20 within the GBR and Torres Strait (Section 7.4). These studies clearly demonstrate that although many fringing reefs can be relatively thin structures and are often relatively young, their growth and development is usually more complex than formerly recognized. Kennedy and Woodroffe (2002) provide an excellent review of geomorphological research on fringing reefs that includes studies on the GBR. The proliferation of fringing reef research on the GBR in the late 1970s and early 1980s was partly due to the relative logistical ease of accessing and working in these nearshore environments compared to those further offshore, but their vulnerability to threats associated with land-based human activities also raised interest. In recent years this interest has intensified as many reefs close to the mainland and islands are reportedly in decline (e.g., Fabricius and Wolanski, 2000; McCulloch et al., 2003b; Wolanski and De’ath, 2005). Many reef scientists now believe that the long-term prospects of all fringing and nearshore reefs on the GBR are poor, and invariably anthropogenic activities are implicated in their demise. These impacts are discussed from a geomorphological perspective in Chapter 13. In this chapter we examine the distribution, structure, Holocene growth, surface morphology, and dominant processes on fringing and nearshore reefs of the GBR, including several from Torres Strait. 7.2 Distribution and settings Hopley et al. (1989) identified 758 fringing reefs in the Great Barrier Reef Marine Park using aerial photographs and satellite imagery, of which 545 have developed recognizable reef flats. The remaining 213 reefs are shore-attached but lack reef flat development; these are referred to as ‘‘incipient’’ fringing reefs in various studies (Chappell et al., 1983; Kleypas, 1996) (Table 7.1). Fringing reefs are relatively abundant on the GBR, but most are quite small (mean: 1 km2). They cover approximately 350 km2 of the inner continental shelf, and comprise just 1.8% of the GBR’s total reefal area (Hopley et al., 1989). The locations of the main fringing and nearshore reefs referred to in this chapter are shown in Fig. 7.1; latitudes and longitudes are given for others. Due to the large number of fringing and nearshore reefs that occur on the GBR, it is not possible to present a map showing them all in this book. The reader is referred to the zoning maps available online at the Great Barrier Reef Marine Park Authority (2004) website for detailed maps of reef locations, including fringing and nearshore reefs.
193
7.2 Distribution and settings
Table 7.1. Fringing reefs and incipient reefs by latitude Latitude (8S)
Number of fringing reefs
Number of incipient reefs
Total number of fringing reefs
108þ –118 118–128 128–138 138–148 148–158 158–168 168–178 178–188 188–198 198–208 208–218 218–228 228–238 238–248 248–248 300 Total
28 22 17 8 38 13 24 15 35 12 206 52 2 73 0 545
0 0 1 0 4 0 0 3 7 16 19 80 76 0 7 213
28 22 18 8 42 13 24 18 42 28 225 132 78 73 7 758
The exact number of nearshore reefs is unknown due to difficulties of classification and recognition. However, an estimate exceeding 100 appears reasonable based on maps recently released by the Great Barrier Reef Marine Park Authority (2004) and low-elevation aerial observations of inshore areas during low spring tides. This estimate includes shoals marked on these maps where inshore or back reef depths are less than 10 m. Many shoals exhibit moderate to high coral cover, and although many are topographically simple (e.g., Shoalwater Bay: Ayling et al., 1998), others develop bathymetric relief (e.g., Paluma Shoals: Smithers and Larcombe, 2003). In this way, they may be synonymous with incipient fringing reefs (Hopley et al., 1989), which may include reefs at a variety of growth stages (Kleypas, 1996). For example, they may be catch-up reefs that will form true fringing reefs with time, they may be reefs in which vertical accretion has temporarily stalled, or they may be coral communities incapable of accumulating sufficient carbonate to produce a coral reef under present conditions. Numerous definitions of coral reefs, coral reef communities, coral communities, etc. exist in the literature (e.g., Buddemeier and Hopley, 1988; Kleypas et al., 2001). We recognize that shoals may not meet the definition of a coral reef preferred by some researchers, but we consider it appropriate given the longer-term geomorphological view presented in this book, and the possibility that these shoals have the potential to ‘‘turn-on’’ (Buddemeier and Hopley, 1988) and grow into ‘‘true’’ coral reefs in the future.
194
Fringing and nearshore coral reefs
Figure 7.1 Fringing and nearshore reefs referred to in this chapter.
The majority of fringing and nearshore reefs on the GBR can be broadly classified as one of four main types on the basis of their coastal setting: (1) headland-attached fringing reefs that develop on rocky headlands, often over steep foundations (Fig. 7.2a);
7.2 Distribution and settings
195
Figure 7.2 (a) Headland-attached fringing reef, Great Palm; (b) bayhead fringing reef, Scawfell Island; (c) narrow beach-base fringing reef, South Myall Reef, Cape Tribulation; (d) nearshore shoal, Paluma Shoals. (2) bayhead fringing reefs that develop in embayments and prograde out from the head of the bay (Fig. 7.2b); (3) narrow beach-base fringing reefs that develop along long, often linear stretches of sandy coast (Fig. 7.2c); and (4) nearshore shoals which can occur in all of the above coastal settings, but are not directly attached to the shoreline (Fig. 7.2d).
This simple classification scheme requires little explanation. However, we emphasize that any particular fringing reef, or section of it, may display characteristics transitional between more than one class, both in space and through time. For example, at Paluma Shoals the northern shoal is attached to the shoreline by intertidal sand flats and would be referred to as a ‘‘beach-base’’ fringing reef, but the southern shoal is presently separated from the shoreline and would be classified as a nearshore shoal (Fig. 7.2d). 7.2.1 Mainland fringing and nearshore reefs Fringing reefs are rare along most of the mainland coast (Hopley, 1982). Fringing reefs with reef flat development do not occur on the mainland coast
196
Fringing and nearshore coral reefs
south of Cape Conway (208 320 S, 1488 550 E), although isolated corals and more rarely coral communities that may be classed as incipient reefs occur around headlands and rocky coasts (e.g., around cliff bases at Water Park Point, north of Yeppoon (228 570 S, 1508 470 E)). In contrast, true fringing reefs are relatively common north of Cape Conway along the Whitsunday coast to Cape Gloucester (208 040 S, 1488 270 E), reaching their greatest development in bayhead settings north of Pioneer Bay (208 150 S, 1488 440 E)(note that rather confusingly, there are two Pioneer Bays in the central GBR – the one referred to here at Airlie Beach along the Whitsunday coast, and another with a wellstudied fringing reef on Orpheus Island (188 360 S, 1468 300 E)), where the influence of large rivers is limited and strong geological structural control has produced embayments confined by steep headlands. On this section of coast, fringing reefs such as that at Hydeaway Bay (208 050 S, 1488 300 E) extend more than 350 m offshore and stretch more than 3 km alongshore. With the exception of a small (0.28 km2) fringing reef at Moonlight Bay inside of Cape Upstart (198 440 S, 1478 460 E), mainland fringing reefs with established reef flats are not found until King Reef (178 460 S, 1468 070 E), located almost 300 km further north at Kurramine Beach. Briefly described by Bird and Hopley (1969), King Reef is considered by Hopley (1982, p. 359) an ‘‘anomalous feature’’ due to its development against a sedimentary rather than rocky shore. However, coral growth at King Reef is clearly confined to the fringes of an extensive outcrop of ‘‘coffee rock’’ – beach barrier sediments indurated by an illuvial ferruginous cement in the weathering profile (Graham, 1993), which has been exhumed at the shoreline as the barrier has retreated. Similar indurated horizons in coastal barriers on the New South Wales coast are typically of Pleistocene age, prompting Bird and Hopley (1969) to infer an equivalent age for the outcrop at Kurramine. Graham (1993) noted that both inner barrier and Holocene beach ridges are ‘‘anchored’’ to a point located in the center of King Reef, suggesting that the coffee rock predates the last interglacial. Small fringing reefs are sporadically developed on sections of rocky coast north of King Reef until Yule Point (168 340 S, 1458 300 E) (Bird, 1970), beyond which an extensive fringing reef extends 6 km north to the southern end of Four Mile Beach. The Yule Point fringing reef is complex geomorphically. First described in detail by Bird (1971a), it is bisected by the Mowbray River. Active reef growth is restricted to the outer fringe that remains submerged during low tide, with much of the outer reef flat dominated by coralline algae. Quartz sand veneers the inner reef flat at the southern end, but closer to the mouth of the Mowbray River, mud dominates. A ribbon of mangroves (Avicennia and Rhizophora) separates the inner reef flat from beach ridges
7.2 Distribution and settings
197
behind. An emerged reef 1.2 m above the contemporary reef flat lies beneath the beach ridges (Fairbridge, 1950; Bird, 1971a), which a radiocarbon date suggests is of mid-Holocene age (4130 110 years BP). Interestingly, Bird (1971a) concluded that this reef was initiated as an offshore patch reef – or nearshore shoal, which later became shore-attached through rapid coastal progradation at the mouth of the Mowbray River (see Fig. 7.7e). The development of the Yule Point fringing reef is thus in many ways similar to nearshore reef development at Paluma Shoals (see Section 7.4, Fig. 7.2d), and demonstrates the utility of considering nearshore and fringing reefs as members of a spectrum of inshore reef types. Mainland fringing reefs are more common but are nonetheless intermittently developed north of Cape Kimberley (168 190 S, 1458 280 E). Early work in this area (Fairbridge, 1950, 1967) overestimated the number of fringing reefs in this region (Hopley, 1982), possibly due to the high inshore turbidities that commonly prevail, and associated difficulties in observation. Mainland fringing reefs around Cape Tribulation (168 060 S, 1458 270 E) (Figs. 2.4, 7.2c, 7.3, 13.2, 13.3, and 13.4) – between Noah Head and Cullal Cullal Reef – are known in most detail. Coral cover and diversity, turbidity and sedimentation, Holocene reef growth, and surficial geomorphology and sediment facies have all been investigated (Ayling and Ayling, 1985, 1999; Johnson and Carter, 1987; Partain, 1988; Partain and Hopley, 1989; Hopley et al., 1990). Hopley et al. (1990) report that the intermittently distributed fringing reefs in this area average 80 m in width, but are quite variable, with some such as Cullal Cullal Reef extending more than 1 km offshore. They typically develop as either narrow fringing reefs attached to exposed rocky headlands, or alternatively as broader cuspate-shaped reefs along beaches and in bays. As discussed in Section 7.3, the latter have been associated with coarse cobble and boulder deltas (Figs. 7.3 and 13.4) that may have provided a suitable colonizing substrate for reef initiation (Partain and Hopley, 1989) (see Fig. 7.7c(iv)). Fringing reefs of similar appearance to those around Cape Tribulation occur further up the Cape York, for example around rocky headlands at Cape Melville (148 370 S, 1448 550 E) and Cape Grenville (118 580 S, 1438 120 E) and along beaches north of Captain Billy Landing (118 350 S, 1428 500 E). Detailed investigations are yet to be carried out on these remote reefs. Nearshore reefs are more evenly distributed along the GBR coast than fringing reefs, but appear particularly common in Halifax Bay north of Townsville, and also south of 218 S (Fig. 7.1) (see Kleypas, 1996, table 1). Many occur in turbid water and adjacent to sedimentary coastlines, both conditions usually considered marginal for reef growth (Perry and Larcombe, 2003). Although the distribution of nearshore reefs may be controlled by many
198
Fringing and nearshore coral reefs
Figure 7.3 Cape Tribulation fringing reefs: location, surface morphology, and internal structure and chronostratigraphy (after Hopley and Partain, 1987).
7.2 Distribution and settings
199
variables, a first-order examination of inshore morphology to identify areas analogous to Halifax Bay, where nearshore reefs are reasonably common, suggests areas offshore from Mossman and Carmila (Fig. 7.1) are most likely to provide substrates suitable for nearshore reef development. However, as noted earlier, nearshore reefs appear relatively well represented along the entire GBR coast on zoning maps recently published by the Great Barrier Reef Marine Park Authority (2004), and as they typically occur in turbid areas, it is possible that their occurrence may be underestimated, particularly those of incipient development. The recent discovery of three submerged coral reefs in turbid waters in the southern Gulf of Carpentaria using multibeam swath mapping illustrates the difficulty in detecting reefs growing in these conditions – cumulatively the three reef platforms cover 80 km2 but were invisible to aerial photography and satellite imagery (Harris et al., 2004). 7.2.2 Fringing reefs of the continental islands Hopley et al. (1989) identified 617 continental ‘‘high’’ islands with reefs and noted that these high islands are not evenly distributed, with more than half occurring between 208 and 228 S, in the Cumberland, Northumberland, and Whitsunday Island Groups. Kleypas (1991) considered that slightly fewer continental islands with fringing reefs occurred between these latitudes (Table 7.1) but confirmed Hopley’s (1982) observation that fringing reef development rapidly diminished over a short geographical distance around 218 S. The Whitsunday and Cumberland Islands north of 218 S support wide and well-developed fringing reefs (Fig. 7.2b) whereas south of this latitude in the Northumberland Group fringing reefs are poorly developed and often incipient. Inshore reef growth in the southern GBR (south of 218 S) is generally poor, with the exception of around the Keppel Islands, where substantive reefs with limited reef flat development are located (Fig. 7.1). Diminished inshore reef growth in the southern GBR appears principally related to turbidity controls on community composition and demography, and on carbonate productivity, preservation, and accumulation (van Woesik, 1992; Kleypas, 1996; van Woesik and Done, 1997). Reduced reef growth toward the southern GBR has also been associated with larger-scale gradients in reef marginality with increased latitude (Harriott and Banks, 2002). North of the Whitsundays most continental islands have well-developed fringing reefs, although some such as Gloucester Island and Hinchinbrook Island do not. Hopley (1982) suggested that high runoff volumes from these larger islands may inhibit fringing reef growth, although he recognized that
200
Fringing and nearshore coral reefs
other large islands had well-developed fringing reefs and thus other explanations may be needed. Terrestrial runoff from the mainland and relatively high inshore turbidity have been suggested as unfavorable for fringing reef development around continental islands that are close inshore, especially south of Cairns where fringing reefs are rather poorly developed around the Family Group and Fitzroy Island. In contrast, the fringing reef at Double Island – just 1 km offshore and 20 km north of Cairns – is over 1.2 km wide, and most high islands, even those very close inshore further up Cape York, have fringing reef development. Especially north of Cape Weymouth, significant fringing reefs may develop around relatively small rocky outcrops. A similar situation occurs at Redbill Reef (Fig. 8.7a) in the southern GBR where a small granitic outcrop (0.016 km2) is surrounded by a reef platform of 8.8 km2 (Hopley et al., 1984), but such development is rare in the southern GBR. Fringing reefs on continental high islands can occupy all of the settings presented in Section 7.2, although those on windward shores are typically narrow and analogous to headland-attached fringing reefs. The relatively broad fringing reef at Iris Point, Orpheus Island is an exception to the generally narrow development observed on windward reefs, extending more than 400 m from the boulder beach at the rear to the reef crest (Fig. 7.1, 7.4, and 7.5).
Figure 7.4 Iris Point, Orpheus Island. Note the boulder beach which extends beneath the reef flat and forms its foundation. See also Fig. 7.5.
7.2 Distribution and settings
Figure 7.5 Iris Point fringing reef. (a) Surface morphology and (b) internal structure and chronostratigraphy (after Hopley and Barnes, 1985). See also Fig. 7.4.
201
202
Fringing and nearshore coral reefs
Fringing reefs attached to leeward northern and western shores are usually broader and developed over detrital spits and thus may be similar to any of the three other settings (bayhead, narrow beach-base, nearshore shoal). 7.3 Fringing reef structure Stratigraphic and sedimentological data suggest that most fringing reefs on the GBR are dominantly detrital, with framework growth generally limited to within 1–2 m of the surface (Hopley et al., 1983). A terrigenous muddy sand is typically found beneath most leeward fringing reefs, grading up into a rubble facies and then to a relatively thin framework cap (see Hopley et al., 1983) (Fig. 7.3). The lower muddy unit may be absent in reefs formed in more exposed locations (Fig. 7.8). Kleypas (1991) noted that fringing reefs in the southern GBR incorporated less terrigenous sediments within their structure than those further north, and suggested that this reflected their greater distance offshore. Chappell et al. (1983) developed a basic structural classification for GBR fringing reefs that differentiates between reefs that form reef flats when patch reefs coalesce over a broad nearshore area and those that build reef flats by prograding out from shore. Reef flat age structure, usually established by dating coral microatolls or material recovered in shallow reef flat cores, is diagnostic in this classification. Hopley and Partain (1987) proposed a more complex structural classification for fringing reefs that takes account of the nature of reef foundations and the relative proportions and positions of framework, detrital, and terrigenous sediments within the reef structure. Fringing reefs on the GBR grow upon a variety of foundations, ranging from hard rock on high island headlands to coastal deposits such as spits and bars. Rock substrates are widely perceived as optimal substrates for reef initiation and growth due to their stability and elevation above the sea floor (Veron, 1995). However, headland-attached fringing reefs on the GBR are relatively rare despite the widespread availability of headlands, and few fringing reefs appear to have initially colonized rocky shores. Only at Scawfell Island on the southern GBR do drill data suggest the possibility of a bedrock foundation (Kleypas and Hopley, 1993) (Table 7.2). Many of the larger fringing reefs of the GBR have developed over sediment deposits formed during the Holocene transgression within well-defined coastal embayments. The data available indicate that Pleistocene reef substrates were generally amongst the earliest colonized during the transgression (even at shallower depths, and thus soon after inundation), with gravel and boulder substrates, and clay and sand substrates typically colonized later (Fig. 7.6). The available data (Table 7.2) suggest that fringing reefs of the northern and central GBR more often establish
3 2 4 3 4 5 7 3 3
108 350
168 020
168 040
168 060
178 580
178 580
188 280
188 360
188 410
3 Hammond Island 4 Emmagen Reef, Cape Tribulation 5 Rykers Reef, Cape Tribulation 6 Myall Reef, Cape Tribulation 7 Lugger Shoals
12 Coolgaree Bay, Great Palm Island b
188420
7
108 120
2 Moa Island
8 Stingaree Reef, Dunk Island b 9 Iris Point, Orpheus Island 10 Pioneer Bay, Orpheus Island 11 Fantome Island
4
98 540
Lat. (S)
1 Yam Island
Reef
Number of holes
2.7
14.0
17.25
8.4
10.6
2.7
8.3
6.95
5.8
9.0
12.2
12.0
1.5
6.0
10.0
1.3
8.7
2.6
7.9
6.0
4.05
6.8
4.6
5.5
Shallowest Maximum Pleistocene/ depth a (m) basement
Clay/sand
Clay/sand
Clay/sand
Boulders
Clay/sand
Clay/sand
Boulders
Boulders
Pleistocene reef Pleistocene reef Pleistocene reef Boulders
n.a.
16
10
12
n.a.
2
4
8
3
5
n.a.
10
Oldest within 1 m of reef flat
539 60
6340 80
n.a.
n.a.
738 34
6150 70
Hopley & Partain (1987) S. Smithers, unpubl. data Hopley et al. (1983)
Hopley & Partain (1987)
Woodroffe et al. (2000) Hopley & Partain (1987)
Woodroffe et al. (2000) Barham (1983)
Reference
n.a.
n.a.
5910 110 5340 80
Johnson & Risk (1987) Hopley et al. (1983)
6610 110 5970 100 Hopley et al. (1983)
7320 125 6260 120 Hopley and Barnes (1985)
n.a.
3993 44
7330 60
7780 260 5410 70
6280 80
6340 80
n.a.
7460 100 6340 80
Number of Oldest Foundation C14 dates Holocene
Table 7.2. Results of shallow and medium-depth drilling on fringing and nearshore reefs of the Great Barrier Reef (also included are the results of other investigations of internal reef structure)
>2
160 4 6 7 2 4 2 2
198 090
198 090
208 050
208 040
208 470
208 520 218 010
218 300
15 Cockle Bay, Magnetic Island 16 Geoffrey Bay, Magnetic Island c 17 Hydeaway Bay
18 Hayman Island
19 Cockermouth Island 20 Scawfell Island 21 Penrith Island
22 Digby Island
23 Middle Percy 218 410 Island 24 High Peak Island 218 580 13.5
7.75
4.1
17.05 27.9
18.35
45.6
9.2
n.a.
7.6
3.15
10.0
10.0
n.a.
0
18.6 8.1
9.9
15.0
1.9
n.a
n.a
3.0
n.a.
Shallowest Maximum Pleistocene/ depth a (m) basement
b
Core depths simply reported here as depth below reef surface. Probe investigations for engineering works. c Jet probe investigation.
a
1
3
198 070
13 Rattlesnake Island 14 Paluma Shoals 1
Lat. (S)
Number of holes
198 020
Reef
Table 7.2. (cont.)
Pleistocene reef
Pleistocene reef Pleistocene reef Bedrock? Pleistocene reef Pleistocene reef n.a.
Clay/sand
Clay/sand
Clay/sand
Clay/sand
Clay/sand
6
5
5 12 (þ1 U series) 1
13 (þ6 U series) 19
7
n.a.
1
9
5
Oldest within 1 m of reef flat
Reference
938 43
Hopley et al. (1983)
110 0.8
<38 090
Kleypas (1991)
Kleypas (1991)
7350 140 5930 100 Kleypas (1991)
3720 70
<38 090
8070 100 3040 80 Kleypas (1991) 7430 120 5650 150 Kleypas (1991)
4806 43
n.a.
Smithers & Larcombe (2003) Spenceley (1980); Foster (1974)
S. Smithers, unpubl. data 9320 730 5990 80 Hopley et al. (1978); Kan et al. (1997) 7880 90 6160 120 Kleypas (1991)
6236 45
n.a.
7320 550 n.a.
1657 83
7010 180 5530 130 Hopley et al. (1983)
Number of Oldest Foundation C14 dates Holocene
7.3 Fringing reef structure
205
Figure 7.6 Fringing and nearshore reef start-up ages by substrate.
over terrigenous foundations, whereas those of the southern GBR more commonly develop over Pleistocene reef, eolianite, or bedrock foundations. Pleistocene reef foundations also underlie all three fringing reefs so far drilled in Torres Strait. Smithers et al. (2006) presented a structural classification for fringing and nearshore reefs on the GBR and in Torres Strait that was based on Hopley and Partain’s (1987) earlier work, but was augmented with additional classes identified or inferred from more recent stratigraphic investigations and included isochrons indicative of possible reef development (Fig. 7.7). The major structural characteristics of the main classes are: (a) Simple fringing reefs formed on inundated rocky foreshores during the postglacial transgression. These reefs are relatively rare on the GBR, with few well-developed reefs grown over rocky headland foundations. The narrow fringing reefs on the windward coast of Great Palm Island are an example (Fig. 7.7a; see also Fig. 7.2a). There are no published Holocene growth histories determined by coring and dating for reefs of this type on the GBR.
206
Fringing and nearshore coral reefs
Figure 7.7 Fringing reef flat structural classification (after Hopley and Partain, 1987). Isochrons are indicative only.
(b) Fringing reefs developed over gently sloping substrates that become outflanked by transgressionary shoreline retreat and are commonly dominated by detrital backfill. On these reefs, shoreline backstepping during the Holocene transgression strands a former fringing reef offshore. Once sea level stabilizes and the reef
7.4 Holocene reef growth
207
reaches sea level, both back reef backfilling and seaward progradation can occur (Fig. 7.7b). The fringing reef at Hayman Island in the Whitsundays is an example of this type, with the substrate being Pleistocene reef (Hopley et al., 1978). (c) Fringing reefs developed over pre-existing positive sedimentary structures including Pleistocene alluvial fans (e.g., Paluma Shoals: Smithers and Larcombe, 2003) (Fig. 7.2d and 7.7c(i)), transgressionary sediment deposits trapped in pronounced embayments (e.g., Pioneer Bay, Orpheus Island: Hopley et al., 1983) (Fig. 7.7c(ii) and 7.8), leeside island spits (e.g., Rattlesnake Island: Hopley, 1982) (Fig. 7.7c(iii) and deltaic gravels (e.g., Myall Reef, Cape Tribulation: Partain and Hopley, 1989) (Fig. 7.3 and 7.7c(iv)). (d) Fringing reefs developed by episodically building a new reef structure parallel to the existing reef front, and then backfilling the intervening space largely with unconsolidated reef sediments (Fig. 7.7d). This mode of fringing reef development has been established for Yam Island, Torres Strait (Woodroffe et al., 2000). (e) Fringing reefs that have developed offshore as nearshore shoals and the terrigenous shoreline has prograded seaward, connecting the reef to the coast, or possibly provided a foundation over which the reef has built shoreward (Fig. 7.7e). Possible examples of this type of fringing reef include Yule Point (Bird, 1971a) and King Reef (Graham, 1993).
7.4 Holocene reef growth Holocene growth has been investigated at 19 of the 758 fringing reefs of the GBR, representing <3% of the total. At two of these reefs, Stingaree Bay at Dunk Island and at Coolgaree Bay on Great Palm Island, no radiocarbon ages are available. Fringing reefs at Yam and Hammond Island in Torres Strait have been cored and dated, and the fringing reef at Moa Island has been cored to Pleistocene foundations, but no radiometric ages have been determined from the core. The fringing reefs for which Holocene growth histories have been investigated all have well-developed reef flats. Although conceptual models of incipient fringing reef growth exist (Chappell et al., 1983), they have been inferred from chronostratigraphic investigations of reefs with reef flats and are yet to be confirmed by coring and dating an incipient fringing reef. The Holocene growth history of only one nearshore reef has to date been published (Smithers and Larcombe, 2003), but additional dates are available from Lugger Shoals (188 590 S, 1468 220 E) and are also reported here. Thus at present information on Holocene fringing and nearshore reef growth on the GBR and Torres Strait is based on a dataset from 24 reefs. This is arguably one of the largest fringing reef datasets in the world, but still obviously represents only a very small subsample of reefs of this type. Current research will improve our understanding of mainland fringing reefs in the Whitsunday
208
Fringing and nearshore coral reefs
A 4
B
200 m
2 0
Modern reef flat Tidal datum
Depth (m)
2 4 6
Terrigenous beach unit Prograding biogenic framework veneer Terrigenous sediments from Interdigitated reef and high islands terrigenous sediments
5320 ± 100
2620 ± 90
5970 ± 100
1630 ± 30 2990 ± 90
Reef front
Beach
Mid-Holocene reef flat
6090 ± 100 prograding biogenic reef detrital unit
3330 ± 20 4060 ± 90 6610 ± 250
8 4460 ± 110
10 12 14 16
Transgressive sand and mud unit (terrigenous)
3 1 2
on t
Coral heads Rubble Biogenic sand and shingle Terrigenous sand and mud Heavy clay Direction of sediment input or reef growth
Reef Fr
Pleistocene heavy clay
Pioneer Bay
Sand Sand, coral debris, dead microatolls Coral debris, some live microatolls Living reef Mangroves Boulder Coral shingle Sediment wave Ebb tide delta (Etd) Transect Drill holes
N A 1 100 m
Etd
2
3
Etd
B
Figure 7.8 Pioneer Bay fringing reef surface morphology, and internal structure and chronostratigraphy (after Hopley et al., 1983).
7.4 Holocene reef growth
209
region, nearshore turbid zone reefs in the central and northern GBR, and fringing reef growth in the Keppel Group at the southern end of the GBR. The drilling data from fringing and nearshore reefs of the GBR and Torres Strait are summarized in Table 7.2. 7.4.1 Initiation of Holocene fringing and nearshore reef growth Most fringing reefs on the GBR initiated quickly after transgressionary seas flooded their foundations, with ‘‘take-off’’ possibly affected by foundation type (Table 7.2). Fringing reefs form over a variety of foundations, unlike shelf reefs which exclusively develop over Pleistocene reef substrates. A radiocarbon age of 9320 years BP from a coral presumed to be in situ (see discussion Section 11.4) above the contact with the Pleistocene reef foundation at Hayman Island is the earliest reported initiation for a GBR fringing reef (Hopley et al., 1978) and the second oldest for the whole of the GBR. However, this age may be more a function of the depth of the substrate rather than its type. Furthermore, as discussed in detail in Chapter 11 and shown in Fig. 11.2, the oldest radiocarbon ages determined for a reef generally do not indicate the initiation of framework growth with the precision often inferred, and where detrital deposits underlie the dated material, the time required to accumulate the detrital facies must be considered. For some cored reefs the detrital unit between the pre-Holocene surface and the oldest ‘‘basal’’ radiocarbon date can be quite thick, and as argued in Chapter 11 (Section 11.4) may represent several thousand years of accumulation. A significant implication (discussed further in Chapter 11) of this lag between inundation, detrital accumulation, and switch to in situ framework growth is that the narrow Holocene recolonization window that has been widely reported for the GBR (Davies et al., 1985), and other reefs (Montaggioni, 1988), becomes less convincing as anything other than an artifact of dating strategy (Fig. 11.2). Nonetheless, as shown in Fig. 11.2, many fringing reefs began growing at approximately the same time as reefs further across the shelf. The oldest basal date from Paluma Shoals, a nearshore reef, is just 1657 years BP (Smithers and Larcombe, 2003), and this is an example of a reef that has initiated well after the purported narrow take-off window. True framework growth was not encountered in cores at Paluma Shoals, which is essentially detrital in structure (see Section 7.4, Fig. 7.12). There can be some confidence that the basal age and start-up date for this reef are in reasonable accord, however, as mangrove muds exposed on the shoreline dated at 3–2.5 ka record shoreline retreat and provide a maximum age for the substrate over which the shoals have developed.
210
Fringing and nearshore coral reefs
7.4.2 Vertical accretion of Holocene fringing reefs Most fringing and nearshore reefs on the GBR have accumulated a Holocene veneer of modest thickness, generally rising 5–10 m above their pre-Holocene foundations. The accumulation and thickness of Holocene reef over the entire GBR is addressed in detail in Chapter 11. The thickness of many fringing reefs is constrained by shallow inshore depths and limited accommodation space. The Holocene reef at Digby Island in the Beverley Group provides an example of extremely limited accommodation space; Pleistocene reef is exposed at the surface and Holocene growth is limited to a thin and patchy veneer (Kleypas and Hopley, 1993). More impressive Holocene fringing reef accumulations do occur; for example, at Scawfell Island the Holocene reef rises from an impenetrable substrate presumed to be island bedrock more than 18 m below MLWN (Kleypas, 1996). A modal framework accumulation rate of 7.5 m ka1 has been calculated for the GBR in Chapter 11, based on 173 radiocarbon dates from 40 reefs of the GBR and Torres Strait, including 35 cores from fringing reefs. This figure represents a net annual rate of accretion for material accumulated over centuries, and thus smooths out distortions due to short-term variability. Minor variations in this rate attributable to head or branching framework types and core location on a reef occur (see Fig. 11.4 and 11.5), but no latitudinal differences in framework accumulation rate can be detected. Detrital accumulation rates are also discussed further in Chapter 11. Partain and Hopley (1989) showed that vertical accumulation rates for fringing reefs outstripped mid-shelf and outer reefs at various paleodepths, and importantly that rapid rates of vertical accretion (>5.5 mm yr1) for fringing reefs are restricted to a narrow depth window at 4 to 8 m below paleo sea level (Fig. 11.7). Reduced vertical accretion rates are common near the surface of reef cores from all over the GBR (Davies and Hopley, 1983). However, higher rates are sustained to depths approaching 16 m at outer and mid-shelf reefs, but only half this on fringing reefs (Hopley, 1989a). This shallower depth envelope of rapid vertical accretion on fringing reefs possibly reflects the higher turbidity and sedimentation experienced at most fringing reefs compared to offshore reefs, and the reduced light-enhanced calcification at greater depths, especially in areas of higher tidal range (Kleypas, 1996; van Woesik and Done, 1997). 7.4.3 Lateral accretion fringing reefs during the Holocene Most reefs on the GBR for which radiometrically dated core material is available accumulated most of their bulk between approximately 8 and 5 ka,
7.4 Holocene reef growth
211
but the trend is especially evident on fringing reefs (Davies and Hopley, 1983; Hopley, 1989a; see Fig. 11.3). Discussions of ‘‘reef growth’’ typically focus on net rates of vertical accretion (e.g., Davies and Hopley, 1983; Davies et al., 1985; Woodroffe C. D. et al., 2005; see Chapter 11), but reefs are three-dimensional structures and lateral growth rates should also be considered. Surprisingly, lateral reef growth has received comparatively little attention in the literature (see Masse and Montaggioni (2001) as an exception). Lateral fringing reef growth on the GBR has been described in most detail by Chappell et al. (1983), who reported the ages of fossil reef flat microatolls (see Section 3.4.2b) across many central GBR fringing reefs. Microatoll age data represent a minimum age for the underlying reef flat, and the spatial distribution of microatoll ages over a reef flat can be used to reconstruct the temporal and spatial pattern of reef flat development. Smithers et al. (2006) have recently examined these and shallow (<1 m) reef core data from fringing reefs from the GBR and Torres Strait. Reef flat progradation rates are affected by various factors that are independent of net carbonate productivity rates, including substrate depth and shape, terrigenous influx, sea-level history, and accommodation space; these factors are discussed later. However, a simple plot of reef flat age against relative location across a reef flat suggests fringing reefs on the GBR have experienced several critical growth phases since the mid-Holocene (Fig. 7.9). These phases are described in detail in Smithers et al. (2006), but are briefly outlined below. (a) Initiation to 5500 Years BP: optimum reef growth and reef flat extension Conditions for both coral calcification and inshore reef growth on the GBR were good prior to 5500 years BP. Most fringing reefs were vertically accreting at close to maximum rates (Fig. 11.3), probably because continuously rising sea level created accommodation space at optimal depths for carbonate production and vertical accretion (Fig. 3.7 and 3.9). Water depths of 4–8 m are sufficiently deep for fragile corals to grow beneath the wave base, allowing a more open framework to develop, but shallow enough for light-enhanced calcification to occur in turbid waters (Fig. 11.7). Dated microatolls and near-surface corals indicate that many fringing reefs were at sea level near the start of the mid-Holocene stillstand approximately 5500 years BP (Section 3.4.3.), and some even before (e.g., a reef top coral at North Iris Point was dated at 6260 120 years BP (Hopley and Barnes, 1985)) (Fig. 11.6). Many fringing reefs have therefore grown up to a rising sea surface for much of their history, and thus experienced optimal conditions for vertical accretion for an extended period.
212
Fringing and nearshore coral reefs
Figure 7.9 Reef flat progradation rates established from published sources. Three broad groups can be distinguished; Sprinters, that had built around 90% of their reef flats by 4800 BP; Strugglers, that built most of their reef flats by around 2500 BP; and Stayers, that have continued to grow to present, but at relatively slow rates during the Late Holocene (from Smithers et al., 2006).
Benevolent coral and reef growth conditions during this period have also been inferred from wide annual bands in massive Porites intercepted in cores drilled through the framework of fringing reefs at Cape Tribulation (Partain and Hopley, 1989). Slightly warmer (1 8C) sea-surface temperatures (SSTs) affected the inner GBR then (Gagan et al., 1998), and SST is strongly and positively correlated with both calcification and linear extension in modern massive Porites (Lough and Barnes, 2000). In an early review Buddemeier and Kinzie (1976) concluded that coral growth rates and reef accretion were not directly linked, but evidence now suggests that lower coral growth rates may limit reef accretion at high latitudes (Harriott, 1999; Lough and Barnes, 2000). It is therefore possible that SSTs favoring higher rates of calcification may also have produced higher rates of reef accretion over this period. Sclerochronological (Gagan et al., 1996) and pollen (Kershaw and Nanson, 1993) records from north Queensland suggest a less variable and intense
7.4 Holocene reef growth
213
Figure 7.10 The distribution and development of the inshore sediment prism (ISP) and nearshore reef growth locations (after Larcombe and Woolfe, 1999a).
rainfall regime occurred around 5500 years BP (Kershaw and Nanson, 1993), which may have further improved carbonate productivity and reef growth. Extreme floods and droughts would have been rare, and good vegetation cover limited catchment sediment yields to the GBR lagoon where they are argued to reduce reef health (McCulloch et al., 2003b). The mid-Holocene ‘‘high-energy window’’ (Hopley, 1984) remained open, possibly allowing greater nearshore flushing and mixing of terrestrial runoff by oceanic swell and waves. Finally, the inshore sediment prism (ISP), a muddy sediment body that extends from near the shoreline to around 15 m depth along most of the coast (Johnson and Carter, 1987), was not significantly developed. The ISP is shore attached in sheltered embayments, preventing coral and reef growth, but is detached from the shore on exposed coasts, such as Halifax Bay north of Townsville, where more energetic waves in coastal shallows (to 5 m LAT) keep fine sediments in suspension (Woolfe and Larcombe, 1998) (see Section 4.6.6) (Fig. 7.10). Resuspension of fine sediments produces turbid conditions in these locations (Larcombe et al., 2001), which can present difficulties for autotrophic calcifiers.
214
Fringing and nearshore coral reefs
(b) Reef flat progradation stalls 5500–4800 years BP Reef flat growth and extension slowed abruptly between approximately 5500 years BP and 4800 years BP in almost half of the fringing reefs for which we have data. Significantly, around 90% of reef flat progradation on these reefs occurred in the millennium preceding 5500 years BP, and only 10% afterwards (Fig. 7.9). Although reef flat progradation dramatically declined over this interval, coral communities in seemingly good condition veneer the contemporary reef fronts of many of these reefs. Examples of this group include the northern part of the reef flat at Iris Point (Hopley and Barnes, 1985) (Figs. 7.4 and 7.5), and the Cape Tribulation fringing reefs (Partain and Hopley, 1989) (Figs. 7.2c and 7.3). The simplest explanation for reduced progradation is that these reefs had outgrown their foundations. Rapid vertical accretion following inundation of suitable substrates in the early to mid Holocene (see Chapter 11) may have quickly exhausted both vertical and lateral accommodation space as reefs rapidly extended to sea level (Fig. 11.6) and completely covered the most favorable foundations. This scenario may apply to the fringing reefs at Cape Tribulation, which grow over fan gravels (Partain and Hopley, 1989) (Fig. 7.3); the fringing reef at North Iris Point which grows over a boulder beach (Hopley and Barnes, 1985) (Fig. 7.5); and possibly also the fringing reefs at Cockermouth and Penrith Islands which have Pleistocene reef and eolianite foundations (Kleypas and Hopley, 1993) (Fig. 7.11). Structural data show that all had substantially covered their present foundations by 5500 years BP. The fringing reefs at Dunk Island and Great Palm extend over low gradient clays and sands and the stall in progradation there is more difficult to explain in this way. However, continued seaward progradation on Hayman Island, where initiation was set back around 100 m from the edge of the Pleistocene foundation suggests that where suitable substrate remained available, lateral growth continued (see fig. 12.6 in Hopley, 1982) (Fig. 7.7b). Woodroffe C. D et al. (2005) recently drew the same conclusion, arguing that prolific midHolocene reef accretion and subsequent late Holocene decline was more likely a function of substrate availability than fluctuations in environmental factors such as SST. Reduced reef flat progradation during this period closely coincides with the mid-Holocene highstand, and direct and indirect consequences of sea-level stabilization and subsequent relative sea-level fall (see Section 3.4.3) may be implicated. Inshore areas of the GBR experienced relative sea levels around 1 m above present around 5500 years BP before falling gradually to present (Chappell et al., 1982; Hopley and Thom, 1983) (see Section 3.4.3). Assuming a
7.4 Holocene reef growth
Figure 7.11 Penrith Island fringing reef internal structure and chronostratigraphy (dates in years BP) (after Kleypas, 1991).
215
216
Fringing and nearshore coral reefs
constant rate of fall (Chappell, 1983), sea levels would have dropped 0.10–0.15 m by 4800 years BP, possibly enough to ‘‘turn off’’ production over many reef flats and restrict healthy reef growth to the outermost reef flats and reef fronts. As sea level dropped, reef flats constructed during higher mid-Holocene sea-level stages became gradually more emerged and back reef areas in particular were modified. Increasingly prolonged subaerial exposure has obvious impacts on most reef flat communities, but those that remain submerged on the reef crest and front may also be badly affected as sediment flux and water properties draining from back reef areas are changed (discussion: Smithers et al., 2006). Rooney et al. (2004) recently reported the collapse of reef-building communities and Holocene reef accretion around 5000 years BP in Hawaii, and speculated that changes in wave climate driven by changes in ENSO frequency and intensity were responsible. Vertical accretion rates calculated for GBR fringing reefs based on radiometrically dated drill cores also declined rapidly and significantly after approximately 5000 years ago (Fig. 11.3). The average reef growth rate for GBR fringing reefs between 5000 and 3000 years ago (2 m ka1) is around half the rate calculated for between 7000 and 5000 years ago (4–5 m ka1). Furthermore, the average rate of vertical accretion on fringing reefs over the last 1000 years has fallen to an even lower rate of 0.75 m ka1, clearly documenting diminishing reef growth capacity since the mid-Holocene. Lateral progradation cannot proceed quickly if vertical accretion reduces, especially under the most common circumstances where fringing reefs prograde into progressively deeper water. It is thus tempting to speculate that the reduced progradation detected for many fringing reefs on the GBR during this period, at least in part, may be a consequence of a broader pattern of diminished vertical accretion and reef vigor. Importantly, apparently healthy coral communities veneer the reef fronts and slopes of many of these fringing reefs, but they have not contributed to reef construction in a significant way over the last 5500 years. Massive Porites and rapidly growing branching Acropora are key reef-builders on inshore reefs in the southern GBR, and reef accretion capacity is reduced if they are absent (van Woesik and Done, 1997). Harriott and Banks (2002) also concluded that latitudinal variation, the presence/absence of key reef-building taxa combined with spatially different calcification rates and disturbance regimes, are the major controls on reef accretion capacity in eastern Australia. Reef cores show that massive Porites colonies and Acropora shingle are the dominant biogenic structural components for most GBR fringing reefs, supporting this conclusion and demonstrating that these taxa have been important reefbuilders since fringing reef growth began. Detailed data on contemporary
7.4 Holocene reef growth
217
community attributes are available for several fringing and nearshore reefs (e.g., Done, 1982; Veron, 1987; van Woesik, 1992; Ayling and Ayling, 1995; van Woesik and Done, 1997; Ayling et al., 1998; van Woesik et al., 1999), and these components are commonly observed on most modern fringing reef slopes, suggesting factors other than the mere absence or presence of these elements determine reef accretion potential. Studies by several researchers have compared contemporary community data with data from fringing reef cores to investigate how paleoecological factors such as community composition, demography, disturbance resilience, calcification rates, and taphonomy might influence reef growth (Kleypas, 1991; van Woesik, 1992; Kan et al., 1997; van Woesik and Done, 1997). Focusing on fringing reefs of the southern GBR, van Woesik (1992) identified three benthic community traits that Kleypas (1991) suggested affect reef growth in that region. Van Woesik (1992) determined that as tide range and turbidity increased: (1) hard corals were generally larger and soft corals smaller; (2) hard and soft coral diversity diminished; and (3) encrusting and plate corals became more common. Near Broad Sound, where the tidal range exceeds 10 m and turbidity is very high, van Woesik and Done (1997) found reef framework corals were significantly reduced in abundance and longevity, retarding reef growth. Branching Acropora sp. were especially rare, and were replaced by an increase in encrusting and foliaceous corals that less effectively build reefs. Similar reductions in growth rates, diversity, and morphology are documented from other reefs affected by high turbidity (e.g., Cortes and Risk, 1985; Rogers, 1990). However, relatively high coral growth rates, coral cover, and diversity are reported from turbid inshore reefs of the central GBR (Veron, 1995; Smithers and Larcombe, 2003), where moderate to high turbidities are commonly experienced. It is likely that the combination of very high tidal range and high turbidity experienced around Broad Sound are exceptionally limiting (Kleypas, 1991) (see Section 4.6). Kan et al. (1997) investigated coral community dynamics and reef structure on the fringing reef at Hayman Island, where framework is the dominant structural component, unlike most fringing reefs on the GBR where detrital accumulation dominates (Hopley et al., 1983). As a result, a better record of community change is preserved in the reef’s structure. Careful examination of reef sections exposed during harbor excavations indicated that large (>2 m) massive Porites colonies are the dominant structural element of most of the reef flat. Significantly for our hypothesis of reduced coral vigor and reef growth after 5000 years ago, Kan et al. (1997) recognized a significant reduction in the dominant size of Porites corals around 3800 years ago to <1 m diameter, and linked this with falling sea level at this time (indicated by a
218
Fringing and nearshore coral reefs
lower terrace on the reef flat). Furthermore, finer-scale observations show community structure changes from a lower unit dominated by fast-growing Acropora to another dominated by slower-growing tabular and massive corals. This change is interpreted as recording the shift from an incipient reef unconstrained by sea level to a shallowing reef near to or at a stable sea level. Unfortunately, however, this technique is less usefully applied on reefs of dominantly detrital structure, and cannot resolve why healthy coral communities now commonly fail to build reefs. (c) Reef flat quiescence between 3000 and 2500 years BP A group can be identified in those reefs that continued to prograde after 5500 years BP that later ceased to actively build out around 3000–2500 years BP (Fig. 7.9). This group includes at least four fringing reefs on the GBR and Torres Strait (e.g., Goold Island: Chappell et al., 1983; Hayman Island: Hopley et al., 1978; and Yam and Hammond Islands in Torres Strait: Woodroffe et al., 2000). Reefs with this history are widely distributed, occurring from Torres Strait to the Whitsundays. This may suggest the influence of broader-scale influences such as continued sea-level fall, or possibly late Holocene climate change. SSTs were cooler at the peak of the mid-Holocene highstand and calcification rates may have fallen (Lough and Barnes, 2000). Rainfall was also becoming increasingly irregular (Gagan et al., 1996), increasing the frequency of droughts, high magnitude floods, and possibly sediment delivery to the GBR, with reductions in reef health and growth as earlier discussed (McCulloch et al., 2003b). Although early data suggested elevated cyclone activity at this time (Chappell et al., 1983) recent work suggests that cyclone frequency in the central GBR has changed little since the mid Holocene (Hayne and Chappell, 2001). However, other evidence does suggest that higher-intensity cyclones occurred more often along the coast between Townsville and Cooktown (Nott and Haynes, 2001; Nott, 2003a) (see Section 4.5), where many fringing and nearshore reefs are located. It is also possible that the turbidity impacts of cyclones on inshore reef growth were heightened at this time, as the significant volume of easily resuspended sediment in the ISP (well developed by 3000 years BP) became accessible and flushing potential was reduced after the Holocene high-energy window closed. Chronostratigraphic evidence suggests that the reduced extension of fringing reefs at Yam and Hammond Islands determined for between 3000 and 2000 years BP occurred because they outgrew their foundations (Woodroffe C. D. et al., 2000). This scenario may also apply at Hayman and Goold Islands, where reef flat progradation also slowed during this
7.4 Holocene reef growth
219
period. Smithers et al. (2006) suggest that progradation into deeper water, with obvious volumetric consequences for reef construction, presented difficulties for continued lateral growth in some of these reefs. Historical variations in vertical accretion rate and paleodepth established from reef cores (see Chapter 11, Fig. 11.8) suggest that where the depth to substrate exceeds 8 m the rate of vertical accretion on fringing reefs of the GBR may rapidly decline. Thus as fringing reefs build out over substrates at greater than 8 m depth their vertical and lateral accretion potential reduced. (d) The present: fringing reef ‘‘decline’’ The Holocene growth histories of fringing reefs on the GBR show that many are merely surviving rather than actively growing, and that this may have been so for millennia (Fig. 7.9), even where apparently healthy coral communities cover reef fronts and slopes. Fewer than half of the fringing reefs with reef top ages investigated have continued to prograde up to the present, most more slowly than in the mid Holocene (Fig. 7.9). Pioneer Bay, Orpheus Island (Hopley et al., 1983) (Fig. 7.8), Fantome Island (Johnson and Risk, 1987), and Magnetic Island (Chappell et al., 1983) are examples in this group. Perhaps surprisingly, all of the reefs with reef flats that have continued to prograde are established over terrigenous clay or sand foundations (e.g., Fantome Island; Pioneer Bay, Orpheus Island Fig. 7.8)), with the exception of the southern part of Iris Point, which overlies detrital sediments shed from the northern part of the same reef (Hopley and Barnes, 1985). This longer-term survival may arise because reefs in this setting have grown over more steadily sloping substrates that do not suddenly drop below the critical 8 m depth. Alternatively, it may reflect the preferential selection of more resilient taxonomic assemblages and/or individuals in these less than ideal settings, but such an hypothesis remains to be tested. Current environmental stresses that may equal or exceed those which accompanied previous declines in fringing reef progradation earlier in the Holocene include: *
*
*
further natural declines in water quality as the inshore mud wedge has continued to accumulate, and north-facing bays which trap a significant proportion of these sediments are filled (Orpin and Ridd, 1996). This may reduce fine sediment deposition in natural depocenters and increase natural turbidity levels; the Holocene high-energy window is now closed. Flushing by waves and wavegenerated currents has diminished relative to that experienced in the mid Holocene (Hopley, 1984); sediment and contaminant delivery to the inshore GBR lagoon is reported to have increased due to land-use changes in adjacent catchments (Furnas, 2003;
220
*
Fringing and nearshore coral reefs
McCulloch et al., 2003b). Contaminants associated with the sediment present additional problems for coral recruitment, resilience and reef growth (Fabricius and Wolanski, 2000; Fabricius et al., 2003); reduced coral cover and calcification due to greenhouse gas emissions which are reported to have raised sea temperatures, increased the frequency of bleaching, and reduced calcification rates (Hoegh-Guldberg, 1999; Kleypas et al., 1999b).
It is also likely that optimal substrates for colonization and reef growth are becoming increasingly rare, as most are already occupied. This may be a significant limitation as conditions for start-up and survival become tougher due to natural and anthropogenically forced declines in inshore water quality. (e) Reefs initiated since 3000 years BP The nearshore reef at Paluma Shoals lies over stiff Pleistocene clay presently around 5 m below LAT (Fig. 7.12), although it is possible that reef growth was initiated on an alluvial gravel fan in a similar fashion to fringing reefs at Cape Tribulation (Partain and Hopley, 1989) (Fig. 7.3). It is less than 2000 years old,
Figure 7.12 Nearshore reefs, Lugger and Paluma Shoals: location, surface morphology, and internal structure and chronostratigraphy (after Smithers and Larcombe, 2003; Perry and Smithers, 2006).
7.4 Holocene reef growth
221
with basal coral near the reef front dated at 1657 years BP (Smithers and Larcombe, 2003). Paluma Shoals is located in a narrow corridor between the inner edge of the ISP and an eroding sandy coastline (Figs. 7.10 and 7.12). Importantly, Paluma Shoals has grown over a substrate that has only recently become available following shoreline retreat, indicated by extensive mangrove sediments of late Holocene age exposed along this coast (Smithers and Larcombe, 2003). This model of development provides an interesting contrast to that proposed for Yule Point, where the coast is argued to have prograded out to the nearshore reef and captured it as a fringing reef (Bird, 1971a). Paluma Shoals, and possibly other nearshore reefs, have managed to recently establish and grow in conditions traditionally considered marginal for reef growth. Turbidity at Paluma Shoals exceeds 40 NTU (nephalometer turbidity units) more than 30% of the time (Larcombe et al., 2001). Nevertheless, the reef was able to initiate and grow because waves close to the shore kept sediments in suspension and appropriate substrate was available at shallow enough depth to allow light-enhanced calcification in turbid water. The inshore zone between the sediment wedge and shore includes substrate at depths within the 4–8 m optimal accretion envelope; at Paluma Shoals the mean tide depth to the initiation substrate is approximately 6 m. Turbid water makes these reefs difficult to observe, but they are reasonably common along most of the GBR coast. Significantly, stratigraphic cores indicate that although the reef flats can be dominated by live coral, the internal structure is dominated by detrital carbonate clasts, often deposited in distinct units and interpreted as signifying the importance of episodic accretion events such as cyclones (Smithers and Larcombe, 2003) (Fig. 7.12). Preliminary investigations at Lugger Shoals reveal an equivalent structure (Perry and Smithers, 2006). Rigidity is mostly provided by a ‘‘capping’’ of detrital carbonate material which forms the substrate for the contemporary reef flat assemblages which include abundant Porites and Goniastrea microatolls. The internal structure does not appear to have been significantly bound by either biological or chemical means and the long-term persistence of individual reefs of this type is unknown. Nearshore reefs of this type are clearly vulnerable to shifts in the position of the shoreline and the inner limit of the nearshore sediment wedge, either of which might shut down production, and their stratigraphy confirms that cyclonic events can erode and redistribute their sedimentary components. Taphonomic examination of detrital clasts from Lugger Shoals shows remarkably little boring and bioerosion, suggesting the possibility that accumulation of these essentially unconsolidated detrital structures is enhanced by the reduced breakdown and export of carbonates produced on the reef. Their largely unbound internal construction and limited
222
Fringing and nearshore coral reefs
thickness due to depth constraints would appear to heighten their vulnerability to disintegration and dispersal if accretion rates stall. Unlike the fringing reefs that accumulated most of their bulk in the mid Holocene, it is possible that little fossil trace of these reefs would persist, other than a temporary shingle strandline on a sandy beach. Perry (2003a, b) has described reefs in a similar turbidity and sedimentary setting in Mozambique, which historical accounts indicate are ephemeral. Turbid patch reefs of Miocene age have, however, been reported (e.g., Wilson, 2005), demonstrating that some reefs of this type are preserved in the fossil record. Paluma Shoals is unusual compared to most nearshore reefs on the GBR in that it has vertically accreted to sea level and has formed a well-developed reef flat. Middle Reef in Cleveland Bay between Townsville and Magnetic Island (198 120 S, 1468 480 E), and Lugger Shoals at Lugger Bay have similarly grown to sea level, but have not yet developed conspicuous reef flats. However, many other nearshore reefs in Halifax Bay, and possibly elsewhere along the GBR coast, are not presently sea-level constrained. Why such differences exist is unknown but may reflect a later start-up, slower accretion, or possibly the impacts of disturbance and destruction. We note that species diversity at the inshore reefs of the GBR (Veron, 1995) and also at the Mozambique fringing reefs described by Perry is relatively high, emphasizing the point that community diversity and reef-building capacity need not be correlated (Perry and Larcombe, 2003). 7.5 Fringing reef morphology and processes Detailed descriptions of the surface morphologies of various GBR fringing reefs are available in several accessible references including Bird (1971a), Hopley (1975), Hopley et al. (1978), Hopley (1982), Chappell et al. (1983), Hopley et al. (1983), Hopley and Barnes (1985), Johnson and Risk (1987), Kleypas (1996), Kan et al. (1997), and van Woesik and Done (1997). Where these fringing reefs have formed reef flats, their surface morphology is usually simple, comprising three main zones – a fore reef, reef crest, and back reef. Although wide reef flats typically form on the windward sides of atolls and platform reefs, leeside bayhead fringing reefs (e.g., Pioneer Bay, Orpheus Island; Fantome Island) are amongst the widest on the GBR (e.g., Hopley et al., 1983; Johnson and Risk, 1987) and the rarer windward fringing reefs tend to be narrow and quite poorly developed (e.g., Wolanski et al., 2005). Biological and morphological zonation is often weakly developed on GBR fringing reefs due to their generally shallow foundations, and especially in leeward or sheltered locations with lower energy regimes (Hopley, 1982).
7.5 Fringing reef morphology and processes
223
Fringing reef morphology can vary, however, over a single reef and from reef to reef, with the greatest complexity usually developed where the reef flat comprises sections formed during and after the mid-Holocene highstand (see Section 3.4.3). As shown in Section 7.4 (see Fig. 7.9), many fringing reefs on the GBR were largely emplaced soon after the mid Holocene, with relatively small additions laterally accreted since (see Figs. 7.3, 7.5, and 7.8). This age structure was established for many fringing reefs of the central GBR by early research by Hopley (e.g., 1975, 1982), and was confirmed as a broader regional trend by Chappell et al. (1983) who dated fossil microatolls across seawardsloping fringing reef flats through the northern and central GBR. The morphological legacy of reef flat construction at higher mid-Holocene sea levels and subsequent emergence remains strongly expressed on many of the larger fringing reefs on the GBR (see discussion later in this section). These fringing reefs are commonly dominated by relatively large back reef areas of dead or sediment-covered reef flat (often with fossil microatolls) elevated up to 1 m above the modern living reef which is confined to a narrow band near the reef edge (Figs. 7.3, 7.5, and 7.8). Typically the elevated older reef grades to a zone of live coral dominated by microatolls formed by Porites, Goniastrea, and various faviids, together with branching Montipora and Acropora, before deepening over the reef slope. The morphology of fringing reef back reef zones can be complicated by the interaction of terrestrial and marine processes. Sediment and runoff from terrestrial catchments can produce deltas and sand bars over the inner reef, as observed at Pioneer Bay, Orpheus Island (Fig. 7.8). At Orpheus and most other continental high islands seasonal streams draining relatively small catchments discharge onto fringing reef flats only during and immediately following heavy rainfall, briefly reducing salinity on the inner reef flat and delivering sediment during short-lived events (Parnell, 1988a). Live corals are scarce on most inner reef flats due to their elevation above the open water tide, and these infrequent freshwater discharges usually have negligible impact on living reef at the edge of broad reef flats. Over time, however, inner reef flats become dominated by terrigenous sediments; this has also been observed on the fringing reef at Geoffrey Bay, Magnetic Island (Smith, 1978; Umar et al., 1998), and also at Yule Point (Bird, 1971a) and Hydeaway Bay (Bland, 2004), both of which are fringing reefs attached to the mainland. At Hydeaway Bay, where the fringing reef is not confined in protected bayhead, terrigenous sediment delivery to the inner reef flat as small creek mouth deltas and its subsequent redistribution alongshore is the subject of an ongoing study. The physical processes acting on GBR fringing reefs were described in detail in Chapter 4. As indicated, consistent reversing tidal streams dominate the
224
Fringing and nearshore coral reefs
circulation patterns experienced by fringing reefs on the mainland and offshore high islands (Hamner and Hauri, 1977; Parnell, 1987, 1989) (Section 4.3.1), and on bayhead fringing reefs these interact with the local morphology (Parnell, 1987). In Pioneer Bay, Orpheus Island, up to 90% of the water in the bay is replaced over a tidal cycle (Parnell, 1987). Current velocities over the reef flats are generally below 0.2 m s1, which is insufficient to entrain sandsized sediment. The interaction of tidal and wave-generated currents has been shown to be important in initiating and maintaining sediment transport over fringing reefs elsewhere (e.g., Ogston et al., 2004; Presto et al., 2006), but in leeward fringing reef settings on the GBR ambient wave energy is very low. The importance of bioturbators for initiating fine sediment transport was discussed in Section 4.3.3, and has been shown to be critical in the development of a band of fine terrigenous sediment across the center of the reef flat at Pioneer Bay (Parnell, 1988a). The low energy experienced under ambient conditions at the rear of many leeward and/or elevated fringing reefs provides a suitable low-energy environment for colonization by mangroves; stands of Rhizophora and Avicennia in particular are commonly established (see Fig. 7.8, Pioneer Bay). Mangrove communities also develop at the rear of exposed mainland fringing reefs; as an example, a broad corridor of mangroves occurs at the back of the fringing reef at Yule Point, where the elevated back reef limits wave energy at most tidal stages and thus provides an environment where mangrove colonization and establishment can proceed with greater success. As discussed in Chapter 4, the passage of wave energy across most reef flats is strongly controlled by the tide (Hardy et al., 1991a, b; Angwenyi and Rydberg, 2005; Kench and Brander, 2006). On most reef flats sediment entrainment and transport occur only within a narrow depth window under ambient conditions, where the water is deep enough to allow significant waves to pass over the reef crest without breaking and dissipating, but shallow enough for these waves to generate significant current velocities at the reef flat surface. On many leeward fringing reefs, where the mean significant wave height can be very small, mean wave energy alone might be too low to mobilize coarser sediments under most conditions. However, fine sediments may be resuspended very quickly by waves (at Nelly Bay, Magnetic Island, turbidity increases of three orders of magnitude up to 100 NTU have been measured in an hour: Orpin et al., 2004) and kept entrained by much weaker currents acting for only short periods (Larcombe and Woolfe, 1999a, b; Larcombe et al., 2001; Wolanski et al., 2005). The dynamics of fine sediment transport and deposition on the windward and leeward fringing reefs of High Island (178 160 S, 1468 010 E), a turbid
7.5 Fringing reef morphology and processes
225
inshore continental high island south of Cairns, have recently been examined (Wolanski et al., 2005). There it was found that fine sediment accumulated over shallower sections of the fringing reefs during calm periods, but during stormier conditions (equivalent to those expected on 20 days per year on average) resuspension occurred to a depth of 12 m on the windward reef slope, and 8 m on the leeward reef slope. Variation in coral cover at shallow depth was noted between the windward (50%) and leeward (20%) reefs, but at both it decreased rapidly with depth, and was replaced with a steep muddy slope below the resuspension thresholds, where flushing is rare and accumulation dominates. In Chapter 11 it is demonstrated that optimal vertical accretion rates on fringing and nearshore reefs on the GBR occurred in paleodepths shallower than 8 m. It is tempting to infer that the interplay of resuspension, sedimentation, and coral and reef growth demonstrated by this recent study at least partly controls this depth distribution. Although fringing reefs behind outer barriers have been described from other locations such as New Caledonia (Cabioch et al., 1995, 1999), most of the fringing reefs studied elsewhere are not as protected as those on the GBR, nor are they located so close to large terrestrial catchments (Kleypas, 1991). Two catchments draining into the GBR exceed 130 000 km2 in area, and the total catchment area draining into the GBR is nearly 425 000 km2 (see Section 4.6.2). Exposure to higher wave energy usually enables greater flushing (Hearn and Parker, 1988; Presto et al., 2006), which can ameliorate the debilitating effects of poor water quality at both regional and across-reef scales. For example, wave-driven cross-reef flow hinders polluted tidal creek discharge onto a fringing reef near Mombasa, Kenya (Angwenyi and Rydberg, 2005), and the dispersal of suspended sediments away from areas of coral growth by wave-driven currents is well documented (e.g., Roberts and Suhayda, 1983; Ogston et al., 2004). Furthermore, as indicated in the introduction to this chapter, fringing reefs are common on uplifting coasts, and uplifted last interglacial and earlier fringing reefs occur behind several well-studied fringing reefs (e.g., Webster et al., 1998; Collins et al., 2004). Where this occurs, streams draining terrestrial catchments pass through the porous reef which filters flow before discharging to the reef flat or lagoon. On the GBR, high terrestrial runoff into a relatively low-energy and protected lagoon routinely exposes inshore corals to ambient levels of turbidity and sedimentation known to be stressful and possibly fatal elsewhere (Pastorak and Bilyard, 1985; Hopley and Choat, 1990), and stratigraphic evidence suggests that inshore and fringing reefs have evolved under naturally turbid conditions (Hopley et al., 1983; Johnson and Risk, 1987; Smithers and Larcombe, 2003). Many corals on these reefs grow vigorously, and have adapted to these turbid conditions (Anthony, 2000; Anthony and Larcombe,
226
Fringing and nearshore coral reefs
2002). Paradoxically, however, the apparent vigor of reef corals does not appear to directly translate to continued reef accretion, as discussed in Section 7.4, most fringing and nearshore reefs have not added much to their structures for several thousand years (Smithers et al., 2006). Evidence of the impact of recent increases in terrestrial runoff on reef growth on the GBR will be discussed in Chapter 13. Although the GBR lagoon is now a relatively low-energy environment, with major sediment transport mostly occurring during higher-energy storms and cyclones (Davies and Hughes, 1983; Done, 1992c; Larcombe and Carter, 2004), this has not always been so. On the central GBR in particular, ambient energy levels were higher than present between around 8500 and 6000 years BP, when the outer barrier had not yet reached sea level and impeded the penetration of ocean waves and swells into the GBR lagoon. Hopley (1984) has termed this period the Holocene high-energy window, and attributed the deposition or reworking of high-energy features such as boulder beaches and spits (Hopley, 1971) to the higher energy levels experienced during this period. Some of these features are preserved at the rear of fringing reef flats; for example, the boulder beach at the rear of Iris Point, and shingle ridges deposited behind the mangroves in the south-eastern corner of Pioneer Bay, and are essentially relic features formed under an earlier process regime (Figs. 7.4, 7.5, and 7.8). Complex morphologies can develop where well-developed fringing reefs occur in more exposed settings, especially where the reef flat contains elements of mid-Holocene and more recent age. This is borne out by descriptions of reef flat morphology at Iris Point, Orpheus Island (Figs. 7.4 and 7.5). Hopley and Barnes (1985) described seven morphological zones on the northern section of the reef flat at Iris Point, the higher inner part of which reached sea level in the mid Holocene (based on a 6250-year-old reef flat faviid). Only five zones were discriminated on the younger (near-surface date: 3780 years BP) southern section of the reef flat (there are eight distinctive morphological zones in total on this reef). Although Iris Point is somewhat unusual for a windward fringing reef on the GBR due to its size, many of the morphological zones identified there are represented on other fringing reefs exposed to moderate energy levels that have experienced a late Holocene sea-level fall. As discussed by Hopley (1982), many exposed fringing reefs exhibit morphological features such as shingle ridges and bassett edges which are common also to low wooded islands (see Chapter 10). The morphological zones depicted in Fig. 7.5 at Iris Point were delineated and described in detail by Hopley and Barnes (1985). Many of these zones also occur on other fringing reefs in the GBR, and they are briefly outlined below:
7.5 Fringing reef morphology and processes
227
(a) High water shingle and debris. An irregular ridge of mainly Acropora shingle to 10 cm length but also including other smaller reefal debris, pumice, and flotsam is deposited at the highest level of contemporary storm activity. This ridge can rise to over 1 m in height, and usually has a steep seaward face. The lee side of the ridge is often less abrupt, and can include small fans or splays. (b) Terrigenous boulder beach. At Iris Point, well-sorted and rounded rhyolite and basalt boulders winnowed from regolith form a boulder beach that lies behind the reef flat and extends beneath it. The upper boulder beach extends to around HAT, and the beach averages around 35 m wide. The lower boulder beach is cemented but the upper beach is not. In situ corals indicative of a higher (1 m) sea level in the mid Holocene have been identified attached to the boulder beach (Hopley and Barnes, 1985). Boulder beaches and spits are common on the islands of the central GBR and Whitsundays, and have been described in detail by Hopley (1971). Most appear to have been reworked by more energetic wave conditions associated with the Holocene high-energy window (Hopley, 1984), when the outer barrier lagged below sea level and oceanic waves and swell penetrated into the inner Central GBR. (c) Inner moat. A moated pool extends from the base of the boulder beach seaward to the shingle rampart. At low tide water to around 0.3 m deep is held within the moated pool, more than 1 m above the open-water low tide level. Porites and Goniastrea microatolls amongst other corals and reef organisms grow in this moat. A veneer of poorly sorted carbonate sediments floors much of this feature, with abundant rhodoliths accumulating at the landward edge. (d) Shingle rampart. A shingle rampart or ridge to 0.3 m high and around 10 m wide was located at the seaward edge of the inner moat when the reef flat was mapped by Hopley and Barnes (1985), located over a broader rubble zone. Parts of this rampart remain, but much of it has been destroyed over the intervening 20 years. Subtle bassett edges are sporadically preserved that document the former extent and position of this feature. (e) Algal terraces. An algal pavement forms a relatively smooth feature that slopes down from the higher inner reef flat and shingle rampart to the lower outer living coral zone along most of the northern part of the reef flat at Iris Point. Localized patches of this pavement are formed by crustose coralline algae, but it is mostly composed of Acropora shingle that has been stabilized with a filamentous turf algae. Within this algal turf matting foraminiferans such as Amphistegina, Calcarina, and Baculgypsina are produced in great abundance. (f) Northern fossil microatolls. A pavement of coalesced fossil microatolls, some reaching 3 m in diameter, is located on the outer northern part of the reef flat at Iris Point. The tops of these microatolls are elevated 0.8 m above their modern open-water counterparts. There is no evidence that these microatolls have been moated, and they are therefore presumed to have formed at a higher sea level. (g) Amorphous southern reef flat. This zone is developed where the reef achieves its greatest width, and essentially consists of a relatively uniform, gently sloping
228
Fringing and nearshore coral reefs
surface of coral rubble bound by pink crustose coralline algae on its outer half and filamentous brown algae on the inner half. Radiocarbon dates indicate that this section of the reef flat was constructed after the mid-Holocene highstand. Hopley and Barnes (1985) speculated that this possibly accounted for the morphological simplicity of this section of the reef flat. (h) Outer living coral zone and reef slope. Live coral cover and biodiversity increases below MLWS on the outer reef flat. Soft corals (Lobophyton, Sinularia) and massive or head corals such Porites and Goniastrea dominate at the rear of this zone, with Acropora becoming more abundant toward the reef crest. Patch reefs seaward of the crest are sporadically distributed along the reef front. Bleaching in 1998 severely affected this zone, but recovery appears to be under way, with new recruits observed on recent visits.
The complexity observed at Iris Point is demonstrative of the significant impact late Holocene relative sea-level fall may have in structuring the morphology of moderately exposed fringing reef flats on the GBR. Hopley and Barnes (1985) recognized this effect at Iris Point, and noted comparable morphological development of the reef flat at Redbill Reef (Fig. 8.7), a platform reef that has experienced a similar late Holocene sea-level history (Hopley et al., 1984). The terrigenous boulder zone described at Iris Point is less commonly represented elsewhere, with shingle or sandy beaches, often with beachrock outcrops, more often located behind fringing reef flats (see Bird, 1971a; Hopley, 1975; Hopley et al., 1983; Johnson and Risk, 1987; Partain and Hopley, 1989). At some locations, such as at Curacoa Island, the backshore is composed of a sequence of storm-emplaced shingle ridges (Hopley, 1968; Hayne and Chappell, 2001; Nott and Hayne, 2001), although attributing similar deposits to storms or higher mid-Holocene sea level can be difficult (see discussion in Hopley, 1982; Chapter 6). As indicated above for Iris Point, a shingle ridge commonly develops near the upper tidal limit, our observations suggesting largely due to swash processes on the shoreface during higher phases of the tide. The development of moated pools over emergent reef flats is common on many fringing reefs (and also low wooded islands) (see Chapter 10), especially in moderately exposed settings where shingle ramparts can be thrown up on the seaward edge by storms (Moorehouse, 1933; Hopley, 1975). These ramparts can become partially cemented and form sills that moat water above the open-water tide (Hopley and Isdale, 1977; Scoffin and Stoddart, 1978). Although moating can occur at any intertidal level, on the GBR most moats pond water below the mean tide level (i.e., usually <1 m above MLWS). Water levels moated behind reef flat ramparts on reef islands visited by the 1973 Royal Society–Universities of Queensland Expedition on the northern
7.5 Fringing reef morphology and processes
229
GBR are unusually high, with a mean maximum elevation 1.8 m above tidal datum, just 0.5 m below MHWS (Stoddart et al., 1978a, b). Relatively luxuriant coral growth may extend to the moated water level, with over 40 species of coral recorded from moats on the GBR (Bennett, 1973). However, as noted earlier for Iris Point (Hopley and Barnes, 1985) and at other fringing reefs such as Holbourne Island (Rainford, 1925; Hopley and Isdale, 1977), the integrity and elevation of ramparts and moated water levels may be modified during high-energy storms and cyclones, lowering moated water levels and causing catastrophic mortality of moated communities. Nearshore reefs occur in a variety of settings and at various stages of development. Published descriptions of their morphology are rare, but a series of broad (400 m windward to shore) coral rubble armored reef flats have developed near LAT at Paluma Shoals (Smithers and Larcombe, 2003). The seaward reef slope is steep but short, rising from the seabed at around 5 m below LAT. Three zones are developed over the reef flat. A narrow (20–50 m) shingle and rubble covered zone borders the reef crest, behind which a zone dominated by Goniastrea sp. microatolls occupies most of the reef flat. The leeward zone is locally indistinct and narrow, and consists almost exclusively of large stands of Galaxea fascicularis. Similar structures to Paluma Shoals can be observed elsewhere along the coast, but relatively few appear to have developed large reef flats (e.g., Ollera Shoals (178 580 S, 1468 050 E), Lugger Shoals (188 590 S, 1468 220 E)). At Lugger Shoals the seaward edge of the reef has formed an incipient reef flat, but to leeward the contiguous surface is replaced by isolated Porites colonies which rise from 1–2 m depth that have reached sea level and have adopted a microatoll morphology. Middle Reef offshore from Townsville is difficult to observe due to high turbidity, but when it is exposed during low spring tides it is apparent that it too has high coral cover, and is composed of a series of patch reefs confined at sea level, analogous to reef patches in Hopley’s (1982) evolutionary classification of reef types (see Fig. 5.6). The reasons for such differences are unclear, but may reflect different ages, different growth modes, different disturbance regimes, alone or in any combination. Ayling et al. (1998) indicate that few of the shoals they surveyed in Shoalwater Bay showed ‘‘true reef development,’’ with most comprising shallow (at low tide – the tidal range here is up to 8 m) rubble banks with low topographic complexity, biodiversity, and coral cover. These reefs were initiated later and achieved sea level after the mid-Holocene highstand, and have not developed reef flats like Paluma Shoals. The physical processes affecting nearshore reefs are very similar to those affecting fringing reefs; excellent accounts of the dominant physical processes at Paluma Shoals are provided by Woolfe and Larcombe (1998) and Larcombe et al. (2001).
230
Fringing and nearshore coral reefs
As noted by many authors, fringing and nearshore reefs are especially vulnerable to catastrophic disturbances due to their proximity to land, such as flood plumes (van Woesik et al., 1999; Devlin et al., 2001) and turbidity and sediment influx (Marshall and Orr, 1931; Wolanski et al., 2005). Others have argued that chronic stresses related to diminished water quality and elevated contaminant and nutrient delivery to inshore waters have directly contributed to such collapses, but have also lowered reef resilience and increased their vulnerability to other events (see Fabricius, 2005). Debate about the impacts of terrestrial exports on the health of the GBR is ongoing (see Chapter 13). It is, however, relevant to emphasize here that many fringing and nearshore reefs are (i) growing at their environmental extremes, and (ii) are relatively old features at the ‘‘mature to senile end of the evolutionary spectrum’’ (see Chapter 5, Fig. 5.6). Although the pressures facing fringing and nearshore reefs may have increased during recent times due to anthropogenic activities, it is critical to note that many fringing reef flats have been largely dead for thousands of years due to late Holocene sea-level fall (see Section 3.4.3), and the net calcium carbonate productivity of most fringing reef reef-slope communities has been insufficient to enable significant lateral accretion for several thousand years (Smithers et al., 2006). Where coral communities flourish over fringing reef flats on the GBR it is usually in moated pools, where as discussed above they may suffer episodic mortalities when cyclones destroy moat sills and lower ponded water levels (Hopley and Isdale, 1977), or where fresh water is ponded over elevated reef flats following heavy rainfall events (Rainford, 1925). The vulnerability of reef flat communities, including those on fringing reefs, to such natural perturbations must be recognized when assessing the significance and causes of changes in reef flat community status. A comparable study of fringing reefs on the GBR is not available, but Connell et al.’s (1997) investigation of long-term dynamics of reef crest corals at Heron Island (238 260 S, 1518 570 E) clearly demonstrates vulnerability of reef flat communities to natural episodic changes such as cyclones. In addition to identifying immediate impacts, Connell et al. also showed how coral cover may gradually decrease over decades when ramparts are locally disrupted, as lower moated water levels over a broad area increase reef flat exposure at low tide. Thus, although studies such as comparisons of historical and recent photographs of fringing reef flats provide an important temporal record of condition, as acknowledged by Wachenfeld (1997), they offer a very coarse view of community health trajectories with a range of limitations. Unfortunately, the visual impact of previously luxuriant and currently dead reef flat communities is far more attractive to less critical users of this information than limitations regarding the interpretation and/or broader significance of such results.
7.6 Conclusion
231
Similar caution is also advised where diminished reef health is inferred from reduced recruitment following disturbances. Where mortalities have occurred because moats have been lowered, significant post-event recruitment will not occur unless moated water levels are re-established, and this may not occur for some time, if at all (Connell et al., 1997). 7.6 Conclusion There has been great interest in the geomorphology of fringing and nearshore reefs in last 30 years, but with only 24 of more than 758 fringing and nearshore reefs in the GBR and Torres Strait investigated in detail, our knowledge of these systems remains incomplete. Fringing and nearshore reefs grow along the mainland coast and adjacent to continental high islands, but are more regularly and better developed about the latter. Indeed, fringing reef growth is generally poor along much of the mainland coast. On both mainland and high island shores fringing and nearshore reefs occur in four basic coastal settings: as headland-attached fringing reefs; bayhead fringing reefs; as narrow beachbase fringing reefs; or as shore-detached nearshore shoals. Coring investigations reveal that fringing reef structures may be complex and diverse, developing over a range of foundations and constructed of a variety of materials deposited in numerous arrangements (Fig. 7.7). Of course, reefs are three-dimensional structures, and significant structural variation may develop at various locations within a single fringing reef. Holocene reef growth histories available for fringing and nearshore reefs of the GBR clearly indicate that most experienced a period of active growth from 7000 to 5000 years ago when they produced most of their structure, and that many have struggled since. The reef flats on most of the fringing reefs for which we have data have laterally prograded by less than 10% of their total width over the last 4000 years. Several periods can be identified when reef flat progradation stalled, suggesting that the impacts of regional-scale factors such as climate change and the evolution of the shelf reefs and sediment bodies may be involved. Clearly, however, many fringing reefs on the GBR are mostly old features formed at higher sea levels, with living coral restricted to a narrow band near the seaward reef edge, and to relatively shallow depths on the reef slope due to turbidity and sedimentation constraints. Although most fringing reefs are morphologically simple, their morphologies may become more complex where they are affected by terrestrial inputs and processes, and where they incorporate reef flats formed at higher midHolocene sea levels. As noted above, this is common on the GBR, where fringing reef back reefs are elevated above the modern living communities at
232
Fringing and nearshore coral reefs
the reef edge. Where coral communities survive on the higher reef flat it is often in moated ponds held above the open-water level at low tide behind ramparts and sills. Such communities have a naturally tenuous existence, being highly vulnerable to catastrophic mortality events caused by moat destruction or moated pond salinity changes. Many researchers now believe that the longterm fate of many fringing and nearshore reefs on GBR is also tenuous due to the proximity to terrestrial and anthropogenic impacts, a debate examined in Chapter 13. The following chapter examines the Holocene growth and evolution of mid-shelf reefs, which are less affected by terrestrial pressures, and have experienced different Holocene sea-level histories.
8 The mid-shelf reefs of the Great Barrier Reef
8.1 Introduction The evolutionary classification for coral reefs of Hopley (1982), although applied elsewhere, was initially devised specifically for the reefs of the Great Barrier Reef (GBR). The classification (Figs. 5.6 and 5.7, Table 5.3) shows antecedent platforms being recolonized, with reef growth during initial juvenile stages being largely vertical, enhancing the original relief of the antecedent surface. A mature phase after reefs reach or catch up with sea level follows, during which both growth and sedimentation infill and obliterate the previously constructed relief. The senile phase consists of flat-topped or planar reefs. At the time the classification was devised there were few drill holes which had penetrated the Holocene reef on the GBR and great reliance was put on seismic refraction survey to identify the depth of the Holocene (Harvey, 1977a, b, 1980; Harvey et al., 1979) and approximately 102 radiocarbon dates from surface or near-surface samples from 24 reefs (Hopley, 1982, table 9.2). A number of intuitive presumptions were also made at this time and these formed many of the questions asked by the drilling projects on the GBR subsequent to about 1980. These studies have built up a considerable data bank on the age and structure of mid-shelf reefs (Table 8.1). Outer shelf or shelf marginal reefs such as the ribbons of the northern GBR, some of the outermost reefs of the central GBR, and the unique Pompey Reefs of the stepped shelf margin of the south central GBR are discussed in Chapter 9 whilst Chapter 11 synthesizes the structural and growth data of all reefs, fringing, mid-shelf, and outer shelf. This chapter focuses on the morphology of the mid-shelf reefs and the way in which this has evolved during the Holocene. The questions which are asked in this chapter come from the presumptions of the early 1980s and include: 233
From shallow drilling.
9 1 1 4 3 3 2 2 2 1 2 3 4 1 2 5 4 3 4 6 3 3
108 130 148 190 148 260 158 220 158 250 158 440 178 420 178 430 178 500 178 530 188 140 188 380 188 480 188 500 198 130 198 180 208 480 208 380 238 200 238 300 238 380 238 520
Warraber (Woodroffe) Stapleton (Thom) Bewick (Thom) Williamson (Davies) Boulder (Davies) East Hope (Davies) Potter (Graham) Ellison (Graham) Taylor (Graham) 17–065 (Graham) Britomart (Johnson et al.) Grub (Hopley) Wheeler (Davies and Hopley) Davies (AIMS) Darley (Hopley) Stanley (Davies and Hopley) Gable (Hopley) Redbill (Hopley) Wreck (Davies) One Tree (Davies) Fitzroy (Davies) Fairfax(Davies)
a
Number of holes
Latitude
Reef and drilling team 10.0 14.6 30.0 21.0 16.5 18.5 29.7 14.3 17.8 9.3 28.0 13.8 20.5 39.05 13.0 24.0 29.5 19.5 10.5 23.0 14.5 13.0
Deepest (m) 5.7 14.6 4.0 17.0 16.0 15.0 25.2 – – – 19.8 – 20.0 25.7 – 15.0 – 12.1 – 13.0 8.0 8.5
Shallowest (m)
Pleistocene
12 3 6 6 8 0 (þ 2 a) 13 3 8 (þ 7 a) 2 18 15 20 (þ 9 a) 0 5 (þ 6 a) 41 (þ 6 a) 22 13 (þ 11 a) 34 62 20 17
Number of 14 C dates
5770 70 3130 80 6380 120 6460 110 5910 190 2570 100 3800 110 3530 60 5130 140 3760 100 5255 65 4100 80 4550 90 – 6070 130 6420 160 4980 90 4570 150 5730 70 5390 180 5740 180 6320 80
Oldest within 1 m of reef flat
Holocene
7190 100 5260 30 6920 130 6910 170 8320 160 2570 10 8200 120 5130 100 6320 160 5350 90 8925 155 6820 140 7110 240 – 6210 140 7990 110 7890 110 7840 80 5970 80 7800 270 7480 140 7270 90
Oldest
Table 8.1. Results of shallow and medium-depth drilling on mid-shelf reefs of the Great Barrier Reef
8.2 The data base *
*
* *
*
235
does the classification of mid-shelf reefs really represent a temporal sequence of morphological development? what influence does the depth of the Pleistocene antecedent surface have on the stage of development of the reef? what influence does the size of the reef have on the stage of development of the reef? what influence does the morphology of the antecedent Pleistocene reef have on the morphology of what are presumed to be younger reefs in the sequence? once the reef has reached modern sea level, where and how does subsequent reef growth take place?
8.2 The data base The data from the 22 reefs, listed in Table 8.1 and located in Fig. 8.1 are used to investigate these questions. These comprise all mid-shelf reefs for which drilling and dating information exists. A brief description of these reefs, from north to south, is given below with greater detail provided for results which have not been previously published. Outlines of all reefs discussed are shown in Fig. 8.2. 8.2.1 Warraber Reef (108 130 S) Warraber Reef and its populated sand cay are situated in central Torres Strait. Most of the reef platform contains elevated corals indicating a sea level approximately 1 m above present 5 ka ago. There is a step down from this elevated reef to a lower reef flat level on the western end of the reef. Eight holes have been drilled into the western half of the reef (Woodroffe et al., 2000) which established a shallow antecedent Pleistocene platform at 5.7 m. Recolonization had taken place before 7 ka ago and the reef had reached sea level shortly after 6 ka ago. Woodroffe et al. (2000) suggest a model of subsequent reef flat development involving growth of reef crest ridges which remain as framework with shingle matrix and infilling of the back reef with sand. 8.2.2 Stapleton Reef (148 190 S) Stapleton and Bewick (see below) were the two reefs drilled on the 1973 Royal Society Expedition (Thom et al., 1978). Only one hole was drilled, on the leeward side of the sand cay, which itself was on the leeward side of the reef. Sand overlies a coarser Holocene sediment unit. The Pleistocene was interpreted to be at 14.6 m, but the leeside position of this hole would suggest a higher level to windward which could equate with the planar stage of this reef.
236
The mid-shelf reefs of the Great Barrier Reef
Figure 8.1 Location of mid and outer shelf drill sites and other locations mentioned in Chapters 8 and 9.
8.2 The data base
Figure 8.2 Outlines of reefs used in the dataset for Chapter 8.
237
238
The mid-shelf reefs of the Great Barrier Reef
8.2.3 Bewick Reef (148 260 S) This is a low wooded island with most of the reef flat covered by mangroves (Fig. 8.3a). One hole was also drilled into the leeside of this reef with the Pleistocene encountered only 4 m below reef flat level. An age of 6920 130
Figure 8.3 (a) Bewick Island and reef, site of one of the 1973 boreholes; (b) East Hope (foreground) and West Hope Reefs.
8.2 The data base
239
years BP above this indicates that vertical growth of the reef closely followed the rise in sea level.
8.2.4 Williamson Reef (158 220 S) Williamson is a crescentic reef near Cooktown into which four holes were drilled (Davies and Hopley, 1983; Davies et al., 1985). The most windward consisted of branching framework and cavity over the Pleistocene unconformity at 17 m. Of the three holes drilled into the reef flat one also reached 17 m, another 21 m, but neither encountered the Pleistocene. Only the windward hole was dated with the oldest date from between 7 m and 8 m being 6910 170 years BP and a date of 6460 110 years BP from approximately 1 m below the reef flat.
8.2.5 Boulder Reef (158 250 S) Boulder Reef has been the site of both three shallow (<16.5 m) boreholes (Davies and Hopley, 1983; Davies et al., 1985) and one of the deep holes drilled by the International Consortium for Great Barrier Reef Drilling in 1995 (International Consortium for Great Barrier Reef Drilling, 2001; Webster and Davies, 2003) (see Fig. 2.11). Details are available for only two of the shallow holes establishing the Pleistocene at 16 m and early reef growth by 8130 180 years BP. As with many other reefs for which data are available, the windward margin is mainly framework (massive corals) and the area to leeward detrital. The reef flat occupies about 85% of its shoal area and is thus classified as a planar reef.
8.2.6 East Hope Reef (158 440 S) Brief details from East Hope Reef were included in Davies and Hopley (1983) and cover two of three holes drilled. On the windward margin the Pleistocene was intercepted at 18 m and 15 m on the reef flat to leeward. This is a relatively small reef (4.4 km2) and has a well-vegetated sand cay on its northern side (Fig. 8.3b). The reef is difficult to classify as it has a reef flat which takes up a high proportion of the total reef area but closer inspection suggests that this is a crescentic reef which has become attached to a large back reef patch. Unfortunately there are no dates available for East Hope apart from those from very shallow drilling (Hopley, 1982).
240
The mid-shelf reefs of the Great Barrier Reef
8.2.7 Potter Reef (178 420 S) Potter, Ellison, Taylor, and Reef 17-065, together with shelf marginal Moss Reef, formed part of a single unpublished cross-shelf study by Graham (1993). Potter is a crescentic reef which has extensive reef flat but has yet to develop a full ‘‘hardline’’ outer edge. The outer edge, facing the south-east, has clearly developed from at least two parallel fronts and in places a lagoon still exists between them. The outer front has not everywhere grown up to sea level. Drilling established the Pleistocene at a depth of 25.2 m (Graham, 1993). 8.2.8 Ellison Reef (178 430 S) Ellison is closer to the mainland than Potter but has a very similar morphology though with a better developed hardline outer edge. Drilling down to 13.8 m did not encounter the Pleistocene foundation (Graham, 1993). 8.2.9 Taylor Reef (178 500 S) Taylor Reef appears to be a more mature reef, with a well-developed crescentic front wrapping around almost 50% of the reef perimeter. It encloses a lagoon with numerous large patch reefs, one of the leeward ones of which has an unvegetated sand cay. Drilling to 17.5 m did not encounter the Pleistocene (Graham, 1993). 8.2.10 Reef 17-065 (178 530 S) Reef 17-065 morphologically is the least mature of all the reefs sampled. It is made up of a series of reef patches some of which appear to have coalesced to form a ragged central reef flat. The windward edge of the reef has not grown to sea level. Unfortunately only one hole could be drilled on this reef and only to 9.3 m all in mainly framework of Holocene age. 8.2.11 Britomart Reef (188 140 S) Britomart is a very large mid-shelf reef, 23 km long and 134.4 km2 in area. Two holes were drilled into the eastern end of the reef (Johnson et al., 1984). One on the windward margin reached 28 m, establishing the Pleistocene at 25.5 m. The second, on a patch reef just behind the windward rim, encountered the Pleistocene at 19.8 m. Although the perimeter is broken on the leeward side, Britomart is classified as an open lagoonal reef.
8.2 The data base
241
8.2.12 Grub Reef (188 380 S) Grub Reef appears to have an anomalous morphology (Fig. 8.4a). It consists of an almost totally enclosed lagoon but cannot be classified as a lagoonal reef as this occupies less than 25% of the total reef platform. To the south-east
Figure 8.4 (a) Grub Reef; (b) Wheeler Reef, planar reef with leeward sand cay.
242
The mid-shelf reefs of the Great Barrier Reef
of the lagoon is an immature crescentic rim which at the southern end of the reef is totally submerged. Behind this complex windward morphology is an extensive field of reef patches some of which are quite large (>50 m in diameter). Data from this reef is extensive but largely unpublished. Three holes were drilled into the reef: on the windward and leeward reef flats of the ‘‘lagoon’’ reef and on the largest leeward reef patch (Fig. 8.9). The maximum depth was 13.8 m and the Pleistocene was not encountered. However, a seismic refraction survey across each of the drill holes gave an indication of the depth and morphology of the antecedent platform. Beneath the windward rim a depth of between 24 and 26 m was suggested, 20 m beneath the leeward rim and 26–30 m below the patch reef. As discussed below, Grub Reef is classified as a crescentic reef but its evolution from reef patches can be demonstrated.
8.2.13 Wheeler Reef (188 480 S) A joint research project by James Cook University and the then Bureau of Mineral Resources studied Wheeler Reef in the 1980s. The summary of results of three of four holes drilled into this very small (1.9 km2) oval reef (Fig. 8.4b) have been published by Davies and Hopley (1983) and Davies et al. (1985). This is a planar reef with a small unvegetated sand cay on its leeward side. The reef flat, however, does not have the normal sediment cover of most planar reefs but, being at an elevation between neap and spring low tides, has numerous living corals even in the center of the reef. Seismic refraction surveys (Harvey, 1980) suggested the Pleistocene was as shallow as 15.5 m near the cay. Of the four holes drilled (Fig. 8.12), only one in the center of the reef appears to have approached the Pleistocene, with calcitic gravels recovered from 20 m, though another sample of a coral clast from the same depth produced a radiocarbon date of 7110 240 years BP.
8.2.14 Davies Reef (188 500 S) Davies Reef was the site of a single borehole by the Australian Institute of Marine Science in 1982 (Grimes, 1982). It was located on the south-eastern (windward) reef flat of the reef, penetrating to 39 m. Holocene reef 26 m thick overlies a Pleistocene paleosol. Unfortunately, no dates are available from this core. The reef is intermediate between a crescentic and lagoonal form, with hardline reef (though including a double reef front in many locations) extending around 70% of the reef perimeter (Fig. 8.5a).
8.2 The data base
243
Figure 8.5 (a) Davies Reef, lagoonal reef with double reef front; (b) Darley Reef, one of the largest reefs used in the dataset.
8.2.15 Darley Reef (198 130 S) With an area of 81.3 km2 and a cross-shelf length of over 20 km Darley Reef is the second largest of the dataset (Fig. 8.5b). The reef has a complex morphology consisting of numerous lagoons and channels which dissect the southern hardline edge. One major channel on the eastern side of the reef is over 4 km
244
The mid-shelf reefs of the Great Barrier Reef
long and 500 m wide but its southern exit is partially blocked by large patch reefs. Darley Reef is difficult to classify as different parts of the reef may be identified as lagoonal and crescentic, and the north-western part even as reef patches. Three shallow holes (see Hopley, 1982; Hopley and Harvey, 1982) and two deeper holes (to 13 m maximum) adjacent to the major channel have been drilled into Darley Reef without encountering the Pleistocene. Seismic refraction profiles suggested a Holocene thickness between 10 and 17 m beneath reef flat level (Harvey, 1980). 8.2.16 Stanley Reef (198 180 S) Stanley Reef was another of the reefs jointly researched by James Cook University and the Bureau of Mineral Resources. Summary results are published in Davies and Hopley (1983), Davies et al. (1985), and Marshall (1985). Results of earlier shallow drilling and seismic profiling were published in Hopley (1982). A 400-m wide reef flat surrounds 45% of the reef and is continued as a submerged feature around a further 30% (Fig. 8.10). Scattered reef patches are found throughout the lagoon, the general depth of which is 5–15 m. A major channel up to 60 m deep transects the reef (Hopley, 1982, fig. 9.6) but is blocked by the reef rim at its southern end. Two holes were drilled into the windward margin, establishing the Pleistocene at 15 m, and three into lagoon patch reefs, only one of which encountered the Pleistocene at 22 m. Stanley Reef is a crescentic reef which is transitional towards an open lagoonal reef. 8.2.17 Gable Reef (208 480 S) Gable Reef1 is a spectacular open lagoonal reef (Fig. 8.6) with a very distinctive double front which is discussed later in this chapter. A continuous reef flat encircles 70% of the reef and reef patches occur over the submerged rim elsewhere. The main lagoon is about 15 m deep and has numerous very large patch reefs apparently formed from the coalescence of smaller features. The outer rim is 100 m wide and separated from the main area by a lagoonal embayment 20–25 m deep. Water depths on the outside of this feature are between 40 and 50 m. This outer front has a recurved spit-like tip and a submerged extension. Four holes have been drilled into this reef (Fig. 8.13), one in the outer front reef flat, one on the inner flat, and two where the outer and inner fronts coalesce. Although penetrating to 30 m in the outer rim Pleistocene was not encountered though is considered to be not far below 1
Note recent name change to Goble Reefs (20–305) by Great Barrier Reef Marine Park Authority, but still Gable Reef on Admiralty Charts.
8.2 The data base
245
Figure 8.6 Distinctive double reef front of Gable Reef.
this level as iron-stained and partially recrystallized coral clasts were recovered from the bottom of the hole. Other drill sites reached 19.4 m without encountering the Pleistocene. 8.2.18 Redbill Reef (208 380 S) Redbill Reef (Fig. 8.7a) has been the site of a shallow drilling and seismic investigation (Hopley et al., 1982) and a deeper drilling project (Fig. 8.11) (Hopley et al., 1984). The reef is unique in having a small (1.6 ha) granitic outcrop on its western side. Redbill is a classic example of a lagoonal reef which has almost completely infilled leaving only two shallow areas in the center of the reef and occupying less that 10% of the reef top area. The deeper drill holes reached depths of between 14.5 and 19.5 m, establishing the Pleistocene at 17.1 m on the leeward rim, at 13.0 m in the center of the reef, and >17.8 m on the windward margin. 8.2.19 Wreck Reef (238 200 S) Wreck Reef is one of four Bunker–Capricorn reefs drilled by the Bureau of Mineral Resources in the late 1970s and 1980s. It is a well-developed planar reef with leeward vegetated sand cay (Fig. 8.7b). Partial results from four drill
246
The mid-shelf reefs of the Great Barrier Reef
Figure 8.7 (a) Redbill Reef, with shallow almost totally infilled lagoon. The granitic outcrop of this mid-shelf reef is seen on the far side. (b) Wreck Reef, a planar reef with vegetated sand cay in the Bunker–Capricorn group of reefs.
holes are contained in Davies and Marshall (1979) and Davies and Hopley (1983). They reached only 10 m and did not penetrate the Pleistocene but do provide a detailed chronology of accretion. Seismic results suggest a depth of between 8 and 17 m for the unconformity.
8.3 Criteria used to classify the selected reefs
247
8.2.20 One Tree Reef (238 300 S) This is one of the most studied reefs of the entire GBR (see Marshall and Davies (1982) and Davies (1983) for drilling results and interpretation together with earlier references). This classic lagoonal reef, with a windward vegetated shingle cay, has been the site of drilling, seismic, sedimentary, and oceanographic research projects (Fig. 8.8a). Six drill holes and ten vibro-cores have been recovered from this reef, clearly establishing the Pleistocene at 13 m toward the windward margin and slightly deeper to leeward. 8.2.21 Fitzroy Reef (238 380 S) Three holes were drilled into Fitzroy Reef of which the summary results are given in Davies and Hopley (1983). This too is a lagoonal reef with only one narrow entrance into the lagoon on the leeward side. On the windward side the Pleistocene was encountered at 10 m. 8.2.22 Fairfax Reef (238 520 S) One of the southernmost reefs of the GBR, Fairfax Reef is almost, but not quite, at the planar stage (Fig. 8.8b). It has a windward shingle cay and leeward sand cay, both vegetated but separated by a small remnant of what has almost certainly been a larger lagoon in the past. Selected results of three holes drilled into this reef were published in Davies and Marshall (1979) and Davies and Hopley (1983). The Pleistocene was encountered only in the windward hole at the shallow depth of 8 m. 8.3 Criteria used to classify the selected reefs In order to test the robustness of the evolutionary classification, the reefs in the dataset need first to be given an identified place within the classification. Identification within the major groups – patch, crescentic, lagoonal, or planar – is relatively straightforward but within each of these it is possible to identify early to late stages. Unfortunately, only one reef (17-065) is classified as a patch reef (i.e., lacking a hardline windward rim) but if others were in the classification, the area of patch reefs as a percentage of the total reef platform area might provide the criteria for a sequence within the patch reef class. Crescentic reefs do have a hardline on their windward side and their place in the classification is provided by the proportion, as a percentage, that they surround the perimeter of the reef platform. In places, reef patches may be
248
The mid-shelf reefs of the Great Barrier Reef
Figure 8.8 (a) One Tree Reef, one of the most studied reefs in the Great Barrier Reef; (b) Fairfax Reef, with large windward shingle cay, and leeward sand cay (photographs: H. Kan).
attached to this outer reef flat, even producing lagoonal structures such as on Grub Reef. However, using the percentage perimeter allows for a smooth transition into open lagoonal reefs, arbitrarily defined as having 75% of the perimeter surrounded by reef flat.
8.3 Criteria used to classify the selected reefs
249
Once the lagoon is more or less closed the next phase of the sequence is provided by the proportion of the reef which remains as lagoon, from high as in the example of One Tree Reef, to low as for Redbill and Fairfax Reefs. Progression to a planar reef at this stage is straightforward but for some planar reefs, such as Boulder or Wreck, there may still be a small area of the reef platform that has not fully grown to sea level. Subsequently the type of reef top cover provides the criteria for final classification, from partial cover of living coral (Wheeler), through sediment covered (often with seagrass beds) with sand cays (such as Coconut Island, Fig. 13.9a), and finally to the low-wooded island type reef tops with a high percentage of mangrove cover, such as Bewick. The majority of the reefs fall easily into this sequence though some qualitative assessment is needed in a few cases, most especially where crescentic reefs have coalesced with possibly earlier patch reefs (Potter, Taylor?). Greatest difficulty occurs with the large reefs such as Britomart and Darley. This was recognized by Hopley (1983c) who, in a revised version of the reef classification, introduced separate systems for small (1 km diameter), medium (2.5 km diameter), and large reefs. The latter class was associated with multiple lagoons and recognized that different sections of the reef may be at different stages of development. Of the reefs discussed above Darley is the most ambiguous, and although classified as crescentic the part of the reef for which data is available, if assessed separately, could quite well be classified as lagoonal. As discussed below, reef size is an important factor in morphological evolution of Holocene reef morphology. Table 8.2 places all the reefs for which drilling data is available in a juvenile to senile sequence. This can be tested against the oldest reef flat age (within either 1 m or 3 m of reef flat) recorded for the reef. Also in this table are what have been considered the most influential factors in placing a reef in this sequence, minimal depth recorded to the Pleistocene and the size of the reef. These relationships are discussed below. 8.3.1 Morphology and reef flat age Ideally the radiocarbon ages for reef flats should reflect the stage that particular reef has reached, i.e., youngest for reef patches, oldest for planar reefs. Reef flat age is shown in Table 8.2 both for the upper 1 m and upper 3 m from drill cores as results frequently show the typical slowing down of reef accretion as the reef approaches sea level (as first demonstrated by Davies and Marshall, 1979). There may also be some mixing of detrital materials in the uppermost part of the cores.
Grub Ellison Williamson Potter East Hope Taylor
2 3 4 5 6 7
Patch
16 17 18 19 20 21 22
Planar
From Harvey (1980).
11 12 13 14 15
Lagoonal
a
8 9 10
Large crescentic or lagoonal
Fairfax Boulder Wheeler Wreck Stapleton Warraber Bewick
Davies Gable Fitzroy One Tree Redbill
Stanley Darley Britomart
17-065
1
Reef stage
Crescentic
Reef (juvenile to senile)
6630 110 6450 290 4650 90 5826 170 5260 130 6580 80 6920 130
7160 170 6210 140 5490 80
6420 160 6070 130 5255 65
6320 80 5910 190 4550 90 5730 70 3130 80 5770 70 6380 120
4920 110 3830 60 6550 140 3920 110 n.a. 5130 140
4100 80 3530 60 6460 110 3800 110 2570 100 5130 140
n.a. 6110 100 6640 100 5880 100 6500 150
3760 100
3760 100
n.a. 4980 90 5740 180 5390 180 4570 150
(Top 3 m)
C age
(Top 1 m)
14
Table 8.2. Classification of reefs and probable causal factors
3.8 12.6 1.9 6.6 4.8 11.9 2.0
13.8 17.5 22.5 18.1 8.8
58.1 81.3 134.3
23.8 13.1 37.1 15.7 4.4 16.5
11.1
Area (km2)
8.5 16.0 20.0 >10.5 14.6 5.7 4.0
25.7 30.0 10.0 13.0 12.1
>17.5 >13.0 19.8
>13.8 >14.3 17.0 25.2 15.0 >17.5
>9.3
Depth to Pleistocene (drilling) (m)
(15–24) (8–17)
(7–23)
(24–30) (9–21) (10–23) (6–18)
(10–19)
(10–19)
(29–30)
Seismics a
8.3 Criteria used to classify the selected reefs
251
Results are far from straightforward. Unfortunately there is data for only one patch reef (17-065), but it is one of the youngest at both 1 m and 3 m depths. The crescentic reefs have an apparently wide spread of ages: 2570 100 years to 6460 40 years BP for the upper 1 m and 2570 100 years to 6550 140 years BP for the upper 3 m. The most anomalous ages are for Williamson Reef, which appears much older than its crescentic morphology suggests, and East Hope Reef which is very young. Although Williamson has a mid-shelf location it is not the usual crescentic reef on the edge of a larger antecedent platform. Instead it appears to rise from a very narrow foundation more akin to the nearby ribbon reefs, which also rise from similar platform depths of 15 m (see Chapter 9) and have similar reef top ages of about 6 ka. The size and depth of the Pleistocene foundations have been considered as influential in determining morphology. It is possible that foundation shape is also a factor, with narrow ribbon like structures able to maintain a high productivity along a relatively narrow rim. Certainly Williamson has accreted at a very rapid rate (>15 m ka1). Unfortunately no other linear mid-shelf reef has been drilled and/or dated though examples do exist. If this interpretation is correct then Williamson may be considered as a planar reef, i.e., occupying most of its antecedent platform, and its age would be highly appropriate for this interpretation. Data from East Hope Reef come not from the deeper drilling but from reef top coring reported in Hopley (1982), which did not penetrate more than 1.5 m. Excluding these two reefs from the crescentic reef data set produces a much narrower age range of 3530 60 to 5130 140 years BP for the upper 1m and 3830 60 to 5130 140 years BP for the upper 3 m. The next group of reefs are the three largest: Stanley, Darley, and Britomart. Although classified as large crescentic or open lagoonal reefs as noted above, they have a complex morphology made up of a series of separate cells each of which could be classified as a more advanced stage than the whole platform. The mid-Holocene ages of between 5255 65 and 6420 160 years BP for the upper 1 m and up to 7160 170 years BP for the upper 3 m would appear to support this interpretation. The lagoonal reefs have a much tighter age range: 4570 150 to 5740 180 years BP for the upper 1m and 5880 100 to 6640 100 years BP for the upper 3 m. The rims of these reefs all appear to have reached sea level within 1000 years of it stabilizing in the mid Holocene. Of the planar reefs, two stand out as anomalously young: Stapleton Reef and Wheeler Reef. Stapleton can be explained from the site which was drilled on the 1973 Royal Society Expedition (Thom et al., 1978). It was located on the sand cay on the far leeward side of the reef, and although it penetrated the
252
The mid-shelf reefs of the Great Barrier Reef
Pleistocene the overlying Holocene deposits are part of the young back reef sand sheet and are not representative of the age of the reef. Wheeler Reef is more difficult to explain, especially as it is the site of four deep and four shallow drill holes with 29 samples dated. The reef top cover with areas of living coral appears younger and lower than other planar reefs and given that this is the smallest reef in the dataset, a much more mature and older reef top would be expected. The Pleistocene unconformity is deeper than beneath other planar reefs, but not as great as below some of the lagoonal and crescentic reefs. The reef also appears to occupy the whole of its antecedent platform, i.e., could not be interpreted as a patch reef on a yet to fully emerge reef platform. Omitting Stapleton and Wheeler from the dataset, the age range for the upper 1 m is 5730 70 to 6380 120 years BP and 5826 170 to 6920 130 years BP for the upper 3 m. Table 8.3 includes a summary of the age data and by omitting the most anomalous reefs of Williamson and Stapleton and reassessing the inclusion of Wheeler it provides an interesting sequence. If the data from the one sample of reef patches can be accepted then an age progression can be seen in the morphological sequence. There may be little difference between the reef top ages of patches and crescentic reefs but using the data from the upper 3 m suggests an age difference of about 700 years. Progression then to a lagoonal reef may be about 1350 years based on the shallower data and 1800 years from
Table 8.3. Mean age, size, and depth to Pleistocene of dataset reefs Mean age Mean size (km2)
Mean depth to Pleistocene a (m)
Reef type
Number of reefs
1 Reef patches
1
3760
3760
11.1
>9.3
2 Crescentic
5
3826
4450 14.7 (excl. Hope)
19.07
3 Lagoonal
4
5170
6282
18.1
4 Planar
6
5776 incl. 6176 incl. 6022 excl. 6481 excl. Wheeler Wheeler
6.42 incl. 10.8 incl. 7.32 excl. 8.55 excl. Wheeler Wheeler
5 Large reefs
3
5915
91.2
a
1m
Where encountered in drilling only.
3m
6286
16.1
17.4
8.3 Criteria used to classify the selected reefs
253
dates in the upper 3 m. Planar reefs are only a little older, perhaps 250 years based on the available data. The rims of the large reefs may be even older, emphasizing the slow rate of infill of lagoons over 50 km2 in area, and the much slower rate of development of larger reefs. Conversely smaller reefs have been suggested as having the most rapid development (Hopley, 1982, 1983c). Although the reef patches and crescentic and lagoonal reefs could come from the same size population sample, the planar reefs in general are only half the size (Table 8.3) and may partially confirm the size control hypothesis. 8.3.2 The influence of the depth of the antecedent platform Tables 8.1 and 8.2 provide the data derived from drilling for the depth to the Pleistocene, or the maximum depth drilled without encountering the Pleistocene, for each of the reefs studied. Mean figures are given in Table 8.3. Where available, the depth of the Pleistocene derived from seismic refraction studies (Harvey, 1980) is also shown though drilling subsequent to 1980 has indicated that in general the seismic methodology has slightly underestimated the depth. The presumption tested here is that reefs growing from deeper foundations will be more juvenile in the evolutionary sequence. Unfortunately again the data for reef patches is very minimal. The Pleistocene is greater than 9.3 m on Reef 17-065. Data for the crescentic reefs suggest depths >15 m and as much as 25.2 m. For the lagoonal reefs the depths are >25 m for the open lagoons of Davies and Gable Reefs but <13 m for the more completely enclosed and infilled Fitzroy, One Tree, and Redbill Reefs. The large lagoonal reefs (Stanley, Darley, and Britomart) have similar depths to their smaller lagoonal counterparts. For planar reefs Wheeler, with a depth of 20 m, again appears to be anomalous as all other reefs have Pleistocene depths of less than 16 m. Based on mean figures alone the depth to the Pleistocene appears to be similar for crescentic and lagoonal reefs, but almost half the depth beneath the planar reefs. The results clearly indicate that planar reefs, as presumed, grow off smaller and shallower Pleistocene foundations. Although there is an age difference between crescentic and lagoonal reefs, size and depth to Pleistocene seem far less influential. However, the data for lagoonal reefs are strongly influenced by the Pleistocene depth on Davies (25.7 m) and Gable (30.0) Reefs. Only one hole was drilled on Davies and, given the variation in the Pleistocene seen on other lagoonal reefs such as One Tree (Marshall and Davies, 1982) and Redbill (Hopley et al., 1984), a variation of at least 5 m might have been expected if more holes had been drilled. The data from Gable Reef is even more biased as it comes from the outer rim probably growing off much younger and deeper
254
The mid-shelf reefs of the Great Barrier Reef
foundations than the main reef, as discussed below. Seismic survey suggested a depth of 24 m below the inner rim. Omitting Davies and Gable, the mean depth to the Pleistocene for the remaining three reefs is only 11.7 m, much closer to that of the planar reefs. As both reef size and depth to Pleistocene have an influence on reef stage, it may not be surprising that exact correlation with either of the controlling parameters individually is not seen. More explanation may be provided by examining the two factors together. For example, crescentic Potter Reef has a relatively small area of 15.7 km2 but is obviously retarded in its development by a deep foundation of 25.2 m. Within the crescentic reefs the one with the largest area, Fitzroy (22.5 km2), has the shallowest foundations (10.0 m). As noted above, the planar reefs have both small size and shallow foundations. Nonetheless, exact relationships do not exist and this is not unexpected. Drill holes are very small sampling points on very large reef surfaces with variations in age and structure expected over short distances. That some relationships at all seem to have been established may be surprising. The way in which individual mid-shelf reefs have evolved is discussed below but there seems to be an increasing amount of evidence for at least parts of reefs being able to keep up with rates of Holocene sea-level rise, whatever the depth of platform from which they are growing. For example, the reef on Tahiti was able to track sea level from a 90-m deep foundation and still reach sea level by about 6 ka ago (Montaggioni et al., 1997; Cabioch et al., 1999). Subsequent growth of windward rims and infilling of lagoons may be very random. Thus drilling into reef patches is much more likely to encounter one of the older parts of this reef type, whilst with the more mature lagoonal and planar reefs locating a drill hole on the oldest part of the reef may be quite arbitrary. These factors alone will produce ‘‘noise’’ in the datasets and tend to compress the age sequence from planar to reef patches more than they should. This too may be important as the dataset above suggests progression from reef patches to crescentic reefs in less than 1000 years, crescentic to lagoonal reefs in about 1500 years, and lagoonal to planar in 250 years. However, the smaller planar reefs may not have passed through a crescentic or lagoonal phase, especially where a 400–500-m wide rim (the width suggested by Kinsey and Davies (1979b) as being the maximum for active growth prior to sediment cover) will cover a very high proportion of the reef platform (Wheeler, Bewick). Thus the time between the crescentic and lagoonal, and lagoonal and larger planar reefs may be more important in determining rates of development of reef morphology. Further information on this is provided by the internal morphology of individual reefs.
8.4 Evidence of morphological evolution from the internal structure of reefs 255
8.4 Evidence of morphological evolution from the internal structure of reefs Individual reefs may provide information on their evolution during the Holocene, but only where dated drill holes come from more than one reef environment. All too often the target for drilling has been the hardline windward margin but a number of the sampled reefs do provide insight into the ways in which they have grown and the rates of change from one morphology to another. A number of the crescentic reefs incorporate large patch reefs but only one, Grub, indicates that the hardline windward rim may be younger than some of the leeward reef flat. As noted above, the complex morphology of Grub Reef (Fig. 8.9) comprises a windward rim which appears to have become attached to a mid-reef linear patch to form a small lagoon. The oldest radiocarbon age from the reef is 6820 140 years BP at a depth of 12.0 m beneath the inner rim. Subsequent to this time the outer reef margin appears to have caught up at about 5 ka ago but the inner reef rim reached modern sea level by about 4 ka ago (an age of 4920 110 years BP is recorded at a depth of only 1.5 m). At this time the outer rim was more than 4 m below sea level catching up only by 3660 110 years BP (age from 0.5 m). Even the large patch reef in the lagoon reached sea level well before the windward rim, an age of 4100 80 years BP being recorded from a depth of only 0.5 m, although this patch reef had been at a depth of 8 m at about 5 ka ago. The evidence suggests that Grub Reef had a reef patches configuration, initially reaching sea level about 5 ka ago. Further patches were added to its morphology prior to 4 ka ago with its windward rim reaching sea level, to form the present crescentic reef only by 3 ka ago. Transition from the juvenile stages of reef patches to crescentic reef has thus taken about 1500 years, probably a more realistic time-span than is provided by comparison of different reefs. Contrasting results are provided by other crescentic reefs. Taylor Reef also appears to have a central large reef patch which has become attached to the outer crescentic rim, but the limited drilling (two deeper and three shallow holes) indicate that both parts of the reef reached sea level at about the same time. Windward rim dates of 4590 130 years BP (0.6 m) and 5760 180 years BP (2.4 m) from shallow drilling (see Hopley, 1982, table 9.2) are more or less matched by a date of 5130 140 years BP from 1.2 m from the central reef flat area (Graham, 1993). The only other crescentic/lagoonal reef for which there is considerable data in the form of six deep and five shallow drill holes is Stanley Reef (Fig. 8.10). Here, the windward rim had formed prior to 6 ka ago with an age of 7160 170 years BP coming from a depth of only 2.2 m (Hopley, 1982, table 9.2) and
256
The mid-shelf reefs of the Great Barrier Reef
Figure 8.9 Morphology, drilling sites, and dating on Grub Reef.
8.4 Evidence of morphological evolution from the internal structure of reefs 257
Figure 8.10 Shallow and deeper drilling and dating results from Stanley Reef for the deeper holes (including data courtesy of P. J. Davies).
258
The mid-shelf reefs of the Great Barrier Reef
6410 160 years BP from only 0.62 m. The oldest of three patch reefs which were drilled reached sea level prior to 5 ka ago, another drilled by the Bureau of Mineral Resources and James Cook University teams was dated at 3390 90 years BP (1.0 m) whilst the third was even younger (3500 100 years BP at 9.0 m). Whilst only two holes were drilled on Britomart (Johnson et al., 1984) it would appear that a lagoonal patch reef reached sea level prior to 5 ka ago about 2000 years earlier than the windward rim which had 9.6 m of water over it when the patch first reached sea level. Further drilling could establish that Britomart has passed through the reef patches and crescentic stages to reach its present morphological classification as an open lagoonal reef. Evidence from other lagoonal reefs can certainly establish an earlier crescentic phase and, in the example of One Tree Reef, an earlier reef patches stage. The detailed analysis of Marshall and Davies (1982) is summarized in their figure 6. After initial drowning prior to 7.5 ka ago, One Tree remained as a submerged reef until about 5.5 ka ago, the first part of the reef to reach sea level being what is now a lagoonal patch reef. Numerous other patch reefs within the lagoon are larger than that sampled and there is a high probability that initially there were several at sea level prior to or simultaneously with the windward margin, which was at sea level by 5 ka ago. The indications are that One Tree was a crescentic reef between 5 ka and 4 ka ago, and possibly an open lagoonal reef (with a lagoon about 11 m deep) between 4 ka and 3 ka ago. Lagoon infilling became significant post 3 ka ago. Redbill Reef (Fig. 8.11) is an even more mature lagoonal reef and it too has a large data bank (Hopley, 1982; Hopley et al., 1982, 1984). However, even though the highest part of the Pleistocene foundation is beneath the central reef flat, a lagoonal morphology appears to have been established as early as 6 ka ago. At this time the lagoon depth was 8 m, only 4 m above the level of the underlying Pleistocene platform. Because the reef morphology was in place prior to the mid Holocene it has responded to a sea level 1.2–1.6 m higher than present, which has been responsible for the construction of high algal rims around the reef, behind which moated corals are growing up to mean sea level at a location with a 5-m tidal range. Lagoon infilling has been slowly taking place since about 4 ka ago probably an indication of the time of the first fall of sea level from the mid-Holocene high and the enclosure of the central lagoon. Redbill’s transition from lagoonal to planar reef is thus taking place over a span of about 4 ka. The construction of isochrons is possible for only two fully planar reefs, Warraber Reef in Torres Strait (Woodroffe et al., 2000) and Wheeler Reef off Townsville. On Warraber Reef establishment was over or close to the shallowest Pleistocene basement. A large proportion of the central area of the
Figure 8.11 Shallow and deeper drilling and dating results from Redbill Reef.
260
The mid-shelf reefs of the Great Barrier Reef
modern reef platform had formed a reef flat by or prior to 5.5 ka ago with an indication that this occupied the western end of the reef only, whilst the southern, and possibly eastern end (for which there are no data) lay in up to 6 m of water. The possibility that this was a crescentic reef oriented towards the north-west is not unreasonable as Torres Strait reefs experience both winter south-easterlies and a summer north-western monsoon and a hardline weather edge is common on the western side of reefs. Extension of the reef margin from the original crescentic reef has occurred through rapid upward growth of coral heads to form a reef crest behind which detrital material has accumulated. This is the pattern that would be expected in the development from a crescentic to planar reef. A lagoonal phase may have taken place but cannot be established without data from the eastern margin of Warraber Reef. As Wheeler is the smallest reef a simple growth pattern might have been expected (Fig. 8.12). However, the isochrons drop away sharply on the windward side from what may have been small patch reefs about 100 m back from the reef front penetrated by drill holes B.M.R.1 and JCU1. At about 4 ka ago these were separated by water depths of about 11 m. The original midHolocene reef appears to have sloped more gently towards the leeward side with no suggestion of a proto-lagoon. Progression from isolated shallow patch reefs to the present small planar reef appears to have taken place between 4 and 3.25 ka ago, with additional reef added subsequently through reef front progradation. 8.5 Evolution of windward reef fronts Once beyond the reef patch stage, the most persistent feature of the reefs is their windward margin, sometimes referred to as hardline. Early work (Davies, 1977; Davies and Kinsey, 1977) suggested that once at sea level reef fronts may then erode. Work on One Tree Reef by Davies and Marshall (1979) produced pre-Holocene radiometric ages for materials taken from reef front depths of 19.8 and 22.0 m. Hopley (1982) suggested that the steepness of the reef front has prevented any addition of interstadial fringing reefs or even erosional notches on which a new reef front may be established. More gently sloping reef fronts such as those found on the mid-shelf reefs especially of the central and northern GBR may have been produced by reef growth at lower sea levels, encouraged by the lower energy levels compared to the exposed One Tree Reef (see for example Fig. 12.7). Thus, whilst high-energy, steeply sloping reef fronts may produce a stable or even eroding hardline reef margin, lower-energy, more gently sloping margins may prograde with secondary or even multiple reef fronts around the windward margin (Fig. 8.5a) (see also Hopley, 1982, fig. 10.1).
8.5 Evolution of windward reef fronts
261
Figure 8.12 Drilling and dating results from Wheeler Reef (including data courtesy of P. J. Davies).
As noted, these are particularly common on the central GBR but may also be found elsewhere as on Warraber Reef where new reef fronts have been added by coral head growth in front of the reef margin, then attached to the main reef by sediment infill of the intervening lagoon as described by
262
The mid-shelf reefs of the Great Barrier Reef
Woodroffe et al. (2000). Wheeler Reef has also prograded on the windward margin. Because of their exposure, and typically a lack of reef flat development, investigations by drilling into a secondary reef front have been difficult. However, Gable Reef in the south central GBR has what is probably the best example of a secondary front almost 100 m wide (Figs. 8.6 and 8.13) on which drilling has been possible. The object was to determine the nature of the foundations of such a secondary front. Drilling of four holes was supplemented by intensive echo-sounding profiling and seismic refraction lines to provide details of the morphology and internal structure of the reef. The second front is on the windward south-eastern side of Gable Reef. It rises from a water depth of between 40 and 50 m with a reef flat level just above MLWS, very similar to the main reef flats to which it is attached at its southwestern end. To the north-east it extends as a submerged feature with a distinctive inward curving hook about 10 m deep. The embayment between the secondary and main front is 20–25 m deep. In contrast the main lagoon of Gable Reef is only 10–15 m deep. Four holes were drilled into Gable Reef, one in the outer front, one in the adjacent inner reef flat, and two where the two reef fronts merge (Fig. 8.13). Twenty-two radiocarbon dates provide a reliable chronology. However, none of the holes, reaching 30 m (outer rim), 20 m (inner rim), 10 m and 14 m (southern site), encountered Pleistocene foundations, although at 30 m beneath the outer front iron-stained calcitic gravel was recovered. A depth of approximately 30 m for the unconformity beneath the outer rim and 24 m beneath the inner rim was indicated by the seismic refraction survey. Holes 3 and 4 to the south show a typical reef framework in hole 3 whilst hole 4 is almost entirely sand and shingle, typical of a detrital back reef slope. The reef here was close to modern sea level shortly after 6 ka ago, the back reef area lagging behind by about 2 ka. Main interest lies in the double front. The lowest unit 30–22 m deep is a coarse sand dominated by Halimeda fragments, with occasional corals, two of which produced dates of just over 7.8 ka ago suggesting that they are not in situ. There is then a 7-m thick unit of fine (3j median size) and well-sorted sand (resembling beach sand) to 15 m. This is topped by shingle dated at 6620 100 years BP at 12 m, then a typical reef framework to the surface. This outer front was within 2 m of modern sea level by 6 ka ago. In contrast the hole to 20 m in the inner reef flat is through reef framework and cemented shingle throughout, with a date of 7560 80 years BP obtained from 17.3 m and 6930 80 years BP from 9.1 m. The reef here was within 2 m of sea level shortly before 5 ka ago. Upward growth at both holes 1 and 2 was very close with hole 2 (the inner reef) being ahead only between 8 and 7 ka ago,
8.5 Evolution of windward reef fronts
Figure 8.13 Morphology, drilling, and dating results from Gable Reef.
263
264
The mid-shelf reefs of the Great Barrier Reef
Figure 8.14 Model for the development of the double front of Gable Reef.
8.6 Rates of geomorphological development: discussion and conclusions
265
at which point the outer reef front changes from being dominated by sand to having a shingle cap and finally reef framework. Analysis of the materials dates and morphology has suggested the following origin (Fig. 8.14): *
*
*
the outer reef front originated as a spit or bar across an embayment in the Pleistocene antecedent reef. Materials were initially reworked from the Pleistocene reef, then, as in many other parts of the GBR, dominated by Halimeda before contributions from the adjacent Holocene reef became dominant the detrital spit was maintained until the main reef platform was drowned, probably reducing the amount of available sediment. By 7 ka ago the spit became the higher of the two reef fronts, and was colonized by corals probably from its outer reef slope, completely encasing it with reef framework whilst in water depths of 5–10 m the outer front was the first to reach sea level shortly after 6.5 ka ago, followed about 1000 years later by the inner margin.
This may be a unique origin for a double reef front and possibly explains its size and, in comparison to double fronts elsewhere, its apparently slow development with a deep lagoon still separating it from the main reef mass. However, it illustrates that features added to the antecedent Pleistocene reefs when they were first drowned and became limestone islands can provide the foundations for significant reef front features. The original hypothesis that multiple reef fronts grow from fringing reef foundations is not nullified by these results. Indeed, they suggest that any positive relief feature added to the antecedent reef before or during the transgression may provide the foundations for windward progradation of reefs with moderate slopes and energy levels, typical of the central GBR.
8.6 Rates of geomorphological development: discussion and conclusions A recent review of the rates of lagoon infilling (Purdy and Gischler, 2005) made the obvious point that the lagoonal stage of reef development is not the end member of reef evolution. Although their conclusions were directed largely at atolls they did consider shelf platform reefs, mentioning both One Tree and Lady Musgrave Reefs from the GBR. They likened lagoonal reefs to ‘‘empty buckets’’ which were slowly infilling though because of the temporal scale of sea-level change the complete infilling of lagoons was unlikely. This is probably true for atolls with deep and extensive lagoons which can be compared to the large lagoonal reefs of the GBR which have large area compared to small peripheral carbonate-producing margins. For shelf reefs this chapter has shown that the lagoonal reef is neither the starting point nor end point in the morphological evolution, but as originally demonstrated by Hopley (1982)
266
The mid-shelf reefs of the Great Barrier Reef
and illustrated by Fig. 5.6, is a single though important stage in reef development. Although Purdy and Gischler (2005) considered size as a factor in their evolutionary sequence, surprisingly, the depth of the antecedent foundations was not given prominence. The filling of the ‘‘bucket’’ was also attributed entirely to detrital movement of sediment aprons. On the GBR growth of lagoonal reefs also plays an important role (see for example, Figs. 5.7e, 5.8a, 8.7a, 8.8, and 13.7). Two factors which these authors present and with which we strongly concur are that uplift, even if slight, can accelerate the lagoon filling whilst on the other hand, gaps in the reef rim can lead to sediment loss and a slowing down of the process. For continental shelf reefs, the analysis in this chapter has given the basis for the timescale involved. Rates of change from one reef type to another are shown by isochrons from within the reefs for which sufficient data is available, and the reef top ages of reefs of different types. Progression from reef patches to crescentic reefs, based on very limited information of reef ages, was about 700 years. Internal data from One Tree Reef suggests about 500 years, for Grub Reef about 1500 years, and for much larger Britomart Reef 2000 years. The reef dates indicated change from crescentic to lagoonal reef over 1350- to 1800-year periods. For One Tree Reef the change occurred over a 1000-year period, with a further 1000 years for evolution into a closed and partially infilled lagoonal reef. Redbill has progressed from a lagoonal reef to a largely infilled near-planar reef in about 4000 years, whilst Warraber has changed from either a crescentic or lagoonal reef to a planar reef in 1000 years. Very small Wheeler Reef has changed from small reef patches to a planar reef in 750 years. Where these morphological changes are compared to detailed rates of reef accretion, as discussed in Chapter 11, it can be seen that the 6500 years of more or less stable sea level since the mid Holocene has been sufficient for reefs to evolve through the full sequence postulated. Carbonate productivity rates as high as 10 kg m2 yr1 and vertical accretion rates exceeding 8 m ka1 can change a reef’s characteristics in a relatively short time. Although a 1000-year period is beyond that of most environmental managers, it should be possible to predict the natural changes to reefs that may take place at this intermediate timescale as they evolve from one type of reef to another. This may be taken into account by management, e.g., a lagoonal reef changing into a planar reef in only a few hundred years may see a reduction in ecological niches and be less attractive for high-level conservation measures than a reef at a more juvenile stage (see Chapter 13). The rates of change are not totally consistent and this would not be expected. Some reefs appear to develop rapidly, possibly missing some early stage, like Redbill (apparently no reef patches or crescentic stages). Other reefs
8.6 Rates of geomorphological development: discussion and conclusions
267
may be better prepared to move from one classification to another. For example Grub Reef, which is currently at the crescentic stage, may evolve rapidly into a lagoonal reef as it already possesses a submerged rim around more than 50% of the reef. Stages in development may also be missed, especially if the reef is small. Wheeler for example appears to have gone from reef patches to planar with no intervening morphology and some crescentic reefs would appear to have the ability to progress to the planar stage if their central reef area possesses massive reef patches, developed either before or after the crescentic hardline. Taylor and East Hope Reefs appear to be this type, and Boulder, now a planar reef, may have evolved in this way in the past, based on the evidence of the small embayment on its lee side. What a reef cannot do is reverse its progress from one type to another, although this was exactly what Maxwell (1968) suggested with his ‘‘resorbed’’ reef type. According to Maxwell, at a late stage in reef evolution restrictive growth and degeneration may take place, a process which may also be responsible for reef patches (Maxwell, 1968, pp. 106–107). It is now clear that these reef types represent an early not a late stage in reef evolution. A further testing of the rates of morphogenesis derived from the internal dating of the reefs can be made by calculating possible rates of change from the carbonate productivity of different reef zones (e.g., Kinsey, 1985) as applied by Kinsey and Hopley (1991). The most easily constrained stages of the evolution are the changes from lagoonal to planar reefs. Using the same assumptions as Kinsey and Hopley, a lagoonal reef is presumed to be producing calcium carbonate at a rate of 10 kg m2 yr1 around its outer reef slopes, 4 kg m2 yr1 over the encircling reef flat (here presumed to be 400 m wide as suggested by Kinsey and Davies (1979b)), and 2 kg m2 yr1 within the lagoon. Assuming a minimal density of 2.89 g cm3 and an average reefal porosity of 50% (Smith and Kinsey, 1976), these rates translate into accretion rates of 7, 3, and 1.5 mm yr1. The dataset consisted of reefs ranging in size from 1.9 km2 (Wheeler) to 18.1 km2 (One Tree) which translates into presumed circular diameters of >1 to approximately 5 km. Figures were calculated for a range of reef sizes from 1–20 km diameter and for lagoon depths (which may reflect the depth to Pleistocene) of 2–20 m. The outer perimeter productivity was presumed to contribute only to reef slope construction or minor reef flat rubble (where its deposition may reduce the endogenous productivity thus producing a neutral result). The 400-m-wide reef flat production at 4 kg m2 yr1 and lagoonal contribution depends very much on reef size. Results for reefs up to 10 km diameter and lagoons 10 m deep are plotted in Fig. 8.15 and are in close agreement with the evolution previously established.
268
The mid-shelf reefs of the Great Barrier Reef
Figure 8.15 Isochrons for lagoonal infilling based on carbonate productivity for reefs between 1 and 20 km diameter and lagoon depths (based on depth to Pleistocene) between 2 and 20 m. Note all planar reefs except Boulder are indicated as needing less than 2000 years to progress from lagoonal to planar and all lagoonal reefs more than 2500 years. Larger lagoonal reefs (Stanley, Darley, Britomart) are indicated as requiring up to 10 000 years for the progression.
Very small reefs may progress from lagoonal to planar in less than 500 years whilst even reefs approaching 20 km diameter (>300 km2 area) and with lagoons up to 10 m deep may infill and become planar in less than 6000 years. Much depends on the morphology from which Holocene growth commences and saucer-shaped Pleistocene foundations may favor rapid evolution whilst irregular pinnacles will develop an advanced morphology more slowly regardless of size and depth of antecedent surface. The rates of morphogenesis calculated in Fig. 8.15 reflect the rates from dating quite closely. For example Redbill was seen to have progressed from lagoonal to almost planar in about 4000 years from dating and in about 3500 years from the carbonate budget calculations. Most of the lagoonal reefs in the data base used in this chapter (e.g., Davies, Gable, Fitzroy) would require more than 5000 years for progression from lagoonal to planar, though One Tree, estimated to require about 3500 years, would appear to be developing more slowly than estimated. Calculations were also made for large lagoonal reefs (20 km diameter), with Pleistocene foundations of about 20 m (e.g., Britomart) and these are shown to
8.6 Rates of geomorphological development: discussion and conclusions
269
require about 10 000 years for progression to the planar stage and it is hardly surprising that they are still at an early stage. Planar reefs such as Wheeler, Fairfax, Wreck, and Warraber are shown to have required between only 1000 and 2000 years for progression from lagoonal to planar morphology (see also Section 10.6 and Fig. 10.16 for complementary discussion of rates of change in relation to island development). These results may help explain some of the complex features of the GBR. Large reefs currently at the lagoonal phase such as those in the Pompey Complex would appear to require a time period longer than recent interglacial highstands to reach the senile phase and the earlier stage morphology is bequeathed from one highstand to another. However, the presence of very large planar reefs, such as those north of Princess Charlotte Bay, e.g., Corbett (207.5 km2), Hedge (152.5 km2), and Magpie (80.8 km2), appear anomalous given the data presented in this chapter. A local area of high Pleistocene reefal foundations seems unlikely but an earlier suggestion (Stoddart et al., 1978c, p. 152) that these reefs may be underlain by sedimentary rocks of the Laura Basin may be worthy of reconsideration. In Chapter 3 evidence was presented for a mid-Holocene sea level up to 1.5 m higher than present. As this is the product of hydro-isostatic processes, it is the inner and sometimes middle shelf that has experienced the higher level. Fringing reefs have been the most likely to retain evidence especially in the form of emerged microatolls. However, some mid-shelf reefs which lie within the zero isobase and which had reached sea level at or prior to the maximum (between 6 and 5 ka ago) may also retain evidence of the higher level. Of the reefs in the data base used in this chapter, two at least have experienced a relative fall in sea level which has greatly accelerated their morphological evolution. These are Redbill and Warraber. It is probable that other GBR mid-shelf reefs have similarly experienced a rapid development with the exposure of reef perimeters, accelerated erosion of emerged areas, and a general shallowing of shoals and lagoons. Notably, in the quantitative regional classification of the GBR by Hopley et al. (1989) areas dominated by planar reefs were the inshore northern GBR and inner Bunker–Capricorn Reefs in which a higher mid-Holocene sea level may be expected. This analysis has largely confirmed the factors which are important in determining the evolutionary stage a reef has reached. Depth of the Pleistocene antecedent surface will determine when a substantial part of the reef will reach sea level although small reef pinnacles may be able to keep up with sea-level rise from whatever the depth of their foundations and form small reef flat areas of anomalously early age. The morphology of the Pleistocene foundation may also be important not only in the location of mid-platform
270
The mid-shelf reefs of the Great Barrier Reef
patch reefs and reef perimeter rims as discussed for example by Harvey and Hopley (1982), but also in determining the distance between a reef perimeter producing calcium carbonate of about 10 kg m2 yr1 at its maximum rate and the interior of the reef with much lower rates of productivity. Reefs narrower than about 1 km appear to be favored in terms of rate of development, a factor that may explain the relatively senile stage of Williamson Reef. Even more important is size. Small reefs develop more rapidly than large ones and for very large reefs (over 100 km2) the rate of evolution may be so slow that they cannot go beyond the lagoonal stage in a single interglacial period. The presumptions made in conjunction with the classification of reefs by Hopley (1982) appear to be confirmed and extended by the additional data that are now available.
9 The coral reefs of the outer shelf of the Great Barrier Reef
9.1 Introduction: shelf-edge morphology The gently sloping continental shelf on which the Great Barrier Reef (GBR) is situated starts to steepen at depths between 40 and 50 m. In the far north the change of slope is rapid and coral reefs are located on the very edge of the dropoff. Along the central GBR the slope is more gentle and this section of the outer shelf is dominated by submerged reefs which only rarely reach the sea surface. The outermost reefs of the south central GBR are the massive Pompey Reefs which appear to be located on the shelf edge but which have been shown more recently (Hopley, 2006) to be on the highest of a series of shelf-edge steps descending to 80 m, at which depth the shelf drops away steeply. Distinctive shelf-edge reefs are not found around the Swain Reefs nor opposite the southernmost Bunker–Capricorn reefs. A feature of all the shelf-edge reefs is that they are linear and parallel to the contours of the shelf. Even the large Pompey Reefs are aligned parallel to the shelf-edge bathymetry, have a series of submerged linear reefs to seawards, and contain linear features which are persistent along the entire complex. The best known of the shelf-edge reefs are the ribbons of the northern GBR. Fairbridge (1950, 1967) suggested that they represent the upgrowth of fringing reefs established at lower sea levels and all the evidence suggests that this is very probable, not just for the ribbons but also for all the shelf-margin linear reefs which are discussed in this chapter. Other reefs included in this chapter are the detached reefs of the far north of the GBR. Some, such as Ashmore, Boot, and Portlock, are more than 40 km from the GBR and will not be considered further. However, off Cape Grenville (118 580 S) are a series of detached reefs less than 5 km seawards of the GBR but separated from it by water depths of several hundreds of meters. They include Raine Island Reef, Great Detached Reef, the two Small Detached Reefs, and Yule Detached Reef. 271
272
The coral reefs of the outer shelf of the Great Barrier Reef
The understanding of these shelf marginal reefs has increased greatly over the last 20 years. Detailed bathymetric survey has identified the extent of submerged, formerly unknown reefs. Continuous seismic profiling (CSP) has provided insight into structural features. Descriptions of the morphological features of several reefs have come from manned submersible dives to depths greater than 200 m and further information has come from cameras on remotely operated vehicles (ROVs). Drilling has penetrated the full extent of the Holocene on several shelf-marginal reefs, including one of the detached reefs (Raine Island), and one of the deep holes through the full extent of the Reef carried out by the International Consortium for Great Barrier Reef Drilling (2001) was on Ribbon 5 Reef (Section 2.6). This information is synthesized in the current chapter. A summary of drilling through the Holocene of the outer shelf reefs is provided in Table 9.1.
9.2 Modes of shelf marginal reef growth and major influences on the growth morphology Five distinctive modes of shelf marginal reef growth are recognized plus the reefless shelf edge off the Swain and Bunker–Capricorn Reefs (Fig. 8.1). In all cases, the transition from one reef type to another is sharp and within each region the morphology is very homogenous. The regions are: (1) (2) (3) (4) (5) (6)
the northern detached reefs the northern deltaic reefs the ribbon reefs the submerged reefs of the central GBR the stepped shelf edge of the Pompey Complex the reefless Swain and Bunker–Capricorn area
Only at the maximum of interglacial highstands when sea levels were at or about present position would it have been possible for the growth of coral reefs over what is regarded as the main reef tract of the GBR. At glacial maxima, with sea level up to more than 120 m below present (see Chapter 3), the shoreline along the entire GBR would have been on the very steeply sloping shelf drop-off and, for morphological reasons, reef growth would have been very limited. Between the two extremes were numerous interstadial levels which would have located the shoreline and presumably reef growth on the shoulder of the shelf. For the last interglacial–glacial cycle Chappell et al. (1996) indicate stillstands at 20, 45, 68, and 70 m (see Chapter 3 for discussion). Sectors of the paleo-Burdekin River (Fig. 6.3) appear to be graded to these
5 3 3 2 4 4 3 3 2 3 3
118 370 148 330 148 360 158 050 158 22
178 570
188 160
188 310 188 500 188 500 208 450
Raine Carter Yonge Lark Pass (Ribbon 8) Ribbon 5 b (P. J. Davies) Moss (T. L. Graham)
b
a
26.5 22.5 2.0 21.0
24.5
8.5
16.6 1.4 19.1 1.3 15.5
c.29.0 22.0 – 17.5
24.0
–
12.0 – 18.2 – 15.0
Deepest Shallowest (m) (m)
All Hopley and colleagues unless otherwise stated. Also site of deep drilling in 1995.
Myrmidon (P. J. Davies) Bowl (P. J. Davies) Viper (P. J. Davies) Viper (D. Hopley) Cockatoo
Number of holes
Latitude
Reef a
Pleistocene
30 19 5 19 (þ3 from shallow drilling)
18 (þ3 from shallow drilling) 21
5 6 19 4 25
Number of C dates (þU series)
14
8630 70 6900 130 2660 90 7330 90
7640 70
5620 140
7040 140 5800 100 6580 120 5910 110 7650 80
Oldest
Table 9.1. Results of shallow and medium-depth drilling on outer reefs of the Great Barrier Reef
3640 60 3610 60 2660 90 4670 80
3890 50
3230 140
4740 190 5420 130 5190 120 5910 110 5940 80
Oldest within 1 m of reef flat
Holocene
274
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.1 Distance apart of the 50 m and 100 m isobaths plotted every 300 latitude and shelf marginal reef type at that latitude.
levels and to levels at 80 m and 90 m (Fielding et al., 2003). Significant interstadial reef growth on the shelf shoulder between 45 and 90 m is thus very probable during the past 120 ka. Interstadial levels may have been similar during earlier cycles, during the past 600 ka. The shelf-edge expression of these reef growth periods relates very strongly to the slope of the shelf between the upper and lower levels achieved during these interstadials (45 to 90 m). To illustrate this the distance between the more convenient levels of the 50 and 100 m isobaths has been plotted at every 300 latitude from 108 to 228 S on Fig. 9.1. The width of shelf between these depths closely correlates with the changes in the total width of the continental shelf (for shelf areas by latitude see Hopley et al., 1989, table iv) (Table 5.7). Deltaic and ribbon reefs occur on the steepest sections of shelf between these critical depths with a horizontal distance of between 100 and 640 m. At 168 S the shelf slope becomes dramatically less steep and the horizontal distance between the 50 m and 100 m isobaths is between 1.6 km and 2.7 km along 800 km of shelf margin. This is the sector where parallel lines of submerged reefs occupy the outer shelf edge. There is a further widening at 208 S to between 3.2 and 4.8 km (and in places up to 20 km) a reflection of the stepped shelf margin in front of the Pompey Reefs.
9.2 Modes of shelf marginal reef growth
275
There is good reason for this relationship between shelf slope and reef growth. Given a reef of about 400 m width (regarded as the equilibrium reef flat width by Kinsey and Davies, 1979b) on the gently sloping shelf discrete and well-separated reefs would result from different interstadial sea levels. On the steep slope the various reefs may not be able to achieve this equilibrium width and will be plastered over each other producing complex carbonate structures at the shelf margin. However, they would provide a substantial foundation for further upward growth during interglacial highstands producing the ribbon or deltaic reef morphology the details of which are dependent on tidal currents (see below). Minor erosion or undercutting during the lowest sea-level period would also produce the near-vertical carbonate wall seen for example in Fig. 9.6. This is not unlike the model of ribbon reef development from low sea-level fringing reefs proposed by Fairbridge (1950, 1967). Complex structural factors have produced these contrasting shelf marginal morphologies. Faulting may explain the steep continental slopes to the north and south. However, the boundary between the shelf-edge ribbon reefs to the north and submerged reefs to the south at 178 S coincides with the northernmost impingement of the offshore Halifax Basin on the Queensland shelf (Symonds et al., 1983) (Figs. 2.1 and 8.1). The basin lies beneath the outer shelf as far south as about 208 S with subsidence indicated at least into the Pleistocene according to results of the Ocean Drilling Program Leg 133 (McKenzie et al., 1993). The shelf margin further south adjoins the Marion Plateau, also affected by subsidence ‘‘pulses’’ until at least the late Pliocene (McKenzie et al., 1993). There is the possibility of continued movement both here and in the Halifax Basin up to the present as mild seismic events on the outer shelf are reported regularly and hydro-isostatic subsidence during the Holocene of at least 1 m on the outer shelf south of Cairns has been demonstrated by Chappell et al. (1982) and Hopley (1982, 1983b). The final influence on reef development appears to be tidal velocity (Section 4.3). Tidal range along most of the GBR outer shelf is approximately 2.5 m but reaches 3 m opposite Torres Strait and exceeds 4 m in the Pompey Complex. Tides are semidiurnal and have relatively unrestricted flow through the open reefs on the Central GBR shelf, where tidal currents rarely exceed 1 m s1. However, tidal currents between the deltaic reefs reach 2.5 m s1 in Yule Entrance (Australian Pilot) and over 3.5 m s1 elsewhere (Veron, 1978). In the channels between the Pompey Reefs currents exceed 4 m s 1 on spring tides. These velocities far exceed the entrainment flows required to move coarse carbonate sands. All sediments from the reefs and channels are swept westwards on the flood tides and eastwards on the ebb. Maxwell (1970, fig. 12) has demonstrated how the higher tidal velocities can produce complex deltaic
276
The coral reefs of the outer shelf of the Great Barrier Reef
patterns in reefs. Jones (1995) from interpretation of seismic survey over similar tidal banks in the constricted passages of Torres Strait has shown how these initially unconsolidated sedimentary structures can become the foundation for subsequent coral growth, a model which may apply to both the deltaic reefs of the northern GBR and the reefs at the exits and entrances to the passages between the Pompey Reefs in the south.
9.3 Detailed structure and evolution of the shelf-edge reefs 9.3.1 The detached reefs These reefs rise steeply from water depths of over 200 m and drop to a further depth of more than 500 m to seawards. They appear to have grown on isolated pinnacles of fragmented continental crust. Seismic survey around Raine Island (Nelson, 1980) indicated faulting around the platform. Smaller detached reefs are planar, with reef flat occurring over most of the platform. Larger reefs such as Great Detached have narrower ribbon-like reefs on their windward edge but with a deeper (35 m) platform, thickly covered by Halimeda sediments (Hopley, 1978b). Raine Island (Fig. 9.2) provides the only significant data for the detached reefs. The reef platform is about 3 km long and on its western end has a large vegetated cay (Stoddart et al., 1981). However, it was severely modified in the 1890s when large quantities of guano were mined from the center of the island, leaving a barren central flat. Prior to this, in 1847, the first light tower was built on Raine, by convict labor, using the phosphatic sandstone as a building material. This is an historic site on the GBR, but in the 1980s undermining of the tower led to the need for some restoration. In 1987 engineering and scientific teams acquired important new information. Part of this came from the drilling at sites on the island, and one on the adjacent reef flat (Fig. 9.3). Holes 1 and 3 were less than 5 m and recovered only phosphate rock, sand, and rubble, not penetrating the underlying reef flat. Hole 2 just penetrated the flat at depths of about 5 m, the flat consisting of cemented rubble and coral over cemented Halimeda. Hole 4 commenced on the reef flat and though going down to 8.6 m was mainly through loose shingle. The most significant hole was number 5 commencing on the cay and intercepting the reef flat (level about LAT) at about 4 m. To 15 m the core was mainly through sand and shingle with occasional corals possibly in situ. At 15 m cemented Halimeda was encountered and continued for at least 1.5 m. Radiocarbon dates were obtained from this core (Fig. 9.3) establishing the cemented Halimeda at 15 m (11 m below reef flat level) as the top of the
Figure 9.2 Raine Island from the north. The bare area in the center of the island was created by guano mining in the 1890s. The light tower built in 1847 is at the far end of the island (photograph: Queensland National Parks and Wildlife Service).
Figure 9.3 Drilling and dating results from Raine Island.
278
The coral reefs of the outer shelf of the Great Barrier Reef
Pleistocene. An anomalous date of 6300 130 years BP comes from coral at 14.0 m but other dates appear reliable and indicate the recolonization of the reef platform well prior to 7 ka ago, reef flat formation some time after 6.3 ka ago, and the initial accumulation of the cay 4740 years BP. These depths and dates for Raine Island are similar to those from the ribbon reefs (see below). Other radiocarbon dates are also available for the cay. All are from Tridacna shells embedded in the phosphatic sandstone and all provide ages of about 1000 years BP (Polach et al., 1978; Limpus, 1987). Given the size of the shells and the clustering of the dates a storm event or stormy period may be indicated. The prominence of Halimeda in the cores may also be significant given the extent of Halimeda gravels found on the submerged part of the Great Detached Reef (Hopley, 1978b). There is the suggestion that conditions during the last interglacial, possibly involving upwelling of nutrients against the continental shelf to produce conditions favorable for Halimeda growth (Section 6.6.4), were very similar to those of today. 9.3.2 The northern deltaic reefs The northernmost 96 km of shelf edge of the GBR province consists of short (up to 4 km length) reefs, parallel to the shelf edge and separated by passages up to 200 m wide and 35 m deep (Fig. 9.4). Water depths at the shelf edge are
Figure 9.4 The deltaic reefs on the shelf edge opposite Torres Strait.
9.3 Detailed structure and evolution of the shelf-edge reefs
279
30–50 m but on the ocean side the drop-off is immediate and steep, down to 700 m or more. These reefs have been termed deltaic reefs by Veron (1978) because of the complex delta-like lobes which have formed on the western side of each passage. He also recognized a less complicated set of ‘‘dissected’’ reefs just to the north which are similar in size and general morphology except for the extent of deltaic-like structures on the western side. There is little quantitative data on which to base the evolution of these reefs. Veron (1978) describes ‘‘saddle’’ or shallow ridges extending across the major channels rising to within 10 m of the surface. This suggests that the outer linear reef may have been much more continuous in the past, perhaps similar to the ribbon reefs of today. However, such a barrier could not have coped with the enormous volumes of water which flood in to Torres Strait on every high tide. The tidal range here is up to 1 m higher than along the ribbons and tidal velocities are increased within the constricted waters of the Gulf of Papua and Torres Strait. Many features of the reef geomorphology reflect this. Flood tides sweep debris from the reefs to form the intricate deltaic structures upon which reefs have subsequently formed as described by Maxwell (1970) and Jones (1995). Many details of these deltaic reefs reflect the velocity of the currents and the debris they carry. Channel floors are worn smooth whilst significant abrasion grooves occur in channel walls and even on the reef flat (Veron, 1978). 9.3.3 The ribbon reefs The ribbon reefs form a classic shelf-edge barrier reef system between 118 and 178 S, a distance of about 700 km (Hopley, 1982). Individual reefs are up to 28 km long and are separated by narrow passages most of which are less than 1 km wide, with reefs on either side curving inwards away from the shelf edge but without the deltaic patterns seen further north (Figs. 2.10 and 9.5). The shelf drop-off is even steeper than in front of the deltaic reefs. From water depths of 40–50 m, depths of 500 m are reached within a few hundred meters of the outermost reefs and 1000 m depth is reached within 1 km. The ecology and geomorphology of the shallower parts (<20 m depth) of the ribbons is well known (e.g., Veron and Hudson, 1978; Hopley, 1982). The high-energy outer edges of the reefs have well-developed spur-and-groove systems which merge into the reef flat initially formed of low encrusting and corymbose corals, with some surge channels incised tens of meters into the reef flat. Zonation of the reef flat is very well developed and consists of an algal pavement of cemented limestone with encrusting algae and occasional large (up to 3 m) reef blocks. The crest of the reef flat is formed by a rubble zone of
280
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.5 The ribbon reefs. Yonge Reef is in the foreground.
Acropora shingle and coarser coral fragments. Behind this, the inner reef flat is what Hopley (1982) termed the aligned coral zone, living coral promontories aligned normal to the reef front and separated by channels up to 3 m deep. The back reef typically has a sand slope and colonies of more fragile corals, with water depths of 30 m reached within 80 m of the back reef zone. The width of the ribbon reefs is typically 600–800 m. The zonation reflects the ambient high-energy conditions that these reefs experience. However, major modifications can still take place during large cyclones when wave heights approach 10 m (Done, 1992c; Puotinen, 2004). The effects of a 960-hPa cyclone, Cyclone Ivor, with wind speeds exceeding 180 km h1, were witnessed by researchers who had just completed a survey of Carter Reef (148 330 S) (Done et al., 1991; van Woesik et al., 1991) Effects of this relatively small cyclone were evident along 80 km of the ribbon reefs with storm waves eroding a 1.5 m step in the reef crest leaving bare substrate and occasional massive colonies. Massive corals were scoured and reef matrix
281
9.3 Detailed structure and evolution of the shelf-edge reefs
peeled off to form rubble banks and a rubble delta at depth. Such events have probably been frequent in the Holocene history of the ribbon reefs. In addition to the information from the 210 m deep borehole discussed in Chapter 2 (International Consortium for Great Barrier Reef Drilling, 2001; Webster and Davies, 2003; Braithwaite et al., 2004) important insights into the evolution of the ribbon reefs has come from manned submersible dives in 1984 off the front of Ribbon 5 (Fig. 2.10) by C. V. G. Phipps, P. J. Davies, and D. Hopley (Fig. 9.6). The living reef front with spurs and grooves extends
Reef flat 0 Living 10 coral
Modern reef front Spur and groove Coral garden
20 30 40 50
Coarse blocks and boulders
UNIT 7 Upper sand slope
Fine to medium Halimeda sand and occasional reef blocks or dead coral heads
50 m reef Largely dead reef Coarse rubble
Deep rubble and sand slope UNIT 6 The brow
Coarse sand in mega ripples 70 m “dead” red reef with numerous large gorgonians and sea whips
80 90 meters
UNIT 8
Modern reef front talus
60 70
UNIT 9 Holocene
UNIT 5 100 Discontinuous ledges, deep ravines, caves, and overhangs
110 120
Small sediment chutes Ubiquitous cover of Halimeda sand
130 140
The wall
UNIT 4 Maximum cave development
150 160
UNIT 3 RIBBON N° 5
170 UNIT 2
180 190 Sand and rubble
Basal sand
UNIT 1
200
Figure 9.6 Profile of the outer side of Ribbon 5 Reef from manned submersible observations. Reef units from Braithwaite et al. (2004) are shown on the right.
282
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.7 (a) The 50 m reef on the front of Ribbon 5; (b) the ‘‘wall’’ on Ribbon 5 at about 180 m.
down to 20 m. Small talus slopes separate dead reefs at 50 m and 70 m beyond which is a vertical drop to at least 200 m forming a wall of bedded carbonate rock (Fig. 9.7). This morphology, repeated in ROV surveys of Hicks and Day Reefs, is not conducive to the building of discrete reefs at lower sea levels, but rather to narrow bands of deposition superimposed or plastered over each other. Undercutting during the lowest sea-level period would also maintain the near-vertical wall. The result is a single multicyclic foundation for shelf marginal reef growth during the high sea levels of the interglacials
9.3 Detailed structure and evolution of the shelf-edge reefs
283
producing the typical ribbon reef (and probably the deltaic reefs further north). Davies et al. (1997) suggested that the physical features of the reef front as shown in Fig. 9.6 reflected the main features of the stratigraphy of the deep borehole. The main units from Braithwaite et al. (2004) have been superimposed on Fig. 9.6. The lowest four units may be expected to outcrop in the vertical wall. It is tempting to allocate the 70 m and 50 m dead reefs to units 5 and 6 but it is more likely that the less steeply sloping upper part of the reef is a composite of several transgressions with features added (and eroded) during each high sea-level stand, including the Holocene. The Holocene evolution of the ribbon reefs can be drawn from several drilling projects and approximately 62 radiocarbon dates (Table 9.1). In addition to the deep borehole on Ribbon 5, Davies had previously drilled four holes to the top of the Pleistocene (Davies et al., 1985). A James Cook University expedition also drilled three holes in Yonge Reef (Fig. 9.8). Previously, near-surface dates had been obtained for Carter and Lark Pass Reefs (Hopley, 1977, 1982). The depth to the Pleistocene on Ribbon 5 was established at 15.0 m (Davies et al., 1985) or 15.85 m (Braithwaite et al., 2004). On Yonge Reef the depth was
Figure 9.8 Yonge Reef drilling and dating results.
284
The coral reefs of the outer shelf of the Great Barrier Reef
19.0 m and consisted of a pure Halimeda grainstone. On Ribbon 5 there was a pale brown weathered surface over a mixture of Halimeda and large corals. The significance of the top of the Pleistocene being composed of cemented Halimeda grains both here on the ribbons and on Raine Island (see above), may suggest similarly favorable conditions for algal growth at the end of the interglacial as well as the start of the Holocene, as was also suggested by Orme and Salama (1988). Holocene growth over the Pleistocene surface had commenced shortly after 8 ka ago with upward growth being rapid at an average rate of about 6 m ka1 (Fig. 9.8). The ribbon reefs appear to have reached modern sea level between 6 and 5 ka ago, though the most leeward hole on Ribbon 5 described by Davies et al. (1985) lagged up to 3 ka behind the more windward locations. The exact location of the deeper hole drilled on Ribbon 5 is not published and its lower section of Holocene material was not recovered, which suggests it may have been sand similar to Hole 2 on Yonge Reef (Fig. 9.8). Between 4 m and 9 m it had a rapid growth rate similar to the more windward sites but slowed significantly in its vertical accretion between 4 ka and 2 ka ago. Braithwaite et al. (2004) suggested that Holocene deposition had been in a shallow reef environment behind or below an active reef edge. Such a higher rim close to the windward margin was suggested originally by Harvey (1977a) using seismic refraction techniques, indicating the Pleistocene at 17 m towards the leeward side of the reef flat on Carter Reef (the next ribbon reef north of Yonge), but rising to 7.3 m near the windward margin. The windward hole, 50 m back from the reef edge on Yonge Reef (Fig. 9.8), penetrated to only 7.2 m and did not intercept the Pleistocene. It consisted almost entirely of cemented shingle and cavity, a mixture of dates between 4.9 ka and 6.1 ka ago suggesting emplacement as a rubble bank. The well-dated Hole 2 may be analogous to the deep borehole on Ribbon 5. The lower section consists of 8 m of sand overlying the cemented Halimeda possibly laid down as a back reef sand slope. It is overlain by cemented shingle and sand and corals with progressively more algal cement towards the surface. The interpreted environment is the aligned coral zone with an algal pavement topping. Undated Hole 3 on Yonge is 200 m back from the reef front and on the seaward edge of the aligned coral zone. It consists almost entirely of back reef sands. The data from the high-energy ribbon reefs indicate shallow Pleistocene foundations which were recolonized at or before 8 ka ago in water depths of about 8 m (Davies et al., 1985). Once the reefs reached sea level at or shortly after 6 ka ago an energy-controlled zonation appears to have formed quickly and subsequently zones have migrated towards the leeward side as can be interpreted from most of the core logs. Comparisons with mid-shelf and more
9.3 Detailed structure and evolution of the shelf-edge reefs
285
southerly reefs are made in Chapter 11 and the part the ribbons have played in the development of the northern GBR in Chapter 12. 9.3.4 The submerged reefs of the central Great Barrier Reef South of Cairns, where the shelf widens, reefs are set back from the shelf edge and water depths of 70–80 m are found beyond these reefs before a more gently inclined slope which reaches only 200 m within 3 km of the shelf edge and exceeds 1000 m beyond 40 km. The shoulder of the shelf on this 800 km section of the GBR, which extends south-east to the Pompey Complex at about 218 S, contains one and up to three lines of submerged reefs (Fig. 9.9). Only occasionally do they approach sea level and some, though not all, are mapped as reefal shoals (e.g., Jenny Louise Shoal, 168 450 S; RAAF Shoal, 178 110 S; Mustard Patches, 178 170 S) or patch reefs (e.g., Eulalie Reef, 198 210 S; Joist
Figure 9.9 Distribution of submerged shelf-edge reefs between 168 and 208 S based on Royal Australian Navy Hydrographic 1 : 250 000 bathymetric charts and Graham (1993).
286
The coral reefs of the outer shelf of the Great Barrier Reef
Reef, 198 280 S). Very few major reefs reach sea level within 8 km of the shelf edge and only one (Myrmidon Reef off Townsville) is found on the edge, its reef front dropping into more than 200 m depth. The reefs that are set back from the shelf edge are mostly small and well spaced with no narrow channels. Tidal currents are generally of low velocity, <0.2 m s1 (see Wolanski and Pickard, 1985). Shoals coming close to the sea surface were first recognized in bathymetric surveys in the early part of the twentieth century, and their scientific significance noted by Naval Surgeon Lieutenant W. E. J. Paradice, who described the shelf-edge shoals at 178 S (Paradice, 1925). Hopley (1982) showed that these submerged linear reefs were much more extensive, stretching from the southernmost ribbon reefs (the Agincourt Complex) at 168 500 S to at least 178 310 S. He showed several echo-sounding profiles displaying the near-surface shoals and deeper reefs growing from a surface with a general level of 73 m. They were interpreted as immature ribbon reefs initiated at lower sea levels. More recent hydrographic surveys (which are in part the basis for Fig. 9.9) and side-scan sonar profiling of areas off Cairns, Townsville, and Hydrographers Passage (208 S) allowed Harris and Davies (1989) to demonstrate the extent of the shelf-edge features in water depths between 25 and 167 m. A series of features including submerged platforms, submerged barrier reefs, pinnacles, lagoons, terraces, and notches were recognized (Fig. 9.10). In a further study, Graham (1993) undertook detailed seismic profiling, depth sounding, and drilling of adjacent reefs in the area east of Innisfail originally highlighted by Paradice. This work was later supplemented by ROV inspection of shelf-edge features down to 100 m (Hopley et al., 1997). The gently sloping shelf-edge shoulder of the central GBR is clearly the influence behind this distinctive morphology. Reef development at lower sea levels has clearly taken place, though the lack of correlation in the depth of features in different areas and with previously proposed stillstand depths prompted Harris and Davies (1989) to suggest that they were composite in nature produced by multiple stillstands. They may also have been eroded an indeterminable amount during each period of exposure and their depth along 800 km of shelf affected by both tectonic and isostatic submergence. Typical of the submerged reefs is the shelf edge in front of Moss Reef at 178 560 S for which echo-sounding, seismic, and ROV observations are available (Figs. 9.11 and 9.14a). A steepening of the shelf takes place in water depths of approximately 80 m, 6 km seawards of the eastern edge of Moss Reef. The sea floor in front of Moss Reef is about 70 m deep and 3.5 km along the echo-sounding transects is a major and persistent submerged feature analogous to that mapped by Paradice (1925). It is up to 1 km wide and in places rises to within a few meters of sea level though nowhere breaking the
287
9.3 Detailed structure and evolution of the shelf-edge reefs
Figure 9.10 Profiles across the submerged shelf-edge reefs Hydrographers Passage (188 500 S) (from Harris and Davies, 1989).
near
surface. In places it is as much as 55 m below sea level. Two kilometres to seawards of this a smaller but also persistent reef occurs rising from a depth of 70 m to approximately 45 m. On the steepening shelf margin other much narrower reefs occur, appearing as pinnacles on the echo-sounding and seismic traverses but shown to be substantial from ROV observations.
288
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.11 Seismic profile across submerged shelf-edge reefs seaward of Moss Reef (from Graham, 1993).
As the shelf slopes away at an average gradient of 1 in 20 other terrace-like features at 90 m, 100 m, and 120 m can be observed. Figure 9.11 is but one of a series of seismic profiles which provide information on the shelf-edge reefs. Most are just north of Moss Reef and the conclusions given below are taken from all data available (Graham, 1993): *
*
*
*
the modern shelf marginal and immediate pre-modern submerged shelf-edge reefs generally grow from more extensive older carbonate platforms actually buried under a shallow sediment cover reef initiation appears to have been favored on the edge of steep drop-offs such as on the margin of older carbonate platforms or erosional features the upper continental slope and shelf-edge reefs display a complex history of both growth and erosional notching and planation, with erosional terraces becoming the preferred site of reoccupation in subsequent sea-level rise events. Shelf-edge reefs are clearly multicyclic. The most recent (Holocene) phase of growth is represented by a veneer of pinnacles, which is probably a characteristic mode of growth for these reefs as they are stranded by rapidly rising sea-level events. interpretation of past sea levels from the morphological and seismic data is extremely complex. Many features are a combination of erosion and depositional episodes with the evidence actually related to a particular sea level often buried beneath the present surface.
9.3 Detailed structure and evolution of the shelf-edge reefs
289
The ROV observations added considerable information on this part of the GBR. The sea floor between Moss and the first submerged reef is a relatively flat sediment-covered surface. Surprisingly, for a feature not previously mapped, the major submerged reef had a significant coral cover unlike the outer reefs which were predominantly dead apart from a few colonies on the crest. All attempts at recovering framework samples for dating were unsuccessful and rubble from either side of the outer reef provided inconclusive dates of between modern and 1130 60 years BP, possibly indicating an origin as cyclone dislodgement material from the reef crest (Hopley et al., 1997). What appeared to be a significant notch at 120 m turned out to be a sandy mound distorted on the echo-sounding profile by foreshortening on the steep slope indicating the care needed in relating such features to past sea level. A manned submersible dive has also been undertaken on the front of Myrmidon Reef, one of the few reefs which come up to sea level on the edge of the shelf on the central GBR. The profile (Fig. 9.12) contrasts greatly with the steeply sloping front of Ribbon 5. Although there are some small vertical steps, most of the slope is about 458. Living corals were encountered to a depth of 100 m and living Halimeda to about 125 m. Between 60 and 80 m depth coral cover (Leptoseris, Pachyseris, and Endophyllia spp.) was almost 100% (Fig. 9.13a). Recent bathymetric surveys show the slope becoming even more gentle beyond 200 m (Fig. 9.13b), the base of the submersible survey. They also suggest that Myrmidon is separated from the main shelf of the central GBR by water depths exceeding 100 m, very similar to the detached reefs further north. Some chronological information for the Holocene evolution of this central part of the GBR shelf comes from the drilling of four of the few reefs that reach sea level. These include Myrmidon, Viper, and Bowl (Davies and Hopley, 1983; Marshall, 1985), and Moss Reef (Graham, 1993) (Figs. 9.14, 9.15, and 9.16). Holes were drilled to over 20 m on Myrmidon, Viper, and Bowl (a maximum of >28 m on Bowl, >24 m on Myrmidon, and >21 m on Viper) and yet the Pleistocene was encountered only in one hole on Myrmidon at a depth of 28 m (Marshall, 1985). Drilling on Moss was to a shallower maximum of 10 m. Using the Holocene sea-level envelopes developed for eastern Australia (Thom and Chappell, 1975; Thom and Roy, 1983) Pleistocene foundations between 25 m and 30 m would have been drowned at between 10.3 and 8.8 ka ago. The basal dates of 8.63 and 8.62 ka for Bowl Reef would seem to confirm fairly rapid colonization (Fig. 9.16). However with a vertical accretion rate of just over 6 m ka 1 these outer reefs would have been trailing the rise in sea level and at about 6.5 ka ago, when sea level first reached its present position, water depth appears to have been ubiquitously greater than
290
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.12 Profile on the windward side of Myrmidon Reef from manned submersible observations.
10 m (Fig. 9.17). Thus the earliest these reefs reached sea level was at about 4.5 ka ago, and for most it appears to have been after 4 ka ago. That these reefs had a considerable water depth over them during their phase of vertical accretion appears to have influenced their Holocene
9.3 Detailed structure and evolution of the shelf-edge reefs
Figure 9.13 (a) 100% cover of Leptoseris sp. at 65 m on the front of Myrmidon Reef; (b) rocky outcrop at 180 m on the front of Myrmidon Reef.
291
292
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.14 (a) Moss Reef; (b) Bowl Reef; (c) Viper Reef.
structure, summarized in Figs. 9.15 and 9.16. Unlike the windward margins of the ribbons, which never had more than a few meters of water over them, the deeper water over the central GBR shelf-edge reefs allowed for a much greater accretion of both massive and branching framework. Within Moss Reef, Graham (1993) reports an average of 20% framework for the four holes drilled. This compares with 6.3% for the three holes on Yonge Reef. The significance of the greater water depth over these reefs in the mid Holocene will be discussed in Chapter 12. What information relevant to the evolution of the submerged reefs comes from the surveys of Moss, Myrmidon, Bowl, and Viper Reefs? Firstly, the greater depth of the Pleistocene may be due to sampling problems or erosion, but the possibility of subsidence, as has been postulated for the outer shelf of the central GBR, may be strengthened by this evidence. However, in all other respects the Holocene reef growth at the central shelf margin appears to have been similar to that elsewhere on the GBR and the present depth of coral growth on Myrmidon equals that of anywhere in the world and extends
9.3 Detailed structure and evolution of the shelf-edge reefs
293
Figure 9.15 Drilling and dating results from Moss and Myrmidon Reefs (data from Graham (1993), and courtesy of P. J. Davies).
below the depth of most of the submerged reefs, the majority of which are presumed dead. Dead shelf edge reefs have been described from other parts of the world (e.g., Adey et al., 1977; Lighty, 1977; Lighty et al. 1978; Macintyre, 1988; Macintyre et al. 1991). The simplest explanation forwarded for the demise of
294
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.16 Drilling and dating results from Bowl and Viper Reefs (data courtesy of P. J. Davies).
295
9.3 Detailed structure and evolution of the shelf-edge reefs ka 0
1
2
3
4
5
6
7
8
1 2 3 4 5 6 32
7 8 9
4
10 11 12 meters
13 14 15 16
1
17 18 19 20
1
21
3
2
22 23 24
1
25 26 27 28
2 Moss Myrmidon Bowl Viper
3
Figure 9.17 Accretion rates for Moss, Myrmidon, Bowl, and Viper Reefs.
the submerged shelf-edge reefs has been drowning during the earlier part of the postglacial transgression when the rapid rise in sea level, greater than 15 mm a year, was too fast for reef growth to keep up (e.g., Carter and Johnson, 1986). In some instances it appears that shallow-water hermatypic corals have been succeeded by deep-water biota dominated by sponges and ahermatypic corals
296
The coral reefs of the outer shelf of the Great Barrier Reef
as described in Barbados by Macintyre et al. (1991). Obviously once this community change has taken place the reef ’s ability to grow upwards is even further handicapped. Other explanations may have a more local application. Goreau and Land (1974) and Blanchon (1995) have suggested that as sea level rises and reefs grow upslope of the shelf-edge structures, sediment from the higher reefs will rain down on the lower reef causing its demise (clastic control). Certainly some of the seismic profiles in Graham (1993) show reef pinnacles clearly being inundated by sediment from upslope. For some higher shelf-edge reefs Blanchon (1995) suggests that the upper limit of growth is provided by the trimming action of low-frequency hurricane waves. Probably the most accepted explanation for the demise of shelf-edge reefs has been a decline in water quality as sea level has crept up over the continental shoulder and onto the shelf. The death of the shelf-edge reefs has variously been referred to as ‘‘being shot in the back by your own lagoon’’ (Neumann and Macintyre, 1985) or ‘‘killed off by inimical bank waters’’ (e.g., Schlager, 1981). Hallock and Schlager (1986) attribute the demise of these reefs almost entirely to eutrophication due to release of nutrients from a submerging shelf at a time when sea level was still rising rapidly. At high latitudes these shelf waters in winter may also have been too cool for reef growth in comparison to the deep oceanic water outside the reefs. However, Marshall (1988) has suggested that changes to upwelling patterns at the shelf edge after the shelf is flooded may also have produced decreases in temperature and increase in nutrients through the rise of deep oceanic water sufficient to kill off these shelf-edge reefs. As sea level flooded the continental shelf it is also probable that much sediment was released with a major decline in water quality within the area on which the presently submerged reefs are now found. Recently Dunbar and Dickens (2003) and Harris et al. (2005) have suggested that there was massive siliciclastic discharge to the continental shelf slope during the latter phase of the Holocene transgression possibly from large storages accumulated during the previous lowstand. Interestingly, the highly viable Myrmidon Reef, partially isolated from the main shelf front during this period when the coastline stood at the shoulder of the shelf, would have been isolated almost 10 km offshore possibly outside the zone of high turbidity. 9.3.5 The south central Great Barrier Reef including the Pompey Complex The unique reefs of the Pompey Complex (Fig. 9.18) occur where the shelf is at its widest, separated from the mainland by the Capricorn Channel. Like
Figure 9.18 Pompey Reefs showing seismic lines, drilling site, and location of seismic profiles seen in Fig. 9.23.
298
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.19 (a) Part of the massive Pompey Complex; (b) tidal scour within the channel between Pompey Reefs.
elsewhere the reefs are parallel to the shelf edge, but are stepped back from it by up to 20 km with lines of submerged reefs on the ocean side. However, the reefs themselves are massive (Fig. 9.19a), quite unlike any other shelf margin and the area of submerged reefs is not gently sloping but stepped. With a tidal
9.3 Detailed structure and evolution of the shelf-edge reefs
299
14
28 Shelf marginal submerged reefs
Major submerged reef
Depth (m)
42
56
70
84
1 km
Figure 9.20 Seismic profile over submerged shelf-edge reefs seawards of the Pompey Reefs.
range exceeding 4 m, tidal currents reach their highest velocities in the whole GBR province (>4 m s1) (Fig. 9.19b) and deltaic structures associated with the narrow (100–2000 m wide) channels which separate the reefs have played a major role in the evolution of these reefs. These channels have been scoured to depths of between 70 and 110 m, 10–40 m deeper than the shelf on either side of the Pompey Reefs. Over a shelf distance of 140 km the Pompey Reefs, many between 50 and 100 km2 in area, form an almost continuous barrier. Major reefs are up to 15 km in length and the entire reef tract is between 10 and 15 km wide. Aspects of their evolution have been described by Hopley (2006). The Pompey Reefs rise from a general shelf depth of 50–60 m. On the ocean side is the stepped platform, up to 20 km wide. The first step down, to a 70 m level, is marked by a significant and apparently persistent linear reef rising in places to within 10 m of the surface and about 1 km in width (Fig. 9.20). In front of it the shelf steps down another 5–10 m to 80 m at the shelf drop-off where a series of up to three lines of reefs may occur rising to within 38 m of the surface (Fig. 9.20). These are not as substantial as the first line of submerged reefs and may not be as continuous along the shelf edge as they did not appear on all survey lines. Beyond the shelf edge is a further downward step of more than 200 m onto the wide Marion Plateau which has an average depth of about 400 m.
300
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.21 (a) ‘‘Blue hole’’ on Molar Reef; (b) ‘‘blue hole’’ on Cockatoo Reef.
Within the Pompey Complex are at least four ‘‘blue holes’’ previously interpreted as collapsed karst dolines (Fig. 9.21) (Backshall et al., 1979). They occur on Molar Reef (32.5 m deep), Cockatoo Reef (40 m), on Reef 20–374 (unsurveyed), and on Reef 20–389 surveyed only by recreational divers but reported as being 90 m deep (Byron, 1985). This compares to an average
9.3 Detailed structure and evolution of the shelf-edge reefs
301
depth of about 10 m in the extensive and intricate lagoons of these reefs. Within many reefs and separating the lagoons are linear reef flat ridges, some of which are parallel to the shelf edge and seem to continue across the deep channels which separate individual reefs (Fig. 9.26a). The most distinctive features of the Pompey Reefs are the reefal ‘‘deltas’’ formed by the bifurcation of the major channels between reefs. The morphology produced is similar to that of the northern deltaics except that the delta lobes are found on both south-western and north-eastern sides, i.e., produced by both ebb and flood currents (Fig. 9.22). Shallower depths of water on the seaward or north-eastern side of the Pompey Complex have permitted accumulation of sediments scoured from the channels on both sides of the main reef tract several kilometers from the reefs, as demonstrated by Maxwell (1970). Seismic surveys in Torres Strait by Jones (1995) on similar reefal structures are helpful in the interpretation of CSP work undertaken in the Pompey Complex (Fig. 9.18). Whilst the seismic signal for the open framework of Holocene reefs shows no internal structure the shoals of reefal debris show a clear structure with a horizontal bedding suggesting sediment accumulation by sand wave activity (Jones, 1995, fig. 4). Seismic surveys undertaken in the Pompey Complex show similar features (Fig. 9.23) on both the open ocean and landward side of the reefs (Hopley, 2006). Sand waves with minimal reef development are found opposite the inter-reefal channels in the form of levee-like banks and deltaic splays up to 2 m in thickness. Where surveys were able to go across the submerged portions of the deltaic reefs, horizontally bedded sediments form the foundation materials. These complex reefal structures also appear to be present in the older deltaic reefs closest to the main Pompey Reefs and their extent suggests that there may be more than one stage in their evolution. Three holes have been drilled into Cockatoo Reef (Fig. 9.24), adjacent to the major channel on the south-eastern side of the reef. The Pleistocene, consisting of recrystallized shingle overlying sand and coral, was encountered in the deepest hole at 18 m. The oldest Holocene date is 7330 90 years BP but this lies 5 m above the Pleistocene surface suggesting a recolonization date closer to 8 ka ago. Vertical accretion in all holes was between 4 and 5 m ka1 with this part of the reef reaching modern sea level by 4.6 ka ago. Sand and shingle appear to make up the base of the Holocene section with robust branches and head corals separated by 2–3 m sequences of mainly coarse shingle and rubble. How representative this channel-side section is, is difficult to assess without further drilling. The reef flat is as well developed along the channels as anywhere within the Pompey Complex. Windward margin sites where fastest accretion rates may be expected in the Pompey Reefs largely consist of what appear to be younger and actively accreting deltaic reefs.
302
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.22 Examples of northern ebb tide delta and southern flood tide delta reefs.
9.3 Detailed structure and evolution of the shelf-edge reefs
Figure 9.23 (a) Seismic profile of flood delta reefs on horizontally bedded sediments; (b) active flood tide sand wave at 30–40 m depth; (c) narrow leveelike formations in an area of horizontally bedded sediments forming the focus for reef development. The location of all sections is shown in Fig. 9.18.
303
304
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.24 Drilling and dating results from Cockatoo Reef.
9.3 Detailed structure and evolution of the shelf-edge reefs
305
The evolution of the Pompey Complex reefs has been the subject of some debate. Maxwell (1970) provided excellent explanations of the deltaic forms, the possible extent of shelf-edge reefs, and the importance of shelf marginal slope in developing the morphology of shelf-edge reef growth. However, his model of lagoon and patch reef development involving a ‘‘resorbtion’’ process (Maxwell, 1968) was dismissed by Hopley (1982, ch. 9). At a time when Purdy’s karstic control of reef growth was highly fashionable (Purdy, 1974) and the blue holes of the Pompey Reefs had just been described (Backshall et al., 1979), Hopley (1982) accounted for much of the Pompey Complex morphology as being determined by karst features produced at low sea-level times. More recently, Hopley (2006) has produced a new interpretation of the evolution of these complex reefs. It incorporates the strong currents produced by the high tidal range of the region and deltaic reef formation, the persistent linear features (Fig. 9.26a) that extend across both reefs and channels, and the linear submerged reefs to seawards. In this new model these complex reefs are considered to have been initiated at an earlier interglacial high sea level as a simple barrier reef on a 50–60 m deep shelf step (Fig. 9.25). Once the barrier had formed the intervening channels were subjected to high-velocity flood and ebb tidal currents, which, because of the relatively shallow depth of shelf on either side, were able to build both leeward and windward complex flood and ebb tide deltas (as per Maxwell, 1970). A line of submerged reefs in front of the initial barrier became incorporated into the proto-Pompey reefs as they grew up to sea level and became attached to the original barrier by these deltaic deposits which became the location of further reef growth. The first lagoons were thus formed between the two linear reef systems. Stabilization of the delta lobes resulted not only from colonization by corals but also by cementation after subaerial exposure during the subsequent glacial low sea level. At the next high sea-level stage, the procedure was repeated with deltaic reefs and channels becoming more permanent. Scouring of major channels would commence. Further over-deepening of the channels by tidal scour possibly at lower or intermediate sea levels as described on the far northern GBR by Harris et al. (2005) also took place as the reef mass developed. As the reef complex became larger, more lagoons were enclosed on either side of the original linear reefs. The cycle may have been repeated one or more times to produce the present morphology of a distinctive line of complex large lagoonal reefs with deltaic reefs now forming over earlier sediments, and with the present deposition of deltaic splays giving the potential for further reef extension. It also explains
WINDWARD SIDE 70 m depth
60 m depth
Ebb tide deltas
50 m depth
LEEWARD SIDE Flood tide deltas 10 km
nel an ch
l ne an g ch
dee
pen in
de
ep en in
g
(a) Penultimate interglacial
10 km
4m
s –1
(b) Last interglacial
curr
en t
s
blue hole
00
deep
cha
nn
el
9
1 0–
blue hole
10 km
(c) Holocene Reef at sea level Submerged reef Deltaic sediments
Back reef and lagoonal sediments Reef patches
Figure 9.25 A model of reef evolution for the Pompey Complex from an original ribbon reef barrier (a), progradation of deltaic reefs and incorporation of a previous line of submerged reefs (b) to form complete lagoonal reefs (c). The time sequence shown is suggestive only.
9.3 Detailed structure and evolution of the shelf-edge reefs
307
why linear reef flats within a reef can extend across inter-reefal channels and continue their line in the next reef and beyond. They may represent the present surface expression of the original ribbon reefs. An early stage of this development may be seen in the area of Moss Reef, described above. Moss and other reefs appear to be forming a shelf marginal barrier and in front of Moss at least, is a further linear reef which comes within a few meters of the surface (Fig. 9.11). Only strong tidal currents to produce deltaic reefs are lacking. Within the Pompey Complex there is an increasing distance between the two linear reef flats towards the south and this seems to affect the stage which the reefs have reached. In the north where they are close together the ‘‘accommodation space,’’ i.e., the amount of reef growth and sediment deposition required to progress from one stage to another, is comparatively small and these reefs have gone through at least the full cycle shown in Fig. 9.25. In contrast the larger accommodation space between the southern reefs has required a greater amount of growth and deposition and the reefs from Olympic Reef southwards (Fig. 9.26b), are at a more juvenile stage, more akin to the second stage of Fig. 9.25. Explanation may also be found for the deltaic morphology of the Pompey Complex being as intricate on the high-energy windward side as it is on the leeside. Wave-dominated fluvial deltas lose their intricate morphology and have smoother, straighter outlines. However, redistribution of the carbonate sediments does not appear to have taken place. In part this may be due to the depth of water (>50 m) in which the sediments have been deposited but seismic reflection survey shows that horizontally bedded sediments beneath the deltaic reefs come within 15 m of present sea level, a depth considered to be within the wave base of cyclonic seas experienced regularly on the outer GBR (e.g., Van Woesik et al., 1991). What may be regarded as currently active ebb tide deltaic deposits were not identified in water depths of less than 40 m. Today, minimal protection is given to the outside of the Pompey Complex by the shelf-edge submerged reefs which only occasionally come within 10 m of the surface. However, at lower sea levels if any of these reefs were already present they would have given greater protection. Even a 10 m sea level would have resulted in significant decreases in wave energy hitting the outer edge of the Pompey Complex and a 20 m level would have given extreme protection. Waves reaching the outer reefs would have developed over a maximum fetch of only 10–20 km within the open lagoon between the Pompey Complex and the shelf-edge reefs further favoring the development of intricate deltaic structures. Further research is needed on the Pompey Reefs to understand fully what has been a very complex evolution. From the data available at present they
308
The coral reefs of the outer shelf of the Great Barrier Reef
Figure 9.26 (a) Reef ridge which runs across inter-reefal channels in the Pompey Reefs; (b) more widely spaced ridges producing more open lagoons and larger accommodation space in the southern Pompey Reefs.
9.4 Conclusion
309
seem to incorporate some features and evolutionary trends from all of the northern deltaic reefs, the ribbons, and the central GBR submerged reefs. 9.3.6 The southernmost Great Barrier Reef The stepped margin oceanwards of the Pompey Complex terminates at the northern end of the Swain Reefs and a steep drop-off to 300 m occurs around the northern part of the Swains (Marshall, 1977, fig. 6). This becomes a gentler slope on the eastern side of these reefs with depths descending from about 60 m just outside of the reefs to 200 m within 20 km. There are no reports of submerged reefs in this area. The Bunker–Capricorn reefs are also set back from the shelf margin. The shelf-break at about 50 m is poorly defined and has been described by Troedson and Davies (1997) as a ‘‘distally steepened carbonate ramp.’’ Some small ‘‘reefal banks’’ have been described for this area seawards of the exposed Bunker–Capricorn reefs (Marshall, 1977). However, south of 248 S corals are not even found in the sediments of the outer shelf which are dominated by foraminifera, molluscks, bryozoans, and calcareous algae (Marshall and Davies, 1978). It is possible that there was no coral growth at this latitude at the height of the glacial periods. 9.4 Conclusion The shelf-edge reefs of the GBR are far more complex and extensive than considered even 20 years ago. They also contain much of the Quaternary history of the GBR. As the continental shelf was flooded only during periods of high interglacial sea levels, accretion of the shelf reefs was possible during only 15% of the total elapsed time since the GBR was initiated. Davies (1988) recognized that for most of the Reef’s history, growth would have been concentrated on the shelf margin. At glacial maxima with sea levels more than 125 m below present (Chapter 3) the shoreline along the entire GBR would have been on the very steep shelf drop-off and for morphological reasons alone, reef growth would have been very restricted, as it would also have been on the very steep drop-off in front of the northern deltaic reefs and ribbon reefs during interstadial periods. There, the style of reef accretion and the probability of periodic undercutting, as for example suggested on the Ribbon 5 profile (Fig. 9.6) may make reef evolution difficult to decipher. In contrast, as has been shown in this chapter, reef growth during interstadials, during transgressive periods leading to interglacial high sea levels, and during regressive sea levels after the interglacials would have been concentrated on the
310
The coral reefs of the outer shelf of the Great Barrier Reef
continental shelf shoulder. On the more gently sloping margin of the central GBR and within the Pompey Complex reef growth may have been possible for up to 85% of the time since the GBR was first established, retaining a much greater proportion of the history of the GBR than the shelf reefs. Only in the south (Bunker–Capricorn Group), where a fall of only 1 or 2 8C in sea surface temperatures may have been sufficient to severely limit coral growth, is a low sea-level record of the GBR lacking.
10 Islands of the Great Barrier Reef
10.1 Introduction Two distinctive types of islands occur on the Great Barrier Reef (GBR), the continental or high islands (hereafter referred to as high islands), and the low or reef islands (hereafter referred to as reef islands). Earlier workers defined high islands as those composed of elevated reef limestones or of non-limestone lithologies (Stoddart and Steers, 1977); on the Great Barrier Reef they are essentially continental outcrops that were separated from the mainland as the Holocene transgression flooded the continental shelf (Chapter 3). The high islands exhibit a range of topography and morphology similar to that observed along the mainland coast, which in large part reflects lithology and structure. Some high islands such as Stone Island (208 030 S, 1498 150 E) are relatively low and gently sloping, but others such as Hinchinbrook Island (188 200 S, 1468 150 E) are high, rugged, and drop steeply to the sea (Fig. 2.6). A good account of the relationships between mainland and high island geology and morphology is provided by Stoddart (1978). Reef islands are, in contrast, usually relatively low-lying accumulations of reef-derived sediments, portions of which may be lithified. On the GBR, reef islands have formed on reef platforms that have reached sea level after the mid-Holocene (see Sections 3.4.3 and 11.7), and which focus incident wave trains at a consistent locus over which deposition and accretion can take place (Fig. 4.11). Although the restricted age, materials, and elevation available on most reef islands might suggest limited potential for the development of great morphological diversity, considerable variation in reef island morphology can be recognized within the GBR. The extent of this diversity was briefly introduced in Chapter 5, and is due to the range of latitude and climate covered by the GBR, the diverse range of reef size and shape over which islands may develop, and the time range over which reef platforms formed and island accumulation began (Hopley, 1997b). 311
312
Islands of the Great Barrier Reef
10.1.1 How many islands are there? Determining the precise number of islands located on the GBR is difficult. Complications arise due to the inconsistent inclusion of features as islands (for example, when is a sand bar an unvegetated cay), the ephemeral nature of some reef islands (Section 10.5), and the variable geographic boundaries of many previous censuses (e.g., the GBR or the Great Barrier Reef World Heritage Area (GBRWHA)). These and other problems of geomorphological mapping and classification on reefs (including islands) have been discussed in Chapter 5. Within the GBRWHA, Hopley et al. (1989) identified 617 continental islands with reefs, and 300 reef islands; continental islands without reefs were omitted from their survey, which only included reefs south of 108 410 S (see Section 5.3). Nevertheless, as discussed in Section 5.5, the gazetteer assessment reported in Hopley et al. (1989) remains the most complete assessment of reef islands in the GBR Marine Park. The number of reef islands of different type (e.g., unvegetated cay, vegetated cay, etc.: see Section 10.2) and their occurrence on different types of reef (fringing, crescentic, etc.: see Table 5.3, Chapter 8) were determined from the gazetteer by Hopley et al. (1989), and are summarized in Table 5.8. As indicated in Chapter 1, it is probable that an additional 750 reefs occur north of 108 410 S in Torres Strait, and thus it is likely that close to 1000 islands are located in the Great Barrier Reef and Torres Strait combined. Those islands referred to in this chapter are shown in Fig. 10.1. 10.1.2 The importance of reef islands Islands in the GBR have long been the focus of human interest. Many have been important sites for indigenous peoples since long before Europeans arrived in Australia; larger high islands with reliable water were more regularly inhabited, whereas those with intermittent water were visited more occasionally (O’Keeffe, 1991). Many reef islands were also visited often, but limited and unreliable water supply precluded permanent settlement. Archeological evidence such as rock shelters, fish traps, middens, hearths, and tools composed of exotic rock materials testify to an extended indigenous history on many GBR islands (see Beaton (1978); and O’Keeffe (1991) for good summaries). Many reef islands remain important story places and part of the ‘‘sea country’’ of various Aboriginal and Torres Strait Islander communities (Aston, 1995). Europeans also took great interest in the islands of the Great Barrier Reef. Guano was mined from several islands (e.g., Raine Island (118 360 S, 1448 010 E) (Fig. 9.2), Lady Elliot Island (248 120 S, 1528 450 E)); navigation aids are located on many (e.g., Low Isles (168 230 S, 1458 340 E) (Fig. 1.2), Lady Elliot Island); and others
10.1 Introduction
Figure 10.1 Location of reef islands referred to in this chapter.
313
314
Islands of the Great Barrier Reef
are used for military bombing practice (e.g., Rattlesnake Island (198 020 S, 1468 300 E), Fairfax Islands (238 510 S, 1528 220 E) (Fig. 8.8)). At various times enterprises ranging from beˆche-de-mer (Lizard Island (148 400 S, 1458 280 E)) and turtle processing (Heron Island (238 260 S, 1518 550 E)) to sizeable pastoral operations on some larger islands have occurred. Magnetic Island (198 080 S, 1468 500 E), just 8 km offshore and only a short ferry trip, is now a suburb of nearby Townsville. In recent years tourism (see Chapter 13 for discussion of some of the associated impacts) has been a major activity on a number of GBR islands. Of course, the islands of the GBR are of great ecological significance, providing permanent and seasonal habitats for a great variety of flora and fauna protected under various Marine and National Park Acts. As an example, Raine Island (Fig. 9.2), an isolated vegetated cay approximately 30 ha in area located around 620 km north-north-west of Cairns, is a significant seabird rookery and the nesting site for the world’s largest remaining green turtle (Chelonia mydas) population (Limpus et al., 2003). Erosion in recent years has compromised the nesting success of turtles on Raine, emphasizing the relevance and importance of understanding the geomorphological evolution and dynamics of these systems. 10.1.3 History of scientific investigation of island geomorphology The history of geomorphological research on the GBR has been covered in Chapter 1, but a brief review of reef island research is warranted. Investigations of the geomorphology of GBR islands may arguably extend back to the descriptions of navigators such as Cook and Flinders, and may include the records of Jukes (1847), MacGillivray (1852), and Saville-Kent (1893). However, the first true scientific investigations of island geomorphology were undertaken by the Great Barrier Reef Committee, which was formed in 1922. The membership and history of this Committee and its endeavors are discussed earlier (see Section 1.4.2). As noted, the expedition was based at Low Isles, but members at various times also visited islands from the Flinders Islands (148 100 S, 1448 150 E) south to the Bunker–Capricorn group of reefs (located on the tropic of Capricorn), allowing a broad range of islands to be documented (Hopley, 1982). The studies of Richards, Hedley, Steers, Spender, Moorehouse, Orr, and Stanley are particularly relevant to island geomorphology; their work considerably advanced knowledge of reef island geomorphology (see references at the end of this book) and highlighted its value for other reef science disciplines. Armed with low-level aerial photography made available following the Second World War, Fairbridge and Teichert (1948) later analyzed some of this early work.
10.2 Classification and geomorphology of reef islands
315
Soon after the establishment of the Heron Island Research Station in 1951, detailed descriptions of the Bunker–Capricorn Group of islands were published (e.g., Fosberg, 1961; Domm, 1971); other ‘‘island’’ research stations are established at Low Isles (University of Queensland), One Tree Island (238 310 S, 1528 080 E) (University of Sydney) (Fig. 8.8a), Orpheus Island (188 360 S, 1468 300 E) and Horn Island (108 360 S, 1428 170 E) (James Cook University), Lizard Island (Australian Museum), and North Keppel Island (238 020 S, 1508 520 E) (Central Queensland University). A second expedition of scientists associated with the Royal Society and Universities of Queensland revisited many of the islands of the northern GBR in 1973, aided by tools and techniques unavailable to the earlier party (see Section 1.4.3). The results of this expedition were published in the Philosophical Transactions of the Royal Society of London in 1978 (series A, volume 291: 1–294; series B, volume 284: 1–162), and provided many insights into reef and reef island development. The 1979 Australian Institute of Marine Science northern cays expedition ventured further north than the 1973 team, and although the primary focus was cay botany, Hekel (1979) opportunistically collected basic geomorphic details from over 40 reef islands. Interest in island geomorphology and evolution has continued, but recognition of cay mobility has seen recent research attention focus on reef island stability and dynamics (e.g., Hopley, 1978c, 1981; Kuchler, 1978; Flood, 1979a, 1984a, 1986, 1988; Gourlay, 1983a, 1988; Aston, 1995; Yamano et al., 2000). This chapter summarizes the major results of reef island research on the GBR and aims to present it in one location. It begins with descriptions of the main types of islands, and a discussion of their distribution and longer-term development. Reef island dynamics are then examined, and the factors that strongly influence cay mobility and stability are discussed. The chapter concludes with a discussion of possible future trends in reef island formation and survival within the GBR, themes that are also discussed in Chapter 13. 10.2 Classification and geomorphology of reef islands As noted above, the diversity and complexity of reef islands on the GBR was quickly recognized. Predictably, therefore, classification schemes soon appeared that attempted to accommodate, and in some cases explain, this variation. Based on his detailed work on the GBR during the 1920s and 1930s, Steers distinguished three reef island types: sand cays, shingle cays, and what he called ‘‘low wooded islands’’ (Steers, 1929, pp. 20–27). A ‘‘general’’ classification of reefs and low islands was proposed by Spender (1930) that recognized five classes of reef, four of which support reef islands. Spender’s classification
316
Islands of the Great Barrier Reef
discerned reefs that supported: (i) a sand cay (Class II); (ii) a sand cay and unvegetated rampart (Class III); (iii) a sand cay and vegetated rampart without extensive mangroves across the reef platform (Class IV); and (iv) a sand cay, vegetated rampart and mangrove forested reef flat (Class V). Spender (1930) termed Class V the ‘‘island reefs,’’ and they are synonymous with the low wooded islands described earlier by Steers. Fairbridge and Teichert (1947) later described these islands as ‘‘low wooded island reefs’’ in an attempt to remove confusion about whether it was the reef or reef island that was being described, but Steers’s ‘‘low wooded island’’ is most widely used (see Section 10.2.4 for description). Fairbridge (1950) reviewed earlier classifications and identified five classes of island, all of which occur on the GBR: unvegetated sand cays (type 1); vegetated sand cays (type 2); shingle cay, with or without vegetation (type 3); sand cay with shingle ramparts, vegetated or unvegetated islands, and with mangrove swamp over reef top (type 4); and island with exposed platform of older emergent reef, sometimes fringed by more recently deposited sediments (type 5). The literature on GBR reef islands and their classification was reviewed in detail by Hopley (1982), who also presented a more recent summary (Hopley, 1997b). Hopley concluded that four criteria form the basis of reef island classification: Criterion 1: Sediment type. Are reef island sediments dominated by sand or shingle, or a mixture of the two? Criterion 2: Island location. Is the reef island located toward the windward or leeward edge of the reef platform? Shingle islands are usually located near the windward reef margin and sand cays to leeward (Fig. 4.11) (see Section 4.4.2 for discussion). Criterion 3: Island shape. Is the reef island linear or compact? This characteristic is largely controlled by the interaction of reef shape and wave refraction and transformation around and across the reef (Fig. 4.11) (Sections 4.4.2 and 10.5). Criterion 4: Stage of vegetation cover. Is the reef island unvegetated or vegetated to some degree? The extent of vegetation is often a function of island size, age, and stability.
The four criteria above can be used to describe the range of reef islands developed on the GBR, which range from small and unstable unvegetated sandy cays to complex low wooded islands consisting of a leeward sand cay, windward shingle island, and central mangrove swamp. However, two of the principal criteria – vegetation cover and island shape – are especially prone to swift adjustment. For example, although early descriptions of Arlington Cay (Steers, 1938) and Pickersgill Cay (Steers, 1929) indicated they were vegetated, neither supported vascular plants in 1973 (Stoddart et al., 1978a); Spender
10.2 Classification and geomorphology of reef islands
317
(1930) had earlier noted that Pickersgill Cay was devoid of vegetation when he visited in 1929. The morphological plasticity of reef islands is demonstrated by relict beachrock outlines of former cay position and shape on several reef platforms (e.g., Sherrard Reef (128 580 S, 1438 380 E) (Hopley, 1982); Ellis Island (138 220 S, 1438 430 E) (Hekel, 1979)) that are discordant with contemporary cay morphology (or even presence). Similarly, historical maps are available for several reef islands which when compared with later versions show significant change over the last 50–100 years (Stoddart et al., 1978b; Flood, 1984a, 1986). The main reef island types found on the GBR are described in the following section (see Tables 10.1 and 10.2 for summaries). The major attributes and features of the major reef islands types are schematically depicted in Fig. 10.2. 10.2.1 Unvegetated solitary islands Unvegetated cays are distinguished by the absence of vegetation; the transitory nature of this condition has been described above. They are generally small, Stoddart et al. (1978a) calculating a mean area of just 0.5 ha for the 18 unvegetated cays surveyed in the 1973 northern GBR expedition, although we note that those also surveyed by Steers had become markedly smaller since 1936. Published descriptions of unvegetated cays in the GBR region are summarized in Table 10.2a. Unvegetated cays may be either linear (elongate) or compact (round to oval) and can be composed of sand and/or shingle. Hopley (1982) differentiated four possible sub-classes of unvegetated solitary reef islands based on these descriptors that are briefly described below. (a) Linear unvegetated sand cays Linear unvegetated sand cays develop on medium to large planar reefs of elongate shape where wave refraction generates opposing wave trains that meet along a central axis where sediments accumulate. Linear unvegetated cays appear to have developed in this way on Linnet and Martin Reefs near Lizard Island (Hopley, 1982), but reef islands of this type are exceptionally unstable (Fig. 10.3). Linear unvegetated cays may also form where winds seasonally reverse, producing an elongate accumulation toward the leeward reef margin. Hopley (1982) considered that the migrating linear sandbanks deposited on the large planar reefs of Princess Charlotte Bay resulted from seasonal shifts between dominant south-easterlies and lighter, more northerly, wet-season winds. Several linear cays in Torres Strait, such as Poruma (Coconut Island) (Fig. 13.9a), have probably also developed in response to seasonal wind shifts (Rasmussen and Hopley, 1995; Hopley and Rasmussen, 1998).
3. Compact sand cays Relatively equidimensional accumulations of unconsolidated biogenic sand, ranging from small ephemeral cays <0.1 ha to larger oval-shaped islands of 1.5 ha 4. Compact shingle cays Range from highly mobile mounds of coral shingle to more complex structures, usually with some degree of cementation
1. Linear sand cays Elongate accumulations of unconsolidated biogenic sand, that range from ephemeral sand bars to features approaching 400 m in length and 120 m width 2. Linear shingle cays Essentially large shingle ramparts
A. Unvegetated solitary islands
Stable in short term. Long-term stability is poor Unstable. Sensitive to weather fluctuations. Cay modification and migration over platform can be rapid
Variable. Simple Uncommon cays less stable than complex
Mostly associated with small May contain recurved reefs with planforms that shingle ridges, some promote centripetal partially cemented. focusing of coarser sediments Beachrock and/or conglomerate on larger examples
Common
Relatively rare
Common
Occurrence
Some of the larger Leeward on planar reefs with planforms that produce islands of this type centripetal focusing of have distinctive swash ridges and some have sediments by wave-generated currents central depressions
Shingle ramparts and tongues. Bassett edges common
Parallel to windward reef front
Stability
Very unstable
Common features
Leeward on large or mediumGenerally small. Many sized planar reefs that are ephemeral. Rare either elongate or experience beachrock seasonal wind reversals. May development, only be centrally located on the on larger examples reef platform
Island class and brief description Typical setting
Table 10.1. Reef island types and characteristics, following the classification of Hopley (1982)
MacGillivray Reef, Twin Cays
Wheeler Reef, Pickersgill Cay, Beaver Cay
Pandora Reef
Linnet Reef, Martin Reef
Example(s)
B. Vegetated solitary island 1. Sand cays Accumulations of biogenic sediment that achieve enough stability to become colonized by vegetation. Vegetated core may range in size from <300 m2 to >1100 m2, and may be more than 7–8 m above LAT 2. Mixed sand and shingle cays Mixture of sand and shingle. Mixing due to seasonal wind reversal, or changes in sediment delivery to the cay, as may occur if storm regimes or reef flat configuration changes through time 3. Shingle cays Accumulations of shingle that are sufficiently stable for vegetation to establish. Mostly compact as linear ramparts are more prone to reworking and less able to retain fresh water necessary for vegetation 4. Mangrove islands Islands formed when mangroves colonize reef platforms that lack the windward shingle ramparts and tongues typical of the low wooded island
Rare. Only three One Tree Island, Lady Elliot islands of this type have been Island identified on the GBR
Five known – four in Torres Strait. Night Island is intermediate between this class and low wooded islands
Once established these appear to be reasonably stable, although they may be destroyed by cyclone activity
Unknown, but likely to be vulnerable to cyclone stripping and instability
Cemented ramparts common
Irregular sand ridges are deposited around the mangrove fringe, with remnants of older ridges preserved in the interior, especially to leeward
Located either on windward margins of larger reefs or centrally on exposed smaller platforms
Sheltered locations with low to moderate tidal range. Usually on high reef platforms that receive terrestrial sediments
Murdoch island
Uncommon on the Lady Musgrave GBR. Some yet Island to be examined cays in Torres Strait may be this type
Core appears relatively stable, although beach is mobile and dynamic
Shingle ridges form the interior, but sandy beach. Beachrock and/or conglomerate platform may outcrop on beach
Stapleton Island, Bushy Island, Heron Island
Lady Musgrave Island is the only known example on the GBR. It is located on the leeward reef flat
Relatively common. 40 identified on the GBR
Vegetated core quite stable, but beach may be highly mobile. Proportion vegetated to unvegetated indicative of stability
Usually possess beachrock; sand dunes may develop on larger examples
Typically on leeward margins of medium to large reefs, or in areas protected from high wave energy
Common features
The mutual location of the components of these islands is characteristic, especially with the mangrove forest over the reef top. Features on the sand and shingle cays similar to those outlined for each above
Usually shingle island near the As for sand and shingle cays windward reef margin and a sand cay near the leeward edge. At Masig–Kodall, Torres Strait, the islands are both sandy. Seasonal wind reversals and wave refraction around asymmetric reef are more likely involved
D. Complex low wooded islands All of these islands are located These islands can achieve a on relatively small reefs that complexity considered unique to the GBR, so only generally appear to have developed over relatively the simplest example is described here. They typically shallow foundations. They occupy a large proportion of consist of a windward shingle island formed of shingle the reef top. The typical locations of the composite ramparts that may encircle elements have been given in much of the reef margin, often with low shingle tongues the description extending some distance onto the reef platform. A sandy cay is situated near the leeward reef edge, and these
C. Multiple islands Where two or more discrete islands form on the reef platform. On the GBR no reef platform supports more than two discrete vegetated islands; typically a vegetated shingle cay at the windward reef margin and a sandy cay near the leeward edge
Island class and brief description Typical setting
Table 10.1. (cont.)
Only three such islands occur, two in the Bunker Group, and Yorke (Masig–Kodall) Island in Torres Strait
Occurrence
Fairfax Island, Hoskyn Island, Yorke (Masig–Kodall) Island
Example(s)
Low Wooded Radiocarbon dating Restricted to the Island, Bewick inner reefs north suggests that these of Cairns, 44 Island islands are relatively islands of this old – suggestive of type occur on considerable stability; however the GBR both geological and mapping evidence suggests considerable mobility in cays on some low wooded islands, and significant
As for sand and shingle cays
Stability
may be of varying size and morphological complexity, some well vegetated and others essentially barren. Mangrove forest occupies a considerable proportion of the reef platform between these two accumulations, typically covering 25–50% of the reef top behind the shingle island
fluctuations in the extent of mangroves across the reef top
1548 33.50
158 510
South Pickersgill (1936) 1468 090 1458 120 1458 120 1458 110 1458 100 1458 38.50 1458 38.50 1448 530
148 440
148 43.50
148 430
148 430
168 070
168 070
148 120
Turtle II (1973)
Turtle IV (1973)
Turtle Midreef (1973)
Turtle Reef (1973)
Undine (1973)
Undine (1936)
Waterwitch (1973) 1448 490 1478 320
0
148 28
188 470
Watson North (1973)
Wheeler (1980)
Sudbury (1973)
0
168 57
1548 33.50
b
1508 050
218 010
158 510
South Pickersgill (1973)
1448 000
138 260
Sandbank No. 7 (1988)
Sandpiper (1986)
1438 090
1548 33.50
158 510
North Pickersgill (1936)
138 290
1548 33.50
158 510
North Pickersgill (1973)
Sandbank No. 6 (1980)
1458 390
168 030
Mackay (1973) b
1448 510
1438 41.50
138 220
Ellis (1973)
1438 090
1438 360
128 530
Chapman West 2 (1973)
128 140
1438 360
128 530
Chapman West 1 (1973)
148 310
1498 340
138 130
Binstead West (1973)
Pickard (1973)
1438 370
118 530
Ashmore (1973)
Sand (1973)
1498 39.50
168 39.50
Arlington (1973) b
Long. (E)
Lat. (S)
collected) a
Island/cay (year data
(a) Unvegetated sand cays
3 667
2 570
27 800
11 200
4 610
210
450
3 280
940
13 950
5 300
490
14 469
41 500
3 500
1 290
4 450
15 100
3 800
9 350
4 000
350
140
920
12 400
4 600
Cay area (m2)
351
120
380
275
220
19
65
110
70
205
75
53
214
430
620
115
250
170
120
190
175
32
17
60
325
120
length (m)
Maximum
133
35
120
50
26
10
18
40
20
105
38
12
86
125
170
27
30
140
40
63
30
19
12
22
60
50
width (m)
Maximum
85
8 600
1 205
1 400
area (m2)
Beachrock
Table 10.2. Published details on unvegetated sand cays, vegetated sand cays, and low wooded islands
1 960
area (m2)
Conglomerate platform
Aston (1995)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Steers (1938)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Steers (1938)
Stoddart et al. (1978c)
Aston (1995)
Aston (1995)
Aston (1995)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Steers (1938)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c), Steers (1938)
References
1518 550 1438 160
128 140
Kay (1973) 1528 230 1458 390 1438 480
248 07
238 540
168 030
118 190
118 51.5
Lady Elliot (1992)
Lady Musgrave (1978)
Mackay (1936) b
MacLennan (1980)
Magra (1973)
1438 170
1528 430
0
0
1458 58.50
168 45.50
238 270
1458 58.50
168 45.50
Green (1973)
Heron 19(91)
1528 280
218 590
Gannet (1984) (1992)
Green (1991)
1528 250
218 44
Frigate (1984) (1985)
1438 430
0
138 390
Fife (1973)
1458 280
158 440
East Hope (1973) 1518 460
1458 230
148 420
Eagle (1982)
238 300
1458 230
148 420
Eagle (1973)
Erskine (1978)
1448 540
148 240
Coombe (1929)
1528 180
1448 540
148 240
Coombe (1973)
238 48
1528 250
218 470
Bylund (1984) (1985)
East Hoskyn (1992)
1508 050
208 57
Bushy Redbill (1986)
0
1518 150
Bell (1992)
0
1438 120
128 110
Beesley (1973)
218 490
1528 230
218 380
Bacchi (1984) (1985)
1498 39.50 1438 370
168 39.50
Arlington (1936) b
Long. (E)
Ashmore Banks (1980) b 118 530
Lat. (S)
collected) a
Island/cay (year data
(b) Vegetated sand cays
Table 10.2. (cont.)
33 470
24 960
23 900
195 198
403 000
4 300
150 753
155 143
139 100
36 740
42 000
71 650
60 083
13 050
35 530
72 800
12 530
49 300
45 700
12 000
76 196
43 762
6 950
17 200
35 800
20 500
Cay area (m2)
Maximum
450
453
385
675
850
185
696
632
690
524
409
580
414
138
270
225
430
530
545
198
446
393
420
350
330
295
Length (m)
130
70
105
375
585
43
276
312
300
100
109
230
185
75
240
130
150
155
155
101
217
126
30
62
160
90
Width (m)
Maximum
3 060
2 680
8 540
6 610
1 445
4 630
6 840
7 250
(m2)
Beachrock area
20 950 (63)
10 480 (43)
1 800 (8)
143 188 (73)
350 500 (87)
285 (7)
93 276 (62)
110 274 (71)
117 420 (84)
9 440 (26)
14 400 (34)
58 130 (81)
14 334 (24)
9 540 (73)
21 900 (62)
34 120 (47)
8 230 (66)
3 920 (80)
27 910 (59)
3 400 (28)
38 366 (50)
16 470 (30)
720 (10)
2 600 (15)
13 640 (38)
500 (2.4)
Stoddart et al. (1978c)
Aston (1995)
Steers (1938)
Aston (1995)
Aston (1995)
Stoddart et al. (1978c)
Aston (1995)
Aston (1995)
Stoddart et al. (1978c)
Heatwole (1986)
Aston (1995), Flood and
Heatwole (1986)
Aston (1995), Flood and
Stoddart et al. (1978c)
Aston (1995)
Aston (1995)
Stoddart et al. (1978c)
Aston (1995)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Heatwole (1986)
Aston (1995), Flood and
Aston (1995)
Aston (1995)
Stoddart et al. (1978c)
Heatwole (1986)
Aston (1995), Flood and
Aston (1995)
Steers (1938)
Vegetated area (m2) (%) References
273 000 272 521
1438 580 1438 500 1528 270 1448 010
118 240
138 550
218 470
118 360
Pandora (1980)
Pelican (1973)
Price (1984) (1985)
Raine (1973)
1528 180 1558 550 1518 580
238 480
238 180
238 200
Wilson (1992)
Wreck (1978)
1458 110
148 440
Turtle III (1973)
West Hoskyn (1978)
70 240
1528 220
218 390
Thomas (1984) (1985)
1458 560
1468 090
168 570
Sudbury (1982)
1518 460
1448 510
148 190
Stapleton (1929)
238 410
1448 510
148 190
Stapleton (1973)
168 480
1438 500
138 570
Stainer (1973)
Tyron (1992)
1438 110
118 420
Saunders (1973)
Upolu (1973)
14 610
1448 000
117 784
18 700
105 625
13 800
16 800
15 262
39 600
46 800
15 336
97 200
65 946
1448 010
118 360
138 220
Raine (1984)
Sandbank No. 8 (1984)
28 400
80 530
72 640
52 187
1 121 920
1528 030
North West (1992)
65 380
16 769
238 30
1438 430
138 300
One Tree (1978)
1458 590
168 35.50
Michaelmas (1982)
Morris (1973)
29 030
468 679
1518 450
1458 590
168 35.5
Michaelmas (1973)
0
1518 440
0
238 320
Masthead (1976)
Cay area (m2)
238 180
Long. (E)
Lat. (S)
collected) a
Island/cay (year data
Table 10.2. (cont.) Maximum
950
184
675
300
594
200
239
287
810
620
235
610
437
831
860
374
430
497
400
1 647
595
348
385
1 403
Length (m)
158
136
188
65
158
150
163
68
96
125
115
215
192
418
420
117
250
197
225
748
170
61
70
425
Width (m)
Maximum
13 230
5 520
860
470
6 800
1 840
980
910
(m2)
Beachrock area
42 111 (36)
9 640 (52)
56 312 (53)
735 (5)
40 200 (57)
6 350 (44)
4 000 (24)
4 012 (26)
16 900 (43)
26 820 (57)
5 130 (33)
64 115 (66)
28 039 (43)
157 959 (58)
163 300 (60)
10 800 (38)
57 100 (71)
46 480 (64)
37 812 (72)
972 160 (87)
47 350 (72)
7 536 (45)
7 580 (26)
289 809 (62)
Aston (1995)
Aston (1995)
Aston (1995)
Stoddart et al. (1978c)
Aston (1995)
Stoddart et al. (1978c)
Heatwole (1986)
Aston (1995), Flood and
Aston (1995)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Aston (1995)
Aston (1995)
Stoddart et al. (1978c)
Heatwole (1986)
Aston (1995), Flood and
Stoddart et al. (1978c)
Aston (1995)
Aston (1995)
Aston (1995)
Stoddart et al. (1978c)
Aston (1995)
Stoddart et al. (1978c)
Aston (1995)
Vegetated area (m2) (%) References
1438 13.50
Piper (Farmer–Fisher)
1448 540
1458 170
1458 170
148 33
158 070
158 070
Three (1973)
Three (1982)
1438 340
128 590
Sherrard (1973)
Sinclair–Morris (1973)
1448 510
0
1448 310
1458 090
148 450
Petherbridge (1982)
148 07.50
1458 150
148 39.50
Nymph (1973)
148 310
1448 550
148 300
Newton (1973)
Pipon (1973)
1438 360
138 170
Lowrie (1973)
Sand (1973)
1458 230
158 05
1458 340
168 230
Low (1991)
Low Wooded (1973)
1458 340
0
1448 400
168 230
148 250
Ingram–Beanley (1973)
148 330
1458 530
148 300
Howick (1973)
Low (1973)
1448 580
148 31.50
Houghton (1973)
Leggatt (1973)
1448 580
148 34
Hampton (1988)
1458 530
1448 530
0
148 25
1448 530
148 340
Hampton (1973)
Ingram (1988)
1448 590
Coquet (1973)
0
1438 13.50
128 14.50
148 32.50
Fisher (1988)
(1973)
1448 260
148 070
128 14.50
Clack (1988)
1438 360
1438 050
118 460
Bird (1973)
128 53
1498 340
138 130
Binstead (1973)
Chapman 1 and 2 (1973)
1448 490
0
1438 21
128 24
148 260
Baird (1988)
Bewick (1973)
Long. (E) 0
Lat. (S)
0
collected)
a
Island/cay (year data
(c) Low wooded islands
Table 10.2. (cont.)
181 063 (13.5)
158 130 (11.8)
19 260 (6.1)
13 870
1 290
7 400 (0.2)
21 820
25 760 (3.6)
2 380
41 600 (4.7)
28754 (2.2)
22 500 (1.7)
20 840 (3.8)
108 720 (1.9)
110 910 (2.1)
45 715
124 600 (9.1)
59 250 (6.9)
4 300 (0.5)
120 700 (19.5)
19 260
61 590
92 812
490
40 310
920 (3.1)
104 580 (5.5)
12 500
(% reef top)
Cay area (m2)
738
715
300
220
115
240
275
395
78
258
240
320
492
490
695
320
600
535
600
17; 19
380
60
600
300
285
112
110
27
65
145
120
42
142
130
125
300
350
230
200
42
201
210
12; 19
150
22
300
70
16 210
2 468
5 124
2 898
6 015
2 312
13 780
8 280
9565
9030
6176
131 400 (83)
11 570 (60)
7 104 (51)
3 850 (52)
10 160 (47)
13 998 (54)
266 (11)
12 869 (45)
14 060 (63)
14 220 (68)
89 810 (81)
28 720 (63)
39 965 (65)
29 250 (73)
104 580 (100)
area (m2) (%)
210
Vegetated
Maximum Maximum Beachrock Length (m) Width (m) area (m2)
Mangrove
81 800
3 010
621 200
76 200
257 000
23 700
319 600
322 900
28 600
8 900
260 000
289 200
57 900
621 200
3 400
1 254 400
area (m2)
Promenode
115 000
9 500
3 600
48 600
241 700
30 500
120 100
29 400
2 300
42 700
115 400
0
84 500
12 100
17 600
8 700
114 200
area (m2)
115 600
31 800
42 900
40 000
7 500
36 500
33 400
19 200
42 900
area (m2)
New rampart
Aston (1995)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Aston (1995)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Aston (1995)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Aston (1995)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Aston (1995)
Stoddart et al. (1978c)
Aston (1995)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Aston (1995)
References
1458 270
1448 290
1458 270
158 010
148 280
158 450
West Hope (1973)
b
a
2 570 (0.4)
194 313 (15.1)
194 600 (15.1)
3 300 (1.1)
940 (0.2)
(% reef top)
Cay area (m2)
120
763
720 35
338
350
20
85
20 140 152 625 (79)
164 450 (85)
area (m2) (%)
length (m) width (m) area (m2)
70
Vegetated
Maximum Maximum Beachrock
174 900
53 300
14 800
2 000
120 000
53 600
area (m2)
Mangrove
79 000
35 500
47 900
5 900
46 800
16 300
88 200
93 200
area (m2)
Promenade
5 500
11 400
73 900
3 200
200
18 300
22 100
35 500
area (m2)
New rampart
Appears in both vegetated and unvegetated cay tables.
Where two separate years are listed for each island/cay the data are a composite from two references, with both being the most complete recent references available.
158 01
Two (1982)
Two (1973)
Watson (1973)
1458 10.50
148 430
Turtle VI (1973)
1458 270
1458 120
0
1458 120
148 43.50
1458 120
148 440
Turtle III (1973)
148 420
1458 120
148 440
Turtle II (1973)
Turtle IV (1973)
1458 110
148 440
Turtle I (1973)
Turtle V (1973)
Long. (E)
Lat. (S)
collected)
a
Island/cay (year data
Table 10.2. (cont.)
References
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Aston (1995)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
Stoddart et al. (1978c)
10.2 Classification and geomorphology of reef islands
327
Figure 10.2 Schematic diagram of major reef island types found on the Great Barrier Reef, showing their characteristic features.
(b) Linear unvegetated shingle cays Linear unvegetated shingle cays are large shingle ramparts that make very unstable reef islands. They are uncommon on the GBR (Hopley et al., 1989) as narrow shingle ramparts seldom achieve a size worthy of cay status due to their vulnerability to reworking and erosion near the reef front. Hopley (1982) describes Pandora Reef as a rare example of an unvegetated shingle cay (Fig. 10.4). This cay extends several hundred meters across the reef flat and reaches a maximum height of 3 m (3.7 m above LAT). The unvegetated shingle cay on Pandora Reef remained reasonably intact from the late 1960s to the early 1980s, including through Tropical Cyclone Althea (1971) – the last
328
Islands of the Great Barrier Reef
Figure 10.2 (cont.)
10.2 Classification and geomorphology of reef islands
329
Figure 10.3 (a) Holmes Cay, a linear unvegetated sand cay located east of the outer barrier. Photograph taken from the north-west in 1989. Weather station at western end of cay. (b) Holmes Cay 1993. Note that weather station located in center of cay in 1989 now isolated on reef flat as cay has changed size and shape (photographs: M. Mabin).
330
Islands of the Great Barrier Reef
Figure 10.4 Pandora Reef from the north-east, showing linear shingle ridges and cay.
severe cyclone to affect the area. However, the presence of bassett edges (the lower cemented portions of shingle ramparts with steep dip away from the reef front that are preserved when the rampart has eroded: Fig. 10.11b) and lower ramparts suggest earlier shingle accumulations have been destroyed, and that unvegetated shingle cays have limited longevity (Hopley, 1982). (c) Compact unvegetated sand cays Compact unvegetated cays are more equidimensional accumulations of biogenic sand or shingle, deposited where strong centripetal sediment transport focuses deposition within a restricted locus. Many are highly mobile and unstable (Section 10.5) (see Taylor, 1924; Hopley, 1978c; Aston, 1995). Based on a survey of 18 unvegetated northern GBR cays, Stoddart et al. (1978a) recognized three categories, two of which may be considered compact. The first category includes smaller (< 0.1 ha) ‘‘ephemeral’’ cays that may be submerged at high tide (e.g., Binstead Cay, Beaver Cay). A second larger (1.0–1.4 ha) but also compact morphology was distinguished which includes generally ovalshaped cays up to 300 m long and 100 m wide (e.g. Sudbury Cay; Wheeler Cay (Fig. 8.4b)). Steep beaches with pronounced swash ridges fringe these islands, and some have central depressions. The third ‘‘type’’ of unvegetated cay
10.2 Classification and geomorphology of reef islands
331
distinguished by Stoddart et al. (1978a) includes cays of variable size (up to 400 m long and 120 m wide), some surrounded by beachrock in configurations that differ from the contemporary cay’s shoreline. Some of these are compact, but others are more linear in planform. (d) Compact unvegetated shingle cays Compact unvegetated shingle cays are generally restricted to small reefs, but vary from relatively mobile mounds of shingle and rubble through to complex structures with some degree of cementation. Hopley (1982) considered that MacGillivray Reef was representative of the former. This oval-shaped cay of around 40 30 m in size is less than 1.5 m high and migrates regularly within an area of 100 50 m near the leeward edge of the reef. Separate shingle cays are also located on two small patch reefs at Twin Cay Reef in the Swain Reefs. A degree of stability is suggested on both these cays by the presence on both of a coarse beachrock (Hopley, 1982). 10.2.2 Vegetated solitary islands Eventually some unvegetated cays achieve sufficient stability for vegetation to establish successfully. The establishment of vegetation on GBR reef islands is well covered elsewhere (Hopley, 1982; Heatwole, 1983, 1984), and is not repeated here. We emphasize again, however, that vegetated cays can quickly become unvegetated through natural agencies such as storms, or human impacts. As for unvegetated reef islands, the sediments of vegetated cays on the GBR form a continuum from pure sand to shingle, but classification schemes such as that of Hopley (1982) usually discriminate four separate sub-classes: vegetated sand cays; mixed sand and shingle cays; shingle cays; and mangrove islands. Published descriptions of vegetated cays in the GBR region are summarized in Table 10.2b. (a) Vegetated sand cays Forty vegetated sand cays were identified on the GBR by Hopley et al. (1989), with considerable variation recognized in size, topography, and proportion of area vegetated. The smallest vegetated cays can be < 1 ha in area (e.g., Kay Islet) and the largest exceed 25 ha (e.g., North West Island; Masthead Island (Fig. 13.10b)). They vary from elongate to oval-shaped, for the same reasons earlier outlined for unvegetated sand cays. Stoddart et al. (1978a) recognized three groups of vegetated cay in the northern GBR, the first comprising smaller (< 4.7 ha) narrow and elongate cays vegetated to varying extents (10–66%) with low scrub and grasses (e.g., Eagle Cay; Stapleton Cay
332
Islands of the Great Barrier Reef
(Fig. 10.5)). Steep beaches are typical, and dunes rise up to 7 m on several (e.g., Milman Island (Hekel, 1979); Stapleton Island (McLean et al., 1978) (Fig. 10.5)). The second group includes larger (mean area: 9.3 ha) oval-shaped islands, usually with a significant (>60%) proportion vegetated with either forest or lower communities. Some of these islands are topographically simple (e.g., Stainer Cay), though many exhibit two distinctive terraces consistently separated by around 1 m on each island (McLean et al., 1978; Stoddart et al., 1978a, b), but at different absolute elevations on different islands (Buckley, 1988) (e.g., Ingram Island (Fig. 10.6); Bewick Island (Fig. 8.3a)). Radiocarbon dating and degree of soil formation indicate that the higher terrace is older than the lower terrace (McLean et al., 1978), suggesting at least two periods of cay accumulation. However, available ages overlap and the significance of these features, which also occur at Green Island (Stoddart et al., 1978a), is yet to be established (Section 10.4). The third group lies within the variation of the second, but they are smaller (3.5–8.0 ha) and more equidimensional in shape (e.g., East Hope Island (Fig. 8.3b)). Vegetated cays are generally more stable than their unvegetated counterparts, but can still change morphology and position (Section 10.5). Hopley (1982) argued that linear vegetated cays are less stable than oval cays, citing as evidence: (i) they usually have a proportionally small vegetated area (e.g., Erskine Island (Fig. 10.7)); (ii) cay-detached beachrock outcrops often occur on the reef top; and (iii) the vegetation on linear cays is often dominated by colonizing species. Although some linear vegetated cays support dunes that rise several meters above this level (e.g., Stapleton Island, Ingram Island: see Figs. 10.5 and 10.6), others are barely emergent at high tide. Oval-shaped islands tend to be more stable, especially along the flanks of the longer axis. Beachrock is able to develop within the beach on these shores, but massive outcrops (up to 30 m wide) are commonly exposed by subsequent erosion. Of the 17 vegetated cays mapped by Stoddart et al. (1978a) (representing 40% of cays of this type on the GBR), beachrock was absent from only one (Upolu Cay). Flood (1977) examined 14 vegetated cays in the Bunker and Capricorn Groups and similarly noted that extensive beachrock development was a common feature (e.g., Heron Island (Fig. 10.8); Wreck Island (Fig. 8.7b) (Davies and Kinsey, 1973)). These southern vegetated cays are generally much larger than those of the northern province, with a mean size three times that of the northern GBR islands (28 ha compared to 5.8 ha: Stoddart et al., 1978a). Moderate though not necessarily extended stability is required for beachrock development (see Section 3.4.2), and massive outcrops can confer additional stability to a cay under erosive conditions. However, as indicated above, beachrock outcrops that are now detached from cays are
10.2 Classification and geomorphology of reef islands
Figure 10.5 (a) Stapleton Island from the south, March 2005; (b) map of Stapleton Island constructed by Stoddart et al. (1978a) during the 1973 Royal Society and Universities of Queensland expedition.
333
334
Islands of the Great Barrier Reef
Figure 10.6 Ingram Island from the north, March 2005, with Beanley Islands in background. Note the more densely vegetated higher cay terrace near the center, and the lower, less densely vegetated terrace in the foreground.
Figure 10.7 Erskine Island, a vegetated cay in the Capricorn Group. See also Fig. 10.14.
10.2 Classification and geomorphology of reef islands
335
Figure 10.8 Massive beachrock outcrops on southern shoreline at Heron Island.
common, and some occur where no cay remains (e.g., Ellis Island). In contrast to the relatively stable lateral flanks, spits forming off the ends of these islands are characteristically mobile, and commonly shift from season to season (Figs. 10.7 and 10.15, Section 10.5). On some vegetated sand cays mineralization of phosphates derived from guano has cemented cay sediments beneath the vegetation cover to form phosphate rock (e.g., Raine Island (Fig. 9.2)), which also increases island stability (Section 10.4.3). (b) Vegetated mixed sand and shingle cays As the name suggests, mixed sand and shingle cays consist of a mixture of sand and shingle. Cays of this type are not common on the GBR, although it has been suggested some of the low wooded island cays may have developed from these islands (Hopley, 1982). Lady Musgrave Island at the southern end of the GBR is perhaps the best-known example, having been described by Steers (1937, 1938), Orme et al. (1974), and Flood (1977). The island has developed over the leeward reef flat (Lady Musgrave reef still has a shallow lagoon). The core of the island consists of shingle ridges, many cemented at the base, but the encircling beaches are sandy. In this regard it is similar to Lady Elliot Island which has been classified as a vegetated shingle island (see following) in the gazetteer (Hopley et al., 1989). Only minor changes were noted by Flood
336
Islands of the Great Barrier Reef
(1977) between his surveys of the island and those of Steers completed four decades earlier, suggesting reasonable stability over this period. Hopley (1982) proposed several ways in which mixed sand and shingle cays may develop. Where wind direction shifts seasonally shingle may be supplied during one season and sand during the other; this process may sensibly account for mixed sand and shingle cay formation in Torres Strait. Alternatively, shingle may be deposited during storms, with sand deposition during regular weather conditions. The latter mode appears applicable to Lady Musgrave Island, where shingle ridges have been episodically deposited by storm waves approaching from a more northerly direction than ambient waves, with sand accumulating around the margins in the interim. Mixed cays also develop where the character of sediments produced on the reef or delivered to the focal point of accumulation change through time, as may occur when relative sea levels change or reef geometry is modified. (c) Vegetated shingle cays Vegetated shingle cays are not common on the GBR (Hopley et al., 1989). One Tree Island (Fig. 8.8a), Lady Elliot Island, and East Hoskyn and Fairfax Islands (Fig. 8.8b), all located at the southern end of the GBR, are among the few cays of this type described. These cays typically develop near the windward margins of larger reef flats, or centrally on smaller ones exposed to high energy. As an example, Lady Elliot Island is located on the central to western side of a kidney-shaped reef platform, is approximately 0.54 km2 in area, and occupies about 20% of the reef platform. It is oval-shaped, with a slightly longer north-east–south-west axis and a small spit extending from the eastern shore. Chivas et al. (1986) described the geomorphology of Lady Elliot Island in detail. Lady Elliot Island was markedly disturbed by guano mining from 1863 to 1873, but near-concentric shingle ridges of coral, Tridacna, and phosphate rock clasts remain visible and dominate the interior of the island, which reaches an average height of 4.5 m above tidal datum. The older ridges are cemented, but those immediately behind the modern sandy beach remain unlithified. Beachrock on the modern beach is not well developed, and a higher-level beachrock is located on the easternmost point of the island. Most vegetated shingle cays are compact in form as linear shingle cays are usually either too narrow to retain adequate fresh water or too mobile for vegetation to establish permanently. Where vegetation survives, the greater stability of the shingle cay allows vegetation succession to proceed until cyclonic disturbance occurs (Hopley, 1982). Hopley (1982) suggested that shingle cays developed from shingle ramparts, with the ‘‘tongues’’ of shingle and rubble that commonly trail leeward from
10.2 Classification and geomorphology of reef islands
337
ramparts also possibly involved. A convincing case can be presented that One Tree Island formed in this way; the island appears to be anchored around a conglomerate platform presumed to have formed by the cementation of a shingle and rubble tongue, to which a series of crescentic very coarse rubble and shingle ridges have been added. The near-concentric shingle ridges at Lady Elliot Island similarly document its formation by the progradation of shingle ridges deposited during episodic storms (Chivas et al., 1986) (Section 10.4). (d) Mangrove islands Mangrove islands commonly develop on reefs lacking windward ramparts in the Caribbean (Stoddart and Steers, 1977), but are rare on the GBR. Murdoch Island (148 370 S) is the only true mangrove island within the GBR, with another four identified in Torres Strait (Hopley, 1997b). Aerial photographs show that mangroves, probably Rhizophora stylosa, densely cover most of the platform at Murdoch Island to within 100 m of the windward margin. Mangroves can only establish in sheltered locations without windward protection; high reef flats in areas of low to moderate tidal range provide suitable conditions. Steers (1938) examined Night Island (138 120 S, 1438 350 E), which has a sandy cay at its rear and extensive mangrove colonization of the reef top to windward. However, Steers noted (p. 122) that ‘‘a marked feature of this island is the absence of shingle ridges on the weather side’’ and that ‘‘the lower platform is well developed but far from being continuous.’’ He subsequently concluded that Night Island represented a form intermediate between mangrove and low wooded islands. 10.2.3 Multiple islands Two long-standing examples of multiple vegetated islands on a single reef occur in the GBR – Fairfax and Hoskyn Islands in the Bunker Group. Both include both a vegetated shingle and vegetated sand cay on a single reef platform (Fig. 8.8b). In the tally of reef island types presented in Table 5.8 each island is counted separately. The development in these cases is caused by sediment (shingle or sand) sorting by wave-generated currents, which are competent to transport sand to the rear of the platform but cannot move shingle far from the reef crest. Both occur on reef platforms described as ‘‘closed-ring reefs’’ by Maxwell (1968), with a continuous reef rim enclosing a partially infilled lagoon. Masig–Kodall (Yorke Island) in Torres Strait (98 450 S, 1438 430 E) is another multiple island, albeit with some differences from those of the GBR. Although detailed sedimentological investigations are incomplete, both cays on
338
Islands of the Great Barrier Reef
Masig–Kodall are sandy, and it is more probable that seasonally reversed waves interacting with an unusually shaped reef have led to the development of two islands here. These islands are deposited on a planar reef top and are connected intermittently by a mobile spit. 10.2.4 Complex low wooded islands The range of complex low wooded islands of the inner reefs north of Cairns is unique to the GBR, and has long been the focus of scientific attention (Section 10.1.3). Indeed both the 1929 and 1973 northern GBR expeditions focused on low wooded islands, and thus many excellent descriptions exist which should be consulted for detail (e.g., Steers, 1929, 1937, 1938; Spender, 1930; McLean et al., 1978; Stoddart et al., 1978a, b). Published details for these and other low wooded islands are summarized in Table 10.2c. The majority of ‘‘classic’’ low wooded islands on the GBR occur on relatively small reefs (e.g., Low Wooded – 87.9 ha; see Table 10.2c), but several on larger reefs are included amongst the 34 low wooded islands examined by Stoddart et al. (1978a) (e.g., West Hope Island – 315 ha; see Table 10.2c, Fig. 8.3b). As indicated earlier, at their simplest these islands comprise a windward shingle cay, a leeward sand cay, and significant mangrove development over the reef top (Hopley, 1997b). This combination of features may occupy a great proportion of the reef top, usually between 25% and 50%, but reaching a maximum of 79.5% at Bewick Island (Stoddart et al., 1978a) (Figs. 5.3, 8.3a, and 10.9). The windward reef on most complex low wooded islands is zoned, with living coral, boulder, algal pavement, and rubble zones typically observed (Stoddart et al., 1978a, 1978b; Hopley, 1982) (Figs. 10.9 and 10.10). The algal pavement slopes upward from the living coral rim, and is often covered with turf algae. Algal ridges of the type developed elsewhere in the Pacific (e.g., Nunn, 1993) are not observed, but shallow algal-rimmed pools were reported on several low wooded islands by Steers (1937, 1938) and Stoddart et al. (1978b). Coral shingle ramparts that grade upward from the reef flat and have steep leeward faces are encountered at varying distances from the windward margin (Fig. 10.11a). The windward boundary frequently parallels the reef edge, but shingle tongues commonly trail from the rampart crest toward the reef interior. Ramparts may be substantial features; Stoddart et al. (1978b) reported widths of 65 m, and a maximum elevation of 3.12 m above tidal datum was measured at Watson Island. Conglomerate promenades of cemented ramparts with one or more planated surfaces form on the seaward margins of many shingle islands, where scarping reveals an internal structure consistent
10.2 Classification and geomorphology of reef islands
Figure 10.9 Geomorphology of Bewick Island (after Stoddart et al., 1978a). Locations of shingle rampart and bassett edges depicted in Figs. 10.11a and 10.11b are shown.
339
340
Islands of the Great Barrier Reef
Figure 10.10 Hannah Island, a low wooded island with a sand cay (top left of photograph) completely encircled by mangroves.
with landward rampart migration over moated reef flat communities (Scoffin and Stoddart, 1978; Stoddart et al., 1978a). On the shingle islands of many low wooded islands a lower promenade occurs near MHWS level (e.g., Bewick, Three Isles, Turtle I: Spender, 1930; Steers, 1937; McLean et al., 1978; Stoddart et al., 1978a) (Fig. 10.11b); an upper platform up to 3.5 m above tidal datum is also developed on several (e.g., Coquet). Promenade widths vary from 10 to 70 m, but average around 30 m. Shingle ridges are deposited over these platforms; for example, two or three separate ridges reaching 4.5 m above tidal datum occur on the shingle islands at West Hope and Low Wooded Island (Stoddart et al., 1978a). Mangrove swamps of varying size lie behind the shingle islands and ramparts. The extent of mangrove colonization was argued to reflect the stage of island development within an evolutionary sequence by some workers (Steers, 1937; Fairbridge and Teichert, 1947), but others contended that it is more reliably a function of the protection afforded by the windward structures (Stoddart, 1965; Stoddart et al., 1978c). Organic mud around 2 m deep is deposited beneath the mangroves on some islands (e.g., Bewick Island) (Figs. 8.3a and 10.9), but
10.2 Classification and geomorphology of reef islands
Figure 10.11 (a) Shingle rampart on south-facing reef flat on Bewick Island. Note steep leeward face and gentle grading from lower seaward reef flat (see Fig. 10.9 for location). (b) Bassett edges and lower platform on south-facing reef flat, Bewick Island (see Fig. 10.9 for location).
341
342
Islands of the Great Barrier Reef
elsewhere mangroves extend over sandy reef flats (e.g., Ingram–Beanley). Fields of fossil microatolls approximately half a meter above modern moated corals are exposed on several islands (e.g., Houghton Island, Bewick Island), documenting the higher mid-Holocene sea level and later regression experienced by the reef platforms on which many low wooded islands have formed (McLean et al., 1978; Chappell et al., 1983) (see Section 3.4.3). The leeward sandy cays are morphologically diverse, possibly more so than ordinary sand cays (Stoddart et al., 1978a). Some are small, ephemeral, and unvegetated (e.g., Watson, Binstead) but larger vegetated cays replete with terraced morphology also occur (e.g., Ingram Island, Bewick Island). Stoddart et al. (1978c) noted that massive beachrock outcropped at the shoreline where the lower terrace was absent, but was only patchily developed where the lower terrace grades gently to the reef flat. The characteristic complexity of low wooded islands ensures that they are a heterogenous group. Stoddart et al. (1978a) distinguished four low wooded island types: (1) Low wooded island with limited reef top mangroves and a separate sand cay. This is the most common type, with a large reef top, well-separated windward shingle and leeward sand cay, and limited mangrove colonization. The sand cay can be relatively large (e.g., Ingram Island (Fig. 10.6)), or very small and rudimentary (e.g., Watson Island, Turtle IV). (2) Low wooded island with reef top mangroves extending between windward shingle and leeward sand cays. Bewick Island is the type-example, with continuous mangrove forest across the reef top (Fig. 8.3a and 10.9). Variations of this general pattern exist; for example, on Newton Island mangroves are well developed between the sand and shingle cays, but a large enclosed lagoon remains near the center. On more elongate reefs such as Coquet, the cay and shingle islands can be linked by a belt of shingle ramparts or conglomerate, and mangroves are confined to protected areas behind these features (Stoddart et al., 1978a). (3) ‘‘Turtle-type islands’’ that have all low wooded island features except the central reef flat. Shingle ramparts and conglomerates on these islands extend to the leeward cays, which are typically dominated by coarser sediments. Turtle-type islands are generally restricted to small reef platforms (< 60 ha) where they occupy a large proportion of the reef flat. This type of low wooded island may also enclose an open or partially filled lagoon (e.g., Nymph, Turtle II). (4) Low wooded islands that do not fit the other three categories. This category includes several of the islands mapped by Stoddart and colleagues that could not be assigned to the groups above. Hannah Island (Fig. 10.10), a sand cay completely encircled by mangroves (Steers, 1937), provides an example.
Radiocarbon dates published by the 1973 expedition provided a chronology for reef flat formation and island accumulation for the complex low wooded
10.3 Island distribution
343
islands (McLean and Stoddart, 1978; Polach et al., 1978; Stoddart et al., 1978c). The dates suggest that: (i) many higher features (e.g., high cay terrace, high beachrock, high promenade, elevated fossil microatolls) formed under higher sea-level conditions that persisted until around 3000 years ago; (ii) the timing and order of shingle and sand cay development varied amongst the low wooded islands examined; and (iii) many low wooded islands have been largely in place for several thousand years (Section 10.4). Importantly, the antiquity established for these islands by radiocarbon dating indicates stability in the longer term, but medium- and shorter-term change is also recognized. Comparison of the detailed maps of Low Isles and Three Isles produced in 1928–29 by Spender (1930) with those constructed 44 years later by Stoddart et al. (1978b) unmistakably show modification of ramparts, mangroves, and cays (Section 10.5). Significant modifications on these islands normally occur during cyclones, although heavier south-easterly weather can produce minor morphological change (Moorehouse, 1933). Conversely, some intense cyclones effect little change; a survey of Bewick and Ingram Islands by a party including two of the authors two months after Severe Tropical Cyclone Ingrid (April 2005) revealed negligible change. 10.3 Island distribution The distribution of islands on the GBR has been briefly covered in Chapter 5 (Section 5.4), where it was established that neither high nor reef islands are distributed evenly, either latitudinally (Fig. 5.8) or across the shelf. More than 50% of the 617 high islands occur between 208 and 228 S. Reef islands are also distributed unevenly, with most (72.7%) associated with planar reefs (Table 5.3) that are concentrated north of 168 S and between 218 and 228 S. As discussed in Chapter 8, planar reefs usually, but not always, develop on relatively small and shallow Pleistocene foundations (Table 8.3), especially where hydro-isostatic uplift has accelerated the rate at which reefs approached sea level (Chapter 3). The relationship between reef growth and reef island formation and distribution is examined in Section 10.4. Clearly, however, reef islands are not restricted to planar reefs; they may develop on all reefs where a platform near sea level and sufficient protection for sediments to accumulate exist (Table 5.3). The distribution of the major reef island types within the GBR is shown in Fig. 10.12; the distribution of reef types is depicted in Fig. 5.10. More than 70% of reef islands classified in the GBR gazetteer (see Chapter 5) are unvegetated cays (many were classified from satellite imagery and aerial photographs, and thus composition was not differentiated), the characteristics of which are described in Section 10.2.1. Most unvegetated
344
Islands of the Great Barrier Reef
Figure 10.12 Distribution of (a) unvegetated sand and shingle cays; (b) vegetated sand and shingle cays and mangrove islands; and (c) low wooded islands and multiple islands (from Hopley, 1982).
cays occur north of Cairns and into Torres Strait. They are also particularly common on the Swain Reefs between 218 and 228 S. A few unvegetated cays are also located on the outer reefs (e.g., Sand Bank Numbers 7 and 8 (138 25.680 S and 138 22.080 S respectively), but normally only where some protection is afforded and accumulation can progress (Hopley, 1982). Perhaps more interesting than the otherwise broad distribution of unvegetated cays is their
10.3 Island distribution
345
complete absence between Wheeler Reef off Townsville south to the northern Pompey Reefs (a distance of 315 km) (Figs. 10.12a). This absence is presumed to reflect the Holocene sea-level history (Figs. 3.8 and 3.9), greater distance of reefs offshore to east of the zero hydro-isostatic isobase (see Fig. 3.9 and Section 3.4.3), high tidal range (Fig. 4.4d), and greater exposure to both normal and cyclonic wave energy on this section of the Reef (Hopley, 1982; Puotinen, 2004) (see Figs. 4.9 and 4.12 and Section 10.4). Although fewer in number, the distribution of vegetated cays, not surprisingly, is similar to that of unvegetated cays. Most occur over planar reefs on the northern GBR, with a second collection of mostly larger vegetated cays located in the Bunker–Capricorn Group (Fig. 10.12b). Vegetated cays also occur in the Swain Reefs and inner reefs in this area (Bell Cay, Bushy Island), but are entirely unrepresented through the central GBR. There are no vegetated cays between Bushy Island (208 570 S, 1508 050 E), located approximately 70 km east of Mackay, and Green Island (168 460 S, 1488 580 E) near Cairns, a distance of more than 600 km. Several vegetated cays are located on the northern outer barrier (Tydeman Cay, an unnamed cay on a reef at 138 220 S, Moulter Cay, and Raine Island), but they are lacking elsewhere. Murdoch Island and possibly Night Island (discussed in Section 10.2.2d) are the only mangrove islands on the GBR, although several are known from Torres Strait. Mangrove islands are restricted to sheltered inshore areas of low to moderate tidal range, where mangroves can establish without windward protection. Multiple islands are also rare. The Fairfax (Fig. 8.8b) and Hoskyn Islands in the Bunker Group, and Masig–Kodall (Yorke) Island in Torres Strait are the only known examples in the region. The distribution of complex low wooded islands is strongly associated with the location of smaller planar reefs with reef platforms of relatively high elevation, accounting for their concentration on the inner shelf north of Cairns (reef platform elevations decrease offshore on the northern GBR). All lie west (inside) the zero hydro-isostastic isobase and have experienced uplift since the transgression (Chappell et al., 1983; Hopley, 1983b). On the GBR, all low wooded islands occur between 118 and 168 230 S, and 94% are within 20 km of the mainland. The distribution of low wooded islands is thus restricted, although Warrior Islet in Torres Strait may be classified as a ‘‘turtletype’’ low wooded island. Neil (2000, p. 292) suggests that Green Island, in Moreton Bay near Brisbane, ‘‘is a low wooded island similar to those of the northern Great Barrier Reef and exhibiting the morphological complexity which sets the north Queensland low wooded islands apart from similar landforms elsewhere in the world.’’ If correct, this Green Island represents a unique subtropical example of a complex wooded island.
346
Islands of the Great Barrier Reef
10.4 Reef island formation The basic processes of wave refraction around and over reefs and the generation of opposing wave trains that converge at a focal point where sediments are deposited and reef islands may form are covered earlier (Section 4.4.2; Fig. 4.11), and are well documented elsewhere (Gourlay, 1983a, 1988). The longer-term aspects of reef island formation are examined here. Darwin (1842) established a paradigm of evolutionary growth and morphological change on coral reefs that has been widely accepted and subsequently applied at various scales. In Chapter 8 the evolutionary classification of GBR reefs developed by Hopley (1982) was tested with a substantive chronostratigraphic dataset and found to be robust. Early researchers also viewed the diversity of reef islands on the northern GBR as part of a developmental progression, ranging from an initial rudimentary unvegetated sand cay through to a ‘‘mature’’ fully developed low wooded island, the perceived climax condition (Spender, 1930; Steers, 1937; Fairbridge and Teichert, 1948; Stoddart et al., 1978c). Clearly parameters other than time also influence whether a reef island will develop on any given reef, and the probability or rate at which an island might progress through the evolutionary sequence. These ‘‘other’’ parameters principally relate to the amount and distribution of energy over a reef platform; they are examined in detail in Section 10.5. 10.4.1 Reef growth, relative sea-level change, and reef islands As previously discussed, stable vegetated reef islands are strongly associated with planar reefs (Table 5.8). With just four exceptions, only relatively unstable unvegetated cays form on reefs yet to reach planar stage. In Hopley’s (1982) evolutionary classification, planar reefs represent a senile stage approaching the end of their growth trajectories. Significantly, most reef islands with any degree of permanence and stability form at this senile stage, when lagoons are largely infilled and sediment-covered reef flats have developed. Thus at the geological timescale the factors that promote the advance of reefs through the reef growth sequence are important drivers of reef island formation. The principal factors have been identified as reef platform size and depth, and relative sea-level history, with the influence of energy factors generally of secondary (in a sequential sense) importance. Generally, reefs that grow from shallow foundations have reached sea level first on the GBR (Chapter 11), and smaller reefs have filled their lagoons sooner than larger ones; few reefs smaller than 1.75 km2 possess lagoons (Hopley, 1982). Shallow pre-Holocene foundations (mean 10.8 m: Table 8.3)
10.4 Reef island formation
347
and small reef size (mean 4.1 km2: Table 5.4) have been established for planar reefs elsewhere in this book (Chapters 5 and 8). Not surprisingly, planar reefs also have the oldest mean reef top age (6022 years, excluding Wheeler Reef: Table 8.3), so that where reef geometry and hydrodynamic conditions concentrate deposition the accumulation of excess sediment as reef islands could have commenced 6000 years ago. The provision of a foundation, usually a reef flat, at or near to sea level is a fundamental condition for reef-island formation, although it need not be extensive (Hopley, 1997b) nor sea-level constrained (Kench et al. 2005b). The relationship between reef flats and sea level on the GBR is complicated by hydro-isostatic adjustment of the shelf to the transgression (Chapter 3, Fig. 3.9), producing varying relative sea-level histories both along and across the shelf. Elevated fossil microatolls and other features have been dated from many of the low wooded islands on the inner shelf north of Cairns, and indicate that sea level was attained there very soon after the mid-Holocene highstand was reached (McLean et al., 1978; Scoffin and Stoddart, 1978; Chappell et al., 1983). Elsewhere, especially on the mid to outer central GBR, reefs reached sea level much later (Chapter 3, Fig. 3.9), where the reef is well east of the zero isobase. Although it is unlikely that reefs within a particular area would reach sea level and begin to form reef flats simultaneously, broad spatial patterns in reef top age can be recognized (Fig. 3.8), and may introduce geographic differences in the duration of reef island formation and possibly, as envisaged by Fairbridge and Teichert (1948), morphological maturity. Spender (1930) noted the concentration of ‘‘mature’’ low wooded islands close to the mainland on the northern GBR and suggested that systematic differences in reef top elevation caused by tilting of the shelf was responsible. Steers (1937) considered that the apparent better development of reef islands closer inshore was simply due to the outer shelf being more exposed to wave energy. For nearly five decades this view was largely accepted (Stoddart, 1965), but few would now discount the importance of hydro-isostatic upwarping of the inner shelf (Chappell et al., 1983; Hopley, 1983b) in accelerating the development of reefs to planar stage, as a consequence improving reef island accumulation and, possibly, preservation upon the emergent reef platforms. 10.4.2 Chronologies of reef island development Somewhat surprisingly given their scientific and general prominence, there have been relatively few attempts to resolve the ages of GBR reef islands. More than 30 years on, the radiocarbon chronologies established for the northern
348
Islands of the Great Barrier Reef
GBR reef islands and platforms by the 1973 expedition remain the most detailed and comprehensive dataset published. Radiocarbon ages of emergent fossil microatolls, elevated beachrocks, the upper promenade, and the upper cay terrace found on many low wooded islands suggests their emplacement at a higher sea level around 1 m above the present from around 6000–3000 years BP (McLean et al., 1978). Although the details vary between different low wooded islands, ages for the storm ramparts that now form the upper promenade cluster between 3600 and 3300 years BP, the upper cay terrace sediments range between 4380 and 3020 years BP, and the higher beachrock formed between 4380 to 2030 years BP (Polach et al., 1978; Scoffin and Stoddart, 1978; Stoddart et al., 1978c). Radiocarbon dating constituent sediments provides a maximum, not necessarily an accurate age for a depositional feature (see discussion of Warraber Island below). Nonetheless, the radiocarbon ages above suggest that the main geomorphic elements of many low wooded islands were largely in place by 3000 years ago (Scoffin et al., 1978). Reef island development continued as relative sea level fell and some platforms emerged; radiocarbon ages suggest that the lower cay terrace (3280–2190 years BP), lower promenade (1480–380 years BP), shingle ramparts and unlithified younger ridges (1500–510 years BP) were all deposited more recently (Polach et al., 1978). McLean and Stoddart (1978) suggested that the shingle islands in particular were considerably enlarged in the last 1500 years. The elevation differences between the older features and their younger equivalents are consistent with the magnitude of sea-level fall suggested for this part of the shelf (Chappell, 1983; Hopley, 1983b) (see Section 3.4.3), although the absolute heights for the cay terraces are problematic in this regard (Buckley, 1988). Similarly, the patchy occurrence of both the upper and lower promenades on otherwise similar low wooded islands confounds their sea-level significance (Stoddart et al., 1978c). Mangrove cover of varying extent occurs on these islands, but organic mud overlying reef top rubble is dated at 2210 170 years BP at Turtle I Island, indicating that mangroves were well established there by then. The ages established for the various features at different elevations appear sensible in the context of the relative sea-level history and the chronological sequence of deposition. However, as noted, radiocarbon dating bulk sediment samples as available in 1973 can occasionally yield a misleading indication of deposit age, as illustrated at Low and Three Isles where beach ridge sediments verified as deposited between 1936 and 1973 yielded radiocarbon ages exceeding 2000 years BP (Stoddart et al., 1978b). The age of bulk sands on the northern GBR often exceeds 2000 years (McLean and Stoddart, 1978); it is not known whether this is a function of sample bias or sediment supply. This problem may now be addressed by
10.4 Reef island formation
349
accelerator mass spectrometer (AMS) radiocarbon dating specific constituents (preferably those where organism age and deposit age are tightly coupled), which requires very small sample mass for high precision. The occurrence of an upper and lower terrace on many of the vegetated sand cays on the GBR suggests that they also experienced at least two phases of accumulation. Published radiocarbon ages are available from the leeward vegetated sand cays of the low wooded islands discussed above, and the chronology is as already described. Vegetated sand cays elsewhere on the GBR exhibit comparable terraces, but ages have not been determined. However, several dates are available for Raine Island, a large vegetated sand cay on the leeward end of Raine Reef, a small detached reef off the northern outer barrier (Fig. 9.2). Cemented coral from just below the level of the present reef flat is dated at 4780 years BP (see Chapter 9, Fig. 9.3) and dates from Tridacna shells within the phosphatized cay range between 1640 and 1040 years BP (Polach et al., 1978; Limpus, 1987). The dates suggest island formation began after 4870 years BP and the beachrock outcrops and phosphate cap have developed in the last 1600 years. Both the development of the reef top and onset of island formation at Raine Island occurred later than on the low wooded islands further inshore. A recent program of AMS dating of specific reef island sediment components on Warraber Island, a vegetated sand cay in Torres Strait, has demonstrated that varying chronologies can be derived by dating different biogenic constituent sediments. Dates from bulk sand suggest that Warraber has a growth chronology similar to many on the GBR, with an initial accumulation phase around 3500 years BP followed by further accretion until 2300 years BP. However, dates from foraminiferans suggest an earlier onset (5290 140) and shutdown (4780 180) of island development, and dates from mollusks which graze on the reef flat and are thus actively being produced suggest the island initiated around 3000 years ago and has built out constantly since (C. Woodroffe, pers. comm.). These variations clearly indicate the importance of selecting, where possible, material that is being produced as sand contemporaneously with island accumulation to develop the most precise and accurate chronologies. Radiocarbon dates are available for two of the three vegetated shingle cays on the GBR. At One Tree Island a radiocarbon age of around 4000 years is available from the cemented cay foundations (Davies and Marshall, 1979). This is marginally older than most of the shingle ridges dated on the low wooded islands (discussed above), indicating considerable longevity and stability. Similar antiquity has been established at Lady Elliot Island, the southernmost reef island on the GBR. The shingle cay there consists of a concentric series of shingle ridges deposited over a shallow reef platform developed by
350
Islands of the Great Barrier Reef
6500 years BP (Flood et al., 1979). Radiocarbon ages of large Tridacna shells within the ridges (large shells were selected to eliminate the pitfalls of bulk sampling and to minimize the possible effects of reworking) suggest island formation began 3200 years BP, and progradation at relatively uniform rates of 60 m ka 1 to leeward and 90 m ka 1 to windward have occurred since (Chivas et al., 1986). Similarly consistent rates of shingle ridge accretion have been reported from elsewhere on the GBR, for example at Curacoa Island where the regular tempo of ridge emplacement has been used to infer a relatively uniform storm regime since the mid Holocene (Hayne and Chappell, 2001; Nott and Hayne, 2001). Unvegetated cays are the most numerous reef islands on the GBR. They are also generally the youngest, although as discussed previously, establishing cay age by dating sediments is problematic. This is exemplified by the 2330 70 years age of coarse sediments collected from the top of an unvegetated sand bank at Pickersgill Cay (Polach et al., 1978), and the age of the recently deposited lower cay terraces at Low Isles and Three Isles (Stoddart et al., 1978c). Sand cays exchange sediments with the surrounding reef flat, contributing sediments to the reef flat during erosion phases and receiving sediments during periods of accretion (Section 10.5). Unvegetated cays are especially dynamic, and will inevitably integrate sediments of varying age as they progress through repeated phases of erosion and accumulation. 10.4.3 Factors affecting reef island development and chronologies The age structure suggested by the available dataset for GBR reef islands generally complies with Fairbridge and Teichert’s (1948) thesis of an evolutionary sequence, but dating problems and variations in physical conditions between reefs limit the resolution with which morphological development can be viewed simply as a function of time (Stoddart et al., 1978c). For example, the temporal sequence of rampart, sand cay, and mangrove development varies on different low wooded islands (Stoddart et al., 1978c; Hopley, 1982), and the extent of mangrove colonization has been shown to vary in response to storms and substrate availability and cannot be reliably calibrated with time (Stoddart et al., 1978b). Nonetheless, several main factors are inferred as important controls on the nature and timing of reef islands formation, and these are briefly addressed below. (a) Sediment supply and delivery The high central core on many leeward sand cays accumulated rapidly around 3500–3000 years ago, with a second later phase on some islands. If
10.4 Reef island formation
351
the radiocarbon chronology is correct, rapid sand production, delivery, and deposition to leeward occurred on many planar reefs between 4000 and 3000 years ago, with only relatively minor additions and modifications since. The greater antiquity and stability of this higher terrace is commonly indicated by greater soil development and the presence of mature vegetation (e.g., Masthead Island (Fig. 13.10b); Douglas Island (Hekel, 1979)). Changes in sea level, hydrodynamic energy, and in the production and delivery of sediments are all potential explanations (Stoddart et al., 1978c), and may all be involved at any location to varying extents. Chappell (1983) suggested that relative sea-level fall since the mid-Holocene highstand was smooth rather than stepped, including during the inferred major periods of reef island accumulation, and this is clearly difficult to reconcile with the formation of two distinct terraces. Buckley (1988) considered that pulsed sediment production and delivery was a more tenable explanation, speculating that higher cay terraces formed from a pulse of sediment produced when falling sea levels met the upward-growing framework and erosion increased. Progressive sea-level fall may also have affected sediment transport efficiency over the reef platform, reducing depth and possibly turning down or off sediment delivery to the leeward cays. Kench and Brander (2006) recently quantified a non-dimensional reef energy window index ( ) that incorporates the effect of reef flat water depth and reef width, which they demonstrated provides a physically and statistically meaningful descriptor of geomorphic processes on reef platforms. Application of this index suggests that as reef platforms shallowed during the late-Holocene regression, accompanied on some reef platforms by an increase in the width of shallow areas, would have declined and sediment transport efficiency and island accumulation rates would have decreased. Although shingle ramparts document the episodic occurrence of cyclones on many reef platforms that support terraced leeward cays, ‘‘peaks of storminess’’ (Stoddart et al., 1978c) cannot easily be linked to terrace formation as shingle ridge sequences from the northern and southern GBR show the storm regimes have not changed since the mid Holocene (Hayne and Chappell, 2001; Nott and Hayne, 2001). The progressive accrual of windward ramparts and shingle ridges would, however, modify wave refraction patterns and disrupt sediment delivery to leeward cays (Stoddart et al., 1978c), and as windward ramparts, mangroves, and cays occupy an increasing proportion of a reef platform they shut down production areas and reduce sediment production (McLean and Stoddart, 1978). How either of these mechanisms would produce the two widely observed distinctive terraces is uncertain.
352
Islands of the Great Barrier Reef
(b) Reef platform elevation The occurrence of 75 unvegetated cays on platforms close to sea level is convincing evidence that emergent reef platforms are not mandatory for reef islands to form as inferred by Spender (1930), but it is likely that higher reef tops confer a stability less often achieved where they are deeper (Kench and Brander, 2006). Nonetheless, just 15% of planar reefs (83 of 544) support vegetated or low wooded island reef islands, and approximately 60% of the GBR’s planar reefs lack reef islands. Although these numbers may vary over short timescales as islands form and disappear, stable reef islands will not form even where a suitable substrate occurs in some locations because other conditions work against their development (see following). (c) Reef geometry and energy exposure The size and shape of reef platforms are important parameters influencing reef island development potential. Cays do not generally form on larger reefs or those with a geometry that impedes the development of centripetal wave refraction, where sediment deposition is dispersed and islands do not accumulate. The formation of stable reef islands in areas of dense reef network is similarly constrained, as low ambient wave energy cannot concentrate the sediments needed to construct larger cays, but episodic storms approaching from variable directions can rapidly redistribute them (Gourlay, 1988). Recovery to a stable condition can be prolonged in these settings, as documented by Flood and Heatwole (1986) for small, grassed cays in the Swain Reefs disturbed by cyclones. In more exposed areas such as on the outer barrier, energy levels, both ambient and extreme, may be too high for accumulated sediments to be stable, except where protection is provided, such as occurs at Raine Island (Hopley, 1982) (Figs. 9.2 and 9.3). The advantages of small platforms for reef island formation have been addressed in detail earlier, but where reefs exposed to large waves are narrower than the surf zone it is likely that sediments will be washed off the platform and not contribute to island formation (Gourlay, 1988). (d) Reef island size Exceptions can be found, but smaller cays tend to be less stable than larger ones. Gourlay (1988) calculated that for a cay to develop on a reef on the GBR where the tidal range is 2 m, the accumulated sediment would be need to be at least 60 m wide, which is approximately twice the width of a stable beach composed of cay sands in this tidal range. Of course, vegetation is indicative of some stability and also may improve or maintain stability in various ways.
10.5 Reef island dynamics
353
Interestingly, the size of reef islands on the GBR lacking vegetation is large compared to other reef areas. For example, Waterwitch Cay has an area of 2.8 ha and was unvegetated when visited in 1973, but on the Belize barrier reef all cays larger than 0.1 ha are vegetated, and at Kapingamarangi Atoll all islets greater than 0.01 ha support terrestrial plants (Stoddart and Fosberg, 1991). It is generally presumed that this difference is a function of the greater instability of reef islands on the GBR due to higher tidal ranges and exposure to cyclonic activity. These constraints are likely to be particularly limiting on the central GBR, where both parameters are regionally high (Chapter 4). (e) Lithification: cementation and induration Many stable reef islands are partially lithified. This is often viewed as indicative of sufficient stability for the lithification process to occur (Steers, 1929, 1937; Stoddart et al., 1978a), and a process that improves stability in the longer term (Scoffin et al., 1978) (Section 10.5). Beachrock (Section 3.4.2d) is common on GBR reef islands, often forming a lithified shoreline resistant to erosion. However, even large islands, including those at least partly stabilized by massive beachrock outcrops, can be destabilized and shift over the reef flat. For example, massive beachrock suggests previous periods of at least moderate stability on Waterwitch Cay, but at present it is undoubtedly mobile, evidenced by its barren state and beachrock outcrops disjunct from the existing cay. In contrast to beachrock which only forms within the shoreface, phosphate rock (formed by the downward mineralization of phosphate solutions derived from guano) usually develops within supratidal reef island interiors. Where phosphate rock is well developed, such as at Raine Island, the island core becomes indurated and prospects for enduring stability are presumably improved. 10.5 Reef island dynamics Where sediment is available, reef island formation is largely controlled by the competency of waves to transport the sediment and the development of a centripetal wave refraction pattern to concentrate its deposition. Under optimal conditions, a reef island will continue to increase in size until it achieves equilibrium, where sediment delivery and removal are balanced (Stoddart and Steers, 1977; Gourlay, 1988). Reef islands are dynamic systems, with their morphology and location sensitive to changes in the intensity and/or direction of wave energy on a reef platform, associated with infrequent extreme events (Section 4.5) or normal variability in ambient conditions (Flood, 1984a, 1986). The latter include minor shifts in predominant wind direction from year to year through to seasonal wind reversals during the
354
Islands of the Great Barrier Reef
northern monsoons. Gourlay (1988) provides an excellent summary of the influence of various climatic fluctuations on GBR reef island development and dynamics. Island responses vary from total obliteration, shifts in size, shape, and position, or even seasonal switches between predictable morphological states. Changes in sediment type, supply, or erosion may also modify reef island morphology (Stoddart et al., 1978c; Yamano et al., 2000). Almost all descriptions of unvegetated cays on the GBR note that they can rapidly and markedly change shape, size, elevation, and position (e.g., Taylor, 1924; Hopley, 1978c, 1982; Aston, 1995). Unvegetated sand cays typically accumulate near the leeward reef margin and shingle cays close to the windward reef edge, reflecting the competency of wave-generated currents to carry sand or shingle away from the reef edge source zone (Fig. 4.11). Unvegetated shingle cays are located in more exposed locations and are themselves often the product of cyclonic storms, but cemented structures such as bassett edges and coarse conglomerates suggest they are more stable than their sandy counterparts. Indeed Steers (1929) could not find beachrock on any unvegetated sand cay, leading him to speculate that vegetation was involved in beachrock formation. Small beachrock and conglomerate outcrops were recorded on only five of 19 unvegetated cays visited by the 1973 expedition to the northern GBR (Stoddart et al., 1978a), with none documented from the smaller ephemeral cays such as Undine (area 4610 m2) and North Pickersgill Cay (area 3800 m2). Compact unvegetated sand cays are especially ephemeral, with impressive mobility documented in several locations. An early account was provided by Taylor (1924) for Beaver Cay, a small cay approximately 180 m long by 90 m wide which migrated 90 m to the south-east during a fortnight of strong north-westerly winds in 1922. Nearby ‘‘Cay’’ cay responded in a similar way. ‘‘Severe’’ erosion was also reported at Beaver Cay during Cyclone Winifred (Blackman et al., 1986). Several examples exist of cays completely disappearing in historical time, with the example of a cay on Pixie Reef, near Cairns, previously reported by Hopley (1982). This cay was 45 m in diameter in 1928, but had transformed into two separate cays 65 m apart by 1929, and in the 1980s not even an ephemeral sandbank was located on the reef platform. Hekel (1979) reported the disappearance of Ellis Island, a cay mapped on nautical charts, whose former presence was indicated in 1979 only by remnant beachrock outcrops. What remains undoubtedly the most detailed account of compact unvegetated cay behavior on the GBR is that of Hopley (1978c) for Wheeler Cay, a normally oval cay approximately 80 50 m in size located over a more extensive area of sanded reef flat (170 250 m), the boundaries of which
10.5 Reef island dynamics
355
document the range of cay migration (Figs. 8.4b and 10.13, Table 13.3). Wheeler Cay is typical of many compact unvegetated cays on the GBR. During the eight years between 1969 and 1977, which included Tropical Cyclone Althea in 1971, it varied considerably in shape, size, and location, moving over an estimated 11 000 m2 of sanded reef flat (Hopley, 1978c). Wheeler Cay’s highest point migrated as much as 13 m in a day that included a 3-hour storm with gusts to 30 ms 1, but the whole cay moved 6 m in 24 hours under light winds of less than 5 ms 1. A later examination of Wheeler Cay by Aston (1995) determined that it had moved and changed shape again by 1980, but was still confined within the original sanded reef flat area. Aston (1995) statistically confirmed the long-assumed relationship between increased vegetated area and decreased cay mobility; various studies demonstrate that the most significant morphological changes on vegetated reef islands normally take the form of swings in the orientation of spits formed at their ends (Fig. 10.14). Gourlay (1983a) suggested that these swings are more vigorous on smaller vegetated cays than larger ones. Flood (1986) clearly showed concordant behavior on different vegetated cays in the Bunker–Capricorn group, most particularly shifts in the orientation of western-end spits that were well correlated with shifts in predominant wind direction between 1960 and 1982. Erskine Island, one of the islands studied by Flood, was also mapped by Steers in 1936, at which time the western end was oriented slightly toward the south-west, intermediate between the two extreme positions mapped during Flood’s 22-year period. The possibility that cay mobility may be related to cyclic weather patterns as proposed for mainland coastal erosion by Thom and Roy (1983) was raised (Flood, 1986), but analysis of longer-term records detected no significant cyclic trend (Flood, 1988). Although vegetated cays are relatively stable, comparison of early and more recent maps of vegetated islands from different locations demonstrates that they remain dynamic and undergo continual change. Within the sample of ten vegetated cays for which maps were prepared by Steers (1938), only one had remained apparently unchanged, four had eroded and decreased in size, four had increased in size, and one had migrated to a new position on the reef platform. Monitoring records extending from 1936 also exist for Green Island near Cairns (Fig. 13.6), a relatively large vegetated sand cay where considerable change in island morphology is also documented (Kuchler, 1978; Beach Protection Authority, 1989). At Green Island the cay has changed shape since the 1930s, with the most significant change being the development and enlargement of a north-west spit since the mid 1970s. Coincident with these conspicuous changes, erosion has occurred elsewhere on the island, and various
356
Islands of the Great Barrier Reef
Figure 10.13 (a) Wheeler Cay from the west in October 1977, and (b) changes in location 1969–80 (after Hopley, 1982; Aston, 1995).
357
10.5 Reef island dynamics (a)
Vegetation
100 m
N
Beachrock 1936*
June 1980
June 1964
Dec 1976
1936*
Mar 1976
Vegetation line 1975
April 1975
Vegetation line 1936*
Beachrock 1975
*Based on pace and compass survey by Steers (1938) - position relative to later surveys estimated. East
(b)
78 77 74 75 73
68
66 69 65 62
72 80 71 76
79
64 63
South
70
2 × 105
0
ΣV3
67
Figure 10.14 (a) Mobility of Erskine Island, and (b) its relationship to wind variations. Annual wind energy resultant vectors from the Heron Island weather station 1962–80. V represents wind speed in knots. Energy is approximately proportional to velocity cubed (after Flood, 1983b, 1986).
358
Islands of the Great Barrier Reef
works have been constructed to protect infrastructure, especially at the western end (see Chapter 13). The greatest changes on some reef islands are associated with tropical cyclones (Stoddart et al., 1978b; Blackman et al., 1986), the effects of which are variable but may be either destructive or constructive (Gourlay, 1983a). Nonetheless, significant sediment interchange around cays, and between cay beaches and the area of sanded reef flat, can occur even during normal weather conditions (Hopley, 1981; Flood, 1986). It is important to recognize that erosion or accretion is the net balance of sediment movement onto or off the cay, but for any stretch of shoreline may be the result of alongshore movement with accretion matched by erosion (or vice versa) elsewhere on the cay. Shifts in wind direction normally force these changes, and can cause cay shorelines to rotate. For example, Masthead (Fig. 13.10b) and Heron Islands rotate in an anticlockwise direction, with accretion on the north-east shoreline and erosion on the north-west, but Green Island shows a tendency to rotate clockwise, with erosion on its north-east and south-west shores and accretion to the north-west (Kuchler, 1978; Beach Protection Authority, 1989). In this regard ‘‘cay erosion’’ is simply the redistribution of the sediments stored in the reef island, and must be clearly differentiated from reductions in island size and particularly volume when the long-term prospects for these islands are considered. The vegetated cays of Torres Strait, such as Poruma (Fig. 13.9a) and Masig–Kodall experience two distinct wind seasons – the dominant southeasterly tradewinds and the north-west monsoon. Waves and currents associated with these seasons move beach sediments in different directions around the cay, a process most clearly manifest by seasonally different accumulation on the eastern and western sides of jetty breakwaters located on the north-west shores of these islands. In the north-west monsoon beachrock is exposed to the east of the breakwaters and the beach builds up to the west, and the pattern is reversed as the seasons switch. The ends of these islands are very dynamic (Fig. 10.15a), with marked variations in shoreline variation occurring due to changes in spit position similar to those described by Flood (1984a) for vegetated cays in the Bunker–Capricorn Group. As much as 60 m of shoreline change occurred at the western end of Poruma between 2004 and 2006. Shoreline movement of this magnitude is of great concern to the inhabitants of these cays, but it is salient to recognize that the dip of beachrock exposed by recent movement on both islands suggests they have prograded across a former shoreline toward the reef edge, and are now moving back toward a previously held position on the reef top (Fig. 10.15b).
10.5 Reef island dynamics
Figure 10.15 (a) Vegetated sand cay dynamics where dominant winds are seasonally reversed: Layoak Islet, Torres Strait. Note conspicuous tails which migrate in response to seasonally variable winds, waves, and currents. (b) Beachrock dipping into the beach, south-eastern end of Masig–Kodall, Torres Strait, documenting that the southern shoreline is presently where the northern shoreline was when the beachrock formed.
359
360
Islands of the Great Barrier Reef
10.6 Discussion: reef island prospects and potentials Reef islands are obviously dynamic features, and current trends of accelerated island mobility and possibly erosion are clearly concerning. The management implications of island shoreline dynamics are dealt with in Chapter 13. Here we aim to examine the future of reef islands on the GBR within a broader geomorphic context. It is useful to begin this discussion with a summary of factors relevant to reef island development and stability on the GBR. (1) A total of 72.6% of reef islands occur on planar reefs, including 97.6% of the vegetated or low wooded islands, which are generally accepted as the most stable and mature reef island type. (2) Only 40% of the available planar reefs support reef islands, 60% have no reef islands at present. (3) At present 62% of reef islands on planar reefs are unvegetated (i.e., only 15% of planar reefs currently support vegetated cays or low wooded islands). (4) Forty-two unvegetated cays occur on reef patches (i.e., 9.4% of reef patches support unvegetated cays). (5) Excluding the ribbons, incipient fringing and true fringing reefs (which are not accommodated as easily within Hopley’s (1982) evolutionary reef growth sequence), about 26% of reefs are planar, 13% are lagoonal, 12% are crescentic, 21% are patches, and 27% are submerged. These reef types are not uniformly distributed over the GBR. (6) Although considerable variation is possible due to variations in size and depth of substrate, hydrodynamic setting, etc., analysis of the dataset presented in Chapter 8 suggests that on average patches transform into crescentic reefs after approximately 700 years, crescentic reefs become lagoonal reefs after around 1350 years, and these become planar reefs in as little as 250 years. (7) Where appropriate conditions exist (energy, reef geometry and exposure, etc.), reef islands can develop quickly after the reef platform reaches sea level.
The points above broadly summarize the current state of reef and reef island development on the GBR, the details of which are presented in more depth elsewhere in this book. Reef islands are clearly concentrated on planar reefs, but not all planar reefs support reef islands, and only a relatively small percentage support islands with vegetation and thus some degree of stability. At least four important questions are raised from the above points: Question 1: What proportion of currently stable islands will remain stable? The challenge in answering this question is to separate variability associated with the natural reef island dynamics from a longer-term trend of decline. To
361
10.6 Discussion: reef island prospects and potentials
Table 10.3. Rates of lagoonal infill and development to planar reef stage; patches and planar reefs of ‘‘average’’ size have been producing excess sediment, available for storage in reef islands, for 4.2 ka and 5.8 ka respectively Patches
Crescentic
Lagoonal
Planar
2
9 100 000 16 800 000 15 700 000 4 100 000 Reef area (m ) Lagoon depth (m) 9.3 19.1 18.1 10.8 84 630 000 320 880 000 284 170 000 44 280 000 Lagoon volume (m3) 26 047 058 Infill per 1000 years (m3) 1000s of years to fill from 7.5 ka 3.2 12.3 10.9 1.7 1000s of years yet to fill 4.2 4.8 3.4 5.8
establish this it must be determined whether the reef island sediments are merely periodically redistributed over a reef top to locations from which recovery is possible, or whether ongoing net sediment loss is under way. This is a difficult task hampered by relatively few long-term datasets, although the GBR dataset is amongst the best available. Stoddart et al. (1978a) showed that the size of two of the three of the northern GBR vegetated cays for which comparisons were possible had diminished since the earlier surveys of Steers, but Stapleton Island had increased by 14% (Table 10.3). Vegetated cays further south show similar variability, with Green Island growing between 1945 and 1978 (Kuchler, 1978), and erosion, accretion, and no change determined for different vegetated cays in the Bunker–Capricorn Group (Flood, 1977). Whether these trends have continued, reversed, or stabilized is unknown. However, cyclic patterns of island change forced by weather (Flood, 1986) and continual sediment exchange between the cays and sanded reef flat have been demonstrated from short-term records (Hopley, 1978c, 1981), and thus detecting significant trends in reef island behavior is difficult. Nonetheless, the stability of a reef island can be more confidently predicted if it is large (Flood, 1986), vegetated (Aston, 1995), cemented or indurated to some degree (Hopley, 1982), occurs on a moderate-sized oval-to lens-shaped reef aligned to the south-east (Gourlay, 1988), and/or where tide range is low (Stoddart and Steers, 1977; Hopley, 1982). The geomorphic consequences of extreme events are variable, but generally cyclones are erosional on sand cays (at least in the short term) and accretionary on shingle cays, so that both erosion and island accretion can occur concurrently on the low wooded islands (Scoffin et al., 1978). Flood (1984a) presents an excellent summary of cyclone impacts on five cays of the Capricorn Group in which both severe erosion involving loss of
362
Islands of the Great Barrier Reef
buildings and trees (North West, Masthead, Erskine) and island accretion (Tyron, Erskine) were recorded during cyclones. Unfortunately the GIS is not yet adequate to calculate the number of vegetated cays on the GBR for which combinations of these parameters apply, and thus the number of reefs meeting these criteria cannot presently be calculated (Chapter 5). Hopley (1982) suggested that the final stage of cay development may be a degenerative phase, forced by changed external conditions (e.g., storm regime or sea level) and/or intrinsic factors such as reduced sediment production and net delivery to the cay due to changes in reef flat morphology and roughness. Much of the sand stored in cays is several thousand years old (McLean and Stoddart, 1978), and relatively little young material is generally included, even on unvegetated and very mobile cays such as Pickersgill and Wheeler (McLean et al., 1978) (Figs. 8.4b and 10.13). This age discrepancy must partly reflect the time necessary for coral framework to erode to sand, but may also point to a reduction of sands and/or sands with durability and hydrodynamic properties necessary for prolonged accumulation. Several reasons for pulsed sediment production and delivery to GBR cays have been proposed, including a enhanced transport due to a ‘‘high-energy window’’ (Hopley, 1984), bursts of framework erosion associated with sea-level fall (Buckley, 1988), progressive change in reef flat roughness (Hopley, 1982), and changes to the dominant contributing organisms (Yamano et al., 2000). A change to highly mobile foraminiferan sands was reported for Green Island (Yamano et al., 2000), which were also shown to be a major component of mobile sands at Bushy Island by Hopley (1981). These sands may be more easily eroded from the cay to the sanded flat, or even off the platform. A link between the expansion of seagrass beds on the reef at Green Island and cay erosion has been identified by several authors (Hopley, 1982; Udy et al., 1999), suggesting that sands are captured on the reef flat and do not replenish the shoreline. Question 2: What proportion of presently unvegetated cays may evolve into more stable islands? It would seem a reasonable assumption that some of the 135 unvegetated cays recorded on planar reefs in the gazetteer would become vegetated. Indeed, some were probably vegetated at classification but not recognized as such from the imagery used. The many unvegetated cays of the northern GBR above 158 S appear to have good prospects for continued development, based on their common close proximity to vegetated cays on reefs of comparable size and geometry (Fig. 10.12). The second concentration of unvegetated cays occurs in the Swain Reefs between 218 and 228 S (Fig. 5.10). The cays here are typically small and ephemeral, and although several have become vegetated, the
10.6 Discussion: reef island prospects and potentials
363
long-term stability of most is questionable. A particular problem for cays developed in the Swain Reefs is that the dense reef network limits wave energy and direction so that sediment transport and concentration is tightly controlled directionally with two major effects. The first is that island size is constrained, thus limiting stability, and the second is that in contrast to ambient cay-building waves, cyclone waves can arrive from any direction and have devastating and long-lasting effects (Flood and Heatwole, 1986). There are relatively few unvegetated cays through the central GBR, reflecting the diminished number of planar reefs and younger reef top ages (Fig. 5.11), but also the higher tidal ranges and energy exposure (including cyclone occurrence in historical times (Puotinen, 2004)). The very few cays that occur in this region, such as Bushy Island, are located inside the zero isobase on elevated reef flats, often with a significant moat, which mediates the destabilizing effects of high tidal range and wave exposure (Hopley, 1982). It is unlikely that the occurrence of vegetated cays will increase markedly in the central GBR within the next few thousand years. Question 3: What prospects are there for islands to form on planar reefs or reef patches currently without reef islands? The theoretical answer to this question is reasonably straightforward, but quantitative answers cannot be derived from the available database. Although mature reef islands are highly concentrated on planar reefs, clearly not all planar reefs can support reef islands. A range of intrinsic and extrinsic factors are likely to prevent their formation even if the reef platform exists, and these have been addressed previously in this chapter (e.g., reef size, shape, location relative to other reefs, tidal range, and exposure to energy). Unfortunately, the GIS database is too coarse and poorly ground-truthed geomorphologically to interrogate such questions in a predictive sense for individual reefs or reef regions (Chapter 5). Question 4: Will the advance of reefs through the evolutionary reef growth sequence produce new reef platforms on which new reef islands may develop? Given stability of sea level and climate, with time it may be expected that more reefs will advance further through the evolutionary sequence, with a greater number reaching the planar stage suitable for reef island formation. A timeframe for progression through this sequence was presented in Chapter 8 based on a sample of radiometrically dated reefs of different size and with preHolocene substrates at different depths, and is summarized in Fig. 10.16. Details of reef type frequency, reef-island occurrence, reef dimensions, and predicted average rates of change through the sequence (based on average size
364
Islands of the Great Barrier Reef
Depth below sea level (m)
Submerged reefs
Reef patches
a Mean age within 1 m of surface:
0
Crescentic
Lagoonal
Planar
3.8 ka
3.8 ka
5.2 ka
5.8 ka
Yet to reach sea level
5 10
9.3 m
15 n.d 20
10.8 m
a
Mean depth to Pleistocene
19.1 m
Time to progress to
b
Mean area
Number of reefs
6.2 km b
Percent of b total number
700 years
??
a
next reef stage
2
9.1 km
2
18.1 m
1350 years 16.8 km
2
250 years 2
2
15.7 km
4.1 km
566
446
254
270
544
19.5
15
8.7
9
18
Percent of b non-fringing reefs
27
21
12
13
26
Unvegetated c reef islands
0
42
14
11
135
Vegetated c reef islands
0
4
2
0
83
Total number of reef islands on this reef typeb,c
0
46
16
11
218
Percent of this type of reef island b,c with reef islands
0
9.4
6
5
40
a
Based on radiocarbon dataset (see Tables 8.1–3) ;
b
c
based on Table 5.3: based on Table 5.8.
Figure 10.16 Schematic diagram summarizing the attributes and frequency of reef types and reef islands on the Great Barrier Reef.
and depth of reefs recorded in the gazetteer) are also presented (Hopley et al., 1989). Using the terminology of Purdy and Gischler (2005), these data were used to calculate the size of the reef ‘‘bucket’’ to be filled to form a planar reef. Sophisticated numerical modeling will soon provide detailed outputs, but a simple model (Table 10.3) clearly reconstructs the current situation effectively, with excess sediment on both reef patches and planar reefs available for reef island formation. The model calculates the time required to move to planar for each of the other reef types, using the average dimensions for each derived from the gazetteer (rather than the radiocarbon-dated dataset established in Chapter 8) and infill rates calculated using data from the planar reefs of the northern GBR where ‘‘mature’’ reef platform ages are best known. As noted in Chapter 8, reef islands may establish on reef patches that coalesce and form a platform rather than progress completely through the sequence, as has occurred at Wheeler (Hopley, 1978c, 1982). It seems unlikely that a vegetated cay will develop at Wheeler Reef, however, as it has completely covered its foundations and thus even though a contiguous platform at sea level is available, its size and shape work against ongoing stability. This may also be the case at other reef patches presently supporting unvegetated cays;
10.7 Conclusion
365
however, the mean patch size is actually larger than for most planar reefs (9.1 km2) and it is probable that many would be of comparable size. More than 40 unvegetated cays exist on the 446 reef patches recorded on the GBR, which have a mean reef top age of just 3800 years. As more reach sea level, and more at sea level coalesce to form more substantive reef platforms, the prospects for the formation of more unvegetated cays, and possibly also more vegetated cays too, should increase. Reef patches are distributed quite evenly over the GBR, with a reasonable proportion located in areas where cays form, especially north of Cairns. Interestingly, it seems the greater dimensions of crescentic and lagoonal reefs will prolong their progress to planar stage (4800 and 3400 years to go respectively), to the extent that it is more likely that many now just at reef patches stage will fully mature more quickly (Fig. 10.16). Moreover, although the majority of cays on reef patches are unvegetated and ephemeral at present, almost 10% of these reefs do support reef islands despite reaching sea level only recently (mean 3800 years ago: Table 8.3). If some of these islands persist as the patches coalesce, an additional model of reef island development in which islands form well before the planar stage of reef growth is reached will need to be developed. 10.7 Conclusion Very few of the reef islands on the GBR have been examined geomorphologically, obviously limiting many of the conclusions drawn above. Of 300 reef islands, geomorphic studies beyond description have been completed for very few – we estimate fewer than 20. Furthermore, the geomorphic detail included in the GIS is inadequate and too inaccurate to be confidently used for detailed investigations or predictions (Chapter 5). Full investigations integrating Holocene evolution (historical changes and dynamics) and process studies, although under way on a few cays, to our knowledge have not been completed for any GBR reef island. Almost 20 years ago Gourlay (1988) summarized the parameters that significantly influence reef island development and dynamics, and lamented that the roles and interactions between many were at that time imperfectly known. Considerable progress has been achieved in some areas, such as establishing reef growth rates (see following chapter) and in understanding hydrodynamic processes on reef platforms (Kench and Brander, 2006) (Chapter 4). However there is still much to be known from a broader range of reef island environments on the GBR, and the necessary integration of geological and hydrodynamic investigations at a range of timescales has not yet been successfully achieved (Gourlay and Hacker (1991) is an exception). Integrated studies of
366
Islands of the Great Barrier Reef
the formation and dynamics of these valuable environments are critical if we are to conserve or manage them effectively, individually and as a habitat per se, in a period of environmental change possibly including increased storm frequency and/or intensity, and lowered carbonate production. Technological and analytical techniques have greatly improved our capacity to collect precise and accurate data (e.g., hydrodynamic instrumentation, AMS radiocarbon dating), and to manage it (GIS). It is hoped that these advantages are used to good effect within the next 20 years. As discussed further in Chapter 13, this is particularly critical for robust predictions to be made regarding the possible impacts of climate and sea-level changes.
11 The accumulation of the Holocene veneer to the Great Barrier Reef
11.1 Introduction In Chapters 7, 8, and 9, the morphology and evolution of reefs in three zones across the Great Barrier Reef (GBR) shelf have been presented. In part the interpretation has come from the numerous drill holes and dated core materials which have become available in the last 25 years (Tables 7.2, 8.1, and 9.1). The examples cited are but a part of the total dataset which now comprises the drilling of approximately 50 reefs (over 160 holes), a total core length of almost 2000 m through the Holocene, and almost 750 radiocarbon dates. Combined, these data are a powerful tool for the interpretation of Holocene growth patterns across the full extent of the GBR. Given the size, latitudinal, and cross-shelf extent of the GBR, results of such a synthesis may be applicable to many reef areas elsewhere for which the information is less comprehensive. Conversely, interpretations from elsewhere in the Indo-Pacific and Caribbean may be helpful in deciphering the Holocene evolution of the GBR. Synthesizing the data, for example into growth rates during the Holocene in different environments, was an early goal of drilling in the 1970s and 1980s (Davies and Hopley, 1983; Davies et al., 1985). The dataset then was much less than now, derived from 68 holes drilled in 22 reefs, and 597 radiocarbon dates. Nonetheless, results as published in 1983 appear to be strengthened by further work and where more raw data have been added the results have not been altered significantly. It is not always possible to use data from all the drilling on the GBR, especially from researchers other than ourselves who may not have presented their results in an appropriate format. The conclusions reached in this chapter, where not quoted from other published sources, use as a minimum the dataset quoted by Davies and Hopley (1983) and in some instances are based on up to 90% of the total information available. The depths referred to are below reef flat level, the approximate level of Mean Low Water Neap Tides. 367
368
The accumulation of the Holocene veneer to the Great Barrier Reef
11.2 The depth to the antecedent surface On the outer and mid-shelf reefs, where the full thickness of the Holocene reef was penetrated, the top of the underlying Pleistocene was reefal limestone. In contrast only six of the fringing reefs investigated have been constructed over prior reefal foundations (Moa, Yam, Hayman, northern transect of Cockermouth, Penrith, and Digby). Pleistocene reef outcrops on the surface of Digby Island’s fringing reef (Kleypas, 1996) and the deepest the Pleistocene foundations have been encountered is 28 m on Myrmidon Reef (Marshall, 1985). Barham (1983) also reports recrystallized reefal limestone at or close to the surface at Saibai Island close to Papua New Guinea. However, a number of holes have been drilled to depths greater than 20 m without identifying the Pleistocene. The extent to which the holes drilled in any one reef are sufficient to identify the depths to the Pleistocene is an important question. On many reefs only two or three boreholes have been drilled and perhaps in only one has the Pleistocene been penetrated. Nonetheless, we believe that there is some significance in the depths established. Most cores have come from windward locations which most prior work has suggested as having the highest antecedent surface (e.g., Purdy, 1974; Harvey and Hopley, 1982; Hopley, 1982; Marshall and Davies, 1982; Davies, 1983). One Tree Reef is probably the most studied of all the GBR reefs and its data clearly support this contention (Marshall and Davies, 1982, 1984; Davies, 1983). Beneath the windward margin the Pleistocene rises to 10.5 m, is 15 m below lagoonal patch reefs, and 13 m below the leeward rim. Together with vibro-coring of lagoonal areas, the total range of Pleistocene on One Tree Reef is from 10.5 m to 23.0 m (Davies, 1983). On Stanley Reef, four out of five holes drilled also reached the Pleistocene (Fig. 8.10). Two on the windward margins reached 15 m and 18 m respectively, whilst at leeward margin sites the depths were 21 m and 22 m (Marshall, 1983a). On only one reef was the shallowest Pleistocene not found on the windward margin. On Redbill Reef (Fig. 8.11) the Pleistocene was clearly encountered at 17.1 m on the leeward algal rim, and was suggested at just below 17.8 m on the windward margin (Hopley et al., 1984). However, the shallowest encounter with the antecedent surface was in the shallow lagoon in the center of the reef where the Pleistocene topped by a partial soil profile was found at 13.0 m. The greater depth to the Pleistocene in Redbill does suggest that there may be some relief in the Pleistocene even on windward margins as a prior seismic refraction survey (Hopley et al., 1982) had indicated a discontinuity at between 10.0 m and 11.5 m.
11.2 The depth to the antecedent surface
369
Figure 11.1 Minimum depth to the Pleistocene plotted by latitude and position on the continental shelf.
Nonetheless the plotting of the shallowest depth to the Pleistocene along the entire GBR (Fig. 11.1) suggests a distinctive geographical pattern. From the shallowest part of the Torres Strait (Woodroffe et al., 2000), where the unconformity is between 5 m and 7 m (and 0 if Saibai further north is included) there is a gradual increase in depth to the southern end of the ribbon reefs at about 168 S. Depths are 11 m at Raine Island, 15 m at Yonge Reef (see Chapter 9), and between 15 m and 17 m across the shelf just north of Cooktown (Davies et al., 1985). One anomaly to this pattern is Bewick Island where Thom et al. (1978) report the unconformity at 4 m below reef flat level. South of Cairns the depth to Pleistocene appears to increase by more than 5 m. At Britomart Reef (Johnson et al., 1984) it is at 19.8 m. Everywhere else, where encountered, it is up to 28 m below present reef flat level as far south at 198 S and not encountered at a number of reefs that were drilled to more than 20 m. At Stanley Reef the depth is 15 m (Marshall, 1985), the same as beneath the fringing reef at Hayman Island just to the south (Hopley et al., 1978). In the cross-shelf transect just south of Mackay the variation is from 0 at Digby Island to between 8.0 m and 13.6 m beneath mid-shelf reefs including fringing reefs on this wide area of shelf, and 17.5 m beneath Cockatoo Reef in the Pompey Complex (Kleypas and Hopley, 1993; Kleypas, 1996). Just inside the Pompey Reefs is Gable Reef and in four holes drilled there to
370
The accumulation of the Holocene veneer to the Great Barrier Reef
between 9.9 m and 29.5 m, the Pleistocene was not encountered, though the deep hole (Fig. 8.13) was in the secondary reef front which may explain its greater depth here (see Chapter 8). In the southern outlier of the Bunker–Capricorn Group the unconformity is again shallow beneath the windward rims of One Tree, Fitzroy, and Fairfax Reefs (8.0 m to 10.5 m) and 13 m beneath the leeward side of Heron Island (Davies, 1974). Davies et al. (1985) suggested that there was no cross-shelf variation in the depth to the Pleistocene and this is certainly true for the transects near Cooktown and Townsville. Some suggestion of a higher surface close inshore near Cape Melville (148 110 S) is given by the relatively high level of 4 m below Bewick Island (Thom et al., 1978) but this low wooded island is near the mainland and nearby Howick Island has continental rocks outcropping to 56 m. This site may be similar to the inshore islands near Mackay, i.e., inshore continental rock foundations appear to favor the preservation of the Pleistocene close to present sea level. However, in the far north there may be a deepening from Torres Strait to the shelf edge and in the south central GBR there is suggestion of a progressive deepening across the wide shelf, from the Cumberland Islands, through the mid-shelf reefs to the outer Pompey Reefs. A possibly more significant pattern is the latitudinal change shown in Fig. 11.1. The Pleistocene foundations are up to 10 m deeper in the central GBR. Perhaps not coincidentally this is also the area of the submerged shelf margin reefs and the possibly still active offshore Halifax Basin (McKenzie et al., 1993). There is also the possibility of some hydro-isostatic subsidence in the same area (Chappell et al., 1982; Hopley, 1982, 1983b) though the morphology and extent of the shelf-edge submerged reefs (Chapter 9, Fig. 9.9) would support a more prolonged and persistent structural movement. 11.3 The fabric of the Pleistocene foundation The Pleistocene surface which the Holocene reefs recolonized was highly variable with perhaps iron staining the only characteristic that is common to all Pleistocene sections. Corals, of course, are common throughout and, as Marshall (1983) noted, in many holes the corals present in the Pleistocene section are similar to those of the Holocene above, except that often the Pleistocene material shows relatively thicker crusts of coralline algae. Beneath many reefs the transition across the unconformity is sharp with the top of the Pleistocene marked with calcite crusts and stringers, as in the Bunker–Capricorn Group of reefs (Marshall, 1983). Elsewhere the boundary is gradational consisting of mud, sand, and gravel probably representing the resorting process when the surface was first re-inundated (e.g., below the
11.3 The fabric of the Pleistocene foundation
371
fringing reefs with limestone foundations in the Cumberland and Northumberland Islands: Kleypas, 1992; Kleypas and Hopley, 1993). A notable feature of the Pleistocene reefs is the abundance of Halimeda in the upper section as first noted in the southern reefs by Marshall (1983). Even where there is an abundance of corals the matrix may be dominated by Halimeda. There may be several meters of this unique limestone at the top of the Pleistocene ranging from 0.8 m thickness on One Tree Reef to 2.0 m on Fitzroy, >5 m on Britomart, and up to 10 m on Ribbon 5 (Marshall, 1983; Johnson et al., 1984; Braithwaite et al. 2004). Beneath Yonge and Raine Island reefs the Halimeda grainstone is almost pure with only very occasional small corals. The thickness here is between 0.8 and >1.6 m (Figs. 9.3 and 9.8). Such accumulations are not common on modern coral reefs (see below) although, as outlined in Chapter 6, Halimeda-dominated banks have accumulated behind the northern ribbon reefs and elsewhere during the Holocene in response to oceanic upwelling and nutrient jetting. On the deep platform behind Great Detached Reef where nutrient upwelling may be still occurring, there is a modern equivalent of Halimeda shoals which may be analogous to those accumulated apparently at the end of the last interglacial and in the early Holocene. Environmental conditions favoring upwelling are suggested at these times. The diagenesis of the Pleistocene material has been examined by Marshall (1983). As expected most aragonite corals and high-magnesium calcitic algae have been altered to low-magnesium calcite although some aragonite may still be found where corals occur at the top of the Pleistocene. There may be considerable infilling with calcite cements, and as noted above, the formation of calcite crusts and stringers. Where discussed (e.g., Marshall, 1983; Kleypas, 1992) most diagenesis is attributed to vertical movement of water from a subaerially exposed surface within the vadose zone. However, Kleypas (1992) recognized freshwater phreatic diagenesis on Penrith and Redbill Reefs. Marshall (1983) also identified the phreatic zone or water table at 17 m to 18 m in the reef rim of One Tree Reef. The phreatic zone was not identified beneath the lagoon suggesting permeability differences between the perimeter and lagoon areas with a more permanent freshwater lens around the perimeter similar to the present-day aquifers on uplifted atolls. Although direct evidence is minimal, the age of the Pleistocene reef upon which the Holocene is founded appears to be last interglacial. A number of minimal radiocarbon dates of >30 ka have been obtained. More revealing are uranium series dates, for example, 125.7 ka from Ribbon 5 (Braithwaite et al., 2004) and dates ranging from 145 ka to 122 ka from the southern GBR (Marshall, 1983). Given the established height of the last interglacial sea level at or just above present (e.g., Chappell et al., 1996) (see Chapter 3 for
372
The accumulation of the Holocene veneer to the Great Barrier Reef
discussion), if it is presumed that the age of the antecedent surface is ubiquitously last interglacial (and if not why did some reefs, especially in the central GBR, not acquire a last interglacial veneer) then the case for shelf subsidence is further strengthened. On the mainland are many locations where the last interglacial shoreline can be seen (e.g., Graham, 1993). At Digby Island (238 300 S) Pleistocene reef outcrops at the surface (Kleypas, 1996). Hopley (1982, ch. 7) discussed why there is the variation in foundation levels, evaluating erosion versus subsidence and dismissing erosion as the main cause due to the rates of karst erosional processes within the time available. The large amount of drilling that has occurred since 1982 still seems to support this conclusion. Some erosion of the surface undoubtedly did take place with development of soil horizons which, perhaps surprisingly, beneath some reefs have survived. These include Fairfax, Fitzroy, Redbill, Hayman, Davies, and Ribbon 5. The hole drilled into the center of Redbill Reef recovered a basic red clay within cavities at the top of the Pleistocene which included quartz, illite, kaolinite, and hydrated kaolinite with more than 80% non-carbonate material. Although this is a mid-shelf reef, it does have a small granite outcrop on its south-western side (Fig. 8.7a) and granitic foundations may not be far beneath this reef elsewhere. This may also help to explain the high level of the Pleistocene in the center of the reef as opposed to the margins as discussed above. Marshall (1983) also reports a soil consisting of quartz sand, set in a humate cement, infilling cavities between coral fragments and lightly cementing them together on Fairfax Reef. How sand-size apparently terrigenous material can reach the surface of a mid-shelf reef is an interesting question. Davies and Hughes (1983) showed how clay-size terrestrial material could be incorporated into an inner shelf reef north of Cooktown (Boulder Reef) during major floods, possibly every five years. Subsequent subaerial exposure, erosion, and soil formation could provide a concentration of this material at the surface of the paleo reef. 11.4 Date of recolonization during the Holocene transgression Using the Thom and Chappell (1975) sea-level curve as a first estimate, one can calculate that the Pleistocene foundations of the GBR should have been first inundated by the postglacial transgression about 10 ka ago, with the highest elevations (excluding Digby Island) going under water about 3 ka later (7 ka ago). However, based on early pre-1984 drilling results, recolonization over the older surface did not appear to be correspondingly spread over this period and, as first noted by Davies and Hopley (1983) and Davies et al. (1985), there appeared to be a distinct gap between inundation and recolonization of 1.2–2 ka with the majority of reefs commencing effective framework growth
11.4 Date of recolonization during the Holocene transgression
373
in what was termed a ‘‘narrow take-off envelope’’ of 800 years between 8.3 ka and 7.5 ka ago. Some support came from Kleypas and Hopley (1993) who suggested earliest reef growth in the south central GBR occurred within a 250year period (8.07–7.84 ka ago) on a cross-shelf transect of six reefs, and, based on 18 drill holes from five Indo-Pacific islands, Montaggioni (1988) suggested the start-up event occurred within the same narrow time-span worldwide. As more data have become available the robustness of these findings may be breaking down. Figure 11.2a plots the oldest radiocarbon date, the depth of the sample, and, as only cores in which the Pleistocene was encountered are used, the depth down to the corresponding antecedent level. The Thom and Chappell (1975) sea-level envelope is also shown with only one site, the fringing reef on Hayman Island, lying above the curve suggesting the materials dated may have been detrital and not in situ coral. The take-off envelope now extends between 8.6 ka and 6.6 ka for outer reefs, 9.9 ka to 6.9 ka for mid-shelf reefs, and (excluding Hayman) 7.8 ka to 6.3 ka for fringing reefs with reefal foundations, certainly wider than the original time-frame suggested by Davies et al. (1985). However, what must also be taken into account is the height of the dated samples above the Pleistocene substrate. Most samples chosen for dating were presumed to be branching or head coral framework whilst the base of the Holocene reefal material almost ubiquitously consists of detrital facies (see below). Thus as shown in Fig. 11.2a there may be several meters (17 m on Viper Reef) between the two levels. It will be argued below that detrital materials such as are found at the base of the Holocene have had average accumulation rates of about 4 m ka1. This would be a typical rate for the take-off period for Holocene reef growth as modeled by Davies and Marshall (1979) and also discussed below. If this is taken into account, then the colonization dates on many reefs are much earlier, as shown in Fig. 11.2.b. They are also very close to the presumed sea-level curve suggesting that for most reefs early colonization took place shortly after inundation in perhaps no more than 4 m of water. A few reefs appear to have had a greater depth when colonized (e.g., Raine 8 m; Wheeler 7 m; Bowl 7 m) but this may be an artifact of the dataset. Stapleton Reef is an anomaly in both plots and in the revised graph Boulder and Williamson Reefs join Hayman questionably above the sea-level envelope. A detrital accumulation rate greater than 4 m ka1 would resolve these anomalies. Regional explanations for colonization delays and clustering as speculated by Davies et al. (1985) included unsuitable oceanographic conditions, absence of larval refuges during the low sea-level stand, and proximity of terrestrial influences. These are no longer required. The location of the majority of boreholes (back from the windward margin of the reef) appear to have been the location of rubble banks formed possibly as storm materials thrown up
374
The accumulation of the Holocene veneer to the Great Barrier Reef
Figure 11.2 (a) Deepest dates from the Holocene reefs and depth down to the Pleistocene which they overlie. (b) Holocene reef take-off time presuming a vertical accretion rate of 4 m ka 1 for the basal rubble facies. The sea-level envelope on both diagrams is from Thom and Chappell (1975).
11.5 Rates of growth and accretion
375
onto the evolving reef flat from an actively accreting reef front. These materials would not have provided a stable enough surface for framework-building corals to colonize until there was either great enough water depth (as for Raine, Wheeler, and Bowl) or some protection was given by a more robust windward margin. On only a very few reefs has it been possible to drill this absolute windward margin and, on One Tree Reef, this type of colonization beneath windward reef flats seems to be suggested. 11.5 Rates of growth and accretion The growth of coral reefs from a geomorphological viewpoint is the net process of biological deposition (by corals, algae, and other organisms), bioerosion of the surface and internal cavities of the reef, mechanical and chemical deposition (as reef detritus and marine cements), and mechanical and chemical erosion (by waves and currents, and in solution). The rates at which the individual processes may operate are highly variable over ecological timescales (see Chapter 1), but over a geomorphological timescale the net rates produced by drilling and radiocarbon dating of cores are highly reliable and produce a picture of how the GBR accumulated during the Holocene. In 1983, using only about 50% of the database now available, Davies and Hopley (1983) were the first to define the growth rates of reefs during the Holocene, from radiocarbon-dated cores. The figures as net rates developed over time intervals of hundreds to thousands of years showed surprisingly close correlation with rates derived from modern community metabolism studies (for review see Kinsey, 1985). Addition of further data has merely strengthened the conclusions of 1983 and the various growth rate strategies that were derived from them (Davies et al., 1985). This section reviews the data in terms of variation over the Holocene, in relation to water depth as the reefs developed, as determined by facies and location on the reef, and with respect to larger-scale location on the GBR. 11.5.1 Rates of accretion throughout the Holocene A hypothetical curve for reef accretion throughout the Holocene was produced by Davies and Marshall (1979). The S-shaped growth pattern has three phases – an early period of slow growth as the Pleistocene is recolonized; a middle period of rapid growth as reef accretion trails the rise in sea level; a final period of slower growth as the reef approaches the stabilized sea level. As noted above the early period of slow growth may be an artifact of the early data used with most accretion immediately above the Pleistocene being detrital.
376
The accumulation of the Holocene veneer to the Great Barrier Reef
When the growth data from all cores are presented together (Fig. 11.3) the pattern that emerges is one which shows all reefs growing at about 5 m ka1 well into the stillstand period. For mid-shelf and outer reefs the highest rate over a 1000-year period is 6 m ka1 between 8 and 7 ka ago. For fringing reefs the highest rate is less than 5 m ka1, between 7 and 6 ka ago. After 5 ka ago rates for all reefs decline steadily to less than 3 m ka1 although a small peak between 3 and 2 ka ago on fringing reefs and 2–1 ka ago on outer reefs can be seen. Variability in reef progradation was discussed in relation to fringing reefs in Chapter 7, but it is uncertain if the variability seen in the last 3000 years in Fig. 11.3 is real or an artifact of the sparser data for back reef areas in which most growth can be presumed after the reefs reached sea level. 11.5.2 Rates of framework and detrital accretion A very clear pattern of framework and detrital growth rates was shown by Davies and Hopley (1983, fig. 5). Figure 11.4 uses the same methodology of identifying the major component within cores between reliably dated sections as either framework or detritus with a growth rate for that section of core based on its length and the bracketed dates. It uses the original data from 23 reefs with additional information from a further 28 holes on eight outer reefs and 35 holes from nine fringing reefs with a total of 173 new radiocarbon dates. These data from north to south include Warraber, Yam, and Hammond Reefs (Woodroffe et al., 2000), Raine and Yonge Reefs (see Chapter 9), the Cape Tribulation reefs (Partain and Hopley, 1989), Ellison, Potter, Taylor, 17-065, and Moss Reefs (Graham, 1993), Fantome Island (Johnson and Risk, 1987), and Cockatoo and the fringing reefs of the south central GBR, on Scawfell, Cockermouth, Penrith, Percy, and High Peak Islands (Kleypas and Hopley, 1993; Kleypas, 1996). The results are plotted in Fig. 11.4. The original results of 1983 are confirmed but the larger data bank allows for some refinement of the interpretation. Framework growth rates range up to 16 m ka1 with a modal value of between 7 and 8 m ka1. A very large proportion of the reef framework accreted within the comparatively narrow range of 3–8 m ka1. Examination of specific core logs indicates that growth rates of above 8 m ka1 are essentially produced by branching corals with a high porosity. Head corals dominate the framework accumulating below 5 m ka1. The very low growth rates of <2 m ka1 are mainly associated with algal crusts, not only at the surface of the reefs, but frequently in relatively thin bands within the cores at intermediate depth. Detrital accretion rates may also be refined by the new plots. Examination of the cores suggests that there may be two overlapping populations with a
11.5 Rates of growth and accretion
Figure 11.3 Vertical growth rates throughout the Holocene based on dated cores.
377
378
The accumulation of the Holocene veneer to the Great Barrier Reef
Figure 11.4 Rates of vertical detrital and framework accretion throughout the Holocene based on dated cores.
lower range of up to 5 m ka1 produced by the rubble banks frequently found immediately over the Pleistocene, and a slightly higher range of 5–10 m ka1 originally identified by Davies and Hopley as steady lagoonal infill, including back reef sand sheets. Rates above 13 m ka1 are considered to be associated
11.5 Rates of growth and accretion
379
Figure 11.5 Vertical accretion rates by environment.
with storm events. Deposition of large amounts of debris which would produce such records has been reported after tropical cyclones have passed over the GBR. A regional survey of impacts of Cyclone Ivor in March 1990 showed patchy damage but in areas where peeling of the superficial matrix to a
380
The accumulation of the Holocene veneer to the Great Barrier Reef
thickness of up to 1.5 m occurred, several meters of back reef gravel patches and reef-front gravel deltas accumulated together with sand wedges up to 2 m in thickness (Done et al., 1991). Framework accretion rates can also be attributed to different locations on the reef. Figure 11.5 is again based on Davies and Hopley (1983) with the additional data from subsequent drilling as above. Framework type has not been identified but again is predominantly branching framework at the upper end of the scale, head dominated around the modal values between 5 and 6 m ka1 and algal crusts at the lower rates. Coral flats have a similar pattern though lack the higher rates. No new data for patch reefs have been added to those of Davies and Hopley (1983), which have the highest growth rates between 7 and 8 m ka1, probably the result of their sheltered location and growth in deeper lagoonal environments for much of their history. Leeward reefs show a spread of growth rates indicative of the mixed energy and zonational patterns in such locations. Greater information is now available for fringing reefs and these too show a great range which is comparable to middle and outer shelf reefs. Davies and Hopley (1983) were not able to identify any latitudinal trends in the accretion rates. This is now confirmed in the studies which extend the dataset northwards into Torres Strait. It is also in accord with modern calcification studies (e.g., Kinsey, 1985) and with drilling results from reefal areas at higher latitudes than the southern end of the GBR (e.g., Collins et al., 1993a; Kennedy and Woodroffe, 2000). 11.6 The timing of reefs reaching modern sea level Modern sea level was first achieved on the GBR about 6.5 ka ago and, on the inner shelf at least, may have been a little higher than present around 5 ka ago (see Chapter 3). The time at which reef flats were first constructed and lateral growth of the reef commenced is a reflection of how well the reef was able to keep up with the final rise in sea level. Reef flat ages vary considerably over the GBR from contemporary with the start of the postglacial stillstand (6.5 ka ago) to very recent (Fig. 3.8). This age is important for a number of reasons: *
*
only when a reef is within the wave base depth range will it start to develop an energy gradient which evolves into a series of windward to leeward reef morphological and ecological zones (Hopley, 1982, Chapter 10) only the older reef flats may retain evidence of the hydro-isostatically induced higher mid-Holocene sea level (which occurred only on the inner shelf) (Chappell et al., 1982; Nakada and Lambeck, 1989)
11.6 The timing of reefs reaching modern sea level *
381
reef islands are not possible until at least some reef flat has formed. This too has ecological implications for species such as turtles and birds. The low wooded islands of the northern GBR appear to be associated with some of the oldest reef flats (>5 ka) and absence of islands on the central GBR appears associated with relatively young reef flat ages (Hopley, 1997b).
Regional variations in reef flat age are apparent and closely correlate with the pattern determined for the depth of pre-Holocene foundations. In general, the deeper the foundation, the younger the reef flat age. Figure 11.6 plots the oldest radiocarbon age at or within 1 m of the reef flat surface at 500-year intervals. The data comes not only from the deeper drilling discussed in this volume but also from the very shallow drilling reported in Hopley (1982, table 9.2) and the fringing reef flat ages in Chappell et al. (1983). The oldest dates are 6460 years BP for Williamson and 6420 years BP for Stanley Reef, both from the mid shelf. In general however, some of the oldest reef flats are those of fringing reefs which were able to continuously grow upwards over the rock foundations of islands or mainland during the latter part of the transgression. Outward growth of reef flat was thus possible as soon as sea level stabilized. Many fringing reefs may have developed reef flat by 6000 years BP. This also appears to be true for reef flats in Torres Strait where ages as old as 6340 years BP have been obtained. The only outer shelf reef (Raine Island) which has been drilled in this region has a reef flat date of 5690 years BP. Further south in the Cooktown region, mainland fringing reefs were developing reef flat by 6000 BP (Cape Tribulation) but so too were mid-shelf reefs (e.g., Bewick, Boulder, Williamson) and the outer ribbon reefs (Ribbon 5). No drilling has been attempted on the fringing reefs of the north central GBR but Chappell et al. (1983) and Graham (1993) have reported ages >6 ka for reef flat microatolls. Dates for reef flats of both mid-shelf (Ellison, Potter) and outer shelf reef flats (Moss, 17-065) are generally younger than 4 ka (Graham, 1993). Further south in the Townsville region inner fringing reef flat ages based on drilling results exceed 5.5 ka. However, reef flat ages also become progressively younger in a cross-shelf direction : 4–4.5 ka for mid-shelf reefs (Grub, Wheeler,) 3.5–4 ka for outer shelf reefs (Bowl, Myrmidon, Viper). As discussed in Chapter 12 this may have allowed for the operation of a midHolocene ‘‘high-energy window’’ to operate on the central GBR, the poor development of outer shelf reef flat until after 4 ka ago allowing larger swell waves to approach the mainland coastline (Hopley, 1984). However a little further south, off Bowen, mid-shelf reef flats are amongst the oldest on the GBR (>6 ka on Darley and Stanley Reefs). In the south central GBR ages at or slightly older than 6 ka have again been obtained for fringing reef flats. However, on this wider section of shelf
382
The accumulation of the Holocene veneer to the Great Barrier Reef
Figure 11.6 Dates at which reefs reached sea level by 500-year intervals and shelf location.
11.7 Reef growth relative to sea-level rise
383
mid-shelf reefs are again younger (4.5–5 ka: Redbill and Gable) with a similar age for the outer shelf Pompey Reefs (Cockatoo). At the southern extremity of the GBR, where the shelf narrows once more, reef flat ages are comparatively old, ranging from 5390 to 6320 years. (One Tree, Fitzroy, Wreck, and Fairfax). Hopley (1982, 1983b), plotted this pattern of reef flat ages over the GBR shelf and this is updated in Fig. 3.8. He concluded that on the central GBR up to 6 m of shelf-edge warping had taken place during the Holocene to explain the younger ages on the outer shelf off Townsville. The extra data now available do not discount this conclusion, the pattern of which, if not the amplitude, correlates well with hindcasting of Holocene isostatic warping patterns of Chappell et al. (1982) and Nakada and Lambeck (1989). However, as indicated in the discussion of shelf-edge reefs (Chapter 9) and the greater depth of the Pleistocene beneath the modern GBR (above) a further factor in producing downward flexing and younger reef flat ages in the central GBR may be recent tectonic activity in the Halifax Basin. 11.7 Reef growth relative to sea-level rise Although the early period of very slow reef growth in the Holocene may not have been as great as suggested by the S model of Davies and Marshall (1979), it is inevitable that some delay in recolonization took place. For the deeper Pleistocene foundations, this was at a time when the rate of sea-level rise was rapid, at least 7–8 m ka1. Any delay would have been difficult to catch up and with the range of framework and detrital accretion rates quoted above it is easy to see how the water deepened over many reefs until sea level stabilized at about 6.5 ka ago. Reefs growing from shallower foundations were inundated at a time when sea-level rise was slower and these were more able to grow upwards with sea-level rise, or catch up at an earlier date. Plots of age against depth for many reef cores prompted Davies et al. (1985) to recognize four basic growth strategies: (1) ‘‘Keep-up’’ strategy, for reefs which managed to keep up with sea-level rise, growing continuously in shallow water and developing reef flats as early as 6 ka ago. Such reefs usually grow from shallow, Pleistocene foundations. No perfect example is known in the GBR but Redbill Reef’s growth pattern (Fig. 8.11) is close. (2) ‘‘Katch-up 1’’ (Catch-up, from herein) strategy 1 for reefs which were left behind by the transgression and had much of their vertical growth in deep water (>5 m) but catching up with sea level before it had stabilized at about 6.5 ka ago. No reefs so far drilled on the GBR appear to have this strategy.
384
The accumulation of the Holocene veneer to the Great Barrier Reef
(3) ‘‘Katch-up 2’’ (Catch-up 2) strategy, for reefs with a similar growth strategy but not able to catch up with sea level until after it had stabilized (e.g., the outer shelf reefs of the central GBR (Figs. 9.15, 9.16, and 9.17)). (4) ‘‘Katch-up 3’’ (Catch-up 3) strategy, for reefs which only commenced growth after the stillstand had commenced (post 6.5 ka ago) and thus grew in an evershallowing environment (e.g., Raine Island (Fig. 9.3)).
Davies, in many private and public presentations, also irreverently described a ‘‘Screwed up’’ strategy for those reefs which ‘‘just died’’ and for which no visible explanation was clear. A very similar set of conclusions was reached simultaneously and quite coincidentally, by Neumann and Macintyre (1985) from Caribbean reef data. They had keep-up and catch-up strategies and they too had a further category, ‘‘give up’’ reefs, which were generally reefs that grew too slowly to keep up with sea-level rise and became drowned though mortality may be due to factors other than just sea-level rise (Macintyre, 1988). The wide range of borehole data analyzed in Chapters 7, 8, and 9 clearly shows that different signatures may be found on the same reef, i.e., different parts of the same reef have responded differently to the sea-level rise. The cause may be very local, such as variations in the level of the antecedent platform or wave exposure factors. However, in many examples, there appears to be an interaction between different parts of the reef, e.g., the need for windward reef growth, even if not at sea level, to provide shelter for the take-off of patch or leeward reefs. It is because of this that each reef has developed in the Holocene in a very individual way and helps explain the present wide range of morphological detail. Variation of growth rates with paleo water depths was also illustrated by Davies et al. (1985, fig. 5). Data were derived from dated core depths in relation to the standard eastern Australia sea-level curve of Thom and Chappell (1975). The plot showed optimum growth rates (>8 m ka1) at depths between 12 and 15 m declining relatively rapidly at greater depth (to 4 m ka1 at 20 m) and more slowly as reefs approached sea level. Davies et al. (1985) also plotted central versus northern GBR data without a significant difference being illustrated. With more core logs now available, the conclusion remains the same: there is little difference in growth rate with depth latitudinally in the GBR. However, Partain and Hopley (1989) and Hopley (1989a) plotted the data in a cross-shelf direction (Fig. 11.7). At paleo water depths greater than 6 m there is little difference between mid and outer shelf reefs but in shallower water depths the growth rate declines from the optimum of about 7 m ka1 to 4 m ka1 within a meter of the water surface, whilst the more sheltered
11.7 Reef growth relative to sea-level rise
385
Figure 11.7 Vertical growth rates (mm year1) plotted against paleo water depth and shelf location.
mid-shelf reefs are able to maintain a high growth rate into shallower water. The pattern for fringing reefs is even more interesting. They too have an optimum growth rate of about 7 m ka1 in a narrow water-depth band of about 4–8 m depth. Growth rates decline rapidly both in greater depth (light attenuation in more turbid waters?) and towards the surface (freshwater and sediment plumes?).
386
The accumulation of the Holocene veneer to the Great Barrier Reef
11.8 Holocene reef structure and facies development 11.8.1 Holocene facies The thickness of the Holocene reef is characteristically within the 10 to 20 m range with some examples of at least 30 m. Within this Holocene veneer can be detected a number of characteristic patterns: stratigraphically within the reef (i.e., representing changes through time), across individual reefs (from highenergy front to low-energy leeward margins), and across the shelf (from exposed outer shelf reefs to more protected inshore reefs, including fringing reefs). A number of descriptive classifications have been used to describe the GBR cores. Combining those of Marshall and Davies (1982) and Davies and Hopley (1983) produces six Holocene bio-lithofacies, three framework, two detrital, and one terrigenous: (1) Coralline algal facies: crusts of corallines around branching and massive corals or laminations many centimeters thick with distinctive faunal associations of vermetid gastropods and encrusting foraminifera. (2) Branching coral facies: a major component of the growth framework and composed mainly of Acropora, Pocillopora, and branching Porites. This facies has a very high primary porosity. (3) Coral head facies: also a major component of the growth framework, comprising mainly Porites, Goniopora, Favia, Platygyra, Symphilia, and massive Acropora. (4) Rubble facies: a poorly sorted unit either cemented or loose and dominated by fragmented branching corals. (5) Sand facies: a better-sorted unit often fining upwards and dominated by Halimeda and foraminiferal constituents. (6) Terrigenous facies: siliciclastic sands forming the foundations of inner fringing reefs, but also including clay infills in inner shelf reefs within the influence of river plumes.
These facies descriptions apply to the drill cores presented in Chapters 7, 8, and 9. 11.8.2 Vertical changes in growth patterns Although great heterogeneity exists in reef core lithology, even in holes only meters apart, some general up-hole trends have been recognized (Davies and Hopley, 1983; Marshall, 1985; Kleypas, 1991, 1996; Kleypas and Hopley, 1993): *
*
on higher-energy reefs coral assemblages may form the basal Holocene unit, head corals to windward, branching corals to leeward elsewhere immediately overlying the Pleistocene the basal unit frequently consists of a coarse branching coral detrital assemblage on high-energy reefs grading to a muddy sand and gravel mix beneath lower-energy or fringing reefs
11.8 Holocene reef structure and facies development *
* *
387
the bulk of the lower 50% of the reef column may be detrital, consisting of gravels then coarse sands fining upwards, but with coarser rubble becoming more prominent again in the upper sections in situ coral framework also becomes more prominent in the upper 50% the coralline algal facies may dominate the uppermost part of the Holocene reef.
Such patterns can be seen in most of the core logs presented in Chapters 8 and 9.
11.8.3 Intra-reef variations Using One Tree Reef as an example Marshall and Davies (1982) recognized three phases of reef development each of which affects the types of material laid down in different parts of the reef. (1) After drowning, an initial phase from 8 to 6 ka ago of mainly vertical growth with framework of encrusted coral heads to windward, encrusted branching corals to leeward. (2) An intermediate phase 6 to 4 ka, during which the coral reef continued to accrete vertically with patch reefs and windward margins developing coralline cappings as they grew up towards the stabilized sea level. Dominantly branching corals developed on the windward reef flat with an encrusted coral head facies developing on the leeward margin. Lagoon sedimentation commenced during this phase. (3) A final phase over the last 4 ka in which most growth is lateral as sand sheets fill in lagoons and back reef areas, lagoonal reefs increase in number and size and reef flat assemblages extend leewards to cap previous sediment accumulation zones.
This facies model is reflected in the evolutionary reef classification scheme of Hopley (1982) (Fig. 5.7, Table 5.3) and the associated discussion of the development of reef zonation through time (Hopley, 1982, p. 313). However, as has been noted in Chapter 8, the depth of the Pleistocene antecedent surface strongly influences the stage that any particular reef has reached. Thus, although many parts of the sequence outlined for One Tree Reef by Marshall and Davies (1982) have more general application, the time-frame is much more flexible and reefs with associated facies development at any stage in the sequence may be found on the GBR. Other variations for the One Tree model have been noted. For example, Marshall (1985) pointed out that in the central GBR reefs (and elsewhere) there is a greater development of the coral head facies beneath the leeward margins, and both here and in the far north (Yonge and Raine Reefs), more low corymbose-type branching corals on the windward side. This appears to contribute to a greater amount of rubble at the base of the Holocene on windward margins on many reefs. Graham (1993) made similar observations
388
The accumulation of the Holocene veneer to the Great Barrier Reef
for a reef transect off Innisfail where Acropora dominated over head corals in cores both on the windward side of individual reefs and on the more highly energetic shelf-edge reefs. A model of cross-reef facies development recognizing these differences is shown in Fig. 11.8. It is based on the growth rates described above. The presumed antecedent surface is at about 20 m, inundated about 9 ka ago. For a high-energy reef a coralgal facies develops over the reef front and even before the crest of the Pleistocene surface is inundated, will be providing rubble to the area behind the crest. As sea level rises, the coralgal zone slowly backsteps over the rubble. Vertical accretion rates are 5 m ka1 for the coralgal material, 4 m ka1 for the basal rubble as indicated by dated cores. Thus by 8 ka ago the sea level was about 12 m, the reef front had lagged behind to be in 3 m water depth and the rubble zone a further 1 m deeper. This possibly allowed some branching coral development, which by 7.5 ka ago is fully developed with the potential to ‘‘catch up’’ with sea-level rise shortly after 6 ka ago. The final reef top veneer may depend on the extension of the coralgal zone into this central or back reef area. Where it does not, the branching corals will form the aligned coral zone commonly found behind a windward margin (see Hopley, 1982, p. 306). After 7.5 ka ago the high-energy reef front will be approaching sea level at a much slower rate, even though the sea-level rise is decelerating. At 7 ka ago it will still be at 9 m and at 6 ka ago, 5 m as its vertical accretion rate slows. The first part of the windward reef to reach sea level may thus be back from what eventually becomes the reef front. Also because the outer front is accreting at a slower rate than the branching coral zone, surplus carbonate productivity over what is required to keep up with sea level may go into the initial formation of the lagoonal (back reef in the case of ribbon reefs) sand slope from about 7 ka ago onwards with the ability to accrete at between 5 and 10 m ka1. For a medium-energy reef, the main difference is that lower ambient energy conditions allow for a proportion of the reef front to consist of branching corals. Accretion rates are higher on this windward margin but as tropical cyclones are still frequently encountered, the reef front is regularly stripped of a higher proportion of its framework. The end result is that the reef is dominated by internal rubble even though it may be capable of a ‘‘keep up’’ strategy. Reef front accretion rates may be as high as 8 m ka1, though part of this will be transported leewards. Rubble facies accretion rates will be at the upper end of the scale – perhaps >5 m ka1. As the reef approaches sea level at about 6 ka ago, across-reef energy gradients are accentuated and the typical zonation patterns develop, producing a coralgal cap on the windward margin, and veneers of head and branching corals to leeward.
11.8 Holocene reef structure and facies development
Figure 11.8 Model of a mid-shelf reef growing from a 20 m Pleistocene surface and in response to sea-level rise in the Holocene.
389
390
The accumulation of the Holocene veneer to the Great Barrier Reef
Further back in the reef, lagoonal reefs may take off at almost any time, taking advantage of highs in the Pleistocene surface or random rubble banks. Dominated by branching corals they may outstrip all other parts of the reef in their upward growth and may be the first to reach modern sea level. Back reef margins develop soon after the windward reef gives some protection and, dependent on the mix of head and branching corals, may reach sea level any time after 6 ka ago, after which the back reef sand sheets expand rapidly. Submarine lithification is also more common beneath windward margins and enhances the facies developed here. For example, in the Bunker–Capricorn Group, cements contemporaneous with reef growth have developed in windward environments characterized by slow accretion rates (Marshall and Davies, 1981). Both crystalline (aragonite and bladed magnesium-calcite spar) and microcrystalline (micrite and pelloids) cements occur within skeletal pores and chambers. On the leeward slopes and within patch reefs, submarine lithification has a more random distribution. Wave-induced or tidal pore water pumping, as described for example by Oberdorfer and Buddemeier (1986), are considered the driving forces for this energy-related cementation pattern. 11.8.4 Cross-shelf variations Energy gradients also exist across the continental shelf and are similarly exemplified in submarine cementation. Marshall (1985) indicates that on the outer reefs such as Viper and Myrmidon cement may be volumetrically greater than framework. Larger mid-shelf reefs have significant cementation only on their windward margins whilst inner reefs show little sign of any cementation. All reefs examined on the outer shelf are framework dominated, especially but not exclusively on the windward margins, with 80–95% framework occurring here on some reefs. Further back on the shelf unconsolidated sediment increases significantly right across the reefs, but especially on the leeward margins (framework only 18–30%). This trend continues to the extent that inner shelf reefs may be described as detrital piles with coral caps. On the central GBR in particular, a steady cross-shelf gradient is exhibited by all these characteristics with relationships between submarine cements and unconsolidated versus framework accretion responding to gradual energy gradients (Marshall, 1985). On the northern GBR where the ribbons occur, there are basically only two environments – the high-energy ribbons dominated by framework and cementation and the much more protected mid- and inner-shelf reefs with rare cementation and high proportion of detrital material and cavities.
11.9 Comparisons with reefs elsewhere
391
11.9 Comparisons with reefs elsewhere The wide range of latitudinal and cross-shelf environments found in the GBR may explain some of the wide variations in, for example, growth rates, facies, and growth strategies, but how representative are these results for reefs elsewhere in Australia and the world? Factors that may produce different results and which were discussed in Hopley (1982, ch. 13) include variations in structure, Holocene sea-level history, biological and ecological factors, and in environmental history during the time the reefs have evolved. Other factors may also affect the growth of reefs. For example, do reefs in areas of high storm activity display greater amounts of detrital material, or do reefs near their latitudinal limit show any decline in growth rates? The following discussion summarizes examples of coral reef drilling results from other parts of the world and makes comparisons with those from the GBR. 11.9.1 Reefs elsewhere in Australia Reefs that have been drilled include the 230-km-long Ningaloo fringing reef of north-west Australia (Collins et al., 2002), the high-latitude (288 450 S) offshore reefs of the Houtman–Abrolhos Group (Collins et al., 1993a, b; Wyrwoll et al., 2006), the even more poleward (338 300 S) fringing reefs of Lord Howe Island (Kennedy and Woodroffe, 2000; Woodroffe et al., 2006), and the nearby atolllike Middleton and Elizabeth Reefs at 298 270 S and 298 560 S (Woodroffe et al., 2004). The latter three sites are all in the Tasman Sea. The Pleistocene was not intercepted at Elizabeth Reef but was at 8.2 m at Middleton. The fringing reef of Lord Howe Island has 5–10 m of Holocene deposition partly over eolianite whilst the major fringing reef of Ningaloo is established over an extensive Pleistocene reef at between 7 and 18.5 m depth. In the Houtman–Abrolhos Islands, Pleistocene reef limestone and eolianite outcrop in the islands but the surrounding Holocene reefs have grown on Pleistocene equivalents, with depths to the Pleistocene varying from 4 to 26 m. These figures are within the range found on the GBR. Dated material near the base of cores at all these sites except Elizabeth Reef comes from just above the Pleistocene and indicates that colonization of the antecedent surface took place shortly after inundation. Oldest uranium series and radiocarbon dates are 9.8 ka (Houtman–Abrolhos), 7.57 ka (Ningaloo), 7.1 ka (Middleton), 6.54 ka (Elizabeth Reef), and 6.24 ka (Lord Howe). Again the figures are in accord with those from the GBR where material from near the base of the Holocene has been dated, and further reinforce the conclusion that there was no narrow take-off window. The Houtman–Abrolhos date from
392
The accumulation of the Holocene veneer to the Great Barrier Reef
Table 11.1. Vertical accretion rates (m ka1) by environment in Australian reefs Ningaloo Windward 1.6 margin Leeward margin Patch reef Lagoon (detrital)
Houtman–Abrolhos Lord Howe Elizabeth
Middleton
3.3–10.2
2.64
2.0–4.2
6.0
0.93 2.5–4.24
7.7 1.0–1.5
1.25
10.0
a depth of 25 m is remarkable as the oldest from a postglacial reef in Australia. The times at which these reefs reached sea level is also within the range recorded for the GBR: 3.2 ka (Ningaloo), 6.5 ka (Houtman–Abrolhos), 4.9 ka (Lord Howe), 4.1 ka (Middleton), and 5.08 ka (Elizabeth Reefs). The Houtman–Abrolhos reef flat was developed as soon as sea level stabilized as it was effectively growing as a fringing reef over the emerged Pleistocene reef which forms the islands. All these reefs are relatively high-energy locations and, as on the outer reefs of the GBR, framework appears to dominate their structure, at least on the outer edge. Table 11.1 indicates the range of accretion from these reefs and again they are within the range for equivalent environments on the GBR. The windward sites produce rates between 0.93 and 4.2 m ka1 with a special exception of the lower part of the Houtman–Abrolhos cores where accretion rates for the lower 10 m of one core exceeded 10 m ka1. These high rates are associated with Acropora-dominated reef structure which Blakeway (2000) and Wyrwoll et al. (2006) suggested grew so quickly and openly that it produced collapse patterns in the Houtman–Abrolhos reefs which have been previously interpreted as karst blue holes. The rate for patch reef growth on Houtman–Abrolhos is also high but is identical to that from the GBR (Fig. 11.5). The wide range of lagoonal sedimentation rates may again reflect the differences between regular lagoonal infill and storm events. 11.9.2 Indo-Pacific atolls A very large proportion of the drilling carried out on Indo-Pacific reefs has been on volcanic islands or atolls overlying volcanic foundations which are part of island chains moving away from mid-oceanic hotspots (active volcanoes) (Scott and Rotondo, 1983). As they move on their oceanic plates they are subjected to a range of vertical movements which have been recorded in the
11.9 Comparisons with reefs elsewhere
393
coral caps of atolls and in reefs around volcanic cores. These movements include cooling subsidence, lithospheric loading, movement over asthenospheric bumps, and subsidence in subduction zones at the margins of the oceanic plates. Other islands, such as New Caledonia, may be unstable even though they are formed of continental crust, as they may also lie close to active plate margins. This is a big contrast to the GBR located in a mid-plate location and questions arise as to the influence on coral reef development of a tectonically more active location. Perhaps surprisingly, for the majority of atolls drilled, the thickness of the Holocene cap (or depth to Pleistocene reef) is in every case within the range found on the GBR, e.g., 8–10 m (Enewetak), 31 m (Funafuti), 6–11 m (Mururoa), 6–12 m (Cocos–Keeling), 8–14 m (Tarawa), and 7–22 m on the Cook Islands (Aitutiki, Pukupuka, and Rakahanga) (McLean and Woodroffe, 1994; Kayanne et al., 2002; Yamano et al., 2002). Unfortunately many of these atolls were drilled for purposes other than elucidating the Holocene record and few have reliable data on which to base growth rates or interpret facies changes. Exceptions are Cocos–Keeling and Tarawa. Cocos–Keeling (Fig. 11.9a) in the eastern Indian Ocean is the site of more than 20 deep boreholes (Woodroffe et al., 1991), seismic profiling (Searle, 1994), vibro-coring of lagoonal sediments (Smithers et al., 1993; Smithers, 1994), and more than 75 radiocarbon dates (Woodroffe et al., 1994). Also available is a comprehensive coverage of the geomorphology and ecology of the atoll (Woodroffe, 1994). The atoll has a characteristic shallow circular lagoon surrounded by a series of reef islands containing conglomerate platforms and in situ microatolls indicating a higher sea level of about 1.0 m 3000 years ago (Woodroffe et al., 1991). The oval atoll has a maximum diameter of about 17 km and an area of 190 km2, i.e., about the same size as the larger of the GBR reefs (Corbett Reef, 207.5 km2). The Pleistocene antecedent surface, dated at about 120 ka, is consistently between 6 m and 16 m below the rim of the atoll and about 35 m depth beneath the lagoon (Woodroffe et al., 1991; Searle, 1994). Holocene reef appears to have been established on the antecedent surface very quickly after inundation with radiocarbon ages of 7480 years BP and 7240 years BP at depths of 14 m and 9 m respectively, just over the Pleistocene. However, later colonization beneath parts of the atoll is also indicated. Vertical reef accretion appears to have been rapid with dates related to modern sea level or above including 4460 years BP for conglomerate platform materials and 4010 years BP for in situ microatolls. Rates of accretion range from 4 to 40 m ka1 in the cores, with the highest rates clearly associated with branching coral assemblages or the detrital facies developed from them.
394
The accumulation of the Holocene veneer to the Great Barrier Reef
Figure 11.9 Indian Ocean reefs. (a) Cocos–Keeling; (b) the extensive fringing reef off the west coast of Rodrigues Island.
McLean and Woodroffe (1994) developed a three-stage model for the Cocos atoll which has similar characteristics to the model of Marshall and Davies (1982) for One Tree Reef: an initial phase of rapid vertical accretion to about 4.5 ka ago, a period of catch-up growth; a period of reef flat development 4.5–3 ka ago; and a final period of island building. They indicated that this model was applicable to other atolls for which similar data were available but noted that the time of ‘‘catch-up,’’ i.e., reef flat development, may vary from
11.9 Comparisons with reefs elsewhere
395
the mid-Holocene to about 2 ka ago, dependent on the depth of the Pleistocene surface but also the timing of the achievement of local sea level which becomes a major factor when data from the wider Indo-Pacific are examined. Hopley (1987) for example shows how the attainment of modern sea level becomes younger across the South Pacific, from Australia (>6 ka ago) to the Cook Islands and Society Islands (6 to 4 ka ago), to the Central Pacific including the Line Islands (less than 4 ka ago). Obviously this will have an influence on reef development giving more opportunity for the ‘‘catch-up 2’’ mode of development in the final vertical growth phase. McLean and Woodroffe (1994) suggest that reefs in a ‘‘catch-up’’ mode are more likely to be dominated by head corals, whilst ‘‘keep-up’’ reefs are predominantly branching framework. Certainly the core data from Cocos atoll suggest this as the shallow depth of the Pleistocene and rapid vertical accretion rates indicate that the reefs were tracking sea-level rise by little more than 2 m for most of the Holocene growth period. Cocos–Keeling also has good records of lagoonal sedimentation rates. Smithers et al. (1993) and Smithers (1994) indicate that lagoonal infill was largely post 4 ka ago with initially shingle accumulating at rates of about 1.5 m ka1 based on their data, i.e., post sea-level stabilization and the reef reaching sea level. Subsequently there have been two types of sedimentation, in the form of a muddy facies in the lee of the islands accumulating at between 0.25 and 0.5 m ka1 and sandy aprons accumulating at between 0.5 and 1.0 m ka1. Relatively low rates of lagoonal infill are reported for other atolls including 1–3 m ka1 for Bikini and Enewetak (Thurber et al., 1965; Tracey and Ladd, 1974). Tarawa, one of the Gilbert Islands, is just north of the equator in the western Pacific. It is triangular in shape, open to the west, and much larger than Cocos–Keeling at 500 km2. Its lagoon has a maximum depth of 20–25 m whilst beneath the islands of the southern and eastern sides of the atoll there is a rim of Pleistocene reef 8 to 14 m beneath present sea level and dated at about 125 ka ago (Marshall and Jacobson, 1985). This is one of the few instances in which Halimeda or Halimeda casts have been recorded. Above the Pleistocene a uranium series age of 8.6 ka was obtained and a radiocarbon date of 7750 years BP. Marshall and Jacobson do not report any reef flat ages but a date of 5680 years BP within 3 m of sea level and 6200 years BP within 4 m suggests that the reef took off soon after the Pleistocene surface was inundated and adopted a ‘‘keep-up’’ strategy, with modern sea level achieved some time prior to 4.5 ka ago. The facies sequence described is relatively simple, a base of mainly coral with some sand and coralline algae 2–12 m thick with an upper cemented cay rock unit or unconsolidated sands and gravels. Accretion rates for the windward
396
The accumulation of the Holocene veneer to the Great Barrier Reef
reef flats are generally between 5.0 and 5.4 m ka1 reaching a maximum of 8.2 m ka1. Marshall and Jacobson (1985) suggest that the top sedimentary unit reflects deposition during a stable sea level and equate the interpretation to some of the early drilling on the GBR, e.g., on One Tree Reef (Marshall and Davies, 1982). 11.9.3 Pacific barrier and fringing reefs A number of sites have been drilled in the Pacific. Four are considered here, the bayhead fringing reefs of Hanauma Bay, Hawaii, the more extensive reefs of New Caledonia, the Darwinian-style barrier reef around the volcanic island of Tahiti, and a similar but possibly more mature barrier reef off Palau. Initially the results from Hanauma Bay (Fig. 11.10a) were interpreted as representing a sea-level curve for the Holocene (Easton and Olson, 1976). Grossman et al. (1998) amongst others rejected this as reef growth probably lagged behind sea-level rise. Grigg (1998) has reviewed the initial interpretation of this reef which was established in a Pleistocene volcanic crater and commenced growth about 7 ka ago (earliest radiocarbon date 7010 years BP). The reef reached sea level about 3 ka ago (earliest reef flat date 3170 years BP). The depth of the Holocene foundations is about 15 m near the outer edge of the reef and <10 m at the bay head. The lower 5–10 m consists of coral, presumed by Easton and Olson (1976) to be largely in growth position. Using their dates this basal unit accumulated at rates between 2 and 8 m ka1. The upper unit is dominated by calcareous algae, as is the present reef crest. Accretion rates for this range from 1.25 to almost 3 m ka1. Grigg (1998) suggests that Hawaiian reefs generally are growing slowly (about 2 m ka1) due mainly to the high ambient wave energy which restricts reef growth around the islands to sheltered locations only (see below). The information for New Caledonia (Fig. 11.10b) comes from single boreholes for eight fringing reefs around the island (Cabioch et al., 1995). In all, 49 cores were recovered, all containing the entire Holocene sequence between 3 and 10.5 m in thickness. The outermost parts of the reef displayed greater proportions of framework and greater lithification. It was here that reef growth first started and greatest recovery made by Cabioch et al. (1995). Their detailed analysis refers to the outermost core from each of the eight sites. Pleistocene fringing reefs appear to provide a foundation for Holocene growth on the south-east and south-west coasts at depths between 3 and 13.5 m. Elsewhere, the foundations are on local bedrock. Inundation took place between 8 ka and 7 ka ago (oldest quoted date 7160 years BP). Reefs were reaching sea level shortly after 5 ka ago (oldest date 4910 years BP),
11.9 Comparisons with reefs elsewhere
Figure 11.10 Pacific Ocean reefs. (a) Hanauma Bay, Oahu, Hawaii; (b) New Caledonia, west coast near Poya (photograph: B. Bouye´); (c) Tahiti.
397
398
The accumulation of the Holocene veneer to the Great Barrier Reef
having adopted either a catch-up 2 or catch-up 3 strategy, i.e., reef flats developing after sea level stabilized, or in some cases the whole Holocene reef taking off and accreting after sea level stabilized. Growth rates ranged from 0.68 to 21.2 m ka1 with mean rates between 1.4 and 21.1 m ka1. Both growth rates and facies distribution are strongly influenced by wave energy. High wave energy sites are dominated by framework, medium energy by detritus, which, as Cabioch et al. (1995) note, is similar to what is found in the GBR. Within the framework facies, branching corals dominate at depth, head corals and crustose corallines closer to the surface, also interpreted as a shallowing upwards sequence. Cabioch et al. (1995) note that early Holocene growth was not synchronous around New Caledonia, being earliest in the south, mainly on karst foundations, but with delays of as much as 2.5 ka elsewhere. They also conclude that reef growth was sudden and quick, without the slow take-off phase previously identified for the GBR (Davies and Marshall, 1979). These results are similar to those from the reanalysis of GBR data discussed above. The results from Tahiti (Fig. 11.10c) are unlike those from anywhere else in the world. The thickness of the postglacial reef is between 85.7 and 92.8 m overlying either Pleistocene reef or volcanic foundations (Montaggioni et al., 1997; Cabioch et al., 1999). Recolonization of this surface had taken place by 13.8 ka ago and up to 11 ka ago may have been in deeper water with the basal facies dominated by Porites and gracile Lithophyllum communities. During the last 11 ka reef framework dominated by Acropora and coralgal assemblages appears to have kept up with the rapid sea-level rise with accretion rates between 9.3 and 20.6 m ka1. Only after the reef reached sea level about 6 ka ago did the adjacent back reef deposits start to accumulate. The continuous growth record of the Tahiti reef can be compared to the similar environment of the barrier reef around Palau. Whilst the Tahiti reef is separated from its volcanic island by a lagoon 2 km wide, the more mature Palau reef has a 15-km-wide lagoon, with a maximum depth of 50 m (Kayanne et al., 2002). The Pleistocene reef which underlies the present barrier is at a depth of 15.7–25.3 m, much shallower than in Tahiti and more comparable to the GBR. Red soil partly fills the cavities at the top of the antecedent surface. Holocene growth had begun by 8.3 ka ago and as in Tahiti was immediate and rapid (basal growth rates of between 3.1 and 14.2 m ka1). The reef had reached modern sea level by 4 ka ago (earliest date 3950 years BP). Initially the reef was dominated by branching Acropora which accumulated at rates exceeding even those in Tahiti (30 m ka1). However, once within the wave zone a more robust Acropora and coralgal assemblage became established accreting at a rate of only 3 m ka1. This catch-up phase appears to have
11.9 Comparisons with reefs elsewhere
399
commenced at about 7.2 ka ago and is related to a local Holocene sea-level rise which slowed significantly at this time and reached modern sea level only between 5 ka and 4 ka ago. This has had a significant influence on reef development. Lagoonal infilling also seems to be related to the barrier reef approaching sea level, with sand and gravel accumulating behind the reef crest from approximately 7.4 ka ago. 11.9.4 Indian Ocean reefs Camoin et al. (1997) have summarized the data from drilling through the reefs of Mauritius, Re´union, and Mayotte and again a contrasting Holocene sea level has had an influence on reef development. A rapid rise took place between 10 ka and 7.5 ka ago, at which time there was a significant slowing down with modern sea level achieved between 3 ka and 2 ka ago. More recently Rees et al. (2005) have reported similar results from Rodrigues Island 500 km east of Mauritius (Fig. 11.9b). Reefs on Mauritius, Rodrigues, and Re´union are fringing, with that on Mayotte being the largest barrier reef in the southwest Indian Ocean. These reefs are also established on Pleistocene reefal limestone at depths of 16.3 m on Mauritius, 21.5 m on Mayotte, and between 20 and 25 m on Re´union. Earliest colonization dates more or less reflect the depth differences (8.2 ka ago Mauritius (though a date of 9.6 ka ago is shown on fig. 3 of Camoin et al., 1997); 9.6 ka ago Mayotte; pre 8 ka ago Re´union). However, all these reefs reached modern sea level only after 2 ka ago, a probable reflection of the local sea-level curve. Average growth rates were about 2 m ka1 though Mauritius and Re´union show the S curve proposed by Davies and Marshall (1979). Initial rates are only 1 m ka1, then reaching a maximum of 2.55–4.73 m ka1 before decelerating to 1.9–2.61 m ka1 as these reefs approached sea level. Growth of the Mayotte barrier reef, however, was sudden and rapid, adopting a keep-up strategy throughout its history. Growth rates to 7 ka ago were about 8.3 m ka1 but then slowing to 1.14 m ka1 in response to rates of sea-level rise. Drilling on the fringing reef of Rodrigues was to only 4 m with the oldest date obtained being 3 ka ago. Average growth rates in this upper section of the reef were relatively slow, between 0.46 and 1.96 m ka1, though the windward side comprises mainly branching corals with significant amounts of coralline algae. Camoin et al. (1997) commented on the variable facies displayed by the reefs noting that as in the GBR the higher-energy reefs were framework dominated. There was little vertical variation in facies at the three sites. Mauritius and Re´union fringing reefs were mainly branching framework though with some
400
The accumulation of the Holocene veneer to the Great Barrier Reef
head coral assemblages in the Mauritius core. Both had an increase in encrusting red algae in the upper portions of the core. The Mayotte core was also dominated by branching corals throughout (mainly Acropora dana/robusta) with some Goniastrea head corals near the base. 11.9.5 The Caribbean There are many contrasts between Caribbean and Indo-Pacific reefs. There is little overlap in species, the number of coral species is significantly lower in the Caribbean, and, from a geomorphological viewpoint, the Holocene sealevel history is in complete contrast, with sea level rising continuously up to the present (for discussion with respect to coral reefs see Hopley, 1982, ch. 13, 1985; McLean and Woodroffe, 1994; Dullo, 2005). Nonetheless, simultaneously Neumann and Macintyre (1985) and Davies et al. (1985) produced almost identical models for Holocene reef development. This final regional synthesis attempts to identify the similarities and differences in reef growth in these two most contrasting environments of the GBR and the Caribbean. Within this large region the foundations of the Holocene reefs, many of which are fringing reefs, are often non-carbonate, but Pleistocene reefal limestone does underlie all types of reefs at depths that equate to reefs elsewhere. It is relatively shallow beneath the atolls off Belize: 9.1–11.7 m at Glovers Reef; 7–9 m at Lighthouse Reef; and 3.1–3.8 m at Turneffe Islands (Gischler and Hudson, 1998; Gischler, 2003). Beneath the nearby reef of Belize (Fig. 11.11a), Pleistocene outcrops at the surface in the north in Ambergris Cay but gradually gets deeper to the south where it is >18 m below the surface (e.g., Macintyre et al., 1982; Shinn et al., 1982). A range of depths to Pleistocene reef are found beneath bank barrier systems, e.g., 10.7 m beneath the Abrolhos Reef in Brazil (Leao and Ginsburg, 1997), 20 m at St Croix (Macintyre, 1988), and in a deepening sequence offshore in Puerto Rico, 12–16 m on the inner shelf, 16–24 m mid shelf, and 28 m at the shelf edge (Hubbard et al., 1997). The deepest Pleistocene reef foundation and thickest Holocene section recorded in the Caribbean is at Alacran Reef, Mexico (Macintyre et al., 1977). The more extensive data now available further strengthen the original reefbuilding model of Neumann and Macintyre (1985). They suggested that ‘‘keepup’’ reefs are established in paleo water depths of < 15 m, when sea level in the Caribbean was beginning to slow down. Whilst an initial rapid phase of growth dominated by Acropora cervicornis may take place, once within 5 m of the surface very soon after initiation, a framework of robust A. palmata
11.9 Comparisons with reefs elsewhere
401
Figure 11.11 Caribbean reef. (a) The Belize barrier reef; (b) Acropora palmata at the reef crest (photographs: I. G. Macintyre).
forms the bulk of the reef (Fig. 11.11b). They may have a rubble cap and a back reef zone consisting of algal or microbialite reefs (Reid et al., 1995, 1999). Steneck et al. (1998) conclude that they represent a late successional stage of Caribbean reef development forming a constructional cap over coral as the ‘‘keep-up’’ reef approaches sea level. This cap can be from 1 to 8.5 m in
402
The accumulation of the Holocene veneer to the Great Barrier Reef
thickness. There is no equivalent to this back reef zone in the GBR. Reefs identified as having a ‘‘keep-up’’ strategy include Galeta Point in Panama (Macintyre and Glynn, 1976), Nonsuch Bay, Antigua, and the main Belize barrier reef (Macintyre et al., 1982). ‘‘Catch-up’’ reefs are dominated by the open framework produced by the rapidly growing A. cervicornis. The Alacran Reef, growing from a 33.5 m foundation, is regarded as the archetype. Its vertical accretion rate reached a maximum of 12 m ka1 until it caught up with the still rising sea level about 3 ka ago, since when its growth rate has been about 2 m ka1 (Macintyre et al., 1977). Neumann and Macintyre (1985) also believe that ‘‘catch-up’’ reefs are not uncommon in shallow water on bank tops or bank margins where the decrease in the rate of sea-level rise from 6 to 3 ka ago allowed time for reef initiation and growth before water depths were excessive. For all intents and purposes these reefs equate to the ‘‘catch-up 3’’ signature of Davies et al. (1985), i.e., reefs that had their full history post the stabilization of sea level, or in the Caribbean examples, post the period of rapid sealevel rise. Examination of other core records from the Caribbean confirms general conservative figures for rates of vertical accretion controlled over the last 6 ka by the decelerating sea-level rise. Only on reefs with deep foundations was there ever enough accommodation space available, at a time when sea level was still rising at rates of 5 m ka1 or greater, for significant deposition at rates at or exceeding 5 m ka1. On the three atolls of Belize, depth of the Pleistocene is clearly reflected in Holocene growth rate. At Glovers Reef (Pleistocene 9.1–11.7 m) the average rates are 1.22–1.49 m ka1, maximum 8.28 m ka1, at Lighthouse Reef (Pleistocene 7–9 m) the rates average 1.28 m ka1 with a short maximum of 22.86 m ka1, and at Turneffe (Pleistocene 3.1–3.8 m) average rate is 0.65 m ka1, maximum 3.7 m ka1 (Gischler and Hudson, 1998). A similar pattern is seen in the progressively deeper foundations offshore from Puerto Rico. Inner shelf rates are largely dominated by non-carbonate deposition but mid-shelf reefs growing from a Pleistocene surface of 16 m to 24 m had average accretion rates of 1.5–3 m ka1, and outer shelf reefs (Pleistocene 28 m) average rates of 2.05–6.1 m ka1 reaching a maximum of 10 m ka1 (Hubbard et al., 1997). Reef flat ages also reflect the regional sea-level curve and are generally less than 3 ka in the upper 1 m of the reef. In contrast, the time of reef initiation has a wide range controlled very closely by the depth to the foundation and time of inundation. Oldest dates came from Puerto Rican reefs (over 9500 years BP) at depths of about 12 m (Hubbard et al., 1997). Beneath the Belize atolls, dates range from 7370 to 5470 years BP, determined very closely by sea level and
11.10 How does the Great Barrier Reef compare?
403
level of antecedent surface. A range of up to 5 ka is in contrast to the narrow range previously envisaged for the Indo-Pacific. This range would be extended even further if the ‘‘give-up’’ reefs in deeper water in the Caribbean were included. Blanchon and Jones (1995) suggest that a large number of reefs started to grow about 11 ka ago at a depth of 30–40 m. These reefs tracked sea-level rise until about 7 ka ago when they were at a depth of about 18 m with a sudden die-off event causing backstepping to new upslope positions. Blanchon attributes this catastrophic event to a sudden sea-level rise associated with ice-sheet collapse (Blanchon and Shaw, 1995). However, Hubbard et al. (1997) could not relate backstepping events in Puerto Rican reefs to sea-level rise but rather favor a sediment stress mechanism. Here too reefs were established by 9.5 ka ago with the major backstepping event from shelf edge to mid shelf also taking place about 7 ka ago. The scale and morphology of these reefs is very different to that of the drowned shelf-edge reefs of the central GBR (Chapter 9), which are located in most instances in much deeper water. Nor are there equivalent structures around the shelf reefs or fringing reefs. The catastrophic event that affected Caribbean reefs about 7 ka ago also seems not to have affected the GBR as this was the time of maximum growth. The Caribbean event would appear to have resulted from regional rather than global causes. 11.10 How does the Great Barrier Reef compare? The cross-shelf pattern of fringing, mid, and outer shelf reefs described in Chapters 7, 8, and 9 and analyzed earlier in this chapter shows a Holocene evolution which is comparable to that in all other regions of coral reef growth. As in the GBR, all reefs other than fringing or bank barrier appear to be underlain by Pleistocene reefal limestones, uranium-series dates on which suggest a last interglacial age (120–140 ka) (Table 11.2). Many fringing reefs are also underlain by reefal limestones as are some in the GBR. Soils and diagenetic characteristics are similar in most locations though caliche or calcrete stringers in the upper layers do not appear to be as common under GBR reefs. However, from available data, a special feature of the GBR is the layer of lightly cemented Halimeda flakes of Pleistocene age found beneath some northern reefs. Just as extensive Halimeda banks may have restricted distribution in the Holocene, so too it seems did similar features during the last interglacial high sea-level stand (see Chapters 6 and 12). The depth to Pleistocene reef on the GBR was established at between 0 and 30 m. Away from regions that are clearly tectonically active and in which
404
The accumulation of the Holocene veneer to the Great Barrier Reef
Table 11.2. Uranium and strontium ages for Pleistocene reefs immediately below the Holocene Location
Age (ka)
Depth (m)
Australia Southern GBR Ribbon 5
145 – 122 125.7
7.4 to 14.3 75 to 8
Houtman–Abrolhos 141.2 1.2 – 117.1 1.6 4.3 to þ2
References Marshall (1983) Braithwaite et al. (2004) Collins et al. (1993a, 2002)
Ningaloo Indian Ocean Cocos–Keeling
134 1.4 – 115 2
8 to 36
123 7 – 118 7
8 to 11
Woodroffe et al. (1991)
Pacific Ocean Tarawa
125 9
10.5 to 17.8
Enewetak Funafuti
136 8 – 128 7 140 150
14 to 20 26.4 to 27.4
Mururoa
120–150
2 to 18
Marshall and Jacobson (1985) Szabo et al. (1985) Assaoui et al. (1990), Ohde et al. (2002) Yokoyama and Nguyen (1980)
Pleistocene reef may be found many meters above present sea level, this depth distribution would appear to incorporate all reefs investigated other than Alacran (33.5 m) and Tahiti (85.7 m). Alacran is only marginally outside the GBR range and is shown by Macintyre et al. (1977) to be the edge of the Isla Perez reef where the Pleistocene slopes away into deeper water. Subsidence of the Campeche Bank since the Pleistocene has also been demonstrated (Macintyre and Aronson, 1997). The sloping volcanic margins of Tahiti appear to have found a substrate appropriate for reef development whatever the rate of sea-level rise 14 ka ago. The depth to the Pleistocene may be an artifact of the site drilled. However, Tahiti is part of a volcanic hotspot chain along which vertical movements may be expected (Scott and Rotondo, 1983). Volcanic loading of the oceanic lithosphere (Lambeck, 1981a, b) is responsible for a significant subsidence rate of 0.4 m ka1. This would lower a lastinterglacial (120 ka old) reef by about 48 m, still only half the depth of the Pleistocene reef beneath the Tahiti barrier, but possibly a significant factor in its current depth.
11.10 How does the Great Barrier Reef compare?
405
The age of recolonization of the GBR has already been shown to be wider than previously calculated, taking place between 9.9 ka and 6.3 ka ago. Similarly wide envelopes are being identified elsewhere: between 9.8 and 6.24 ka ago elsewhere in Australia; between 13.08 and 7.01 ka ago in the Pacific; between 9.6 and 8.2 ka ago in the limited sample from the Indian Ocean; and between 9.5 (older if ‘‘give-up’’ reefs are included) and <4 ka ago in the Caribbean. The major determining factor for the age of recolonization is the depth to the underlying foundation whether it be reefal or other lithologies. The correspondence between depth of antecedent surface and age of take-off is not perfect but this would not be expected as the timing of the event may be as much dependent on the local Holocene sea-level transgression curve as is the timing of reef flat initiation (see below). Data on sea level for the start of the Holocene, however, are not as reliable as they are for the mid Holocene. If Australia and the Caribbean are taken as the extreme for non-tectonically active tropical locations and the sea-level curves of Thom and Chappell (1975) taken as representing Australia and Lighty et al. (1982) and Toscano and Macintyre (2003) representing the Caribbean, it can be seen that at the start of the Holocene (10 ka ago) the difference may be minimal, in the order of 27.5 m to 30 m. However, whilst modern sea level was achieved by 6.5 ka ago in Australia, in the Caribbean it was still 8 m below present at the same time. Timing of reef flat development and the characteristics of the facies found in the upper levels of the reefs of the two areas are considerably different (Fig. 11.12). Reefs in other areas of the Indo-Pacific where the sealevel signature may be intermediate between the two extremes similarly show reef flat and facies characteristics reflecting this intermediate position. For the GBR, the reef flat age range was 6.46 to 3.04 ka and for other Australian reefs 6.5 to 3.2 ka. In the Pacific the range is similar from 6.0 to 3.1 ka but careful examination of reef core and supratidal deposits related to a higher Holocene sea level shows regional patterns related to large-scale global sea-level variations (Hopley, 1987; Grossman et al., 1998) and more local hotspot loading anomalies (Pirazzoli and Montaggioni, 1988) to which reefs clearly responded. Comparison of growth rates is more difficult as not all reports differentiate between framework construction and detrital deposition (e.g., Dullo, 2005). Most results are within the ranges established for the GBR with the reef at Tahiti showing the most rapid framework accretion until it was approaching the sea surface. Rates ranged from 9.3 to 20.6 m ka1. In the Caribbean the most rapid accretion rate for the thick Holocene reef at Alacran was 12 m ka1. High rates (>15 m ka1) in other reefs all appear to be related to detrital, probably storm deposition, although rates of 30 m ka1 for the base of the Palau reef are identified as being associated with both rapidly growing
meters
2
3
4
5
6
30
28
26
24
22
20
8
8.5
8.0
7.0
CATCH-UP REEFS
3.0 5.0 6.0
10 ka
ll (1 happe 83) and C Thom and Roy (19 Thom 975)
Robust branching corals (A.palmata)
3
4
5
6
7
8
10
Rapid growing
Inundated about 8–9 ka
Pleistocene
Basal rubble zone whilst still close to S. L.
Acropora cervicornis dominated
CATCH-UP REEFS
9
ka
7.0
6.0
5.0
4.0
3.0
3.0
1.5
Figure 11.12 Contrasting sea-level curves of (a) Australia and (b) Caribbean, reef response, and subsequent reef structure.
Rubble
2
Microbialite reefs Shallow reefs flats with KEEP-UP REEFS vertical accretion continuing and prograding algal reefs Acropora palmata dominated
1
Fragile branching corals incl. A.cervicornis
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
(b) 0
Head and encrusting corals
Algal crust
Formation of basal rubble zone whilst still close to S. L.
Fragile branching framework High accretion rate
Pleistocene Inundated about 8.5–9.0 ka
9
velope S. L. en tralian s u A t s Ea
18
16
14
12
10
8
6
7
Rubble as reef approaches S. L.
Lateral reef growth Algal crusts 2 wide sediment Robust head covered reef and encrusting 4 flats post 5 ka corals
1
meters
(a) 0
) (2003 tyre acin ) M and l. (1982 ano ta osc ghtly e T i e L v cur curve L. S. S. L.
11.10 How does the Great Barrier Reef compare?
407
Acropora and bioclastics (Kayanne et al., 2002). Other conclusions that can be made from the datasets and which are shared by the GBR include: *
*
*
*
*
*
It is possible that more reefs display a rapid take-off mode once the antecedent surface is inundated than the original S curve model of Davies and Marshall (1979). Examples of the former strategy include the Houtman–Abrolhos reefs, Cocos–Keeling, New Caledonia, Palau, and Mayotte. Slow take-off strategies have been identified at Alacran, Mauritius, and Re´union. Most rapid framework growth is associated with branching corals growing at depths greater than 6 m below the paleo sea level. Not only do these have the maximum calcification rates of about 10 kg m2 yr1 (Kinsey, 1985) but their structure is very open. As reefs approach sea level their accretion rate may slow to <2 m ka1. Calcification rates may still remain high but the framework laid down may be far more dense, consisting of head corals and coralline algae in the Indo-Pacific and changing from Acropora cervicornis to A. palmata in the Caribbean. Also some of the productivity may be transposed laterally to infill lagoons or back reef sediment aprons. Reef size is important in determining the detrital components of reefs, especially away from windward margins. On a small reef the ratio of productive margin to volume of lagoon or back reef is high and the distance from reef margin to lagoon is small. Sedimentation rates are thus high (>5 m ka1) and materials laid down generally dominated by coarse gravels. On very large reefs, typified by the atolls, the ratio of margin to lagoon is low, sedimentation rates are slow (<1 m ka1), and the material laid down consists of sand or mud. These patterns may apply to some of the larger (>100 km2) reefs of the GBR. Wave energy has a very strong influence on rate and type of reef growth. The pattern of high-energy areas producing framework-dominated and mediumenergy detrital-dominated reefs (as shown from the GBR) appears to apply to many other reefs of the world. However, there may also be another category of ultra high energy associated either with persistent large swell environments or areas hit frequently by cyclones or hurricanes. In such areas reefs may be largely absent (even though coral growth is present) or limited to sheltered environments and very slow (2 m ka1) growth rates as in Hawaii (Grigg, 1998), or reefs may be prevented from reaching sea level (Macintyre and Adey, 1990). Around Grand Cayman, hurricanes may not only determine the location of fringing reefs but also the internal rubble facies (Blanchon et al., 1997). To date such a scenario has not been identified on the GBR. Reefs growing from shallow foundations, especially in the Caribbean, are more likely to display a ‘‘keep-up’’ strategy whilst those from deeper foundations are more likely to ‘‘catch up’’ with sea level after it has slowed or stabilized. Tahiti is a possible exception but during its rapid growth phase it was never within 6 m of sea level and in strict terms, did not actually catch up until 6 ka ago.
408
The accumulation of the Holocene veneer to the Great Barrier Reef
11.11 Conclusion At the start of Section 11.9 four factors were identified as possibly influencing the Holocene growth history of coral reefs around the world: structure, sea level, ecology, and environment. Two of these factors emerge as being extremely important, namely structural variations and sea-level history, but the remaining two have also played their part in determining the morphology and internal structure of modern reefs. The depth of the antecedent reefal surfaces from which the majority of nonfringing reefs grow has such a range (almost 100 m as identified here) that explanations such as differences in the height to which last interglacial reefs grew or differences in the amount they have been lowered by erosion during 100 ka of exposure are unlikely. It is possible that these antecedent reefs belong to one of the interstadials of the last glacial but where dated, most are clearly of last interglacial age (Table 11.2), and why an interstadial reef would apparently not take advantage of prior reefal foundations is difficult to explain. As McLean and Woodroffe (1994) noted ‘‘differences in the elevation of the last interglacial surface between islands in a similar climatic setting (i.e. Aitutaki and Atiu which are <200 km apart), must result from vertical movements and cannot be entirely ascribed to solution’’ (McLean and Woodroffe, 1994, p. 289). Moreover, over the Pacific in particular, the variation in the level of the last interglacial reef, whether submerged beneath modern reefs or emerged as ‘‘makatea’’ islands, fits logical patterns related to flexure of the ocean lithosphere in response to, for example, volcanic island loading (Lambeck, 1981a, b). In the Caribbean, the depth of the Pleistocene beneath the Belize barrier reef deepens from <5 m in the north to over 25 m in the south and is clearly linked to subsidence (Purdy, 1998; Purdy et al., 2003), and as elsewhere, the depth of the antecedent surface demonstrably affects modern reef morphology and internal structure. These conclusions would appear to support a tectonic subsidence hypothesis for the greater depth to Pleistocene foundations in the central GBR. Whilst structure may affect the deeper architecture of Holocene reefs by determining their take-off rate and timing, the facies, growth rates, and ages of the upper 10 m or so are clearly influenced by the relative sea-level curve. At the start of the Holocene 10 ka ago, the difference between the level of the sea in Australia and the Caribbean may have been relatively small but as the global isostatic effects of deglaciation became sharper (Nakiboglu et al., 1983), the differences became greater and by 6 ka ago the Caribbean lagged behind by as much as 8 m, producing the differences discussed above.
11.11 Conclusion
409
Differences in the environmental histories of reefs around the world are likely to emerge as more paleoenvironmental analysis of coral cores takes place. In this review the apparently widespread Caribbean collapse of reefs at about 7 ka ago was obviously a major event in the Holocene history of the reefs of that region. Similarly, the effects of the synchronous quiescent periods for fringing reef growth on the GBR at around 5.5 and 3 ka ago discussed in Chapter 7 can be detected in modern reef flats. An interesting environmental contrast occurs in the central Pacific. Whilst reef growth from about 14 ka ago was continuous and prolific in Tahiti at 178 300 south of the equator, similarly located Hawaii (straddling 218 N) had a very different environmental and reef growth history. Inter-island reefs which may have existed at 14 ka ago (Grigg et al., 2002) were drowned because they could not keep up with sea-level rise (13 m ka1). Subsequent growth on higher antecedent karst pinnacles was similarly drowned by sea-level rise (9 m ka1) between 8 and 6 ka ago, and as already identified Grigg (1998) believes that any further Holocene accretion is limited to a constantly recycled Holocene veneer capable of only 2 m ka1 vertical accretion rate produced by the high wave energy of the islands. More recently, Rooney et al. (2004) believe that mid-Holocene termination of reef growth was caused by an increase in wave energy at that time related to a strengthening of the El Nino–Southern Oscillation. Rooney et al. (2004) suggest ˜ that ‘‘ENSO-related variations may be an important consideration in evaluating Quaternary reef accretion and palaeo-climatology in Hawaii and other Pacific islands and warrant further investigation’’ (Rooney et al., 2004, p. 321). Although changes to the strength of ENSO have been demonstrated on the GBR, a recognizable effect on reef accretion, with the possible exception of the quiescence in fringing reef flat progradation, has yet to be demonstrated. Biological differences appear to have little influence on geomorphological evolution of reefs worldwide. Whilst taxonomy and biodiversity may be very different especially in the Caribbean in contrast to the Indo-Pacific, ecological and geomorphological zonation is essentially the same where similar environments are encountered. For example the microbialite/algal ridges on top of the Caribbean reefs are more an expression of the slow and constant sea-level rise than any major biological contrast to the Pacific. Possibly one difference is the presence and importance of Acropora palmata in Caribbean reefs. It grows quickly, can withstand high wave energy conditions, and is a major contribution to reef framework. In the Indo-Pacific other Acropora species fill this ecological niche, e.g., A. robusta in Tahiti, but elsewhere different species are found as reef exposure or water depth change during reef evolution. It is not surprising that as the GBR comprises only 3.25% of the world’s reefs (see figures in Kinsey and Hopley, 1991), there are many features in the
410
The accumulation of the Holocene veneer to the Great Barrier Reef
Holocene sections from elsewhere in the world that may direct future questions for the Australian reef. Given the global importance of structural control of antecedent foundations and differences in sea-level histories, further research into the probable shelf marginal subsidence in the central GBR and hydroisostatically induced differences in mid to late Holocene histories between inner and outer shelf may be warranted. Major storm rubble facies in the GBR reefs may also be more important than previously thought, especially as cyclone disturbance appears to be a major control of reef ecology at the present time (Puotinen, 2004). Major interruptions to reef growth in the Holocene seem to have occurred in many locations but except for interruptions to progradation of late Holocene reef flats, not on the GBR. These and other questions raised by research into the evolution of coral reefs worldwide may help to further refine the Holocene history of the GBR.
12 The Holocene evolution of the Great Barrier Reef province
12.1 Introduction In previous chapters the characteristics and development of different sectors of the Great Barrier Reef (GBR) have been discussed largely on a spatial division – inter-reef shelf (Chapter 6), fringing and nearshore reefs (Chapter 7), mid-shelf reefs (Chapter 8), shelf-edge reefs (Chapter 9), and reef islands (Chapter 10). Chapter 11 derived rates of process from the data of previous chapters and provided some of the basics of the chronological framework used in this chapter, which attempts to provide a holistic rather than sectorial evolution of the GBR. Different processes clearly will be occurring simultaneously on different parts of the reef which may interact with each other. The driving force for the development of the reef, and the coast and shelf in general has been the postglacial rise in sea level, the basics of which were established in Chapter 3. This chapter looks at the changing geomorphology as sea level rose from its absolute low of about 125 m 20 000 years ago at what are regarded as critical stages in the transgression. Initially sea level rose against a steeply sloping shoreline. Subsequently the antecedent Pleistocene reefs of the outer shelf became limestone islands before becoming submerged and developing their Holocene veneer. A completely different geomorphology existed on what is now the wide gently sloping shelf between the outer reef and mainland. Rates of shoreline change were rapid until the stillstand of the last 6.5 ka. This again has allowed another scenario of catch-up reefs and prograding shorelines to evolve (see Hopley (1994) for initial discussion). At most of these stages the geomorphology was very different from that of today. To illustrate this modern analogies are used (mostly from tectonically active areas so as to reproduce emergence or submergence), to give some insight into the geography of the region at a time when it was occupied by early Australians. Finally the approach allows for some interpretation of the changing biogeography as for example the extent of mangroves and other coastal habitats and the 411
412
The Holocene evolution of the Great Barrier Reef province
availability of ecological niches for what are presently regarded as some of the GBR’s most important species such as turtles, dugong, and seabirds. 12.2 The glacial maximum low sea level – 20 000 years BP With sea level 120 m or more below its present level at 20 ka ago, the geography of the coastline of north-east Queensland was very different from what it is now. The shelf marginal topography discussed in Chapter 9 gives some indication of the coastal geomorphology of this time and indicates some regional contrasts. 12.2.1 The northern coastline Outside what are now the deltaic and ribbon reefs (Chapter 9) there is an almost vertical drop-off. This is shown by bathymetric maps and echo-sounding profiles but is best exemplified by the submersible dives in front of Ribbon 5 Reef in 1984 (Fig. 9.6) and more recent remotely operated vehicle (ROV) surveys at Hicks and Day Reefs. From levels of 50 m to 70 m (related to modern sea level) the coastline consisted of a vertical cliff about 60 m high with outcropping horizontally bedded limestone. This would have extended a further 70 m into the ocean above a basal scree slope (Fig. 9.6). Close scrutiny of the entire outer reef from the far north to 178 S shows that there would have been little variation in this morphology which may have been analogous to the present cliffs of western Victoria or even the Great Australian Bight (Fig. 12.1). Small notches are apparent in the cliff face of Ribbon 5 but development of both shore platforms and fringing reefs would have been very difficult if not impossible on the vertical slope. Present detached reefs described in Chapter 9 would have existed as offshore islands with cliffed limestone coasts. The closest modern analogy may be uplifted reef islands and atolls close to plate boundaries in the Pacific such as Makatea in French Polynesia and Walpole, Ouvea, Lifou, and Mare´ in the Loyalty Islands (Fig. 12.6a). 12.2.2 The central Great Barrier Reef Between 178 and 218 S the bathymetry around the 120 m mark is much more gently sloping, typically between 1: 15 and 1: 30. The bases of submerged shelfedge reefs described in Chapter 9 are almost totally above 120 m and the isobath is remarkably un-indented. Detailed echo-sounding, ROV, seismic, and sediment sampling studies across the 120 m isobath at about 188 S (Hopley et al., 1997) showed a shelf edge lacking any erosional notching but with
12.2 The glacial maximum low sea level – 20 000 years BP
413
Figure 12.1 Vertical cliffs cut in calcareous sandstone, western coast of Victoria, resemble the ocean side of the ribbon reefs at the absolute low sealevel stage.
significant depositional features at about 90 m, 100 m, and 120 m depth comprised of largely non-carbonate sediments. Echo-sounding and seismic transects greatly exaggerate the relief of these features and of the shelf margin itself. At the maximum low sea-level stage the coastline would have consisted of a moderately sloping plain possibly bordered by narrow sand barriers similar in many respects to parts of the present coastline of north Queensland where the Pleistocene fan deposits at the base of the coastal range have a similar slope and are fronted by narrow beach ridge sequences. 12.2.3 South central Great Barrier Reef and Swain Reefs At the maximum low sea-level stage the Pompey Complex (Chapter 9) would have formed a dissected limestone plateau bordered on the seaward side by parallel ridges, the antecedent foundations of the shelf marginal reefs (Fig. 9.20). From about 80 m the shelf margin drops off steeply to well beyond the 120 m mark and the low sea-level coastline appears to have consisted of steep limestone cliffs about 50 m high. These became far less
414
The Holocene evolution of the Great Barrier Reef province
steep around the margin of the Swain Reefs where the low sea-level shoreline may also have consisted of a gently sloping coastal plain beneath a limestone scarp which formed the margin of the Swain limestone plateau. 12.2.4 The Capricorn Channel A 120 m sea level narrows the Capricorn Channel to a width of only 100 km. It is very gently sloping at this level and although no surveys have confirmed paleo-drainage channels in the area it is likely that Queensland’s largest river, the Fitzroy, passed out to sea via the Channel at 20 ka ago (Chapter 6). A low high-energy coastline with sand barriers (and possibly dunes as occur on the adjacent coastline today) and estuaries forming a refuge for mangroves and associated species, which would have been very restricted on most of the GBR 20-ka shoreline, probably characterized this coastline. Further south, to the east of what is now the Bunker–Capricorn Group, although the slope is greater than in the Capricorn Channel, the lowstand coastline was an alluvial plain across which siliciclastic sediments were delivered to the nearshore zone and shelf upper slope by way of fluvial channels (Troedson and Davies, 1997). The hinterland from these contrasting shorelines consisted of what is now the main reef tract, a series of limestone mesas rising up to 50 m from the surrounding plain. The largest area of exposed limestone occurred in what is now the Pompey Complex and although channels through these reefs reach over 90 m depth and probably existed as typical karst gorges, the complex as a whole formed a drainage divide for major streams and contained karst features (now forming the blue holes: Fig. 9.21). In the north the shelf marginal ribbon reefs would have formed a similar continuous positive relief feature. No Quaternary analogy of this scale is known but the Canning Basin in Western Australia contains a well-preserved Middle to Upper Devonian barrier reef which has been exhumed as a series of limestone ranges 350 km long (Playford, 1980). The reef tract is up to 50 km wide and the barrier stands tens to hundreds of meters above what was the surrounding sea floor (Fig. 12.2). The reefs, including patch reefs and reef pinnacles, are remarkably well preserved and certainly provide a picture of what the northern GBR looked like when totally exposed. Landwards of what are now the reefs, the exposed shelf was a gently sloping coastal plain with entrenched rather than incised major drainage channels, as discussed in Chapter 6. Many, including the Burdekin River (Fig. 6.3), can be traced to about the 80 m isobath beyond which the last glacial channels are not clear. Fielding et al. (2003) noted that beyond this depth is a shallowing of
12.2 The glacial maximum low sea level – 20 000 years BP
415
Figure 12.2 The exhumed Devonian ‘‘Great Barrier Reef’’ of the Canning Basin, Western Australia, looking north-west over Windjana Gorge (photograph: P. E. Playford).
the shelf which required the Burdekin to aggrade to a point where it was able to spill over the barrier. However, the barrier is almost certainly part of the shelfedge reefs (Chapter 9) and although very persistent, gaps do occur through them. Beyond this it would have been possible for deltas to form on this less steeply sloping shelf margin, which, as suggested by Harris et al. (1990), were eroded during the early part of the transgression, adding to the supply of terrigenous sediment to the lower shelf slope. Within this reconstruction of the 20-ka shoreline there would have been remarkably little opportunity for the growth of coral reefs. In the far north and adjacent to the Pompey Complex the near-vertical slope would have precluded reef development. On the central GBR it is possible that rivers were delivering at least as much sediment to the paleo-coastline as today with corresponding nearshore turbidity extending into depths too great for coral growth. In the far south, the lack of evidence for any low sea-level coral growth even in the sediments of the shelf suggests that at these latitudes there were no reefs during the height of the glacial periods. Some refuges may have been provided by the detached reefs in the north, and by Myrmidon Reef on the central GBR which would have been isolated from the ‘‘mainland’’ and the high-turbidity nearshore waters. However, the general consensus has been that the reefs of the
416
The Holocene evolution of the Great Barrier Reef province
Coral Sea plateaux formed the refuges for corals and the source for recolonization as sea level rose. It is also probably that the extent of mangroves was very much reduced during the low sea-level period. Refuges may have been in the far north (Papua New Guinea) and the far south (the Capricorn Channel) but for the intervening 2000 km suitable habitats would have been very rare. The biogeographic implications for species depending on coral reefs and coral cays, mangroves, and tropical shallow waters would have been drastic. Present distributional and dispersal patterns for species such as turtles, dugongs, and crocodiles have developed entirely within the last 12 ka or less as will be shown by the changing geographical patterns during the transgression. 12.3 The early transgression to 12 000 years BP From an initial steady rise in sea level of about 5 m ka1 the transgression accelerated to as much as 50 m ka1 at about 14 ka ago (see Section 3.4). However, until sea level reached 60 m there were few modifications to the geomorphology of the coastline along the GBR. Adjacent to the ribbon reefs and the Pompey Complex the cliffed coastline persisted and off the central GBR the probable barrier beaches and wave-dominated deltas were merely transposed up the gentle slope of the shelf shoulder. By about 12 ka ago, however, the sea had transgressed over parts of the shelf margin and standing at 50 m major geomorphological changes were taking place. Adjacent to the ribbon reefs the coastline remained a relatively straight one of low limestone outcrops similar to many uplifted reef coastlines of today. To seawards fringing reefs were being added to the more gentle slope above the vertical wall, notably at 70 m and 50 m (see Fig. 9.6). South of the ribbons for about 150 km off Cairns the indented coastline consisted of rocky headlands which were formed from agglomerations of the outer reefs with chains of limestone islands to seawards (Fig. 12.3). Estuaries (now the major reef openings such as Trinity Opening and Grafton and Flora Passages) penetrated the coastal plain by as much as 25 km. Further south again, on the central GBR, by 12 ka ago it is possible that the shoreline was landwards of the main reef tract. Apart from occasional limestone headlands (the innermost reefs) it was low and probably swampy, sheltered by the archipelagos of limestone islands just offshore (Fig. 12.4). These may have resembled the limestone islands of Palau (Fig. 12.5) or raised ‘‘makatea’’-type islands such as Walpole (Fig. 12.6a). In the south central GBR the low shoreline was up to 50 km inside the reefs from which it was separated by shallow waters only 10 m deep.
12.3 The early transgression to 12 000 years BP
417
Figure 12.3 The coastline near Cairns at the 50 m sea-level stage. The mainland consists of low-lying coastal plains, estuaries, and mangroves with intervening limestone headlands. Offshore are emerged limestone islands and a discontinuous barrier reef.
The Pompey Complex and Swain Reefs remained largely dry though some of the channels between the Pompey Reefs may have been drowned forming deep ria-like gorges analogous to parts of the Kimberley coast in Western Australia. In the Capricorn Channel was a deep embayment close to the Pompey Reefs and receiving numerous drainage channels from the west. An estuarine environment
418
The Holocene evolution of the Great Barrier Reef province
Figure 12.4 The coastline near Townsville at the 50 m sea-level stage. The shoreline here is largely inland of the main reef tract which forms chains of offshore islands. The mainland coastline would have been very low energy and possibly fronted by mangrove along its entire length.
ideal for mangroves is indicated. There appear to have been few changes to the paleogeography in the vicinity of the Bunker–Capricorn Group apart from a westward movement of the low sandy coastline. The period at about 12 ka ago appears to have been an important one for the GBR as it represented the first period of reef growth after the last glacial.
12.3 The early transgression to 12 000 years BP
419
Figure 12.5 The ‘‘rock’’ islands of Palau suggestive of more complex reefs of the Great Barrier Reef at low sea-level stages (photograph: Patricia Davies, C3).
Initially most prolific was the recolonization of the linear reefs which now form the submerged barriers of the central GBR (Fig. 9.9) and east of the Pompey Reefs. Although no dates are available for any of these reefs, which Graham (1993) and Harris and Davies (1989) considered to be multicyclic, all evidence suggests that their last period of growth was during the last transgression and depth relationships suggest the period post 14 000 years BP. However, once the transgression rose above the shelf edge, which because of varying bathymetry would have been at different times on different parts of the GBR, reworking of subaerially deposited sediments would have greatly increased both turbidity and nutrient levels as discussed in Chapter 9, leading to the demise of these reefs. A period of less prolific reef growth may have then occurred as the mainland shoreline migrated westwards through the main reef tract, especially of the central GBR. Potential recolonization sites would also have to have contended with fresh fluvial sediments coming from river mouths which were in close proximity. Although the start-up date for recolonization of the shelf reefs may have been over a longer period than previously hypothesized (Chapter 11) all the data used for reconstruction have come from the drilling of shallow reef flats. Reef terraces similar to those in front of the ribbon reefs do exist and when
420
The Holocene evolution of the Great Barrier Reef province
Figure 12.6 (a) Tectonically uplifted Walpole Island, New Caledonia suggestive of simpler Pleistocene reefs of the Great Barrier Reef at low sealevel stages (photograph: B. Bouye´). (b) Notch and visor cut into the emerged barrier reef of Sabari Island, Papua New Guinea, a model for the ribbon reefs during the transgression.
12.3 The early transgression to 12 000 years BP
421
0 Modern reef front spur and groove 10 20
BOWL REEF
Sand flat Coral slope
meters
30 40 50 60 70
Branching and head corals
35 m reef
Coralline algae
42 m terrace Platey corals, 48 m terrace sponges Bare sand Halimeda “forest”
50 m reef Dead Basal slope corals, gorgonians 5° loose 3° blocks and Halimeda coarse sand “meadow”
2° hummocks and mega ripples Halimeda, bryozoans, and sponges
Figure 12.7 The front of Bowl Reef from manned submersible observations showing a series of terraces, probably of multicyclic origin.
conditions became suitable, e.g., the reef became sufficiently far from the coastline or from major river mouths, these may represent the first addition to the shelf reefs in postglacial times. Many of the terraces are at about 50 m and, if originating as fringing reefs constructed during the transgression, their age would be about 12 ka or less. An example is Bowl Reef (Fig. 12.7). Submersible observations indicated a 50 m largely dead reef as the deepest of a series of reef flat terraces. If coral growth first took place when sea level was about 50 m then Bowl Reef at this time was a limestone island (which also incorporated Yankee, Arc, and Coil Reefs) about 25 km from the mainland shoreline. Thus, although the paleo Burdekin River had previously passed just to the east of Bowl Reef (Fig. 6.3), its mouth was now some 40 km away, probably on the downdrift side of the coastline. Reef growth could now commence without environmental hindrance. However, some caution is needed in attributing these reef front terraces entirely to the Holocene transgression. Time for their construction was very limited and many may be multicyclic, formed at interstadial sea-level periods or during earlier transgressive or regressive phases. As well as being the period when reef growth commenced, 12 ka ago was also the time when the distribution of mangroves would have become far wider. V-shaped estuaries were cut into the shelf near Cairns and it is notable that the deepest and oldest mangrove peats reported by Grindrod et al. (1999) are associated with these, especially Flora Passage (Fig. 6.4). The oldest date from 64.5 m is 10 050 190 years BP and although the depth and date do not match the presumed sea level, very similar geomorphology would have existed 2000 years earlier. Mangrove peats have also been reported on the shelf off Townsville (Chapter 6) and the shoreline of the south central GBR and
422
The Holocene evolution of the Great Barrier Reef province
Capricorn Channel would have been ideally suited to mangrove growth in the numerous estuaries of these areas. The increase in sediments and nutrients that was so detrimental to coral growth around 12 ka ago may have played a part in the great increase in the growth of Halimeda from this time. Scoffin and Tudhope (1985) report the high abundance of Halimeda gravels in water depths between 60 m and 100 m on the central GBR. Whilst the surface over which the Halimeda banks is somewhat shallower and the basal age about 2 ka younger (see Chapter 6) the upwelling and nutrient jetting so important for the development of the banks may have been commencing about 12 ka ago. This is discussed in the next section. 12.4 The start of the Holocene – 10 000 years BP By 10 ka ago sea level had risen to about 30 m, the extra 20 m of water depth producing major changes in the coastal geomorphology. Because of bathymetry and contrasts in the antecedent reef topography, changes occurred at different times on different parts of the reef. In the north flooding of the outer half of the shelf occurred for the first time and the geomorphology with one exception was similar to that of the central GBR 2 ka earlier. Limestone headlands (the mid-shelf reefs) were separated by presumably sandy beaches and estuaries whilst offshore were lagoon limestone islands around which some fringing reef growth may be presumed. The contrast to the earlier morphology of the central Reef was the outer barrier which at this stage was as continuous as it is today, with similar narrow passes giving access to the outer shelf for deep upwelling water. However, the barrier would have stood as linear limestone islands at 10–20 m above sea level. There are a few analogies today such as the tectonically uplifted 7-km-long Sabari Island on the barrier reef surrounding the Louisiade Archipelago in eastern Papua New Guinea (Fig. 12.6b). However, there is nothing to compare with the 700-kmlong chain of linear islands which stood off the northern GBR 10 ka ago. To the south of the ribbons the shoreline was now a relatively straight one indented only by mangrove-lined estuaries. Because of the low gradient, the transgression moved rapidly over the coastal plain and, in the south in particular, incorporated some of the outer high continental islands such as the Northumberland Group initially as headlands but before long becoming isolated islands. It was at this time that some of the more prominent headlands, such as Cape Upstart, Cape Gloucester, and Cape Clinton, became part of the coast. Offshore the main reef tract was just being inundated. Some reefs may still have been emergent, others were shallow carbonate platforms, though as
12.4 The start of the Holocene – 10 000 years BP
423
yet not developing any significant coral reef growth. The shelf-edge reefs would also have been close to the surface but may have been in a state of decline, as shedding of siliciclastic material from the shelf was reaching its maximum at this time (e.g., Dunbar and Dickens, 2003). The Pompey and Swain Reefs at 10 ka ago were a very complex group of closely spaced limestone islands and narrow channels within which deepening by high-velocity tidal currents was taking place (Fig. 12.5). The shelf-edge reefs may have lasted longer here as terrigenous sediments made up only a small proportion of the outer shelf cover (Fig. 6.1). However, it would appear that many were now being left behind by the sea-level rise and were 10 m or more below the sea level of the time. To the west, the Capricorn Channel was more like it is today, an open water area extending northwards to the coastal GBR rather than being a discrete bay or series of estuaries as it was at lower sea levels. Sea level was also approaching the base of the Bunker–Capricorn Reefs. Coral growth at 10 ka ago was still very minimal. The oldest Holocene date obtained is 9320 730 years BP (Hopley et al., 1978) and that is from a fringing reef (Hayman Island), though it is probable that some coral growth had taken place prior to that time, especially in the form of fringing reefs around the base of the outer reefs (Fig. 12.3). What is more certain is that it was about this time that tidal jetting of deep upwelled water onto the continental shelf between the ribbon reefs was commencing. A peat layer from near Lizard Island lying over the Pleistocene and overlain by carbonate sands and Halimeda gravels was dated at 10 070 180 years BP (Orme, 1985) and it is from about this time that the Halimeda banks began to accumulate (see Marshall and Davies, 1988, fig. 8). Coral growth may have been minimal, with Halimeda growth just commencing, but the start of the Holocene was the peak period for mangroves (Fig. 12.8). On the shelf of northern Australia in general about 9 ka ago was the time of peak mangrove growth (as indicated by deep-sea cores) declining to about their present level by 6.5 ka ago (Grindrod et al., 1999). Grindrod et al. relate this to the availability of suitably developed estuaries in drowning coastal settings. Low sea-level channels were transformed into deep estuaries during the transgression with the gradient of the shelf and a high tidal range extending mangrove growth more than 100 km upstream on the shelf of northwest Australia. Inside the central GBR the estuaries may have been shorter but Grindrod et al.’s data show extensive estuarine mangrove growth across the shelf in the Cairns to Cardwell section of the GBR (Fig. 6.4), and further research is likely to discover similar deposits over a wider area of the GBR shelf (see Chapter 6).
424
The Holocene evolution of the Great Barrier Reef province
Figure 12.8 Mangroves in low-energy Bowling Green Bay, suggestive of much of the mainland coast of the middle shelf during the transgression.
12.5 The final 2000 years of the transgression – 7000 years BP The rate of sea-level rise slowed considerably during the last 2 ka of the transgression between about 8.5 ka ago (sea level 10 to 15 m) and 6.5 ka ago when modern sea level was first achieved, as discussed in Chapter 3. In terms of vertical growth and calcium carbonate production this was the most important time for the modern GBR. The mainland shoreline had now retreated away from the reefs and except in the far north lay tens of kilometers away from the main reef tract. Although this may have been a period of increasing rainfall, causing a change from sclerophyll woodland to a simple notophyll vine forest on the Atherton Tableland (Kershaw, 1978; Moss and Kershaw, 2000), the increased protection given by the vegetation may have reduced rapid runoff and sediment yield from the mainland to the GBR lagoon. Some of the main island groups such as the Palms and Cumberland and Northumberland Groups were isolated by 8 ka ago and were commencing fringing reef growth. Indeed, the oldest Holocene date so far achieved on the GBR of 9320 730 years BP comes from the fringing reef of Hayman Island (Hopley et al., 1978).
12.5 The final 2000 years of the transgression – 7000 years BP
425
Mangroves would still have dominated much of the mainland coastline and as estuaries became inundated would have been buried within the paleochannels (e.g., Grindrod et al., 1999; Hull, 2005). Mainland headlands may not have stood out from the coast as prominently as they do today, allowing for a more continuous northwards movement of sediment which was being transported shorewards by the transgression, becoming trapped only around some of the newly formed island groups such as the Whitsundays (Heap et al., 2002). However, it was on the outer shelf that most changes were taking place during this period. Whilst some coral growth may have commenced earlier around the bases of the reefs, the Pleistocene foundations of most reefs were being totally inundated between 8 and 7 ka ago, especially on the central GBR where some subsidence may have been responsible for deeper foundations (see Chapter 9). Reanalysis of core data suggests that once inundated, regolith removal was quick and recolonization took place rapidly on most reefs (Section 11.4) as opposed to the narrow time envelope previously suggested (e.g., Davies et al., 1985). Certainly the re-establishment of larval dispersal patterns, most probably from the Coral Sea plateaux reefs, may have been a factor in delays in the earliest part of the Holocene. On the northern reef where the exposed ribbon reefs would have formed a very effective barrier to the passage of Coral Sea waters carrying coral planulae, the inner shelf reefs would have had a restricted larval source. However, as sea level continued to rise and reef growth in most cases took on a ‘‘catch-up’’ mode, lagging behind that rise, the open circulation associated with the open window period (see below) would have encouraged widespread dispersal of coral spawn. With a depth of less than 18 m, Torres Strait came into existence some time after 8 ka ago completing the physical conditions required for the modern-day circulation patterns of the Western Coral Sea. The end result was that between 8 and 6 ka ago the highest vertical growth rates for the GBR were achieved and up to 50% of Holocene reefal accretion took place within this 2000-year period (Fig. 11.3). Maximum accumulation rates of calcium carbonate over reefs is about 10 kg m 2 yr 1 (Kinsey, 1985), limited to areas of almost total coral cover. The inundated surfaces of the reefs between 8 and 6 ka ago were ideal accumulators, contrasting with today’s reefs where such accumulation rates are limited to reef slopes and margins. No dates are available for the lower sections of Halimeda banks (Chapter 6), but they too may have flourished in this period. However, the tidal vortices which had previously delivered nutrient-rich water from below the thermocline and which appear to be an essential process for their growth, for a time may have been less pronounced, as during this period, when the ribbon reefs
426
The Holocene evolution of the Great Barrier Reef province
were inundated and before they caught up with sea-level rise, there would have been a less restricted tidal flow. As noted above, fringing reefs commenced growth as soon as their foundations were inundated (Chapter 7). As these vary in depth and have different types of substrate the range of colonization dates extend from >9 to <6.5 ka ago with even mainland fringing reefs (Cape Tribulation) growing by about 7780 260 years BP (Partain and Hopley, 1989). As sediments may have accumulated around island groups (Heap et al., 2002) and even up against the mainland during the final part of the transgression, the water quality in which the fringing reefs grew may already have been poorer than that on the outer shelf. A possible result of this is the common occurrence of alcyonarian spiculite (Konishi, 1982) formed from the soft coral Sinularia and forming the basal section of many fringing reefs in the Palm, Cumberland, and Northumberland Islands (Kleypas, 1992, 1996). Nearly all components of the modern GBR were thus in place by 6.5 ka ago though the appearance of the Reef at this time was very different to that of today. Few reefs would have developed reef flat and reef islands would have been even rarer. Instead, the reef tract would have consisted of reefal shoals with occasional pinnacles reaching the surface. Especially in the central GBR, where the Holocene reef was growing from much deeper Pleistocene foundations than elsewhere, at 6.5 ka ago the reef was typically 10 m or more below sea level (Fig. 9.17), and little protection would have been provided to the mainland (see below). This was an important factor for reef and coastal development during the early part of the stillstand period. 12.6 The mid to late Holocene The stabilization of the regional sea level about 6.5 ka ago has permitted the subsequent recognition of relatively subtle hydro-isostatic warping of the GBR shelf involving vertical movements of only a few meters (see Chapter 3). This was one of the most important processes of the mid to late Holocene, superimposing submergence, or at least stability, on the outer shelf and emergence of one to two meters on the inner shelf, with important influences on reef development as discussed in Chapters 7 and 9. Inshore reefs were helped to catch up with sea level and developed reef flat earliest, especially the fringing reefs (see Fig. 11.6). They therefore recorded the subsequent history of falling sea level post 5.5 ka in the form of emerged microatolls and descending reef flat steps (see Chapters 3 and 7). Mid-shelf reefs were also able to catch up with sea level within 1–2 ka and are therefore dominated by the mature to senile morphologies (Fig. 8.1). The exception was
12.6 The mid to late Holocene
427
in the central GBR where deeper foundations and possibly greater subsidence has resulted in reefs catching up with sea level only in the last 4000 years and many of these reefs are of the juvenile forms (patch and crescentic). The development of the reef islands also reflects this pattern (Chapter 10) with the oldest reefs displaying the most advanced forms of island development such as the low wooded islands of the northern GBR, whilst the central GBR between Green Island (168 460 S) and Bushy Island (208 580 S) has not a single vegetated reef island. Island development reflects a change in process on the reefs over the latter part of the Holocene, from mainly vertical accumulation during the transgression at accumulation rates approaching 10 kg m2 yr1 (Kinsey, 1985) with vertical growth rates of up to 7 or 8 m ka1, to horizontal movement of sediments towards the rear of the reef with reef flat accumulation rates less than half those during the transgression. Also as the reefs have come into the wave zone they have become more complex ecosystems as the reduction in wave energy from reef front to back reef is largely responsible for modern coral-reef zonation. As the reefs continued to develop after the Holocene stillstand, the mainland also experienced significant changes after 6.5 ka ago. Initially, the energy levels reaching the coastline may have been the highest recorded in the Holocene. In the early Holocene the exposed reefs on the shelf limited the available fetch and gave great protection. Since the reefs reached sea level they have again provided protection as is illustrated by the contrast between the open surf beaches of southern Queensland and those of northern Queensland from which open oceanic swells are excluded. In between was a ‘‘high-energy window’’ as originally discussed by Neumann (1972) and applied to the GBR by Hopley (1984). The concept relies on shelf reefs adapting the ‘‘catch-up’’ mode and allowing the freer passage of waves between the time of initial drowning of the antecedent platforms and the reefs reaching sea level. On the northern and southern GBR this was a relatively short period between about 8 and 5.5 ka ago (note that more data are now available than when Hopley (1984) first discussed the concept and the ‘‘open window’’ dates quoted here are a little different). However, on the central GBR because of the greater depth of the Pleistocene foundations, the high-energy window may have been open for considerably longer. The effects may have had an impact on the mainland as the higher wave energy levels would have aided the onshore transport of coarser sediments from the shelf. This was a process which had been operating on the shelf throughout the latter part of the transgression, but within the shallow area of what is now the GBR lagoon, the shoreline was linear and northward drift
428
The Holocene evolution of the Great Barrier Reef province
of sediments was a major process (e.g., Harris et al., 1990; Larcombe and Woolfe, 1999b) except where intercepted by what are now offshore island groups (Heap et al., 2002). As the shoreline approached its present position, numerous headlands became part of the coast dividing it into discrete sediment compartments and trapping the sediment transported from the shelf. This coarser fraction was molded into the beach barriers and occasional dunes of the mainland coast with some barriers containing up to 100 ridges of Holocene age (e.g., Hopley, 1970; Bird, 1971b) (Fig. 2.5a). Soil profiles and vegetation maturity on many of these beach ridge sequences together with occasional radiocarbon dates suggest that the majority of these shoreline features were deposited in the few thousand years immediately after the end of the transgression and that there has been minimal addition more recently (the contribution of fluvial sediments especially in the 200 years since European settlement is discussed in the next chapter). The significance of the opening and closing of the Holocene high-energy window for the emplacement of beach barriers (and other features as discussed by Bird, 1970) may be debatable (Graham, 1993) but certainly appears to have had an influence on the timing of deposition. Similar to the outer shelf reefs, fringing reefs reached the peak of their development at or before the Holocene transgression (Fig. 11.3). As described in Chapter 7, they have subsequently been subjected to an array of environmental conditions which have affected their rates of lateral growth. These have included a fall in sea level of a meter or more (Chapter 3), and the concentration of a shore-attached or nearshore mud and silt wedge as described in Chapter 6 (see also Hopley, 1994; Larcombe and Carter, 1998; Larcombe and Woolfe, 1999b). This reaches a maximum thickness of 20 m and may extend up to 20 km offshore, i.e., out to at least the innermost of the high islands with fringing reefs. It is resuspension of this material accumulated over thousands of years, rather than the increased sediment supply over the last 200 years due to anthropogenic activity, that may be the limiting factor in the establishment, survival, or demise of nearshore reefs (Larcombe and Woolfe, 1999b; Smithers and Larcombe, 2003). The nearshore sediment wedge plays another part in the health of the GBR. It has been noted that the passage of cyclones leads to shelf-wide resuspension of nutrient-bearing sediments on the inner shelf. Subsequently short-lived periods of algal blooms have been noted around the reefs (Furnas and Mitchell, 1986). However, the shore-attached sediment wedge is not the only store of nutrients. Seagrass beds and estuarine systems, most notably the mangroves, have also stored land-derived nutrients since the mid Holocene and natural and man-made disturbances may also release these to the wider environment, including the reefs.
12.7 Conclusion
429
12.7 Conclusion There was an Aboriginal presence in the GBR region throughout the Holocene; indeed archeologists have hinted that tribal territories that once existed on the continental shelf have been squeezed up against the mainland as the early people were forced westwards by the rising sea levels (e.g., Beaton, 1978; Campbell, 1980; Hughes and Lampert, 1982). Within the relatively short period of 12 ka the geography of the area that currently contains the GBR went through some remarkable changes. Probably the least hospitable conditions for human occupation of the coast occurred at the maximum low sealevel stage when a cliffed coast facing the open ocean occurred along both the northern and southern reef. Even the suggested coastal barriers of the central GBR would have been high-energy environments with mangrove and seagrass beds at minimal extent. Changes were rapid as sea level rose over the shoulder of the continental shelf and, although the northern GBR remained cliffed for a much longer period, the presence of archipelagos of limestone islands, sheltered estuaries, and extensive mangrove and potential seagrass areas provided a range of rich habitats not only for early Aboriginal peoples but for many of the plants and animals which currently inhabit the area. As Grindrod et al. (1999) have pointed out, mangroves, so important as the habitat for a wide range of species and the nursery ground for many others, reached their maximum extent during the early Holocene. The stillstand period of the last 6.5 ka has allowed the stabilization of ecological communities but is probably not the richest period in the recent geological past. The advent of European settlement of Australia has undoubtedly had many impacts on the GBR (e.g., Hopley, 1988, 1989b) but many of these such as ship groundings and anchor damage are localized. The potential for more widespread effects from factors such as increased sedimentation and nutrient input are considered in the next chapter. However, these need to be assessed against naturally driven declines in water quality over at least the last 3 ka (Hopley, 1994). These include: turbidity levels resulting from the nearshore sediment wedge (Larcombe and Woolfe, 1999a, b), release of nutrients from the natural erosion of mangroves or by the stirring up of sediments during cyclonic storms (Furnas and Mitchell, 1996), and shelf margin nutrient enhancement as the result of upwelling associated with the El Nino ˜ phenomenon, strong enough to be picked up by geochemical signals in the coral skeletons (Rasmussen et al., 1993a, b). The future of the GBR is dependent on a wide range of variables. Most immediately, it may depend on the degree to which anthropogenic impacts can be curtailed. At a longer, geomorphological timescale, it may depend on when
430
The Holocene evolution of the Great Barrier Reef province
(or if) the next glacial phase is experienced by the planet, with yet another major fall in sea level, exposure of the GBR, and a probable repeat of the processes described in this book. On a geological timescale Harris et al. (1990) presumed a continued high sea level to extrapolate sedimentation patterns into the future for comparison with ancient sequences. Prograding terrestrial sediments are shown first to slow down reef growth, then, when combined with shelf subsidence in about 700 ka, to terminate all reef growth. In about 1 Ma reef submergence and burial will have transposed the shoreline close to the edge of the continental shelf at which time sediment loss down submarine canyons will balance sediment input. Whether or not the GBR ever reaches these phases is debatable, but within a geomorphological timescale, it is highly probable that further dramatic changes in the geography of the Reef and its adjacent coastline will take place.
13 Geomorphology’s contribution to the understanding and resolution of environmental problems of the Great Barrier Reef
13.1 Introduction In Chapter 1 geomorphology was presented as a spatial science with a unique temporal scale which lies between ecological and geological studies. This book has adopted a spatial scale related to a cross-shelf gradient with the two main driving forces behind the present geomorphology of the Great Barrier Reef (GBR) being sea-level change which approaches a geological timescale and oceanography which is essential for understanding at an ecological scale. Geomorphological processes can be derived either from analysis of the evolution of landforms, largely the focus of this book and summarized in Chapter 11, or from measurement and monitoring of processes operating at the present time. Unfortunately, such monitoring, for example of sedimentation rates, has had only a short history on the GBR and on reefs worldwide in comparison to the measurement of terrestrial, coastal, and fluvial processes. Reefs respond rapidly to environmental change and the measurement of present geomorphological processes may incorporate much variability for which the longer record from dated sedimentary sequences can provide a benchmark. This is particularly important at a time when anthropogenic impact on the Reef is being widely reported (e.g., Bellwood et al., 2004; Fabricius, 2004). If geomorphological studies of the development of the GBR over the last 10 ka, and especially over the last 6.5 ka since sea level stabilized, can provide this benchmark against which changes can be measured, then they have a vital role in contributing to the management process. This final chapter examines five themes centered around recognized problems for the GBR to illustrate a geomorphological perspective to management problems and solutions. 431
432
Geomorphology’s contribution to the problems of the Great Barrier Reef
13.2 Sediments and reefs The GBR has always had to contend with terrigenous sedimentation throughout its >700-ka history. The continental shelf on which it rests is built largely of offlapping sediments derived from the land (Fig. 2.8), the Holocene transgression pushed large sediment bodies across the shelf and against the present mainland coast (Chapter 12), and in pre-European times even the most undisturbed catchments yielded sediment to the coast. Harris et al. (1990) even suggest that if sea level were to remain close to its present level, within 700 ka the Reef would be completely buried by terrigenous sediments. However, at the present time terrestrial runoff and declining water quality are recognized by the Great Barrier Reef Marine Park Authority (GBRMPA) as a major threat affecting the GBR. This section provides a geomorphological perspective to the possible sedimentation problems on the GBR. 13.2.1 Tolerance of corals to sediments Sediments affect corals by reducing light in the water column and thus affecting photosynthesis in the symbiotic zooxanthellae (Yentsch et al., 2002), and by direct settlement and smothering. Some sedimentation can be tolerated as corals have a mucus coating and have ciliary cleaning mechanisms. Sedimentation also affects coral indirectly by reducing coral recruitment and survival (for general discussion see Tomascik et al., 1997). A presumed lack of coral reefs in nearshore areas has long been attributed to sedimentation (e.g., Stoddart, 1969; Guilcher, 1988). However, many of these conclusions were made before quantitative information on turbidity and sediment settlement rates was available and, as Tomascik et al. (1997) have pointed out, in Indonesia there are many thriving reefs in highly turbid environments which are often ignored and excluded from management initiatives. An excellent example in Australia is the recent discovery of thriving patch reefs in turbid waters of the Gulf of Carpentaria (Harris et al., 2004). Discussion of nearshore reefs in Chapter 7 would suggest the same situation may exist on the GBR. Response to suspended sediment loads is seen in the limitation to which corals will grow with depths as shallow as 8 m in highly turbid areas (e.g., Perry and Larcombe, 2003). Original measurements of turbidity used a Secchi disk, which for example in the muddy waters of Bantem Bay (West Java) disappeared in less than 2 m (Tomascik et al., 1997). However, more recently measured sediment loads within the water column have become available, including for the GBR. Several experimental and field studies showed that corals could withstand suspended sediment levels of up to 100 ppm at least for
433
13.2 Sediments and reefs
short periods (e.g., Dodge and Vaisnys, 1977; Dodge, 1982; Szmant-Froelich et al., 1982). Two studies carried out in the late 1980s on the inner reefs of the GBR (Cape Tribulation and Magnetic Island) have indicated that reefs can withstand even higher turbidity without detriment. The study at Cape Tribulation (Figs. 2.4, 13.1, 13.2, 13.3, and 13.4) was in connection with the development of a new road bulldozed through the rainforest, on steep slopes and only a few hundred meters from the sea. The study (Hopley et al., 1990) compared results from three contrasting areas, one adjacent to the new road, one adjacent to relatively flat land cleared in the 1950s, and the northernmost area where the new road was directed inland with a watershed between it and the coast. This latter area acted as a control. Water samples were taken above and below the road, at stream mouths, and over the reef. Sediment traps were also deployed over the reef. Suspended sediment loads within the catchments and over adjacent reefs are summarized in Table 13.1. Stream data came from above and below the new road and the control area. Difficulties in accessing the reef in poor weather conditions limited the data but show that even minor runoff conditions produce turbidity levels of between 14.0 and 210 mg l1, the result of resuspension by waves and tidal currents of the pre-existing sediment wedge (Johnson and Carter, 1987). A healthy reef has survived these conditions for several thousand years. However, the rates over 1000 mg l1 after heavy rainfall adjacent to the highly disturbed catchments would probably not be tolerated for periods of more than a few days. Only the rapid dispersal of the highly turbid plume under the rough conditions following the tropical storm which caused the high rate of runoff is allowing the reef to survive (see below). A parallel study was carried out on Magnetic Island adjacent to bayhead fringing reefs in association with the baseline study for development of a Table 13.1. Suspended sediments (mg l1) from Cape Tribulation streams and reef in contrasting weather conditions
Nil or low rainfall ‘‘Normal’’ rainfall event (100 mm in 24 hours) Heavy rainfall event (300 mm in 24 hours)
Undisturbed catchment
Disturbed/cleared catchment
Over reef
5–20
10–50
14–70
100–200
200–500
65–210
180–260
500–19 000
534–1711
434
Geomorphology’s contribution to the problems of the Great Barrier Reef
Figure 13.1 The coastline north of Cape Tribulation showing fringing reefs and unsealed road just inland from the coast.
13.2 Sediments and reefs
435
Figure 13.2 (a) The Cape Tribulation road in 1985 bulldozed through the rainforest and yielding large amounts of sediment to the adjacent coastline; (b) part of the Cape Tribulation fringing reef with sediment traps.
marina basin. Turbidity levels varied between 35 and 115 mg l1 determined more by wind speed, wave action, and turbid currents than runoff (Hopley and van Woesik, 1988). These figures are in accord with those reported by Larcombe and Woolfe (1999b) of about 50 mg l1 around inshore isolated reefs such as Paluma Shoal and Middle Reef near Townsville (see Chapter 7).
436
Geomorphology’s contribution to the problems of the Great Barrier Reef
Figure 13.3 Sediment plume engulfing Cape Tribulation reefs after Cyclone Manu in April 1986.
Figure 13.4 Small gravel delta delivered to the coast from short steep stream in the Cape Tribulation area. In the past these have formed the foundations for fringing reef growth.
13.2 Sediments and reefs
437
There is much contradictory evidence about the measurement and effects of sediment on coral reefs. Sediment traps are normally used to obtain data which is reported as mg cm2 d1. However, there are no set standards for sediment trap design and this in part may be responsible for varying results around the world (for discussion see Hopley et al., 1990). Early measurements came from Hawaii (Maragos, 1972), Florida (Griffin, 1974; Marzalak, 1982), Jamaica (Aller and Dodge, 1974; Dodge et al., 1974), Barbados (Ott, 1975), the Red Sea (Loya, 1976), Guam (Randall and Birkeland, 1978), and Puerto Rico (Morelock et al., 1979). These and other studies including laboratory experiments were synthesized by Pastorok and Bilyard (1985) who suggested that sedimentation rates of 1–10 mg cm2 d1 produced only slight to moderate effects on coral reefs, 10–50 mg cm2 d1 moderate to severe effects, and >50 mg cm2 d1 severe to catastrophic effects. These limiting levels were more or less repeated by Rogers (1990) in her review of reef sedimentation and have been widely quoted ever since (e.g., Stafford-Smith, 1993; Fabricius, 2004) without a critical and certainly geomorphological examination of the evidence. However, more recently there has been some skepticism about the threshold 10 mg cm2 d1 settlement rate previously accepted (e.g., Macdonald and Perry, 2003; Ogston et al., 2004). Even in 1986 Cortes and Risk reported rates an order of magnitude higher than those quoted with figures ranging from 12.8 to 1179.9 mg cm2 d1 adjacent to cleared catchments in Puerto Rico. They considered that levels of >30 mg cm2 d1 would produce stress if applied for long periods. The possible reason for the very low thresholds suggested by Pastorok and Bilyard comes from the sites sampled and the associated geomorphology. One site on Guam (Ylig Bay), Barbados, and the Florida Keys are largely limestone with very reduced particulate sediment yields compared to solute loads. The Red Sea, because of its climate, would also not normally experience high sediment yield from the land. In Kaneohe Bay, Hawaii, sediment stress was combined with enhanced eutrophication. There is now much evidence to suggest that corals become adapted to their ambient environments, including sediment regime, and regions with normally high yields will have reefs with much greater tolerance. On the GBR it has been shown that corals from inshore turbid environments are far more tolerant of high suspended sediment levels than exactly the same species on mid- and outer-shelf reefs (Anthony, 2000). As noted by Dikou and van Woesik (2006), the effects of sediment load on coral reef communities should be evaluated on a case-by-case basis. Thus the rates of sediment settlement reported from inshore reefs of the GBR will be equally inapplicable to outer reefs or to many reefs elsewhere in the world. Results from the three contrasting areas of Cape Tribulation
438
Geomorphology’s contribution to the problems of the Great Barrier Reef
Table 13.2. Sediment trap results (mg cm2 d1) from three contrasting areas, Cape Tribulation reefs
Control (northern area) Heavily impacted (new road area) Lightly impacted (southern cleared area)
Mean
Maximum
Sand
Mud
26 146 88
38 216 131
11 129 67
15 17 21
collected over a seven-month period are presented in Table 13.2. These figures include both mud (<63 mm) and sand (>63 mm) fractions with the amount of sand deposited being largely a function of the local source. The finer sediments are more representative of what is normally falling from suspension with mean rates of about 20 mg cm2 d1 and maximum rates of about 25 mg cm2 d1, with no differences between the three contrasting areas. Other results from the GBR confirm these conclusions. Collection of sediment from traps on Magnetic Island 10 km offshore from Townsville showed average daily rates ranging from 6.6 mg cm2 d1 to 114 mg cm2 d1 (Mapstone et al., 1989). At Daydream Island in the Whitsunday Passage rates of 5–15 mg cm2 d1 were obtained from a control site (A. D. L. Steven and R. van Woesik, pers. comm.). What may be regarded as ‘‘normal healthy inshore reefs’’ are found in all these areas where apparently high sedimentation rates have been recorded. At Cape Tribulation a study of coral cover using five permanent transects from the three contrasting regions used in the sedimentation study showed that initially in 1985 hard coral cover ranged from 33.2% to 62.6% (Ayling and Ayling, 1987). One year later there was a 24% decline but this was due almost entirely to physical damage produced by small tropical cyclone Manu (27 April 1986) (Fig. 13.3) and subsequent reports have indicated a return to former cover, as predicted by Ayling and Ayling (1987). Similarly, with respect to larval recruitment rates, there appeared to be no measurable impact of the extra sedimentation (Fisk and Harriott, 1989). Coral cover figures for the reefs of Magnetic Island are very similar to those of Cape Tribulation with reef slope cover ranging from 24% to 40% and reef flats from 12% to 23%, (Mapstone et al., 1989). A number of factors significant for management result from this assessment of susceptibility of corals and reefs to sediments. Firstly, sedimentation is a geomorphological/sedimentological process which requires in-depth knowledge to fully understand. There are very distinctive geographical differences in tolerance levels and just as an ecological sequence in the Indo-Pacific may not be applicable to the Caribbean, nor are the physical processes in some
13.2 Sediments and reefs
439
instances. Climate and lithology both have influences on rates of sedimentation and thus on ambient conditions to which the individual corals may be attuned. A further factor discussed by Hopley et al. (1990) was the contrasting Holocene sea-level history of the GBR and the northern Pacific and Caribbean (see Chapter 11). For isostatic reasons the relative Holocene sea-level history of different parts of the world varies (Chapter 3). The 6.5 ka of relative stability in Australia has allowed the build-up of the significant terrigenous sediment wedge (Chapter 6) consisting largely of fine sediments regularly resuspended by even moderate wave activity. In contrast, the Caribbean has seen a continuously rising sea level throughout the Holocene; the shoreline has not been stable in its present position but has continued to move inland. Sedimentation in the nearshore zone has been spread over a much wider area and finer sediments available for resuspension are much less. A different ambient environment exists for corals of this region. Tomascik et al. (1997) state that a clear understanding of the processes involved in producing sedimentation rate figures is required for the formulation of conservation and management processes. They indicate that where resuspension of sediment is involved an overestimation of the sediment problem may result as the impact may be of short duration and within the Holocene limits of the community. Larcombe and Woolfe (1999b) also take up the same theme and stress the need to differentiate clearly sediment accumulation (the increase in thickness of a sediment body), a regional geological phenomenon which has probably influenced the location of reefs over the Holocene, and turbidity, which is transient both spatially and temporally. Larcombe and Woolfe suggest that in the GBR, because of the large amounts of sediment accumulated nearshore during the Holocene, turbidity is not sediment limited but instead related to the physical forces acting on the seabed. Both phenomena are thus beyond the normal temporal scale of ecology, and failure to acknowledge this and the processes involved may lead to the overestimation of anthropogenic impacts, as taken up in the next section. 13.2.2 Enhanced sediment yield to the Great Barrier Reef in the last 200 years A significant increase in sediment yield to the GBR lagoon in association with changing land-use patterns producing detrimental effects for nearshore reefs in particular has been presumed in many assessments of the status of the GBR. For example: ‘‘Changing sediment discharge patterns associated with land runoff and coastal developments have been identified as a major threat to the Reef’’ (Crossland, 1997).
440
Geomorphology’s contribution to the problems of the Great Barrier Reef
However, as well as estimating any increase in sediment yield to the GBR lagoon, the particle size of the sediments, their transport paths, and eventual settling areas are also required to assess any impact on the Reef. Sediment input to the GBR was discussed in Section 4.6 but based mainly on the review by Furnas (2003) can be summarized as follows. Terrestrial input to the GBR lagoon is about 14.4 million tonnes, 42% coming from the large basins of the Fitzroy and Burdekin Rivers. Similar calculations estimate the pre-1850 input at 4.4 million tonnes, an increase of 3.27 times and less than previously published estimates (Great Barrier Reef Marine Park Authority, 2001). The estimates are based almost entirely on the suspended silt and clay sediments carried by rivers. Furnas (2003) acknowledged that some bedload mainly in the form of sand did reach the coast but was quickly deposited in deltas or similar river mouth features. Other studies have shown that as a result of tidal asymmetry and ineffective capacity of fluvial processes in estuaries of the drier tropics that open into the GBR lagoon, net transport of sand-size sediment is landwards, not out to sea (e.g., Bryce et al. 1996, 1998). Only small steep catchments deliver sand and even gravel to the coast (such as those of Cape Tribulation) but this too is quickly deposited (Fig. 13.4). Once the suspended sediment reaches the sea, especially during floods, it may extend tens of kilometers seawards and even greater distances up the coast (usually northwards) as river plumes (Devlin et al., 2001). However, the sediment concentration even a few kilometers from the river mouths is usually within the range of 1–20 mg l1 and only in the upper few meters (Furnas, 2003) well within the range of normal resuspension turbidity, and, as indicated in the previous section, well below the tolerance thresholds of the corals of the inshore reefs. It is not often that there is opportunity for a long-term comparison of sedimentation rates on the GBR (or reefs anywhere), but on the 1928–29 Royal Society Expedition (see Chapter 1) an experiment was designed to collect sediments weekly over a 25-week period from Low Isles reef (Marshall and Orr, 1931). There has been a significant increase in the amount of land clearing on the adjacent mainland around Cairns since 1928 and in 1991–92 the experiment was replicated using fiberglass containers designed to be identical dimensions to the glass storage jars used in 1928, stationed at exactly the same reef flat locations over the same period of the year (December–June) and with sediments also collected over weekly intervals (Johnston, 1996). Weather records were kept during both studies and were shown to be statistically similar over the two time periods over the 63-year interval. Perhaps surprisingly given the amount of catchment modification of the adjacent Barron River during the twentieth century a significantly smaller
13.2 Sediments and reefs
441
Figure 13.5 Results of weekly sediment collection in 1928–29 and 1991–92 at Low Isles, with weekly rainfall totals. Although there is some correlation between the two parameters Marshall and Orr (1931) believed that afternoon winds from the north-east were the main driving force by creating sediment re-suspension (from Johnston, 1996).
amount of sediment was collected in 1991–92 than 1928–29, individual trap totals (calculated from the 44-cm2 opening) ranging from <0.1–90 mg cm2 d1 in 1928–29 to <0.1–18 mg cm2 d1 in 1991–92 (Fig. 13.5). The greatest changes were in the coarser carbonate fraction and appeared to be the result of changes to reef flat morphology. Slightly more finer sediment was collected in 1991–92 but it was not statistically significant. Results of both studies suggested that winds from the north were most influential in producing high sedimentation rates, probably resuspending material previously deposited in the lee of the reef. Low Isles reef is 13 km offshore and is clearly within the influence of the plume of the Barron and possibly Daintree Rivers, yet appears not to have experienced decline in water quality as the result of turbidity and suspended sediments. This concurs with the conclusions of Larcombe and Woolfe (1999b) that turbid plumes are not a threat to mid- and outer-shelf reefs. These authors state than in comparison to the amount of sediment delivered to the nearshore zone and resuspended by wave and current activity, the amount of new sediment delivered in the last 100 years is insignificant. Furnas (2003, p. 258) also concluded that the additional sediment input since 1950 (<500 million tonnes) is small relative to the large pre-existing sediment stocks (more than 10 000 million tonnes) in the nearshore zone. Especially taking into account the higher threshold sedimentation rates outlined in Section 13.2.1, there appear to be only limited impacts on GBR reefs of increased sediment yield alone since 1850. Sediment plumes are restricted to nearshore areas where water depths are generally shallow and, as discussed in Section 4.6.6, resuspension is frequent at depths <10 m. However, if reefs
442
Geomorphology’s contribution to the problems of the Great Barrier Reef
do extend into deeper water then deposition especially of flocculated ‘‘marine snow’’ may cause reef decline. This has been illustrated recently for the fringing reefs of High Island located south of Cairns and only 5 km offshore (Wolanski et al., 2005). The reef extends down to 20 m but resuspension in all except cyclonic conditions is possible only down to 12 m. Below this depth fine sediments are accumulating at a rate of 2000 mg cm2 yr1 and as corals appear to have previously grown to depths of 20 m, Wolanski et al. (2005) attribute the increased rate of sedimentation to soil erosion from the adjacent high rainfall catchments. Only on reefs growing in greater depths (not widespread for nearshore reefs which, as illustrated in Section 11.7, decline rapidly in growth rates below 8 m) and adjacent to severely modified catchments, will sedimentation be the only cause of reef decline. Nonetheless, there remain many reports of reef decline due to sedimentation (e.g., Zann, 1996; Williams, 2001). Undoubtedly there are reefs on which the coral cover has diminished in the last 100 years but in most instances the initial mortality has been the result of fresh water (e.g., the reefs of the Whitsunday Islands and Edgecombe Bay in 1918 after a severe cyclone) possibly with subsequent recovery retarded by nutrient enhancement. Unless the reef recovers quickly then the natural low rates of sediment settlement will slowly infill the reef giving the false impression that it is an increased sediment load which is responsible. The geomorphological context of some of the examples quoted must also be taken into account. For example, the death of reef flat corals on Holbourne Island 30 km off Bowen was originally attributed to runoff associated with the 1918 Mackay cyclone, but subsequent analysis showed that it was the result of a shingle rampart being breached by the cyclone and partially draining a reef flat lagoon (Hopley and Isdale, 1977). North of Cairns, burial of a fringing reef at Yule Point (Bird, 1971a) may be at least in part due to a change in the location of the mouth of the nearby Mowbray River. Under natural conditions the reefs of the nearshore turbid zone have experienced many perturbations but have subsequently recovered. A much greater threat than sediment or fresh water at the present time may be eutrophication.
13.3 Nutrient excess and the Great Barrier Reef 13.3.1 Nutrient enhancement and ecological response Nutrients have played an important role in the Holocene evolution of the GBR both positively and negatively. Positively, nutrient jetting through the passes between the ribbon reefs has been responsible for the Halimeda bank growth (Chapter 6). Negatively, nutrients either upwelled from the shelf margin or
13.3 Nutrient excess and the Great Barrier Reef
443
reworked by the rising sea level from the shelf edge may have played a part in the demise of the central GBR’s submerged shelf-edge reefs (Chapter 9). Today, an increase of nutrients from the land may again be playing a major negative role, the effects of which are being recorded on the mid- if not outershelf reefs. Although there may be some discussion regarding the effects of nutrient enrichment on coral reefs (e.g., Szmant, 2002) there is little doubt that over the past 150 years the nutrient levels entering the GBR lagoon have increased substantially. Nutrient input and its effects were outlined in Chapter 4 (Sections 4.6.3, 4.6.4, and 4.6.6). Summarizing the figures from Furnas (2003), pre 1850 the mainland is calculated to have exported 23 000 tonnes of nitrogen and 2400 tonnes of phosphorous annually. Today these figures stand at 43 000 tonnes nitrogen and 7070 tonnes phosphorous, increases of 1.8 and 2.9 times respectively. The large catchments of the Burdekin and Fitzroy contribute 32% terrestrial nitrogen and 38% phosphorous. The much smaller catchments of the wet tropics with 29% of runoff volume contribute 19% riverine nitrogen and 12% phosphorous. The type of nutrient and its amount depends to a large extent on land use but both sugar-cane cultivation and cattle grazing have contributed to the enhanced levels, especially over the last 50 years with further land clearing, introduction of drought-resistant cattle, and increased use of chemical fertilizers (Brodie, 2002; Furnas, 2003). Bioavailable dissolved nutrients delivered to the GBR lagoon are generally low in quantity and quickly removed by phytoplankton, algae, and bacteria. Much of the nutrient load is delivered attached to sediment particles thus emphasizing the importance of flood plumes to nearshore and mid-shelf reefs (Devlin et al., 2001). Within the plumes nitrogen and phosphorous levels may be an order of magnitude greater than normal ambient levels. As the plumes spread levels decline as the result of dilution, sedimentation, biological uptake, and chemical reactions (Furnas, 2003). Some storage of nutrients may take place as sediments drop out of suspension. Normally corals and related organisms live in a low nutrient environment and although actual evidence and observations of large-scale reef decline reported to the Productivity Commission’s 2002 Report were few, sometimes contradictory and often anecdotal (Productivity Commission, 2002), there are acknowledged changes to coral metabolism and community structure which may make recovery from natural or other acute stresses more difficult (e.g., Fabricius, 2004). Also, more recently, research along eutrophication gradients in the Whitsunday Islands has shown reductions in coral cover, species richness, and abundance combined with increased coral recruit mortality (van Woesik et al., 1999).
444
Geomorphology’s contribution to the problems of the Great Barrier Reef
An early sign of eutrophication is an increase in chlorophyll in coastal waters. Phytoplankton are very efficient extractors of nutrients and their growth supports other organisms which compete for space with corals. Macroalgae also rapidly take up and store nutrients especially in brief intervals such as during flood times. They may rapidly overgrow corals and reduce the level of light (already reduced by the phytoplankton in the water column) reaching the coral substrate, though there may be other factors involved in algal blooms related to fish herbivory (McCook, 1999). At a smaller scale, nutrient enhancement may affect the density of the corals’ symbiotic zooxanthellae (Marubini and Davies, 1996), affecting skeletal growth rates (Rasmussen, 1988; Ferrier-Page`s et al., 2000) and calcification (Marubini and Atkinson, 1999). Overall there may be a reduction in density and whilst the total cover of soft and hard corals may not change, there may be a shift to more resistant species (Fabricius and De’ath, 2001). Other detrimental effects that have been reported include increases in disease (Bruno et al., 2003) and internal bioerosion (Chazottes et al., 2002; Carreiro-Silva et al., 2005). 13.3.2 A geomorphological perspective of eutrophication Geomorphologists in cooperation with a wider range of earth scientists have been involved in research to delineate the geographical extent of nutrient enhancement and other pollution on the GBR using corals and geochemical analysis. More importantly paleo-oceanographic techniques have permitted retrospective analysis using the annual rings laid down within massive corals (e.g., Barnes and Lough, 1996; Swart and Grottoli, 2003). For example, using mid-Holocene (6.2 ka old) and modern coral cores, Gagan et al. (2002) were able to show that recent deforestation for agriculture may have led to higher land–sea hydraulic gradients and greater groundwater discharge to the GBR today. A result has been a raising of the partial pressure of carbon dioxide in sea water especially after strong summer monsoons. Consequently modern nearshore corals may now be subjected to lower aragonite saturation states which is possibly reducing coral calcification and contributing to recent coral degradation on nearshore reefs. Another example from the most recent historical past relates to the intensive period of gold mining post 1869 (Walker and Brunskill, 1997). The period 1870 to 1880 was particularly wet and resulted in the rapid flushing of mercury used in the gold extraction process. Combined sediment and coral cores indicated a 25-fold increase in mercury delivered to the GBR at this time. It is also apparent that fertilizers in agricultural runoff can leave a record in the coral skeletons. Using coral cores from Green Island Reef, offshore from
13.3 Nutrient excess and the Great Barrier Reef
445
Cairns, Rasmussen (1991) showed that both geochemical and geomechanical changes occurred in response to anthropogenic influences (see also Rasmussen et al., 1993a). Annual bands within the corals enabled the study to be extended back to 1877. The natural chemical and morphological signals showed a rhythmic periodicity until the 1950s when clear perturbations in the records took place, especially in the calcium : strontium ratio and skeletal voids. This coincided with a major period of agricultural expansion on the adjacent mainland (sugar on the coastal plain and mixed crops and dairying on the Atherton Tableland in the immediate hinterland). In particular the use of agricultural fertilizers correlated very strongly with the increased magnesium concentration post 1960. Weakening of the skeleton was a particularly noticeable change. Green Island also provides a smaller-scale example of the impact of eutrophication on coral reefs (Fig. 13.6). An initial report on the island itself (Kuchler, 1978) attributed severe erosion to the rapid expansion of seagrass beds around the island which increased from about 900 m2 in 1945 to over 130 000 m2 in 1978 (see also Hopley, 1982). The reason, subsequently confirmed by current flow studies (van Woesik, 1988), was leakage from a pipe carrying sewage across the reef flat. Under natural conditions there is a seasonal and tidal exchange of sand between the island and the reef flat (e.g., Hopley, 1981) but the binding and baffling effect of the seagrass blades interrupted this exchange, starving the island of sediment and resulting in probably unnecessary major engineering works. The link between eutrophication and seagrass expansion at Green Island has been confirmed more recently by Udy et al. (1999). A further paper (Yamano et al., 2000) has commented on the high contribution of benthic foraminifera (30%) to the beach sands of Green Island. These authors link the accumulation of these sediments to the long-term evolution of the island since the mid Holocene, and to a sea-level fall (for which there is no evidence on Green Island as it lies outside the zero isobase for hydro-isostatic uplift). Earlier studies, including a visit to Green Island by one of the present authors in the late 1960s and van Woesik (1988), make no mention of this high abundance of foraminifera, unlike, for example, on Bushy Island in the southern GBR (Hopley, 1981). In contrast, the seagrass blades have a high abundance of the foraminifera Yamano et al. (2000) report. It is interesting to postulate that this component of the cay sands is a recent addition coinciding with the expansion of seagrass beds. As Yamano et al.’s samples were collected only from the surface of the beach, sampling from deeper within the island may resolve this small geomorphological paradox. The broader geomorphological concerns for the GBR as a consequence of enhanced nutrients and their effects on the inner reefs relate to the laying down
446
Geomorphology’s contribution to the problems of the Great Barrier Reef
Figure 13.6 Green Island near Cairns in 1978. The dark areas on the reef flat are seagrass beds which had expanded to about 130 000 m2 from only 900 m2 in 1945. Sand trapped by the beds was the cause of erosion resulting in the construction of rock walls and groynes (bottom right) and artificial replenishment which caused the build-up of the lightly vegetated area (bottom left). The island has been re-engineered in the last ten years but this western end of the island still needs protective walls (photograph: D. Kuchler).
of a more fragile framework, further weakened by increasing bioerosion, especially as a rising sea level as part of climate change may increase wave energy on some reefs. A reduction in calcification rates will obviously affect the rate of reef evolution, and the degree to which this is slowed down or even terminated possibly dependent on the frequency of flood plumes. On the wet tropical coast where in addition to high rainfall, the mid-shelf reefs come
13.4 Geomorphological assessment for conservation
447
within 25 km of the coastline, the problem may become chronic producing the community changes described by Fabricius et al. (2005). The majority of mid-shelf reefs of this area are at a relatively juvenile stage (patch, crescentic, and open lagoonal), which in an unperturbed state would be producing rates of framework building and lagoonal infill at the upper end of the scale reported in Chapter 11. Initially, as communities change and framework becomes more fragile, there may be an increase in sedimentation, for example lagoon infilling, but this would not be long lasting. However, resolution of the water quality issues is entirely within the jurisdiction of Australian governments, unlike the problems associated with climate change. The implementation of the Great Barrier Reef Water Quality Protection Plan as part of a national Water for a Healthy Country Mission aims to achieve a ten-fold increase in the social, economic, and environmental benefits by 2025 (Great Barrier Reef Marine Park Authority, 2001). 13.4 Geomorphological assessment for conservation Geomorphology has contributed much to the management and conservation of the GBR Marine Park since its declaration in 1975 through workshops (e.g., Fringing Reefs Workshop, Great Barrier Reef Marine Park Authority, 1987), contractual research (such as that on Cape Tribulation described in Section 13.2.1), and provision of the initial set of 1 : 250 000 maps and accompanying gazetteer (see Chapter 5) upon which the initial zoning plans of the GBR were based (Hopley et al., 1989). However, the use of the wide range of knowledge on the geomorphology of the GBR in the management of the conservation process has been disappointingly low. In terrestrial National Parks, the geomorphology is often seen as being one of the major factors in identification and subsequent management of conservation areas. For example, Yellowstone and Yosemite National Parks in the U.S.A. are identified by their glacial and volcanic geomorphological features. Whilst the whole of the GBR may be regarded as a ‘‘geomorphological feature’’ the low esteem given to geomorphology can be illustrated by the recent rezoning of the Reef to achieve 33% as a conservation no-take zone (Day et al., 2002; Fernandez et al., 2005). This was achieved through a Representative Areas Programme (RAP) to which over 60 scientists around Australia with research experience in the GBR contributed. Although information on geomorphology was collected, as far as we are aware, no geomorphologist had direct input into the classification phase. Nonetheless, the end result of 30 reef bio-regions and 32 non-reef areas perhaps not surprisingly has a strong resemblance to previously published regionalizations based on
448
Geomorphology’s contribution to the problems of the Great Barrier Reef
geomorphology (e.g., Hopley, 1982, fig. 9.14; Hopley et al., 1989). The earlier subjective classification had only ten regions based on reef types but many of the boundaries coincide with those of the RAP reef bio-regions. The 1989 classification, based on cluster analysis of reef types within 0.50 latitude and longitude squares at the 10-cluster level, produced an even stronger regionalization based on reef morphology which may have helped in the definition of some of the ‘‘fuzzy’’ boundaries described by Day et al. (2002). The expected strong relationship between reef geomorphology and ecology is most easily illustrated by the outer reefs. In Chapter 9 the regional breakdown was into the far northern deltaic reefs, the ribbon reefs, the central GBR submerged reefs, the Pompey Complex, Swain Reefs, and Bunker–Capricorn Group. The only difference when compared to the reef bio-regions is that the latter introduces an area of strong tidal outer-shelf reefs (region RA4) between the Pompey Reefs and the central submerged reefs. Despite the scientific input and a very large public participation program (Great Barrier Reef Marine Park Authority, 2003) smaller geomorphological features of importance were not given the protection they deserved in the final plan. For example, submissions were made for all four of the blue holes described in Chapter 9 on Molar, Cockatoo, and Reefs 20–374 and 20–389 to be given a high protection status. For the GBR these are unique features and still require further scientific exploration. In the final plan, those on Reefs 20–374 and 20–389 were given a ‘‘green zone’’ status, but these have not been explored scientifically. In contrast, the features on Molar and Cockatoo Reefs for which there are published data (Backshall et al., 1979) have been given a much lower protection status. A major theme of this book has been the nature and rates of change that take place in coral reefs within geomorphological timescales. Chapter 8 discussed rates of gross geomorphological development and Chapter 11 rates of framework accretion and sediment accumulation. These rates over a period of 1000 years would certainly change the ecology of many of the reefs of the GBR, but is this the timescale for management? It may be more realistic to examine change within a timescale normally accepted by engineering, say over 100 years. Even over this shorter span, which represents only three times the period that Great Barrier Reef Marine Park has been in existence, significant ecological changes may take place, with patch reefs accreting up to 1 m, lagoon infilling by about the same amount, and fringing reefs (if still healthy) prograding by about 2 m. In Chapter 8, reef size and depth to the Pleistocene were shown as being most influential in determining the stage that reefs have reached, with smaller reefs evolving more rapidly. Lagoon reefs are ecologically one of the richest types on the GBR because of the high proportion of reef
13.4 Geomorphological assessment for conservation
449
Figure 13.7 Showers Reef (198 300 S), a small 8.8 km2 lagoonal reef which is rapidly infilling from sand aprons and lagoonal patch reefs and which will lose many of its ecological habitats in 1000 years or so.
slope environments that they contain. Karlson et al. (2004) have shown that such environments have a species richness twice as large as reef flats. Our data have shown that a small reef with only a shallow lagoon (<5 m depth) may develop into a planar reef with far fewer ecological niches in less than 1000 years (Fig. 13.7). As the average size of lagoonal reefs on the GBR is only 15.7 km2, those at the smaller end of the scale may be regarded as ‘‘endangered’’! If sea level and other environmental factors remain stable, reef evolution towards more depauperate reef systems is inevitable and there is a case in the conservation program for focusing on larger and more juvenile reefs for greatest protection so as to sustain the greatest number of habitats. Also relevant to this discussion is what defines a healthy reef, and where are corals most likely to survive any major perturbations such as those associated with global climate change (Section 13.6). The inshore reefs described in Chapter 7 have never had a high conservation standing, indeed many believe that reefs such as Middle Reef near Townsville and Paluma Shoals in Halifax Bay are largely dead. There may be many other inshore reefs yet to be identified because of high ambient turbidity levels. Like the reefs of Cape Tribulation, they may have a diversity of coral species as great as that of the outer shelf reefs (Veron, 1987). The conservation importance of these reefs may be particularly
450
Geomorphology’s contribution to the problems of the Great Barrier Reef
high at a time of global climate change. These reefs are protected from harmful UVB and shorter UVA radiation which causes coral bleaching. Although the larval stages may still be susceptible after spawning, these reefs could be important refugia from which damaged reefs can be reseeded in the future. None of these inshore reefs so far identified in the Halifax Bay–Townsville region have been given the highest ‘‘green zone’’ conservation status in the 2004 Great Barrier Reef Marine Park Authority’s rezoning program. Further research is required on the distribution and resilience of these reefs. Similar conclusions have been reached by Harris et al. (2004) for the recently discovered submerged coral reefs of the Gulf of Carpentaria. Here the reefs are protected not only by high turbidity levels, but also by their depth of >25 m. There are possibilities that the submerged reefs of the shelf edge of the central GBR may similarly act as refuges if they have any coral cover at present. Visual sampling of these reefs is limited to one or two sites only (Chapter 9) and some coral cover has been found. Most are still within the photic zone with corals growing down to –100 m on nearby Myrmidon Reef. At least those close to the surface may be aided by shelf-edge cool water upwelling. The inimical conditions that killed off or at least slowed down vertical growth to below the rate of sea-level rise have now gone and at the very least the potential for coral growth is there. As Harris et al. (2004) state: ‘‘mapping of submerged reefs should be a priority for government agencies responsible for the management and protection of these important living marine resources.’’
13.5 Management of reef islands 13.5.1 The use of the islands The reef islands of the GBR played a very important role in the development of geomorphological ideas from the mid nineteenth century onwards (Chapter 1). Before the time when underwater survey was possible and remote sensing gave a holistic view, the islands provided small ‘‘windows’’ into the evolution of the reef, fully exploited by geomorphologists such as Alfred Steers (Section 1.4.2). Chapter 10 showed that the cays and low wooded islands are amongst the youngest features of the GBR, none being more than 6000 years old, being dependent on the formation of reef flat after sea level reached its present position about 6500 years ago (Chapter 3) and when there was sufficient accumulation of sediments on the reef top for waves to concentrate them in specific areas (for earlier review of island types and formation see Hopley, 1982, 1997b). The islands are important for both ecology and human use. Some 215 species of birds are attracted to the GBR with 22 species of seabirds and 32
13.5 Management of reef islands
451
Figure 13.8 Michaelmas Cay near Cairns, a lightly vegetated sand cay and third most important breeding site on the Great Barrier Reef for birds, including 20 000 sooty terns and 8000 common noddies.
species of shorebirds nesting or roosting on the islands, mainly on the islands between Bramble and Michaelmas Cays, the Swain Reefs, and the Bunker–Capricorn Group (Fig. 13.8). Also found are six species of turtles, with three (green, loggerhead, and hawksbill) nesting almost exclusively on the coral cays. The use of the cays by these birds and turtles is an adaptation of the last few thousand years only, as earlier than 6 ka ago no reef islands would have existed not only in GBR waters but throughout the southern hemisphere where there was a similar relative sea-level history for the Holocene (Chapter 3). Post European settlement of the islands dates back to the nineteenth century, when many, such as Raine Island, were mined for their phosphatic guano deposits (Stoddart et al., 1981). Also in the same period, islands such as Green Island were used as bases by beˆche-de-mer fishermen. Between 1904 and 1932 turtle-canning factories were set up on a number of the Bunker–Capricorn islands, including Heron Island (P. Olgilvie, pers. comm.). As this industry came to a close in 1932, various other extractive industries such as fish canning and drying were suggested, but in 1936 the Heron Island lease was taken over by the Poulsen family and a tourist resort established, helped by the island being declared a National Park in 1943. Green Island had a similar history and was used by Cairns residents as a fishing base from 1889 onwards (Green
452
Geomorphology’s contribution to the problems of the Great Barrier Reef
Island Management Committee, 1980). It became a recreational reserve in 1906 based mainly on day visitors. Its National Park status was declared in 1937, and the lease for a tourist resort applied for in 1938. These resorts remain today with only one further addition on a reef island, on Lady Elliot Island where the resort was built after 1970. All other resorts are on high continental islands. However, with tourism booming on the GBR many islands are visited on a day basis though under strict control regarding carrying capacity and environmental protection. For example, at Michaelmas Cay near Cairns, one of the most important bird nesting sites on the GBR (Fig. 13.8), the maximum number in tourist groups has been limited to 15 people and vessels mooring at the reef have to be less than 35 m. One of the oldest uses of the islands has been as sites for navigational beacons. One of the earliest was the light tower built in 1844 from phosphatic cay sandstone on Raine Island (Stoddart et al., 1981). The main priority was the safety of the inner passage, between the mainland and the main reef tract, which north of Cairns can be as narrow as 3 km. A key figure in the establishment of many of the beacons was George Poynter Heath (1830–1921) who was appointed Marine Surveyor for Queensland in 1859 and subsequently Portmaster of Brisbane and Chairman of the Marine Board of Queensland (1862–90). He worked tirelessly in improving navigational aids, especially in the far north. He became known as the ‘‘father of Queensland lighthouses’’ (Bowen and Bowen, 2002). The first lighthouses on many of the reef islands belong to this period. 13.5.2 Mobility of the islands As the islands are composed of unconsolidated sediments (plus some beachrock and other cemented deposits) and their position, determined by wind and weather patterns (Section 4.4.2 and Fig. 4.11), can change over time, even vegetated cays can be subjected to erosion, aggradation, and changing orientation. Changes to the reef top sediment budget can also have dramatic effects on the adjacent island as illustrated by Green Island (Section 13.3.2). Geomorphological surveys documenting changes to a number of the 300 reef islands of the GBR (Hopley et al., 1989) have been made at temporal scales from 45 years (comparing mapping of the northern islands on the 1928 Royal Society Expedition and subsequently by Steers in the 1930s with surveys on the same islands on the 1973 Royal Society–Universities of Queensland Expedition (Stoddart et al., 1978a, b)) to single tidal cycles (Hopley, 1981, 1982). Most other surveys have depended on aerial photography and thus commence, at their earliest, from the 1940s. Islands for which several surveys have been
453
13.5 Management of reef islands
possible include Coconut Island in Torres Strait (Hopley and Pichon, 1994; Hopley and Rasmussen, 1998), Warraber Island, also in the Torres Strait (Rasmussen and Hopley, 1995), Raine Island (Gourlay and Hacker, 1991), Low Isles (Rasmussen, 1986), Green Island (Kuchler, 1978; Beach Protection Authority Queensland, 1989), cays of the Swain Reefs (Flood and Heatwole, 1986) and in a series of long-term studies the islands of the Bunker–Capricorn Group, including North Reef, Erskine, Wilson, Tryon, Heron, and Masthead, by Flood (1977, 1979a, b, 1983b, 1984a, b, 1985, 1986, 1988). Many of the results have been synthesized by Aston (1995). Island type, size, and vegetation cover all have major and obvious controls on rates of change, the amplitude of which varies with the time interval of the monitoring. For example, at the shortest end of the scale Hopley (1981, 1982, ch. 11) used beach profile measurements and sediment traps on the adjacent reef flat to estimate the amount of diurnal change on three contrasting cays: unvegetated Wheeler Cay (Fig. 8.4b), maturely vegetated Bushy Island on Redbill Reef (Fig. 8.11), and what may be referred to as a senile vegetated cay on Three Isles, a low wooded island. Results are summarized in Table 13.3. Unvegetated Wheeler Cay has a mobility an order of magnitude greater than the other islands (Fig. 10.13). An annual cycle of change is illustrated by Coconut Island (Fig. 13.9) where surveys were undertaken at two-monthly intervals at 21 sites in 1996–97 (this long narrow island is approximately 40 ha in area). Most mobile was the western spit of the island, a feature recorded on almost all cay surveys. An
Table 13.3. Contrasting beach changes, reef-flat sediment movement, and cay migration
Range of mean wind speeds (m s1) Mean daily beach change (cm) Maximum mean daily beach change (cm) Mean daily sediment trap recovery (g) Maximum mean daily sediment trap recovery (g) a Recorded long-term migration (m) Period a
Wheeler Cay
Bushy Island
Three Isles
2.0–7.5 17.46
2.5–10.5 2.74
7.5–13.0 1.51
24.95
3.76
1.98
531.1
204.2
22.8
793.1
344.4
39.7
110 1975–77
40 1936–74
35 1929–73
This is the maximum mean recovery from all traps set. Individual traps have recorded up to 4063 g on a single day on Wheeler Cay.
454
Geomorphology’s contribution to the problems of the Great Barrier Reef
Figure 13.9 (a) Coconut Island (in Torres Strait) showing the mobile spit at the western end of the cay and the proximity of the island to deep water. (b) Coconut Island harbor walls acting as groynes and steering sand moving anticlockwise around the island (left to right on photo) into deeper water and starving the downdrift beach which is now almost entirely beachrock.
annual pattern of short-lived erosion during the summer north-westerly monsoon and return of sediment over the mid year (winter period) reinforces the pattern of interchange between beach and reef flat. However, during the period of survey a net loss of about 20 000 m3 from the island was recorded, equivalent to about 1% of the total (Hopley and Rasmussen, 1998).
13.5 Management of reef islands
455
Longer-term surveys using aerial photography and/or ground survey produce results of similar dimension. On Raine Island Gourlay and Hacker (1991) reported erosion of up to 20 m and accretion of up to 40 m between 1967 and 1990 as the cay became more elongate. Annual sand movement was about 5000 m3. On Warraber Island maximum shoreline movements were about 60 m over a 20-year period (Rasmussen and Hopley, 1995) and at Green Island about 50 m between 1945 and 1978 (Kuchler, 1978). Flood’s measurements along fixed transects on cays of the southernmost reef showed maximum movement on profiles of 116 m (average 53 m) on Erskine Island, 88 m (average 28 m) on Tryon Island, 110 m (average 23 m) on Masthead, and 69 m (average 14 m) on Heron Island between 1972 and 1986 (Flood, 1988). 13.5.3 Causes of island changes As reef islands are formed by wind and wave processes moving reef sediments to fixed points on the reef flat, and determining the shape of the depositional area, any change to weather patterns may be expected to produce a response in the cay geomorphology. Indeed, the daily changes measured at Bushy and Wheeler Cays were linked to weather and tide. However, long-term changes in weather may result in unidirectional changes to the islands, an important consideration at a time when global climate change appears to be accelerating (Section 13.6.3). Two analyses of weather conditions and their correlation with beach changes have been undertaken on the GBR, one in the far north at Warraber Island (Rasmussen and Hopley, 1995) and the other on Erskine (Fig. 10.14) and Heron Islands in the southern end of the Reef (Flood, 1986). For Warraber, the analysis period was 1951–92 the analysis being for nearby Thursday Island. The analysis showed that the pattern for the north-westerly summer monsoon has changed little over the period but that there is a statistically significant increase in winter winds from the south-east and a general increase in wind speeds between 7 and 10 m s1, the change taking place between 1975 and 1977. The result appears to be a realignment of the island in a clockwise direction producing small changes to its oval shape. The analysis of wind data for Heron Island was for the period 1962–80 and indicated that the annual wind energy vector had oscillated within a 458 arc from south-south-east in the early 1960s to east-south-east in the 1970s (Flood, 1986). Shoreline changes at uninhabited Erskine Island reflected this change and influenced Flood to conclude that erosion on Heron Island may not be caused by the engineering works associated with the tourist resort at this site.
456
Geomorphology’s contribution to the problems of the Great Barrier Reef
Superimposed on the long-term changes are the effects of tropical cyclones which may pass within the vicinity of coral islands. Several studies have documented such changes. The period of Flood’s monitoring of Erskine Island included three cyclones: Emily 1972, David 1976, and Simon 1980. The effects were variable depending on the path and intensity of the storm and the stage of the tide. The net effects may be erosional if water levels are high (Flood and Jell, 1977) or constructional if water levels are low (Flood, 1980, 1981). Examples include approximately 27 m of rubble prograding the south-eastern side of One Tree Island during Cyclone Emily in 1972 (T. J. Done, pers. comm.) or erosion of up to 60 m on the unstable spit of Green Island during Cyclone Winifred in 1986 (Muir, 1986). An interruption to the sediment budget of a coral cay may also cause erosion as illustrated by the reef flat sediment trapping on Green Island as the result of eutrophication (Section 13.3.2). Sand may be lost off the lee side of the reef, especially if the mobile island spit is close to the reef edge (Figs. 13.9a and 13.10a). Erosion of the cay will result if this is not replaced by new sand from the reef flat. Many reef flats with islands on the GBR are becoming increasingly senile with organisms producing only low amounts of new carbonate sediments and the height of the reef flat on inner reefs at least, as the result of relative sea-level fall not allowing efficient transport of sediment by transmitted waves beyond the reef edge on anything except the highest tides (see Section 13.6.3). Sediment budgets, especially where there are distinctive seasonal movements of sand, may also be affected by man-made structures (Gourlay and Flood, 1981; Gourlay, 1983b; Hopley, 1988). Examples include the placing of various groynes on Green Island between 1939 and 1980 (Beach Protection Authority Queensland, 1989), all of which interrupted the clockwise movement of sediment around the island and caused severe downdrift erosion at the site of the tourist resort and original jetty abutment. Attempts at artificial replenishment by pumping 18 250 m3 of sand from the adjacent boat basin between 1973 and 1975 were equally unsuccessful as they were carried out without an appreciation of basic geomorphological processes. The sand was too fine to remain where it was wanted and merely moved around to its equilibrium area on the north-western side of the island (Fig. 13.6). A concrete retaining wall now surrounds the western end of the island and forms the abutment for the new jetty. Heron Island has had similar problems. A channel dredged through the reef flat in about 1945 has proved to be a conduit for sand loss from the island. Its deepening in 1966 to form a harbor and placement of the dredged sand on the adjacent beach merely accelerated the sand loss. Coconut and Warraber
13.5 Management of reef islands
Figure 13.10 (a) Vertical air photograph of MacLennan Cay (118 190 S). Though lightly vegetated, this is one of the most unstable cays of the Great Barrier Reef, characteristically long and narrow with highly mobile spit very close to deep water (photograph: Queensland National Parks and Wildlife Service). (b) Masthead Island in the Bunker–Capricorn Group is one of the most stable cays due to its size (32 ha) shape and location on the reef. Nonetheless some instability is indicated by the younger vegetated area on the near right of the image (photograph: H. Kan).
457
458
Geomorphology’s contribution to the problems of the Great Barrier Reef
Islands in Torres Strait have similar loss of sand because of harbor developments. In these examples the retaining walls direct the sand to the reef edge where it is totally lost from the system (Fig. 13.9b). 13.5.4 The management problem At least in the near future there is a problem on the cays only where they are being used for human activities. In addition to concerns for tourist resort infrastructure, the effects of cay mobility on navigational structures have been a cause of concern for even longer periods. For example Gleghorn (1947), a Queensland lighthouse inspection engineer, reported a beacon mast with a 450-lb screw pile attached was carried about 90 m from its site during a cyclone in 1943 on Fisher Reef north of Lizard Island. Such forces and accompanying changes to the cays are not always predictable. In November 1975 an engineer inspecting a weather station on Gannet Cay in the Swain Reefs reported ‘‘the cay is high level and grassed, so piled foundations were not required’’ (Gourlay, 1990). By 1979 it became necessary to construct a new weather station before the original one collapsed into the sea as the cay moved away from it (see also Fig. 10.3b). In some instances, the problem results from failure to recognize the oldest and therefore the most stable part of reef islands, which can be done on the basis of soils and vegetation (Figs. 10.6 and 13.10b). However, management decisions such as the location of navigational structures would be very much helped if an index of mobility of reef islands were available. Research by Aston (1995) used data from islands for which results of monitoring of movement over a significant period were available (such as Green, Low, Heron, and Wheeler) and related morphological and dynamic data to the degree of island change over various timescales. Multivariate statistical analysis techniques were used to discuss patterns in the dependent variable (mobility) resulting from the influence of the various primary independent (form and process) variables. Morphological variables included cay area, vegetated area and percent vegetated, perimeter length, major and minor axes, centroid, shape, and volume. Dynamic variables included wind and cyclone data. Predictably, the complexity of the cays and their mobility precluded any monotypic description of mobility for different cay types. The most important factors were: * * *
unvegetated cays: major and minor axes and ellipticity index vegetated cays: percentage of vegetated area vegetated cays on low wooded islands: a volume equation for an ellipsoid shape.
13.6 Global climate change, geomorphology, and coral reefs
459
The model was applied to the other cays of the GBR, initially testing it on islands such as Erskine, Wilson, Tryon, and North Reefs for which a smaller data bank existed. Observations from about 100 cays were then tested against the refined model data. Unvegetated cays such as MacLennan (Reef 11-070) (Fig. 13.10a), Undine (Reef 96–020), and Sudbury (Reef 17-001) not surprisingly came out the most unstable. Of the vegetated sand cays tested, elongate islands such as Wreck Island (Reef 23-051) were the most mobile whilst nearoval cays, including Bushy (Reef 20-310) and Wilson (Reef 23-050) Islands, and the largest cays, such as North-west Island (Reef 23-049) and Masthead Island (Fig. 13.10b) were predicted as the most stable. Heron Island was also in this group, and was indicated as far more stable than Green Island, perhaps suggesting that the effects of the harbor are a significant factor in its recent erosion history. At a time when sea-level rise associated with climate change is seen as a threat to the islands of the GBR, the availability of these data may help in decisions on which islands may best benefit from higher conservation status. The question of climate change, coral reefs, and geomorphology is taken up in the next section.
13.6 Global climate change, geomorphology, and coral reefs 13.6.1 Projected changes and major impacts Coral reefs are one of the ecosystems for which greatest concerns are held in relation to global climate change (Buddemeier et al., 2004). Warming of ocean temperature resulting in coral bleaching, in which the coral expels its zooxanthellae, has been reported worldwide, and in the GBR especially in 1998 and 2002. Recovery may or may not take place but is impeded by chronic anthropogenic stresses. In the latest projected changes an increase in sea surface temperatures of only 1–3 8C may not seem high but it is the extremes that are imposed upon this steady increase which have the greatest impact. Corals are acclimatized to their ambient ocean temperatures and whilst similar species may live in a wide range of temperatures (the Red Sea is often quoted as an example of the extreme temperature within which corals may live), at any particular site there may be a relatively narrow acceptable range (Kleypas et al., 1999a). Some adaptation may be possible. For example, the adaptive bleaching hypothesis (Buddemeier and Fautin, 1993) suggests that corals may host different types of zooxanthellae and that during a bleaching episode those least suited to the applied stress are lost in favor of less vulnerable types. However, recent increases in the frequency and intensity of conditions that contribute to bleaching may have outpaced the compensating mechanisms.
460
Geomorphology’s contribution to the problems of the Great Barrier Reef
In Australia a number of regional assessments of projected changes have been made, the most recent by the Australian Greenhouse Office (2005). Projections include: *
* * * *
*
an increase in average temperature, of between 0.4 and 2.0 8C by 2030 and of between 1.0 and 6.0 8C by 2070. more periods of extreme heat more frequent El Nino–Southern Oscillation (ENSO) events ˜ rainfall increases across much of the tropical north more severe wind speeds in cyclones and associated storm surges being progressively amplified a possible change in ocean currents affecting coastal waters.
All these changes have the potential to impact the GBR. Whilst bleaching affects the coral tissue projected increases in atmospheric carbon dioxide may drive a reduction in ocean pH to levels not seen for millions of years (Caldeira and Wickett, 2003), which affects the rates of calcification in corals and other reef organisms (Langdon, 2002). Laboratory experiments show coral calcification rates decreasing by 11–37% and calcareous algae by 16–44%. Increased atmospheric carbon dioxide concentrations also result in decline in reef calcification (Kleypas et al., 1999b). Decline in calcification rates results in either a slower skeletal extension rate or the laying down of a less dense skeleton. If the former, corals may lose ability to compete for space on the reef. If the latter, then they are less resistant to breakage and more susceptible to bioerosion. Future changes to sea-water chemistry may also lead to increases in calcium carbonate dissolution once carbon dioxide levels in the atmosphere reach double the pre-industrial levels (Halley and Yates, 2000). Sea-level rise resulting from the combined effects of thermal expansion of sea water and melting of glaciers and ice caps will also have an important effect on coral reefs, perhaps one of the few changes which may have some beneficial effects though possibly not for low-lying ocean island nations (see below). The latest projection is for a rise between 0.1 and 0.9 m by 2100, a rate which was exceeded at times during the postglacial transgression. Although these changes are referred to as ‘‘projected’’ rather than ‘forecast’ the growing weight of environmental data showing that climate changes are already taking place gives confidence to the projections. These include: *
*
the atmospheric concentration of carbon dioxide has increased 31% since 1750 and from about 310 ppm in 1950 to 372.3 ppm in 2002 (Carbon Dioxide Information Analysis Center, 2005) global average surface temperature has increased by 0.6 0.2 8C over the last century; the 1990s was the warmest decade on record with 1998 the warmest and
13.6 Global climate change, geomorphology, and coral reefs
*
*
461
2001 the second warmest on record (Intergovernmental Panel on Climate Change, 2001; Tiempo, 2002). In Australia 2005 was the warmest year ever recorded (data from Australian Bureau of Meteorology) average sea level has risen between 10 and 20 cm during the twentieth century, mainly as a result of thermal expansion (International Panel on Climate Change, 2001) other changes have included decreases in snow cover, changes to rainfall patterns, and, since 1920, the occurrence of ENSO episodes (International Panel on Climate Change, 2001).
Although there is much discussion (and a small amount of action) to cut greenhouse gas emissions, it must be remembered that it has taken at least 250 years of anthropogenic activity to reach the present situation and even the most rigorous abatement measures are going to take decades if not a hundred years or more to take effect. In the meantime, an understanding of the environmental responses including geomorphological process will help in putting in place the most effective mitigation planning. 13.6.2 Geomorphological responses When climate change first became a major environmental focus in the late 1980s, many extravagant claims were made about the future geomorphology of coral reefs. Even though the sea-level rise was only a meter or so it was suggested that barrier reefs would be drowned and their breakwater effect for coastlines previously protected by the reefs completely removed. Contours were merely lifted around reef islands many of which were forecast to disappear completely. Some beneficial effects were seen in new coral growth occupying reef flats and, with higher temperatures, a poleward extension of reefs (even though suitable temperatures are not the only factor in reef ‘‘turn-on’’ (Buddemeier and Hopley, 1988) and the fact that today’s coastlines are heavily populated and already under stresses which would not make coral migration easy). Sea-level rise, if rapid enough (see below), will revitalize reef flats of the GBR which have been at sea level for up to 6000 years and even experienced a small fall in level (Chapter 3). These senile flats (Fig. 13.11) may have considerable amounts of sediment resting upon them, but water depth is sufficient for wave action to move that sediment only on the highest tides. For example, Kench and Brander (2006) show that on the wide Warraber Reef flat in Torres Strait, waves greater than 0.05 m occurred for less than 30% of a spring neap tidal cycle on the outer reef and waves of 0.1 m occurred only 19% of the time, and for most of the time the reef flat is geomorphologically inert. A rise in sea
462
Geomorphology’s contribution to the problems of the Great Barrier Reef
Figure 13.11 Reef flats which have been at sea level for up to 6000 years or even experienced a small fall in level may have a significant sediment cover but are geomorphologically inert for most of the tidal cycle. Coconut Island Reef, Torres Strait.
level will increase reef flat wave energy but, with the noted changes in ocean chemistry, the increase in atmospheric carbon dioxide and further eutrophication in the increased runoff resulting from higher rainfall, any new coral growth is likely to be much more fragile. Combined with an increase in cyclonic storm intensity and storm surge height, the end result may not be aesthetically pleasing living coral flats, but reef tops and leeward zones dominated by rubble banks. A close analogy may have been the reef tops when the Pleistocene foundations were first submerged in the early Holocene (Chapter 11, Fig. 11.8), when similar conditions produced the basal rubble seen within many GBR reefs. A similar result may follow widespread mortality after bleaching. Sheppard et al. (2005) have shown that the breakwater effect of fringing reefs in the Seychelles has broken down since the 1998 bleaching event, and subsequent coral mortality and disintegration. Higher wave energies are now reaching the beaches and large rubble banks are accumulating, a situation that may be more widespread in the future. There is the possibility that if reefs remain healthy this renewal of growth and calcification may accelerate the rates of reef evolution discussed in Chapter 8. Kinsey and Hopley (1991) estimate that the current calcification
13.6 Global climate change, geomorphology, and coral reefs
463
rate for the GBR is about 50 million tonnes yr1. Recolonization of presentday reef flats could increase this production substantially by about 40% to 70 million tonnes yr1. Unfortunately, initial hopes that increased coral growth would act as a global sink for carbon were dismissed when it was shown that the immediate effect of CO2 precipitation is to raise PCO2 of the surface oceans, i.e. to encourage CO2 efflux to the atmosphere (Ware et al., 1992). 13.6.3 Response of reef islands to sea-level rise ‘‘It is estimated that a 60 cm rise around the Maldives in the Indian Ocean would cover these coral islands and displace 177 000 people’’ (Falk and Brownlow, 1989, p. 65). Many statements such as this, made around 1990, took little note of the geomorphological processes involved in coral island formation, and merely moved the contours up the island beach. Hopley (1981) was one of the first to monitor sand movement around cays and came to the conclusion that water depth was the critical factor in transporting sediment to the beach. This has been confirmed by Gourlay (1990) and most recently by Kench and Brander (2006) using more sophisticated equipment. Sediment movement is restricted to less than 50% of the time when water levels over the reef flat are sufficiently deep to allow waves of sufficient size and transportational capacity to pass over the reef. This is particularly significant in an area of mesotidal range such as the GBR. As many of the reefs, especially on the inner shelf, have experienced a Holocene fall in sea level of a meter or more (Chapter 3), or have been at sea level for several thousands of years, reef flat height is an impediment to sediment movement but conversely has allowed for build-up of reef flat sediments more or less immobile in all except major storms. Woodroffe (2002b) has shown that Warraber Island in Torres Strait and other islands in the IndoPacific ceased to accumulate sediments about 2000 years ago, indicative of a falling sediment supply from the adjacent reef flat, possibly related to the ineffectiveness of reef flat waves to move sediment towards the island. A rise in sea level of even 0.5 m may unlock these sediments and move them towards the focal point of wave refraction, the reef island (Hopley, 1993). Build-up of the cay itself will depend on the characteristics of the waves that break on the cay beach. The height of the beach berm is dependent on wave run-up. Numerous studies (e.g., Coastal Engineering Research Center, 1984) show that run-up height rises with wave height, wave steepness, beach slope, shape of the beach profile, and roughness and permeability of the beach material. Although most beach models and empirical formulae suggest that
464
Geomorphology’s contribution to the problems of the Great Barrier Reef
given adequate sediment supply (not a problem on many GBR reef flats), there have been few applications to coral islands. However, Gourlay and Hacker (1991) working on Raine Island found that the relative wave run-up height varied in a consistent manner with the ratio of the breaker height to water depth over the reef flat, consistent with the fact that the wave heights are limited by shallow water breaking conditions. They indicated that the height of the beach berm is determined by the run-up height of the dominant wave action. This could be expected to recur at the highest spring tides. A berm elevation of 4 m could be built by small flat waves of 0.5 m height breaking directly onto the beach at a tide level as low as 2.3 m. They also showed that similar heights could be attained by the maximum breaking waves of 1.6 m at an extreme tide level of 3 m. Thus a small rise in sea level without any responding build-up of reef flat level would result in the attainment of greater berm height under most weather conditions, i.e., a build-up of the island by an amount which could exceed the amount of increase in water level. For example, at Raine Island, Gourlay and Hacker (1991) suggest that with a 0.6 m rise in sea level, the larger 1.6 m waves would increase berm height by a factor of 0.8 m and 0.5 m waves would increase the height by 1.2 m, i.e., to 4.8 m and 5.2 m from the initial 4.0 m height. Reef geomorphologists who have worked on reef island processes generally agree that rising sea level of the magnitude projected over the next 100 years will produce substantial reworking of surficial sediments, enough to at least maintain the island mass (e.g., McLean, 1989; Parnell, 1989; Hopley, 1993, 1997a, b; Kench and Cowell, 2002). As McLean (1989) suggests this will be aided by present beachrocks and other cemented materials on the reefs. As cementation is a rapid process, new beachrocks and conglomerate may form at higher levels and help in island stabilization. Once the presently immobile reef flat sediments are exhausted as they are moved to the island or lost over the edge of the reef, the amount of new sediment available will depend on the amount of sea-level rise. Hopley (1997a) modeled an atoll reef flat responding to a 0.5 m and 1.8 m rise by 2100 to produce contrasting budgets of carbonate production. Known productivities of each zone were entered into the model, together with their widths. The results showed that at the more modest rise, almost all reef zones would be able to accrete vertically and reef morphology would not change. However, at the higher rate of rise, the algal zone would convert to a reef cover and the shallow reef flat limited at present by water level would be able to grow and calcify at much higher rates. Accepting that at least a proportion of their new carbonate productivity would be converted into sediments it would seem that paradoxically a higher rate of sea-level rise may be more beneficial for reef islands.
13.6 Global climate change, geomorphology, and coral reefs
465
Table 13.4 models a typical GBR reef flat and coral cay in the same way that atoll budgets were calculated. The zone widths are calculated based on a number of Bunker–Capricorn reefs, especially Masthead Island, and calcification rates from Kinsey (1985). Results show that the more modest sea-level rise of 0.5 m, which is much closer to the projected level in 100 years time than the 1.8 m rise projected in the early 1990s (Intergovernmental Panel on Climate Change, 1990), reduces carbonate productivity by almost 40%. Changes to reef roughness (Kench and Brander, 2006) may also become a factor which in the long term (>50 years) could reduce the effectiveness of wave transport and result in the shallow reef flat lagoon becoming a sediment sink again. Increased cyclone intensity is not thought to be a negative factor. As has been discussed above, cyclones may add to rather than erode coral cays and storm surge levels on mid-shelf reefs are far lower than on the mainland where topography and shoaling enhances surge levels. For example Cyclone Emily in 1972 produced surge levels of 2 m on the mainland near Gladstone but in spite of passing directly over Heron Island, with a central pressure of 985 hPa, produced a surge of less than 0.8 m (Hopley, 1972). Some characteristics of the islands will change. Any movement of cay location or overtopping with new sediments will remove the oldest soils and most mature vegetation, and vegetation succession may have to recommence from early stages. This may also release nutrients stored in the soil and especially in guano deposits and phosphatic cay sandstone, the normal leaching of which may already be producing eutrophic conditions (Chen and Krol, 1997). Loss of Pisonia grandis, the climax vegetation of many GBR cays, would have serious implications for many seabirds, e.g., terns on islands such as Heron. However, Michaelmas Cay near Cairns is at an early stage with only five species of vascular plants but is still one of the major nesting and roosting sites on the GBR. Also producing changes to the vegetation is the possible rise in saline groundwater levels. Again a full understanding of the physical processes involved produces a different prognosis compared to some of the more simplistic approaches using the Ghyben–Herzberg model (e.g., Miller and MacKenzie, 1988). This model assumes homogeneous materials and that the outflow, or loss of fresh water, occurs at island margins in, or below, the intertidal zone producing a predominantly horizontal flow. Although some of the fresh water remains above sea level, the greater part of it (40 units of depth for every unit of head) will reside below sea level. The model also produces a great depletion in the potential for groundwater resources if island size is reduced. Freshwater lenses may not occur on islands that are less than 300 m wide. However, the more recently developed layered-aquifer model (Wheatcraft and Buddemeier, 1981; Herman et al., 1986; Oberdorfer
Outer coral Algal ridge Coral–algal zone Shallow reef flat Rubble/sanded reef flat Total
Zone 10 4 5 1.5 0.5
0 þ0.5 0 0.5 þ0.25
200 3445
Present CaCO3 production (kg m 2 yr 1)
65 80 100 3000
Nominal width (m)
Nominal height in relation to low water mark (m)
0.3
7 3 3.5 1
Equivalent upward growth rate (mm yr 1)
100 2020
650 320 500 450
Present CaCO3 production (kg yr 1 along 1-m wide transect)
1000 7950
650 800 1000 4500
1.8 m SL rise
300 3120
650 320 500, perhaps less 1350
0.5 m SL rise
Production by 2100
Table 13.4. Calcium carbonate productivity changes on a typical reef flat in the Great Barrier Reef
13.7 Conclusion
467
and Buddemeier, 1986; Buddemeier and Oberdorfer, 1990) has much greater applicability to real-world situations though it does suggest that the overall freshwater resource may be less than in the Ghyben–Herzberg model. The layered-aquifer model presumes two basic geological layers possessing distinct porosities: a surficial layer, of Holocene age, of low permeability overlying deposits of high permeability, of Pleistocene age, separated by a solution unconformity at relatively shallow depths of 7–25 m (Fig. 4.8). Primary mechanism for loss of fresh water is not outflow at island margins, but loss to degradation by downward mixing into the saline water in the Pleistocene deposit below. This creates a broad transition zone of brackish water. However, this model is far less sensitive to island size, and a threshold island width figure of 120 m has been suggested for retention of the freshwater resource. Moreover, if the island size remains constant, Oberdorfer and Buddemeier (1990) have suggested that rising sea level has a counter-intuitive effect on total freshwater resource for islands possessing a layered aquifer. An increase in sea level makes available more of the low-permeability aquifer for retention of fresh water, increasing the total freshwater resource. Thus, a rise in sea level may not be disastrous for island groundwater resources. Indeed, if accompanied by an increase in island size as seems likely and also by increases in rainfall as is predicted for some areas of low latitudes, groundwater resources may actually increase. 13.7 Conclusion The five themes chosen in this final chapter illustrate that reef geomorphology is an important physical science which contributes essential understanding of the way in which coral reef ecosystems work. This can be through the measurement of processes operating today or, as indicated in Section 13.1, over timescales as long as or longer than the Holocene. This type of approach, which has been the focus of this book, may allow short-term perturbations in process rates to be filtered out or for important changes which create impacts more permanent than those of, for example, an individual storm or bleaching event to be identified. Short-term monitoring, for example of sedimentation rates, can lead to the definition of important environmental thresholds. Longterm studies produce rates of growth and change at the reef rather than colony level. Both are important components of information needed for management purposes. In an increasingly changing environment, recognition of the most appropriate reefs and islands for special protection is a management priority and examples of applied geomorphological data have been given in this chapter. High-turbidity reefs which may have protection from bleaching are
468
Geomorphology’s contribution to the problems of the Great Barrier Reef
thought to have a wider distribution than is currently recognized and identification of such reefs where they are not exposed to other anthropogenic stresses would be an important step in their protection as future refuges. The submerged reefs are in a similar position and they could be major refuges over the next century before climate mitigation programs take effect, but much more information about them is required. Reef islands may be under less threat than is generally thought, but identification of those most suitable for special protection definitely requires a geomorphological input. The GBR is the best-managed coral reef in the world, yet there are predictions that it could be completely degraded by the middle of this century (e.g., Wolanski and De’ath, 2005). A major drawback for management is that the political focus and timescale for dealing especially with climate change and the Reef is totally inappropriate. ‘‘The period through to 2030 and to a lesser extent 2050 is the one that is most relevant today for decisions about adaptation strategies. This is because most discussions that could be affected by climate risks involve assets and business systems whose economic life falls within or near this time horizon.’’ (Australian Greenhouse Office, 2005, p. vii) Ecological timescales through which reefs are perceived as ‘‘fragile’’ and geoscientific timescales through which reefs have been seen as ‘‘robust’’ (Section 1.2) lie either side of this economic line, the application of which could lead to the fragility of the system being exposed without its robustness being allowed to take over if sufficient mitigation action is carried out quickly enough. Just as the study of reefs needs to be multidisciplinary combining biological and physical sciences, so too is there a need for understanding of the various time-period scales that the individual disciplines bring to the table, but these should not be dictated by economics or politics. ‘‘Coastal geomorphology is built on a social, empirical and descriptive foundation. It has been at the forefront of development of ideas in the earth sciences. Paradigms have shifted and we now have powerful conceptual frameworks within which to understand variability on the coast. The tools available to future coastal scientists promise unprecedented sophistication in terms of research and monitoring at spatial scales and over time frames never before anticipated.’’ (Woodroffe, 2002a, p. 497) These comments are particularly applicable to reef geomorphology and it would be unfortunate if the GBR was to suffer irreparable harm before these new tools were fully applied. This book has tried to summarize the geomorphological advances on the GBR over the 25 years since Hopley’s 1982 Geomorphology of the Great Barrier Reef was written. The present pace of change to both geomorphology and the environment suggests that more frequent reviews will be required in the future.
References
Abel, K. M. and Drew, E. A. 1985. Response of Halimeda metabolism to various environmental parameters. Proc. 5th Int. Coral Reef Congr., Tahiti 5, 21–26. Adey, W. H. 1978. Coral reef morphogenesis: a multidimensional model. Science 202, 831–837. Adey, W. H., Macintyre, I. G., Stuckenrath, R., and Dill, R. F. 1977. Relict barrier reef system of St. Croix: its implications with respect to late Cenozoic coral reef development in the Western Atlantic. Proc. 3rd Int. Coral Reef Symp., Miami 2, 15–21. Agassiz, A. 1898 A visit to the Great Barrier Reef of Australia in the steamer Croydon during April and May, 1896. Bull. Mus. Comp. Zool. Harvard Coll. 28, 95–148. Aharon, P. and Chappell, J. 1986. Oxygen isotopes, sea level changes and the temperature history of a coral reef environment in New Guinea over the last 105 years. Palaeogeog. Palaeoclimatol. Palaeoecol. 56, 337–379. Aharon, P., Chappell, J., and Compston, W. 1980. Stable isotope and sea-level data from New Guinea supports Antarctic ice-surge theory of ice ages. Nature 283, 649–651. Ahmad, W. and Neil, D. T. 1994. An evaluation of Landsat Thematic Mapper (TM) digital data for discriminating coral reef zonation: Heron Reef (GBR). Int. J. Remote Sensing 15, 2583–2597. Aller, R. C. and Dodge, R. E. 1974. Animal–sediment relations in a tropical lagoon, Discovery Bay, Jamaica. J. Mar. Res. 32, 209–232. Alonghi, D. M. and McKinnon, A. D. 2005. The cycling and fate of terrestriallyderived sediments and nutrients in the coastal zone of the Great Barrier Reef shelf. Mar. Poll. Bull. 51, 239–252. Andrefouet, S., Berkelmans, R., Odriozola, L., et al. 2002. Choosing the appropriate spatial resolution for monitoring coral bleaching events using remote sensing. Coral Reefs 21, 147–154. Andrefouet, S., Kramer, P., Torres-Pulliza, D., et al. 2003. Multi-site evaluation of IKONOS data for classification of tropical coral reef environments. Remote Sens. Environ. 88, 128–143. Andrews, E. C. 1902. Preliminary note on the geology of the Queensland coast. Proc. Linn. Soc. N.S.W. 2, 146–185. Andrews, E. C. 1922. Contributions to the hypothesis of coral reef formation. J. Proc. Roy. Soc. N.S.W. 56, 10–38. Andrews, J. 1983. Lagoon–ocean interactions. In Baker, J., Carter, R., Sammarco, P., and Stark, K. (eds.) Proceedings of the Great Barrier Reef Conference. Townsville, James Cook University, pp. 403–408. 469
470
References
Andrews, J. C. and Bode, L. 1988. Tides of the central Great Barrier Reef. Continent. Shelf Res. 8, 1057–1085. Andrews, J. C., Dunlap, W. C., and Bellamy, N. F. 1984. Stratification in a small lagoon in the Great Barrier Reef. Austral. J. Mar. Freshw. Res. 35, 273–284. Angwenyi, C. M. and Rydberg, L. 2005. Wave-driven circulation across the coral reef at Bamburi Lagoon, Kenya. Estuar. Coast. Shelf Sci. 63, 447–454. Anthony, K. R. N. 2000. Enhanced particle-feeding capacity of corals on turbid reefs (Great Barrier Reef, Australia). Coral Reefs 19, 59–67. Anthony, K. R. N. and Fabricius, K. E. 2000. Shifting roles of heterotrophy and autotrophy in coral energetics under varying turbidity. J. Exp. Mar. Biol. Ecol. 252, 221–253. Anthony, K. and Larcombe, P. 2002. Coral reefs in turbid waters: sediment-induced stresses in corals and likely mechanisms of adaptation. Proc. 9th Int. Coral Reef Symp., Bali 1, 239–244. Asano, D. 1942. Coral reefs of the South Sea Islands. Tokyo Imp. Univ. Geol. Palaeo. Inst. Rept. 39, 1–19. Assaoui, D. M., McNeil, D. F., and Kirschvink, J. L. 1990. Magnetostratigraphic dating of Mururoa Atoll and global eustasy. Earth Planet. Sci. Lett. 97, 107–112. Aston, J. P. 1995. The relative mobilities of coral cays on the Great Barrier Reef can be modelled. M.Sc. thesis, James Cook University, Townsville. Atkinson, M. J., Smith, S. V., and Stroup, E. D. 1981. Circulation in Enewetak Atoll lagoon. Proc. 4th Int. Coral Reef Symp., Manila 1, 335–338. Australian Government Mission and the Maldives Marine Research Centre. 2005. An Assessment of Damage to Maldivian Coral Reefs and Baitfish Populations from the Indian Ocean Tsunami. Canberra, Commonwealth of Australia. Australian Greenhouse Office. 2005. Climate Change: Risk and Vulnerability. Canberra, Australian Greenhouse Office. Australian Institute of Marine Science. 2001. Big Bank Shoals of the Timor Sea. Available online at http://www.aims.gov.au/pages/reflib/bigbank/pages/66-04 Australian Institute of Marine Science. 2005. Ancient mangrove forests discovered under reef. Available online at http://www.aims.gov.au/news/pages/mediarelease-20050217.html Ayling, A. M. and Ayling, A. L. 1985. A Preliminary Survey of Coastal Reefs in the Cape Tribulation Region. Townsville, Great Barrier Reef Marine Park Authority. Ayling, A. M. and Ayling, A. L. 1987. Is silt run-off affecting coral communities on the Cape Tribulation fringing reefs? In Baldwin, C. L. (ed.) Fringing Reef Workshop: Science Industry and Management. Townsville, Great Barrier Reef Marine Park Authority, pp. 83–85. Ayling, A. M. and Ayling, A. L. 1995. A Preliminary Survey of Benthic Communities on Fringing Reefs in the Middle Cairns Section. Townsville, Great Barrier Reef Marine Park Authority. Ayling, A. M. and Ayling, A. L. 1999. Medium-Term Changes in Coral Populations of Fringing Reefs at Cape Tribulation. Townsville, Great Barrier Reef Marine Park Authority. Ayling, A., Ayling, A., and Berkelmans, R. 1998. Shoalwater Bay Fringing Reef Resource Assessment. Townsville, Great Barrier Reef Marine Park Authority. Backshall, D. G., Barnett, J., Davies, P. J., et al. 1979. Drowned dolines: the blue holes of the Pompey Reefs, Great Barrier Reef. Bureau Min. Resources J. Austral. Geol. Geophys. 4, 99–109.
References
471
Baker, R. G. V. and Haworth, R. J. 2000a. Smooth or oscillating late Holocene sea-level curve? Evidence from cross-regional statistical regressions of fixed biological indicators. Mar. Geol. 163, 353–365. Baker, R. G. V. and Haworth, R. J. 2000b. Smooth or oscillating late Holocene sea-level curve? Evidence from the palaeo-zoology of fixed biological indicators in east Australia and beyond. Mar. Geol. 163, 367–386. Baker, R. G. V., Haworth, R. J., and Flood, P. G. 2001a. Inter-tidal fixed indicators of former Holocene sea levels in Australia: a summary of sites and a review of methods and models. Quatern. Int. 5, 257–273. Baker, R. G. V., Haworth, R. J., and Flood, P. G. 2001b. Warmer or cooler late Holocene marine palaeoenvironments? Interpreting southeast Australian and Brazilian sea-level changes using fixed biological indicators and their delta O-18 composition. Palaeogeog. Palaeoclimatol. Palaeoecol. 168, 249–272. Bard, E., Hamelin, B., and Fairbanks, R. 1990. U–Th ages obtained by mass spectrometry in corals from Barbados: sea level during the past 130 000 years. Nature 346, 456–458. Bard, E., Hamelin, B., Arnold, M., et al. 1996. Deglacial sea-level record from Tahiti corals and the timing of global meltwater discharge. Nature 382, 241–244. Barham, A. 1983. Preliminary Report to the Research and Exploration Committee of the National Geographic Society on Part 1 (Reef and Mangrove Coring Expedition) of the Torres Strait Palaeo-Environmental Research Program, July–Sept, 1983. London, Department of Human Environment, Institute of Archaeology, and Department of Geography, University College London. Barnes, D. and Chalker, B. 1990. Calcification and photosynthesis in reef-building corals and algae. In Dubinsky, Z. (ed.) Ecosystems of the World, vol. 25, Coral Reefs. Amsterdam, Elsevier, pp. 109–131. Barnes, D. J. and Lough, J. M. 1996. Coral skeletons: storage and recovery of environmental information. Glob. Change Biol. 2, 569–582. Barnes, R. D. 1984. Morphogenesis of a nearshore fringing reef. Hons. thesis, James Cook University, Townsville. Bassett, S. E., Milne, G. A., Mitrovica, J. X., and Clark, P. U. 2005. Ice sheet and solid earth influences on far-field sea-level histories. Science 309, 925–928. Beach Protection Authority Queensland. 1989. Green Island Data Report. Brisbane. Beaman, R., Larcombe, P., and Carter, R. M. 1994. New evidence for the Holocene sea-level high from the inner shelf, central Great Barrier Reef, Australia. J. Sed. Res. A: Sed. Petrol. Processes 64, 881–885. Beaton, A. M. 1978. Archaeology and the Great Barrier Reef. Phil. Trans. Roy. Soc. Lond. B 234, 141–147. Bell, A. 1982. LANDSAT looks at the Great Barrier Reef. Ecos 31, 3–8. Bellwood, D. R., Hughes, T. P., Folke, C., and Nystrom, M. 2004. Confronting the coral reef crisis. Nature 429, 827–833. Belperio, A. P. 1978. Inner shelf sedimentation model for the Townsville Region, Great Barrier Reef. Ph.D. thesis, James Cook University, Townsville. Belperio, A. 1979a. Negative evidence for a mid-Holocene high sea level along the coastal plain of the Great Barrier Reef province. Mar. Geol. 32, M1–M9. Belperio, A. P. 1979b. The combined use of wash load and bed material load rating curves for the calculation of total load: an example from the Burdekin River, Australia. Catena 6, 317–329. Belperio, A. 1983a. Late Quaternary terrigenous sedimentation in the Great Barrier Reef lagoon. In Baker, J., Carter, R., Sammarco, P., and Stark, K. (eds.)
472
References
Proceedings of the Great Barrier Reef Conference. Townsville, James Cook University, pp. 71–76. Belperio, A. P. 1983b. Terrigenous sedimentation in the central Great Barrier Reef lagoon: a model from the Burdekin region. B.M.R.J. Austral. Geol. Geophys. 8, 179–190. Benbow, D. D. 1980. The petroleum prospects of the Great Barrier Reef region. Austral. Petrol. Explor. Ass. J. 20, 159–175. Bennett, I. 1973. Brief notes on the zonation and intertidal invertebrate fauna of Low Isles and Heron Island reefs. Unpublished field notes for the 2nd Int. Coral Reef Symp. Bird, E. C. F. 1970. The steep coast of the Macalister Range, north Queensland, Australia. J. Trop. Geog. 31, 33–39. Bird, E. C. F. 1971a. The fringing reefs near Yule Point, North Queensland. Austral. Geog. Stud. 9, 107–115. Bird, E. C. F. 1971b. Holocene shore features at Trinity Bay, North Queensland. Search 2, 27–28. Bird, E. C. F. and Hopley, D. 1969. Geomorphological features on a humid tropical sector of the Australian coast. Austral. Geog. Stud. 7, 89–108. Birkeland, C. (ed.) 1997. Life and Death of Coral Reefs. New York, Chapman and Hall. Black, K. P. and Moran, P. J. 1991. Influence of hydrodynamics on the passive dispersal and initial recruitment of larvae of Acanthasterplanci (Echinodermata, Asteroidea) on the Great Barrier Reef. Mar. Ecol. Progr. Ser. 69, 55–65. Black, K. P., Gay, S. L., and Andrews, J. C. 1990. Residence times of neutrallybuoyant matter such as larvae, sewage or nutrients on coral reefs. Coral Reefs 9, 105–114. Blackman, J. G., Winter, J. W., and King, B. R. 1986. Effects of Cyclone Winifred on coastal and island fauna. In Dutton, I. (ed.) Workshop on the Offshore Effects of Cyclone Winifred. Townsville, Great Barrier Reef Marine Park Authority, pp. 60–70. Blakeway, D. R. 2000. Geomorphology and ecology of cellular reefs in the Houtman Abrolhos Islands, Western Australia. Ph.D. thesis, University of Western Australia, Perth. Blanchon, P. 1995. Control on modern reef development around Grand Cayman. Ph.D. thesis, University of Alberta, Edmonton. Blanchon, P. and Jones, B. 1995. Marine-planation terraces on the shelf around Grand Cayman: a result of stepped Holocene sea-level rise. J. Coast. Res. 11, 1–33. Blanchon, P. and Shaw, J. 1995. Reef drowning during the last deglaciation: evidence for catastrophic sea-level rise and ice sheet collapse. Geology 23, 4–8. Blanchon, P., Jones, B., and Kalbfleisch, W. 1997. Anatomy of a fringing reef around Grand Cayman: storm rubble not coral framework. J. Sed. Res. 67, 1–16. Blanchon, P., Jones, B., and Ford, D. C. 2002. Discovery of a submerged relic reef and shoreline off Grand Cayman: further support for an early Holocene jump in sea level. Sed. Geol. 147, 253–270. Bland, L. N. 2004. Surficial sediments of a fringing reef complex, Hydeaway Bay, Great Barrier Reef. M.Sc. thesis, University of Auckland, Auckland. Bloom, A. L. 1978. Geomorphology: A Systematic Analysis of Late Cenozoic Landforms. Englewood Cliffs, NJ, Prentice Hall.
References
473
Bowen, J., and Bowen, M. 2002. The Great Barrier Reef: History, Science and Heritage. Cambridge, Cambridge University Press. Braga, J. C. and Aguirre, J. 2004. Coralline algae indicate Pleistocene evolution from deep, open platform to outer barrier reef environments in the northern Great Barrier Reef margin. Coral Reefs 23, 547–558. Braithwaite, C. J. R., Dalmasso, H., Gilmour, M. A., et al. 2004. The Great Barrier Reef: the chronological record from a new borehole. J. Sed. Res. 74, 298–310. Brander, R. W., Kench, P. S., and Hart, D. 2004. Spatial and temporal variations in wave characteristics across a reef platform, Warraber Island, Torres Strait, Australia. Mar. Geol. 207, 169–184. Briggs, J. C. 1992. The marine East Indies: centre of origin? Glob. Ecol. Biol. Lett. 2, 149–156. Briggs, J. C. 1999. Coincident biogeographic patterns: Indo-West Pacific Ocean. Evolution 53, 326–335. Brinkman, R., Wolanski, E., Deleersnijder, E., McAllister, F., and Skirving, W. 2001. Oceanic inflow from the Coral Sea into the Great Barrier Reef. Estuar. Coast. Shelf Sci. 54, 655–668. Brodie, J. 2002. Keeping the wolf from the door: managing land-based threats to the Great Barrier Reef. Proc. 9th Int. Coral Reef Symp., Bali 2, 705–714. Brodie, J., Steven, A., and Baer, M. 1997. The extent of river plumes associated with cyclone Sadie rainfall. In Steven, A. (ed.) Cyclone Sadie Flood Plumes in the Great Barrier Reef Lagoon: Composition and Consequences. Townsville, Great Barrier Reef Marine Park Authority, pp. 27–34. Broecker, W. S. 1998. The end of the present interglacial: how and when? Quatern. Sci. Rev. 17, 689–694. Broecker, W. S. and Henderson, G. M. 1998. The sequence of events surrounding Termination II and their implications for the cause of glacial–interglacial CO2 changes. Paleoceanography 13, 352–364. Bruno, J., Petas, L. E., Harvell, D., and Hettinger, A. 2003. Nutrient enrichment can increase the severity of coral diseases. Ecol. Lett. 6, 1056–1061. Bryan, W. H. 1928. The Queensland continental shelf. Repts. GBR Comm. 2, 45–50. Bryant, E. A. and Nott, J. 2001. Geological indicators of large tsunami in Australia. Nat. Hazards 24, 231–249. Bryce, S., Larcombe, P., and Ridd, P. V. 1996. Sediment transport in the Normanby River estuary, northern Great Barrier Reef, Australia. In Larcombe, P., Woolfe K. J., and Purdon, R. G. (eds.) Great Barrier Reef: Terrigenous Sediment Flux and Human Impact. Boca Raton, FL, CRC Press, pp. 40–44. Bryce, S., Larcombe, P., and Ridd, P. V. 1998. The relative importance of landdirected tidal sediment transport versus freshwater flood events in the Normanby River estuary, Cape York Peninsula, Australia. Mar. Geol. 149, 55–78. Buckley, R. 1988. Terrace formation on sand cays of the northern Great Barrier Reef: regional or reef-scale switch? Search 19, 289–292. Buddemeier, R. W. and Fautin, D. G. 1993. Coral bleaching as an adaptive mechanism: a testable hypothesis. BioScience 43, 320–326. Buddemeier, R. W. and Hopley, D. 1988. Turn-ons and turn-offs: cause and mechanisms of the initiation and termination of coral reef growth. Proc. 6th Int. Coral Reef Symp., Townsville 1, 253–261. Buddemeier, R. W. and Kinzie, R. 1976. Coral growth. Oceanogr. Mar. Biol. Ann. Rev. 14, 183–225.
474
References
Buddemeier, R. W. and Oberdorfer, J. A. 1990. Climate change and island groundwater resources. In Pernetta, J. C. and Hughes, P. J. (eds.) Implications of Expected Climate Changes in the South Pacific Region, UNEP Regional Seas Reports and Studies 128. Nairobi, United Nations Environmental Programme, pp. 56–67. Buddemeier, R. W., Kleypas, J. A., and Aronson, R. B. 2004. Coral Reefs and Global Climate Change: Potential Contributions of Climate Change to Stresses on Coral Reef Ecosystems. Arlington, VA, Pew Center on Global Climate Change. Bunt, J. S., Williams, W. T., and Bunt, E. D. 1985. Mangrove species distribution in relation to tide at the seafront and up rivers. Austral. J. Mar. Freshw. Res. 36, 481–492. Bureau of Meteorology. 2005. Tropical Cyclone Ingrid. Available online at http:// www.bom.gov.au/inside/services_policy/tc_ingrid/index.shtml Burrage, D. M., Black, K. P., and Steinberg, C. R. 1995. Long-term sea-level variations in the central Great Barrier Reef. Continent. Shelf Res. 15, 981–1014. Burrage, D. M., Steinberg, C. R., Skirving, W. J., and Kleypas, J. A. 1996. Mesoscale circulation features of the Great Barrier Reef region inferred from NOAA Satellite imagery. Remote Sens. Environ. 56, 21–41. Burrage, D., Steinberg, C., Bode, L., and Black, K. 1997. Long-term current observations in the Great Barrier Reef. In Wachenfeld, D., Oliver, J., and Davis, K. (eds.) State of the Great Barrier Reef World Heritage Area Workshop. Townsville, Great Barrier Reef Marine Park Authority, pp. 21–45. Burrage, D. M., Heron, M. L., Hacker, J. M., et al. 2002. Evolution and dynamics of tropical river plumes in the Great Barrier Reef: an integrated remote sensing and in situ study. J. Geophys. Res. Oceans 107, 12. Byron, T. 1985. Blue holes: land of coral rivers, raging currents and whirlpools. In Scuba Divers’ Guide to the Whitsunday Islands. Sydney, Aqua Sports Publications, pp. 82–87. Cabioch, G. and Ayliffe, L. K. 2001. Raised coral terraces at Malakula, Vanuatu, Southwest Pacific, indicate high sea level during marine isotope stage 3. Quatern. Res. 56, 357–365. Cabioch, G., Montaggioni, L. F., and Faure, G. 1995. Holocene initiation and development of New Caledonian fringing reefs, S.W. Pacific. Coral Reefs 14, 131–140. Cabioch, G., Camoin, G. F., and Montaggioni, L. F. 1999. Postglacial growth history of a French Polynesian barrier reef tract, Tahiti, Central Pacific. Sedimentology 46, 985–1000. Caldeira, K. and Wickett, H. E. 2003. Anthropogenic carbon and ocean pH. Nature 425, 365. Camoin, G. F., Colonna, M., Montaggioni, L. F., et al. 1997. Holocene sea-level changes and reef development in the southwestern Indian Ocean. Coral Reefs 4, 231–246. Campbell, J. B. 1980. Human adaptation in the Quaternary. In Henderson, R. A. and Stephenson, P. J. (eds.) The Geology and Geophysics of North-Eastern Australia. Brisbane, Geological Society of Australia, pp. 402–407. Cann, J. H., Harvey, N., Barnett, E. J., Belperio, A. P., and Bourman, R. P. 2002. Foraminiferal biofacies eco-succession and Holocene sealevels, Port Pirie, South Australia. Mar. Micropaleontol. 44, 31–55. Carbon Dioxide Information Analysis Center (CDIAC). 2005. Current greenhouse gas concentrations. Available online at http://cdiac.esd.ornl.gov/pns/ current_ghg.html
References
475
Carreiro-Silva, M., McClanahan, T. R., and Kiene, W. E. 2005. The role of inorganic nutrients and herbivory in controlling micro-bioerosion of carbonate substratum. Coral Reefs 24, 214–221. Carter, R. M. and Johnson, D. P. 1986. Sea level controls on the post-glacial development of the Great Barrier Reef, Queensland. Mar. Geol. 71, 137–164. Carter, R. M., Carter, L., and Johnson, D. P. 1986. Submergent shorelines in the SW Pacific: evidence for an episodic postglacial transgression. Sedimentology 33, 629–649. Carter, R. M., Johnson, D. P., and Hooper K. G. 1993. Episodic post-glacial sea-level rise and the sedimentary evolution of a tropical continental embayment (Cleveland Bay, Great Barrier Reef shelf, Australia). Austral. J. Earth Sci. 40, 229–255. Chalker, B. E. and Dunlap, W. C. 1983. Primary production and photoadaptation by corals on the Great Barrier Reef. In Baker, J., Carter, R., Sammarco, P., and Stark, K. (eds.) Proceedings of the Great Barrier Reef Conference. Townsville, James Cook University, pp. 293–298. Chappell, J. 1974. Geology of coral terraces, Huon Peninsula, New Guinea: a study of Quaternary tectonic movements and sea level changes. Geol. Soc. America Bull. 85, 553–570. Chappell, J. 1980. Coral morphology, diversity and reef growth. Nature 286, 249–252. Chappell, J. 1983. Evidence for smoothly falling sea level relative to north Queensland, Australia, during the past 6000 yr. Nature 302, 406–408. Chappell, J. 1994. Upper Quaternary sea levels, coral terraces, oxygen isotopes and deep-sea temperatures. J. Geog. Japan 103, 828–840. Chappell, J. 2002. Sea level changes forced ice breakouts in the Last Glacial cycle: new results from coral terraces. Quatern. Sci. Rev. 21, 1229–1240. Chappell, J. and Grindrod, J. 1984. Chenier plain formation in northern Australia. In Thom, B. G. (ed.) Coastal Geomorphology in Australia. Sydney, Academic Press, pp. 197–231. Chappell, J. and Polach, H. 1991. Postglacial sea-level rise from a coral record at Huon Peninsula, Papua New Guinea. Nature 349, 147–149. Chappell, J. and Shackleton, N. J. 1986. Oxygen isotopes and sea level. Nature 324, 137–140. Chappell, J. and Veeh, H. H. 1978. Late Quaternary tectonic movements and sea-level changes at Timor and Atauro island. Geol. Soc. America Bull. 89, 356–368. Chappell, J., Rhodes, E. G., Thom H. G., and Wallensky, E. 1982. Hydroisostasy and the sea level isobase of 5500 BP in North Queensland Australia. Mar. Geol. 49, 81–90. Chappell, J., Chivas, A., Wallensky, E., Polach, H. A., and Aharon, P. 1983. Holocene palaeo environmental changes, central to north Great Barrier Reef, inner zone. B.M.R.J. Austral. Geol. Geophys. 8, 223–235. Chappell, J., Omura, A., Ezat, T., et al. 1996. Reconciliation of late Quaternary sealevels derived from coral terraces at Huon Peninsula with deep sea oxygen isotype records. Earth Planet. Sci. Lett. 141, 227–236. Chazottes, V., Le Campion-Alsumard, T., Peyrot-Clausard, N., and Cuet, P. 2002. The effect of eutrophication-related alterations to coral reef communities on agents and rates of bio-erosion, Reunion Island, Indian Ocean. Coral Reefs 21, 375–390. Chen, D. and Krol, A. 1997. Hydrology of Heron Island, Great Barrier Reef, Australia. Devel. Sedimentol. 54, 867–884. Chen, J., Curran, H., White, B., and Wasserburg, G. 1991. Precise chronology of the last interglacial period: 234U–230Th data from fossil coral reefs in the Bahamas. Geol. Soc. America Bull. 103, 82–97.
476
References
Chivas, A., Chappell, J., Polach, H., Pillans, B., and Flood, P. 1986. Radiocarbon evidence for the timing and rate of island development, beachrock formation, and phosphatization at Lady Elliot Island, Queensland, Australia. Mar. Geol. 69, 273–287. Chongprasith, P. 1992. Nutrient release and nitrogen transformations resulting from resuspension of Great Barrier Reef shelf sediments. Ph.D. thesis, James Cook University, Townsville. Church, J. A. 1987. East Australian current adjacent to the Great Barrier Reef. Austral. J. Mar. Freshw. Res. 38, 671–683. Church, J. A. 2001. Climate change: how fast are sea levels rising? Science 294, 802–803. Church, J. A. and White, N. J. 2006. A 20th century acceleration in global sea-level rise. Geophys. Res. Lett. 31, L01602. Church, J. A., Andrews, J. C., and Boland, F. M. 1985. Tidal currents in the Central Great Barrier Reef. Continent. Shelf Res. 4, 515–531. Church, J. A., White, N. J., Coleman, R., Lambeck, K., and Mitrovica, J. X. 2004. Estimates of the regional distribution of sea level rise over the 1950–2000 period. J. Climate 17, 2609–2625. Church, J. A., White, N. J., and Arblaster, J. M. 2005. Significant decadal-scale impact of volcanic eruptions on sea level and ocean heat content. Nature 438, 74–77. Clark, P. U., McCabe, A. M., Mix, A. C., and Weaver, A. J. 2004. Rapid rise of sea level 19 000 years ago and its global implications. Science 304, 1141–1144. Clarke, J. A., Farrell, W. E., and Peltier, W. R. 1978. Global changes in post-glacial sea level: a numerical calculation. Quatern. Res. 9, 265–287. Coastal Engineering Research Center (CERC). 1984. US Army Corps of Engineers Shore Protection Manual, vols. 1 and 2. Vicksburg, MS, CERC. Collins, J. D. and Walker, T. A. 1985. A Drift Card Study of the Great Barrier Reef. Townsville, Great Barrier Reef Marine Park Authority. Collins, L. B., Zhu, Z. R., Wyrwoll, K. H., et al. 1993a. Late Quaternary evolution of coral reefs on a cool-water carbonate margin, the Abrolhos carbonate platforms, South-west Australia. Mar. Geol. 110, 203–212. Collins, L. B., Zhu, Z. R., Wyrwoll, K. H., et al. 1993b. Holocene growth history of a reef complex on a cool-water carbonate margin: Easter Group of the HoutmanAbrolhos, eastern Indian Ocean. Mar. Geol. 115, 29–46. Collins, L. B., Zhu, Z. R., Wyrwoll, K. H., and Eisenhauer, A. 2002. Geological evolution of the northern Ningaloo Reef System during the late Quaternary. Proc. 9th Int. Coral Reef Symp., Bali 1, 231–237. Collins, L. B., Zhu, Z. R., Wyrwoll, K. H., and Eisenhauer, A. 2004. Late Quaternary structure and development of the northern Ningaloo Reef, Australia. Sed. Geol. 159, 81–94. Colorado Center for Astrodynamics Research. 2006. Global near-real-time altimetry data viewer: South East Pacific. Available online at http://argo.colorado.edu/ realtime/gsfc_global-real-time_ssh/ Colquhoun, D. G. 1979. World Shorelines Map: Pacific Indian Ocean Sector. Columbia, SC, INQUA Commission on Quaternary Shorelines. Connell, J. H., Hughes, T. P., and Wallace, C. C. 1997. Long-term dynamics of reef crest corals on Heron Island. In Wachenfeld, D., Oliver, J., and Davis, K. (eds.) State of the Great Barrier Reef World Heritage Area Workshop. Townsville, Great Barrier Reef Marine Park Authority, pp. 114–123. Cook, P. J. and Mayo, W. 1977. Sedimentology and Holocene history of a tropical estuary (Broad Sound, Queensland). Bureau Min. Resources J. Austral. Geol. Geophys. 170.
References
477
Cook, P. J. and Polach, H. A. 1973. A chenier sequence at Broad Sound, Queensland and evidence against a Holocene high sea level. Mar. Geol. 14, 253–268. Cook, P. J., Colwell, J. B., Firman, J. B., et al. 1977. The late Cainozoic sequence of southeast South Australia and Pleistocene sea-level changes. Bureau Min. Resources J. Austral. Geol. Geophys. 2, 81–88. Cortes, J. N. and Risk, M. J. 1985. A reef under siltation stress, Cahuita, Costa Rica. Bull. Mar. Sci. 36, 339–356. Coventry, R. J., Hopley, D., Campbell, J. B., et al. 1980. The Quaternary of northeastern Australia. In Henderson R. A. and Stephenson, P. J. (eds.) The Geology and Geophysics of North-Eastern Australia. Brisbane, Geological Society of Australia, pp. 365–417. Crossland, C. 1997. Preface. In Proc. Natl Conf. The Great Barrier Reef: Science, Use and Management. Townsville, Great Barrier Reef Marine Park Authority, pp. 1–2. Crowley, G. M. and Gagan, M. K. 1995. Holocene evolution of coastal wetlands in wet-tropical Northeastern Australia. Holocene 5, 385–399. Cutler, K. B., Edwards, R. L., Taylor, F. W., et al. 2003. Rapid sea-level fall and deepocean temperature change since the last interglacial period. Earth Planet. Sci. Lett. 206, 253–271. Daly, R. A. 1915. The glacial control theory of coral reefs. Proc. Am. Acad. Arts Sci. 51, 155–251. Daly, R. A. 1919. The coral reef zone during and after the Glacial period. Am. J. Sci. 48, 136–159. Daly, R. A. 1934. The Changing World of the Ice Age. New Haven, CT, Yale University Press. Darwin, C. R. 1838. On certain areas of elevation and subsidence in the Pacific and Indian Oceans as deduced from the study of coral formations. Proc. Geol. Soc. Lond. 2, 552–554. Darwin, C. R. 1842. The Structure and Distribution of Coral Reefs. London, Smith, Elder and Co. Davies, P. J. 1974. Subsurface solution unconformities at Heron Island Great Barrier Reef. Proc. 2nd Int. Coral Reef Symp., Queensland 2, 573–578. Davies, P. J. 1977. Modern reef growth: Great Barrier Reef. Proc. 3rd Int. Coral Reef Symp., Miami 2, 325–330. Davies, P. J. 1983. Reef growth. In Barnes, D. J. (ed.) Perspectives on Coral Reefs. Townsville, Australian Institute of Marine Science, pp. 69–106. Davies, P. J. 1988. Evolution of the Great Barrier Reef: reductionist dream or expansionist vision. Proc. 6th Int. Coral Reef Symp., Townsville 1, 9–17. Davies, P. J. and Hopley, D. 1983. Growth facies and growth rates of Holocene reefs in the Great Barrier Reef. Bureau Min. Resources J. Austral. Geol. Geophys. 8, 237–251. Davies, P. J. and Hughes, H. 1983. High energy reef and terrigenous sedimentation, Boulder Reef, Great Barrier Reef. Bureau Min. Resources J. Austral. Geol. Geophys. 8, 201–209. Davies, P. and Kinsey, D. W. 1973. Organic and inorganic factors in recent beach rock formation, Heron Island, Great Barrier Reef. J. Sed. Petrol. 43, 59–81. Davies, P. J. and Kinsey, D. W. 1977. Holocene reef growth: One Tree Island, Great Barrier Reef. Mar. Geol. 24, M1–M11. Davies, P. J. and Marshall, J. F. 1979. Aspects of Holocene reef growth: substrate age and accretion rate. Search 10, 276–279.
478
References
Davies, P. J. and Marshall, J. F. 1985. Halimeda bioherms, low energy reefs, northern Great Barrier Reef. Proc. 5th Int. Coral Reef Congr., Tahiti 5, 1–7. Davies, P. J. and Martin, K. 1976. Radial aragonite ooids, Lizard Island, Great Barrier Reef, Queensland, Australia. Geology 4, 120–122. Davies P. J. and McKenzie, J. A. 1993. Controls on the Plio-Pleistocene evolution of the north-eastern Australian continental margin. In McKenzie, J. A., Davies, P. J., Palmer-Jackson, M. A., and Sarg, J. F. (eds.) Proceedings of the Ocean Drilling Program, vol. 133, Scientific Results, Northeast Australian Margin. College Station, TX, Texas A and M University, pp. 755–762. Davies, P. J. and Montaggioni, L. F. 1985. Reef growth and sea level change: the environmental signature. Proc. 5th Int. Coral Reef Congr., Tahiti 3, 477–511. Davies, P. J. and Peerdeman, F. 1998. The origin of the Great Barrier Reef: the impact of Leg 133 drilling. Spec. Publ. Int. Assoc. Sedimentol. 25, 23–38. Davies, P. J. and West, B. 1981. Suspended-sediment transport and water movement at One Tree Reef, Southern Great Barrier Reef. Bureau Min. Resources J. Austral. Geol. Geophys. 6, 187–195. Davies, P. J., Stewart, D., Thom, G., McIntosh, E., and Kores, A. 1979. A rock and sediment drill for use on coral reefs. Bureau Min. Resources J. Austral. Geol. Geophys. Rpt. 1979/21. Davies, P. J., Marshall, J. F., and Searle, D. E. 1981. Shallow inter-reefal structure of the Capricorn Group, southern Great Barrier Reef. Bureau Min. Resources J. Austral. Geol. Geophys. 6, 101–105. Davies, P. J., Cucuzza, J., and Marshall, J. F. 1983. Lithofacies variations on the continental shelf east of Townsville, Great Barrier Reef. In Baker, J. T., Carter, R. M., Sammarco, P. W., and Stark, K. P. (eds.) Proceedings of the Great Barrier Reef Conference. Townsville, James Cook University, pp. 89–93. Davies, P. J., Marshall, J. F., and Hopley, D. 1985. Relationships between reef growth and sea level in the Great Barrier Reef. Proc. 5th Int. Coral Reef Congr., Tahiti 3, 95–103. Davies, P. J., Symonds, P. A., Feary, D. A., and Pigram, C. J. 1987. Horizontal plate motion: a key allocyclic factor in the evolution of the Great Barrier Reef. Science 238, 1697–1700. Davies, P. J., Symonds, P. A., Feary, D. A., and Pigram, C. J. 1989. The evolution of the carbonate platforms of north-east Australia. In Crevello, P. D. (ed.) Controls on Carbonate Platform and Basin Development, Society of Economic Paleontologists and Mineralogists Special Publication 44. Tulsa, OK, S.E.P.M., pp. 233–258. Davies, P. J., Briggs, J. C., Webster, J., and Montaggioni, L. 1997. Stratigraphy and lithofacies variations of the drill holes at Boulder and Ribbon 5. In Abstracts of 10th Edgeworth David Symposium, Funafuti to Mururoa: A Century of Reflections on Carbonate Reservoirs. Sydney, Sydney University, p. 37. Davis, W. M. 1899. The geographical cycle. Geog. J. 14, 481–504. Davis, W. M. 1917. The Great Barrier Reef of Australia. Am. J. Sci. Ser. 4, 44, 339–350. Davis, W. M. 1928. The Coral Reef Problem. New York, American Geographical Society. Day, J., Fernandes, L., Lewis, A., et al. 2003. The representative areas program for protecting biodiversity in the Great Barrier Reef World Heritage Area. Proc. 9th Int. Coral Reef Symp., Bali 2, 687–696.
References
479
De Vaugelas, J. 1985. Sediment reworkings by Callianassid mud-shrimp in tropical lagoons: a review with perspectives. Proc. 5th Int. Coral Reef Congr., Tahiti 6, 617–622. Debenay, J. P., Guiral, D., and Parra, M. 2002. Ecological factors acting on the microfauna in mangrove swamps: the case of foraminiferal assemblages in French Guiana. Estuar. Coast. Shelf Sci. 55, 509–533. Devlin, M. J. 1997. Offshore measurements late in the river plumes associated with cyclone Sadie. In Steven, A. D. L. (ed.) Cyclone Sadie Flood Plumes in the Great Barrier Reef Lagoon: Composition and Consequences. Townsville, Great Barrier Reef Marine Park Authority, pp. 45–54. Devlin, M. J. and Brodie, J. 2005. Terrestrial discharge into the Great Barrier Reef Lagoon: nutrient behavior in coastal waters. Mar. Poll. Bull. 51, 9–22. Devlin, D., Waterhouse, J., Taylor, J., and Brodie, J. 2001. Flood Plumes in the Great Barrier Reef: Spatial and Temporal Patterns in Composition and Distribution, Research Publication 68. Townsville, Great Barrier Reef Marine Park Authority. Dight, I. J., Bode, L., and James, M. K. 1990a. Modeling the larval dispersal of Acanthaster planci. I. Large-scale hydrodynamics, Cairns Section, Great Barrier Reef Marine Park. Coral Reefs 9, 115–123. Dight, I. J., James, M. K., and Bode, L. 1990b. Modeling the larval dispersal of Acanthasterplanci. II. Patterns of reef connectivity. Coral Reefs 9, 125–134. Dikou, A. and van Woesik, R. 2006. Survival under chronic stress from sediment load: spatial patterns of hard coral communities in the southern islands of Singapore. Mar. Poll. Bull. 52, 7–21. Dodge, R. E. 1982. Effects of drilling muds on the reef-building coral Montastrea annularis. Mar. Biol. 71, 141–147. Dodge, R. E. and Vaisnys, J. R. 1977. Coral populations and growth patterns: responses to sedimentation and turbidity associated with dredging. J. Mar. Res. 35, 715–730. Dodge, R. E., Aller, R. C., and Thomson, J. 1974. Coral growth related to re-suspension of bottom sediments. Nature 247, 574–577. Domm, S. B. 1971. The uninhabitated cays of the Capricorn Group, Great Barrier Reef, Australia. Atoll Res. Bull. 142, 1–27. Done, T. J. 1982. Patterns in the distribution of coral communities across the Central Great Barrier Reef. Coral Reefs 1, 95–107. Done, T. J. 1983. Coral zonation: its nature and significance. In Barnes, D. (ed.) Perspectives on Coral Reefs. Townsville, Australian Institute of Marine Science, pp. 107–147. Done, T. J. 1991. The debate continues: robust versus fragile reefs. Reef Encounter 9, 5–7. Done, T. J. 1992a. Phase shifts in coral reef communities and their ecological significance. Hydrobiologia 247, 121–132. Done, T. J. 1992b. Constancy and change in some Great Barrier Reef communities: 1980–1990. Am. Zool. 32, 655–662. Done, T. 1992c. Effects of tropical cyclone waves on ecological and geomorphological structures on the Great Barrier Reef. Continent. Shelf Res. 12, 859–872. Done, T. J., Ayling, A. M., and Van Woesik, R. 1991. Broadscale Survey of Impacts of Cyclone Ivor on Coral Reefs, Research Publication 24. Townsville, Great Barrier Reef Marine Park Authority. Dong, K. 1988. El Nino ˜ and tropical cyclone frequency in the Australian region and the Northwest Pacific. Austral. Meteorol. Mag. 36, 219–225.
480
References
Douglas, B. C. 2001. Sea level change in the era of the recording tide gauge. In Douglas, B. C., Kearney, M. S., and Leatherman, S. P. (eds.) Sea Level Rise: History and Consequences. New York, Academic Press, pp. 37–64. Drew, E. 2001. Ocean nutrients to sediment banks via tidal jets and Halimeda meadows. In Wolanski, E. (ed.) Oceanographic Processes of Coral Reefs: Physical and Biological Links in the Great Barrier Reef. Boca Raton, FL, CRC Press, pp. 255–267. Drew, E. A. and Abel, K. M. 1983. Growth of Halimeda in reefal and inter-reefal environments. In Baker, J. T., Carter, R. M., Sammarco, P. W., and Stark, K. P. (eds.) Proceedings of the Great Barrier Reef Conference. Townsville, James Cook University, pp. 299–304. Drew, E. A. and Abel, K. M. 1985. Biology, sedimentology and geography of the vast inter-reefal Halimeda meadows within the Great Barrier Reef province. Proc. 5th Int. Coral Reef Congr., Tahiti 5, 15–20. Drew, E. A. and Abel, K. M. 1988. Studies of Halimeda. I. The distribution and species composition of Halimeda meadows throughout the Great Barrier Reef province. Coral Reefs 6, 195–205. Driscoll, E. M. and Hopley, D. 1967. Coastal development in a part of tropical Queensland, Australia. J. Trop. Geog. 26, 17–28. Dubinsky, Z. (ed.) 1990. Ecosystems of the World, vol. 25, Coral Reefs. Amsterdam, Elsevier. Dullo, W. C. 2005. Coral growth and reef growth: a brief review. Facies 51, 33–48. Dunbar, G. B. and Dickens, G. R. 2003. Massive siliciclastic discharge to slopes of the Great Barrier Reef platform during sea-level transgression: constraints from sediment cores between 158 S and 168 S latitude and possible explanations. Sed. Geol. 162, 141–158. Easton, W. H. and Olson, E. A. 1976. Radiocarbon profile of Hanauma Reef, Oahu, Hawaii. Geol. Soc. America Bull. 87, 711–719. Edwards, R. J. 2001. Mid- to late Holocene relative sea-level change in Poole Harbour, southern England. J. Quatern. Sci. 16, 221–235. Edwards, R. L. and Horton, B. P. 2005. Developing detailed records of relative sealevel change using a foraminiferal transfer function: an example from North Norfolk, UK. Phil. Trans. Roy. Soc. Lond. A 364, 973–991. Edwards, R. L., Chen, J. H., Ku, T. L., and Wasserburg, G. J. 1987. Precise timing of the last interglacial period from mass spectrometric determination of thorium230 in corals. Science 236, 1547–1553. Edwards, R. L., Beck, J. W., Burr, G. S., et al. 1993. A large drop in atmospheric ‘‘SUP 14’’ C/ ‘‘SUP 12’’ C and reduced melting in the Younger Dryas, documented with ‘‘SUP 230’’ Th ages of corals. Science 260, 962–968. Elvidge, C. D., Dietz, J. B., Berkelmans, R., et al. 2004. Satellite observation of Keppel Islands (Great Barrier Reef) 2002, coral bleaching using IKONOS data. Coral Reefs 23, 123–132. Emery, K. O., Tracey, J. I., Jr., and Ladd, H. S. 1954. Geology of Bikini and nearby atolls, Marshall Islands. U.S. Geol. Surv. Prof. Pap. 260A, 1–265. Endean, R., Stephenson, W., and Kenny, R. 1956. The ecology and distribution of intertidal organisms on certain islands off the Queensland coast. Austral. J. Mar. Freshw. Res. 7, 317–342. Environmental Protection Agency. 2005. Interim report on severe tropical cyclone Ingrid. Available online at http://www.epa.qld.gov.au/publications/ p01584aa.pdf
References
481
Esat, T. M., McCulloch, M. T., Chappell, J., Pillans, B., and Omura, A. 1999. Rapid fluctuations in sea level recorded at Huon Peninsula during the penultimate deglaciation. Science 283, 197–201. Fabricius, K. E. 2004. The Effects of Terrestrial Runoff of Sediments, Nutrients and Other Pollutants on Coral Reefs, Briefing Paper 3. Townsville, International Society for Reef Studies. Fabricius, K. E. 2005. Effects of terrestrial runoff on the ecology of corals and coral reefs: review and synthesis. Mar. Poll. Bull. 50, 125–146. Fabricius, K. E. and De’ath, G. 2001. Biodiversity on the Great Barrier Reef: large scale patterns and turbidity related local loss of soft coral taxa. In Wolanski, E. (ed.) Oceanographic Processes on Coral Reefs: Physical and Biological Links in the Great Barrier Reef. Boca Raton, FL, CRC Press, pp. 127–144. Fabricius, K. E. and Wolanski, E. 2000. Rapid smothering of coral reef organisms by muddy marine snow. Estuar. Coast. Shelf Sci. 50, 115–120. Fabricius, K. E., Wild, C., Wolanski, E., and Abele, D. 2003. Effects of transparent exopolymer particles and muddy terrigenous sediments on the survival of hard coral recruits. Estuar. Coast. Shelf Sci. 57, 613–621. Fabricius, K., De’ath, G., McCook, L., Turak, E., and Williams, D. McB. 2005. Changes in algal, coral and fish assemblages along water quality gradients on the inshore Great Barrier Reef. Mar. Poll. Bull. 51, 384–398. Fairbanks, R. G. 1989. A 17 000-yr glacio-eustatic sea level record: influence of glacial melting rates on the Younger Dryas event and deep ocean circulation. Nature 342, 637–642. Fairbridge, R. W. 1950. Recent and Pleistocene coral reefs of Australia. J. Geol. 58, 330–401. Fairbridge, R. W. 1961. Eustatic changes in sea level. Phys. Chem. Earth 4, 99–185. Fairbridge, R. W. 1967. Coral reefs of the Australian region. In Jennings, J. N. and Mabbutt, J. A. (eds.) Landform Studies from Australia and New Guinea. Canberra, Australian National University, pp. 386–451. Fairbridge, R. W. and Teichert, C. 1947. The rampart system at Low Isles, 1928–1945. Repts. GBR Comm. 6, 1–16. Fairbridge, R. W. and Teichert, C. 1948. The Low Isles of the Great Barrier Reef: a new analysis. Geog. J. 111, 67–88. Falconer, R. A. and Mardapitta-Hadjipandeli, L. 1986. Bathymetric and shear-stress effects on an island’s wake: a computational model study. Coast. Engin. 11, 57–86. Falconer, R. A., Wolanski, E., and Mardapitta-Hadjipandeli, L. 1986. Modeling tidal circulation in an island’s wake. J. Waterway Port, Coast. Ocean Engin. 112, 234–254. Falk, J. and Brownlow, A. 1989. The Greenhouse Challenge: What’s to Be Done? Ringwood, Penguin. Falkland, A. C. 1993. Hydrology and water management on small tropical islands. Proc. Yokohama Symp. on Hydrology of Warm Humid Regions, 263–303. Falkowski, P. G., Jokiel, P. L., and Kinzie, R. A. 1990. Irradiance and corals. In Dubinsky, Z. (ed.) Ecosystems of the World, vol. 25, Coral Reefs. Amsterdam, Elsevier, pp. 89–107. Feary, D. A., Symonds, P. A., Davies, P. J., Pigram, C. J., and Jarrard, R. D. 1993. Geometry of Pleistocene facies on the Great Barrier Reef outer shelf and upperslope seismic stratigraphy of Sites 819, 820 and 821. In McKenzie, J. A., Davies, P. J., Palmer-Julson, A. A., and Sarg, J. F. (eds.) Proceedings of the Ocean
482
References
Drilling Program, vol. 133 Scientific Results, North-East Australian Margin. College Station, TX, Texas A and M University, pp. 535–541. Ferland, M. A., Roy, P. S., and Murray-Wallace, C. V. 1995. Glacial lowstand deposits on the outer continental shelf of southeastern Australia. Quatern. Res. 44, 294–299. Fernandes, L., Day, J., Lewis, A., et al. 2005. Establishing representative no-take areas in the Great Barrier Reef: large-scale implementation of theory on marine protected areas. Conservat. Biol. 19, 1733–1744. Ferrier-Page`s, C., Gettuso, J.-P., Dallot, S., and Jaubert, J. 2000. Effect of nutrient enrichment on growth and photosynthesis of the zooxanthellae coral Stylophera pistillata. Coral Reefs 19, 103–113. Fielding, C. R., Trueman, J. D., Dickens, G. R., and Page, M. 2003. Anatomy of the buried Burdekin River channel across the Great Barrier Reef shelf: how does a major river operate on a tropical mixed siliciclastic/carbonate margin during sea level lowstand? Sed. Geol. 157, 291–301. Fisk, D. A. and Harriott, V. J. 1989. The Effects of Increased Sedimentation on the Recruitment and Sediment Dynamics of Juvenile Coral at Cape Tribulation, North Queensland, Technical Memoir 20. Townsville, Great Barrier Reef Marine Park Authority. Fleming, K., Johnston, P., Zwartz, D., et al. 1998. Refining the eustatic sea-level curve since the last glacial maximum using far- and intermediate-field sites. Earth Planet. Sci. Lett. 163, 327–342. Flood, P. G. 1977. Coral cays of the Capricorn and Bunker Groups, Great Barrier Reef Province, Australia. Atoll Res. Bull. 195, 1–24. Flood, P. G. 1979a. Geomorphology of Tryon Island and reef. Queensland Nat. 22, 113–125. Flood, P. G. 1979b. Heron Island erosion problems. Reeflections 3, 4. Flood, P. G. 1980. Cyclone ‘‘Simon’’ changes cays. Reeflections 6, 4. Flood, P. G. 1981. Coral cays and cyclones. Beach Conservat. 42, 6. Flood, P. G. 1983a. Holocene sea level data from the Southern Great Barrier Reef and Southeastern Queensland: a review. In Hopley, D. (ed.) Australian Sea Levels in the Last 15 000 Years: A Review, Monograph Series, Occas. Paper 3. Townsville, Department of Geography, James Cook University, pp. 85–92. Flood, P. G. 1983b. Climatically included changes to the shape of coral cays, southern Great Barrier Reef, Australia. In Baker, J. T., Carter, R. M., Sammarco, P. W., and Stark, K. P. (eds.) Proceedings of the Great Barrier Reef Conference. Townsville, James Cook University, pp. 379–384. Flood, P. G. 1984a. Changes in the Shoreline Position on Six Coral Cays, Capricorn Section, Great Barrier Reef Marine Park: A Record to 1984. Armidale, University of New England. Flood, P. G. 1984b. Variability of shoreline position in five uninhabited islands of the Capricorn Section, Great Barrier Reef Marine Park. In Ward, W. T. and Saenger, P. (eds.) The Capricornia Section of the Great Barrier Reef: Past, Present and Future. Brisbane, Royal Society of Queensland and Australian Coral Reef Society, pp. 17–24. Flood, P. G. 1985. Changes in Shoreline Positions on Coral Cays, Capricornia Section, Great Barrier Reef Marine Park. Townsville, Great Barrier Reef Marine Park Authority. Flood, P. G. 1986. Sensitivity of coral cays to climatic variations, southern Great Barrier Reef, Australia. Coral Reefs 5, 13–18.
References
483
Flood, P. G. 1988. Shoreline changes on coral cays, Capricornia section, Great Barrier Reef Marine Park, Australia. Proc. 6th Int. Coral Reef Symp., Townsville 2, 219–224. Flood, P. G. and Frankel, E. 1989. Late Holocene higher sea level indicators from eastern Australia. Mar. Geol. 90, 193–195. Flood, P. G. and Heatwole, H. 1986. Coral cay instability and species turnover of plants at Swain Reefs, Southern Great Barrier Reef, Australia. J. Coast. Res. 2, 479–496. Flood, P. G. and Jell, T. S. 1977. The effect of cyclone ‘David’ (January 1976) on the sediment distribution patterns on Heron Reef, Great Barrier Reef, Australia. Proc. 3rd Int. Coral Reef Symp., Miami 2, 119–125. Flood, P. G. and Orme, G. R. 1988. Mixed siliciclastic/carbonate sediments of the northern Great Barrier Reef Province, Australia. Devel. Sedimentol. 42, 175–205. Flood, P. G., Orme, G. R., and Scoffin, T. P. 1978. An analysis of the textural variability displayed by inter-reefal sediments of the impure carbonate facies in the vicinity of the Howick Group. Phil. Trans. Roy. Soc. Lond. A 291, 73–83. Flood, P. G., Harjanto, S., and Orme, G. R. 1979. Carbon-14 dates, Lady Elliot Island, Great Barrier Reef., Queensland Govt. Min. J. 444–447. Foreman, M. G. G. 1977. Manual for Tidal Height Analysis and Prediction, Pacific Marine Science Report 77–10. Sidney, British Columbia, Institute of Ocean Science. Fosberg, F. R. 1961. Description of Heron Island. Atoll Res. Bull. 82, 1–4. Foster, D. F. 1974. Geomechanical properties of coral rock. B. E. Hons. thesis, James Cook University, Townsville. Frankel, E. 1974. Recent sedimentation in the Princess Charlotte Bay area, Great Barrier Reef Province. Proc. 2nd Int. Coral Reef Symp., Queensland 2, 355–369. Frith, C. A. 1982. Circulation in a platform reef lagoon, One Tree Reef, southern Great Barrier Reef. Proc. 4th Int. Coral Reef Symp., Manila 1, 347–354. Frith, C. A. 1983a. Some aspects of lagoonal circulation and sedimetation One Tree Reef, Great Barrier Reef. Bureau Min. Resources J. Austral. Geol. Geophys. 8, 211–221. Frith, C. A. 1983b. Windward reef circulation, Davies Reef, Central Great Barrier Reef. In Baker, J. T., Carter, R. M., Sammarco, P. W., and Stark, P. (eds.) Proceedings of the Great Barrier Reef Conference. Townsville, James Cook University, pp. 435–440. Frith, C. A. and Mason, L. B. 1986. Modelling wind driven circulation One Tree Reef, Southern Great Barrier Reef. Coral Reefs 4, 201–211. Furnas, M. 2003. Catchments and Corals: Terrestrial Run-Off to the Great Barrier Reef. Townsville, Australian Institute of Marine Science. Furnas, M. J. and Mitchell, A. W. 1986. Phytoplankton dynamics in the central Great Barrier Reef. I. Seasonal changes in biomass and their relation to intrusive activity. Continent Shelf Res. 6, 363–384. Furnas, M. J. and Mitchell, A. W. 1996. Nutrient inputs into the central Great Barrier Reef (Australia) from subsurface intrusions of coral sea waters: a two-dimensional displacement model. Continent. Shelf Res. 16, 1127–1148. Furnas, M. and Mitchell, A. W. 2001. Runoff of terrestrial sediment and nutrients into the Great Barrier Reef World Heritage Area. In Wolanski, E. (ed.) Oceanographic Processes Of Coral Reefs: Physical and Biological Links in the Great Barrier Reef. Boca Raton, FL, CRC Press, pp. 37–51. Gagan, M. K., Sandstrom, M. W., and Chivas, A. R. 1987. Restricted terrestrial carbon input to the continental shelf during Cyclone Winifred: implications for terrestrial runoff to the Great Barrier Reef Province. Coral Reefs 6, 113–119.
484
References
Gagan, M. K., Johnson, D. P., and Carter, R. M. 1988. The Cyclone ‘Winifred’ storm bed, central Great Barrier Reef shelf, Australia. J. Sed. Petrol. 58, 845–856. Gagan, M. K., Chivas, A. R., and Herczeg, A. L. 1990. Shelf-wide erosion, deposition, and suspended sediment transport during cyclone Winifred, central Great Barrier Reef, Australia. J. Sed. Petrol. 60, 456–470. Gagan, M. K., Chivas, A. R., and Isdale, P. J. 1994. High-resolution isotopic records from corals using ocean temperature and mass-spawning chronometers. Earth Planet. Sci. Lett. 121, 549–558. Gagan, M. K., Chivas, A. R., and Isdale, P. J. 1996. Timing coral-based climatic histories using 13C enrichments driven by synchronized spawning. Geology 24, 1009–1012. Gagan, M. K., Ayliffe, L. K., Hopley, D., et al. 1998. Temperature and surface–ocean water balance of the mid-Holocene tropical Western Pacific. Science 279, 1014–1018. Gagan, M. K., Ayliffe, L. K., Opdyke, B. N., et al. 2002. Coral oxygen isotope evidence for recent groundwater fluxes to the Australian Great Barrier Reef. Geophys. Res. Lett. 29, 1–4. Gallup, C. D., Cheng, H., Taylor, F. W., and Edwards, R. L. 2002. Direct determination of the timing of sea level change during termination II. Science 295, 310–313. Gardiner, J. S. 1898. The building of atolls. Proc. Int. Congr. Zool. 119–124. Gehrels, W. R. 1994. Determining relative sea-level change from salt-marsh foraminifera and plant zones on the coast of Maine, USA. J. Coast. Res. 10, 990–1009. Gehrels, W. R., Roe, H. M., and Charman, D. J. 2001. Foraminifera, testate amoebae and diatoms as sea-level indicators in UK saltmarshes: a quantitative multiproxy approach. J. Quatern. Sci. 16, 201–220. Gibb, J. G. 1986. A New Zealand regional eustatic sea level curve and its application for determination of vertical tectonic movements. Bull. Roy. Soc. New Zealand 24, 377–395. Gill, E. D. and Hopley, D. 1972. Holocene sea levels in eastern Australia: a discussion. Mar. Geol. 12, 223–242. Gischler, E. 2003. Holocene lagoonal development in the isolated carbonate platforms off Belize. Sedi. Geol. 159, 113–132. Gischler, E. and Hudson, J. H. 1998. Holocene development of three isolated carbonate platforms, Belize, Central America. Mar. Geol. 144, 333–347. Gleghorn, R. J. 1947. Cyclone damage on the Great Barrier Reef. Repts GBR Comm. 6, 17–19. Goodwin, I. D. 1998. Did changes in Antarctic ice volume influence late Holocene sea-level lowering? Quatern. Sci. Rev. 17, 319–332. Goreau, T. F. and Land, L. S. 1974. Fore reef morphology and depositional processes, North Jamaica. In Laporte, L. F. (ed.) Reefs in Time and Space, Society of Economic Paleontologists and Mineralogists Special Publication 18. Tulsa, OK, S.E.P.M., pp. 17–89. Gourlay, M. R. 1983a. Accretion and erosion on coral cays and some consequent implications for management of marine parks. In Baker, J., Carter, R., Sammarco, P., and Stark, K. (eds.) Proceedings of the Great Barrier Reef Conference. Townsville, James Cook University, pp. 475–482. Gourlay, M. R. 1983b. Interaction between actual processes and engineering works on the leeward side of a coral cay: a case study of Heron Island on the Great Barrier
References
485
Reef. Proc. Int. Conf. Coastal and Port Engineering in Developing Countries, Colombo, 1468–1482. Gourlay, M. R. 1988. Coral cays: products of wave action and geological processes in a biogenic environment. Proc. 6th Int. Coral Reef Symp., Townsville 2, 491–496. Gourlay, M. R. 1990. Wave set-up and currents on reefs: cay formation and stability. Proc. Engineering in Coral Reef Regions Conf., Townsville, 163–178. Gourlay, M. R. 1993. Wave set-up and wave-generated currents on coral reefs. Proc. 11th Australasian Conf. Coastal and Ocean Engineering, Townsville, 479–484. Gourlay, M. R. 1994. Wave transformation on a coral reef. Coast. Engin. 23, 17–42. Gourlay, M. R. 1996a. Wave set-up on coral reefs. I. Set-up and wave-generated flow on an idealized two dimensional horizontal reef. Coast. Engin. 27, 161–193. Gourlay, M. R. 1996b. Wave set-up on coral reefs. II. Set-up on reefs with various profiles. Coast. Engin. 28, 17–55. Gourlay, M. R. and Colleter, G. 2005. Wave-generated flow on coral reefs: an analysis for two-dimensional horizontal reef-tops with steep faces. Coast. Engin. 52, 353–387. Gourlay, M. R. and Flood, P. G. 1981. Impact of coastal engineering works upon a coral cay: Heron Island. Conf. Environ. Engineering, Townsville 6, 159–163. Gourlay, M. R. and Hacker, J. L. F. 1991. Raine Island Coastal Processes and Sedimentology, Department of Civil Engineering Report CH40/91. Brisbane, University of Queensland. Graham, T. L. 1993. Geomorphological response of continental shelf and coastal environments to the Holocene transgression: Central Great Barrier Reef. Ph.D. thesis, James Cook University, Townsville. Great Barrier Reef Marine Park Authority. 2001. Great Barrier Reef Catchment Water Quality Plan: Report to Ministerial Council on Targets for Pollutant Loads. Townsville, Great Barrier Reef Marine Park Authority. Great Barrier Reef Marine Park Authority. 2003. Basis for Zoning Decisions Report: Describing the Issues, Public Comments and Management Responses for the Proposed Zoning. Townsville, Great Barrier Reef Marine Park Authority. Great Barrier Reef Marine Park Authority. 2004. Great Barrier Reef Marine Park Zoning Plan 2003. Townsville, Great Barrier Reef Marine Park Authority. Available online at http://www.gbrmpa.gov.au/ Green Island Management Committee. 1980. Green Island Management Plan Queensland. Brisbane, National Parks and Wildlife Service. Griffin, D., Middleton, J., and Bode, L. 1987. The tidal and longer-period circulation of Capricornia, Southern Great Barrier Reef. Austral. J. Mar. Freshw. Res. 38, 461–474. Griffin, G. M. 1974. Dredging in the Florida Keys: Case History of a Typical Dredge–Fill Project in the Northern Florida Keys – Effects on Water Depth, Sedimentation Rates and Biota. Fort Pierce, FL, Harbor Branch Foundation. Grigg, R. W. 1992. Coral reef environmental science: truth versus the Cassandra syndrome. Coral Reefs 11, 183–186. Grigg, R. W. 1994a. Science management of the world’s fragile coral reefs. Coral Reefs 13, 1. Grigg, R. W. 1994b. Coral reef environmental science: truth versus a false dichotomy. Reef Encounter 15, 9–10. Grigg, R. W. 1998. Holocene coral reef accretion in Hawaii: a function of wave exposure and sea-level history. Coral Reefs 17, 263–272. Grigg, R. W. 2002. Coral reef evolution: short-term instability versus evolutionary stasis. Integr. Coast. Zone Mgmt Dec. 2002, 65–68.
486
References
Grigg, R. W., Grossman, E. E., Earle, S. A., et al. 2002. Drowned reefs and antecedent karst topography, Au’au Channel, S. E. Hawaiian Islands. Coral Reefs 21, 73–82. Grimes, K. G. 1980. The Tertiary geology of North Queensland. In Henderson, R. A. and Stephenson, P. J. (eds.) The Geology and Geophysics of Northeastern Australia. Brisbane, Geological Society of Australia, pp. 329–347. Grimes, K. G. 1982. Drill site report, AIMS Davies No. 1. Geol. Surv. Queensland Rec. 18. Grindrod, J. and Rhodes, E. 1984. Holocene sea-level history of a tropical estuary: Missionary Bay, North Queensland. In Thom, B. (ed.) Coastal Geomorphology of Australia. Sydney, Academic Press, pp. 151–178. Grindrod, J., Moss, P., and van der Kaars, S. 1999. Late Quaternary cycles of mangrove development and decline on the north Australian continental shelf. J. Quatern. Sci. 14, 465–470. Grossman, E. E., Fletcher, C. H., and Richmond, B. M. 1998. The Holocene sea-level highstand in the equatorial Pacific: analysis of the insular paleo sea-level database. Coral Reefs 17, 309–327. Guilcher, A. 1988. Coral Reef Geomorphology. Chichester, John Wiley. Guinotte, J. M., Buddemeier, R. W., and Kleypas, J. A. 2003. Future coral reef habitat marginality: temporal and spatial effects of climate change in the Pacific basin. Coral Reefs 22, 551–558. Halley, R. B. and Yates, K. K. 2000. Will reef sediments buffer corals from increased global CO2? Proc. 9th Int. Coral Reef Symp., Bali, Abstr. 248. Hallock, P. and Schlager, W. 1986. Nutrient excess and the demise of coral reefs and carbonate platforms. Palaios 1, 389–398. Hamner, W. M. and Hauri, I. R. 1977. Fine-scale surface currents in the Whitsunday Islands, Queensland, Australia: effect of tide and topography. Austral. J. Mar. Freshw. Res. 428, 333–359. Hamner, W. M. and Hauri, I. R. 1981. Effects of island mass: water-flow and plankton pattern around a reef in the Great Barrier Reef lagoon, Australia. Limnol. Oceanogr. 26, 1084–1102. Hardy, T. A. and Young, I. R. 1991. Modelling spectral wave transformation on a coral reef flat. Proc. 10th Australasian Conf. Coastal and Ocean Engineering, Auckland, 345–350. Hardy, T. A., Young, I. R., Nelson, R. C., and Gourlay, M. R. 1991a. Wave attenuation on a coral reef. Austral. Civ. Engin. Trans. CE33, 17–22. Hardy, T. A., Young, I. R., Nelson, R. C., and Gourlay, M. R. 1991b. Wave attenuation on an offshore coral reef. Proc. 22nd Coastal Engineering Conf., Delft 1, 330–344. Hardy, T. A., Mason, L. B., and McConochie, J. D. 2000. A wave model for the Great Barrier Reef. Ocean Engin. 28, 45–70. Hardy, T. A., McConochie, J. D., and Mason, L. B. 2003. Modeling tropical cyclone wave population of the Great Barrier Reef. J. Waterway, Port, Coast. Ocean Engin. 129, 104–113. (Model available online at http://mmu.jcu.edu.au) Hardy, T. A., Mason, L. B., and Astorquia, A. 2004. The frequency of surge plus tide during tropical cyclones for selected open coast locations along the Queensland east coast. Unpublished report. Townsville, Marine Modelling Unit, James Cook University. Harriott, V. J. 1999. Coral growth in subtropical eastern Australia. Coral Reefs 18, 281–291.
References
487
Harriott, V. J. and Banks, S. A. 2002. Latitudinal variation in coral communities in eastern Australia: a qualitative biophysical model of factors regulating coral reefs. Coral Reefs 21, 83–94. Harris, P. T. 1988. Sediments, bedform and bedload transport pathways on the continental shelf adjacent to Torres Strait, Australia – Papua New Guinea. Continent. Shelf Res. 8, 979–1003. Harris, P. T. 1991. Sedimentation at the junction of the Fly River delta and northern Great Barrier Reef. In Lawrence, D. and Cansfield-Smith, T. (eds.) Sustainable Development for Traditional Inhabitants of the Torres Strait Region. Townsville, Great Barrier Reef Marine Park Authority, pp. 59–85. Harris, P. T. 1995. Muddy waters: the physical sedimentology of Torres Strait. In Recent Advances in Marine Science and Technology, Proc. Pacific Congress on Marine Science and Technology, pp. 149–160. Harris, P. T. 1999. Discussion: sequence architecture during the Holocene transgression – an example from the Great Barrier Reef shelf, Australia: comment. Sedi. Geol. 125, 235–239. Harris, P. T. and Davies, P. J. 1989. Submerged reefs and terraces on the shelf edge of the Great Barrier Reef, Australia. Coral Reefs 8, 87–89. Harris, P. T., Davies, P. J., and Marshall, J. F. 1990. Late Quaternary sedimentation on the Great Barrier Reef continental shelf and slope east of Townsville, Australia. Mar. Geol. 94, 55–77. Harris, P. T., Baker, E. K., and Cole E. R. 1993. Late Quaternary sedimentation at the Fly River–Great Barrier Reef junction (North-eastern Australia). Proc. 7th Int. Coral Reef Symp., Guam 2, 1147–1156. Harris, P. T., Heap, A. D., Wassenberg, T., and Passlow, V. 2004. Submerged coral reefs in the Gulf of Carpentaria, Australia. Mar. Geol. 207, 185–191. Harris, P. T., Heap, A., Passlow, V., et al. 2005. Tidally incised valleys on tropical carbonate shelves: an example from the northern Great Barrier Reef, Australia. Mar. Geol. 220, 181–204. Hartcher, M. and Shearin, J. 1996. Developing a Corporate-Wide Network for GIS. Townsville, Great Barrier Reef Marine Park Authority. Harvey, N. 1977a. The identification of subsurface solution disconformities on the Great Barrier Reef, Australia, between 148 S and 178 S using shallow seismic refraction techniques. Proc. 3rd Int. Coral Reef Symp., Miami 2, 45–51. Harvey, N. 1977b. Application of shallow seismic refraction techniques to coastal geomorphology: a coral reef example. Catena 4, 333–339. Harvey, N. 1980. Seismic investigations of a pre-Holocene substrate beneath modern reefs in Great Barrier Reef Province. Ph.D. thesis, James Cook University of North Queensland, Townsville. Harvey, N. and Hopley, D. 1982. The relationship between modern reef morphology and the pre-Holocene substrate in the Great Barrier Reef Province. Proc. 4th Int. Coral Reef Symp., Manila 1, 549–554. Harvey, N. and Searle, D. E. 1983. Seismic investigations of Late Quaternary reefal and inter-reefal sediments of the Great Barrier Reef province. In Barnes, D. J. (ed.) Perspectives on Coral Reefs. Townsville, Australian Institute of Marine Science, pp. 154–163. Harvey, N., Davies. P. J., and Marshall J. F. 1979. Seismic refraction: a tool for studying coral reef growth. Bureau Min. Resources J. Austral. Geol. Geophys. 4, 141–147.
488
References
Harvey, N., Belperio, A., Bourman, R., James, K., and Brunskill, G. 2001. New evidence contributing to the debate on the Holocene high sea-level stand in north east Queensland. Proc. New Zealand Geog. Soc. and Austral. Geog. Conf., Dunedin, 177–184. Haslett, S. 2001. The palaeoenvironmental implications of the distribution of intertidal foraminifera in a tropical Australian estuary: a reconnaissance study. Austral. Geog. Stud. 39, 67–74. Haworth, R. J., Baker, R. G. V., and Flood, P. G. 2002. Predicted and observed Holocene sea-levels on the Australian coast: what do they indicate about hydroisostatic models in far-field sites? J. Quatern Sci. 17, 581–591. Hayne, M. and Chappell, J. 2001. Cyclone frequency during the last 5000 years at Curacoa Island, north Queensland, Australia. Palaeogeog. Palaeoclimatol. Palaeoecol. 168, 207–219. Hayward, B. W., Grenfell, H. R., and Scott, D. B. 1999. Tidal range of marsh foraminifera for determining former sea-level heights in New Zealand. J. Geol. Geophys. 42, 395–413. Heap, A. D., Dickens, G. R., Stewart, L. K., and Woolfe, K. J. 2002. Holocene storage of siliciclastic sediment around islands on the middle shelf of the Great Barrier Reef platform, north-east Australia. Sedimentology 49, 603–621. Hearn, C. J. and Parker, I. N. 1988. Hydrodynamic processes on the Ningaloo coral reef, Western Australia. Proc. 6th Int. Coral Reef Symp., Townsville 2, 497–502. Hearn, C. J., Atkinson, M. J., and Falter, J. L. 2001. A physical derivation of nutrientuptake rates in coral reefs: effects of roughness and waves. Coral Reefs 20, 347–356. Heatwole, H. 1983. Interaction of plants and animals in development of coral cays. In Baker, J., Carter, R., Sammarco, P., and Stark, K. (eds) Proceedings of the Great Barrier Reef Conference. Townsville, James Cook University, pp. 385–390. Heatwole, H. 1984. Terrestrial vegetation on coral cays, Capricornia Section, Great Barrier Reef Marine Park. In Ward, W. T. and Saenger, P. (eds.) The Capricornia Section of the Great Barrier Reef: Past, Present and Future. Brisbane, Royal Society of Queensland and Australian Coral Reef Society, pp. 87–140. Hedley, C. 1925a. The natural destruction of a coral reef. Repts. GBR Comm. 1, 35–40. Hedley, C. 1925b. A raised beach at the North Barnard Islands. Repts. GBR Comm. 1, 61–62. Hedley, C. 1925c. Coral shingle as a beach formation. Repts. GBR Comm. 1, 66. Hedley, C. 1925d. A disused river mouth at Cairns. Repts. GBR Comm. 1, 69–72. Hedley, J. D., Mumby, P. J., Joyce, K. E., and Phinn, S. R. 2004. Spectral unmixing of coral reef benthos under ideal conditions. Coral Reefs 23, 60–73. Hekel, H. 1979. Geological Report on the AIMS Northern Cays Expedition 1979. Townsville, Australian Institute of Marine Science. Hendon, H. H. and Liebmann, B. 1990. A composite study of the onset of the Australian summer monsoon. J. Atmos. Sci. 47, 2227–2240. Hendon, H. H., Davidson, N. E., and Gunn, B. 1989. Australian summer monsoon onset during AMEX, 1987. Monthly Weather Rev. 117, 370–390. Hendy, E. J., Gagan, M. K., and Lough, J. M. 2003. Chronological control of coral records using luminescent lines and evidence for non-stationary ENSO teleconnections in northeast Australia. Holocene 13, 187–199. Herman, M. E., Buddemeier, R. M., and Wheatcraft, S. W. 1986. A layered aquifer model of atoll island hydrology: validation of a computer simulation. J. Hydrol. 84, 303–322.
References
489
Hill, D. 1974. An introduction to the Great Barrier Reef. Proc. 2nd Int. Coral Reef Symp., Queensland 2, 723–731. Hill, T. M., Brooks, G. R., Duncan, D. S., and Medioli, F. S. 2003. Benthic foraminifera of the Holocene transgressive west-central Florida inner shelf: paleoenvironmental implications. Mar. Geol. 200, 263–272. Hillis, L. 1997. Coralgal reefs from a green alga perspective and a first carbonate budget. Proc. 8th Int. Coral Reef Symp., Panama 1, 761–766. Hillis-Colinvaux, L. 1986. Halimeda growth and diversity on the deep fore-reef of Enewetak Atoll. Coral Reefs 5, 19–21. Hillis-Colinvaux, L. 1988. Characteristics of Halimeda meadows, with emphasis on a meadow near Eniwetok Islet, Eniwetok Atoll, Marshall Islands. Proc. 6th Int. Coral Reef Symp., Townsville 3, 119–125. Hillis-Colinvaux, L. and Orme, G. R. 1988. Trends in research on the biology and geology of Halimeda. Proc. 6th Int. Coral Reef Symp., Townsville 1, 193–198. Hine, A. C., Hallock, K. P., Harris, M. W., et al. 1988. Halimeda bioherms along an open seaway: Miskito Channel, Nicaraguan Rise, SW Caribbean Sea. Coral Reefs 6, 173–178. Hochberg, E. J., Atkinson, M. J., and Andrefouet, S. 2003. Spectral reflectance of coral reef bottom-types worldwide and implications for coral reef remote sensing. Remote Sens. Environ. 85, 159–173. Hoegh-Guldberg, O. 1999. Climate change, coral bleaching and the future of the world’s coral reefs. Mar. Freshw. Res. 50, 839–866. Hoegh-Guldberg, O., Fine, M., Skirving, W., et al. 2005. Coral bleaching following wintry weather. Limnol. Oceanogr. 50, 265–271. Holmes, A. 1944. Principles of Physical Geology. London, Thomas Nelson. Hopley, D. 1968. Morphology of Curocoa Island spit, North Queensland. Austral. J. Sci. 31, 122–123. Hopley, D. 1970. The Geomorphology of the Burdekin Delta, North Queensland, Monograph Series, Occas. Paper 1. Townsville, Department of Geography, James Cook University. Hopley, D. 1971. The origin and significance of North Queensland island spits. Z. Geomorphol. N.F. 15, 371–389. Hopley, D. 1972. The storm surge associated with Cyclones Althea and Emily. In Trollope, D. H. (ed.) Cyclone Althea Part II: Storm Surges and Coastal Effects. Townsville, James Cook University, pp. 4.1–4.29. Hopley, D. 1974a. The cyclone Althea storm surge. Austral. Geog. Stud. 12, 90–106. Hopley, D. 1974b. Australian weather example No. 2: storm surge. Austral. Geog. Stud. 12, 462–468. Hopley, D. 1975. Contrasting evidence for Holocene sea levels with special reference to the Bowen–Whitsunday area of Queensland. In Douglas, I., Hobbs, J. E., and Pigram J. J. (eds.) Geographical Essays in Honour of Gilbert J. Butland. Armidale, Department of Geography, University of New England, pp. 51–84. Hopley, D. 1977. The age of the outer ribbon reef surface, Great Barrier Reef, Australia: implications for hydroisostatic models. Proc. 3rd Int. Coral Reef Symp., Miami 2, 23–28. Hopley, D. 1978a. Sea level change on the Great Barrier Reef: an introduction. Phil. Trans. Roy. Soc. Lond. A 291, 159–166. Hopley, D. 1978b. Geomorphology of the reefs and reef islands north of Lizard Island. Proc. Workshop on the Northern Sector of the Great Barrier Reef. Townsville, Great Barrier Reef Marine Park Authority, pp. 218–252.
490
References
Hopley, D. 1978c. Wheeler Reef: cay mobility. In Hopley, D. (ed.) Geographical Studies of the Townsville Area Monograph Series, Occas. Paper 2. Townsville, Geography Department, James Cook University, pp. 55–58. Hopley, D. 1980. Mid-Holocene high sea levels along the coastal plain of the Great Barrier Reef Province: a discussion. Mar. Geol. 35, M1–M9. Hopley, D. 1981. Sediment movement around a coral cay, Great Barrier Reef, Australia. Pacif. Geol. 15, 17–36. Hopley, D. 1982. The Geomorphology of the Great Barrier Reef: Quaternary Development of Coral Reefs. New York, John Wiley. Hopley, D. 1983a. Evidence of 15 000 years of sea level change in tropical Queensland. In Hopley, D. (ed.) Australian Sea Levels in the Last 15 000 Years: A Review, Monograph Series, Occas. Paper 3. Townsville, Geography Department, James Cook University, pp. 93–104. Hopley, D. 1983b. Deformation of the North Queensland continental shelf in the late Quaternary. In Smith, D. E. and Dawson, A. G. (eds.) Shorelines and Isostacy. London, Institute of British Geographers, pp. 347–366. Hopley, D. 1983c. Morphological classification of shelf reefs: a critique with special reference to the Great Barrier Reef. In Barnes, D. J. (ed.) Perspectives on Coral Reefs. Canberra, Brian Clonston Publisher, pp. 180–199. Hopley, D. 1984. The Holocene high energy window on the Central Great Barrier Reef. In Thom, B. G. (ed.) Coastal Geomorphology in Australia. Sydney, Academic Press, pp. 135–150. Hopley, D. 1985. Geomorphological development of modern coastlines. In Pitty, A. F. (ed.) Themes in Geomorphology. Beckenham, Croom Helm, pp. 56–71. Hopley, D. 1986a. Corals and reefs as indicators of paleo-sea levels, with special reference to the Great Barrier Reef. In Van de Plassche, O. (ed.) Sea-Level Research: A Manual for the Collection and Evaluation of Data. Norwich, Geo Books, pp. 195–228. Hopley, D. 1986b. Beachrock as a sea-level indicator. In Van de Plassche, O. (ed.) SeaLevel Research: A Manual for the Collection and Evaluation of Data. Norwich, Geo Books, pp. 157–173. Hopley, D. 1987. Holocene sea level changes in Australia and the southern Pacific. In Devoy, R. J. N. (ed.) Sea Surface Studies. London, Croom Helm, pp. 375–408. Hopley, D. 1988. Anthropogenic influences on Australia’s Great Barrier Reef. Austral. Geog. 19, 26–45. Hopley, D. 1989a. Coral reefs: zonation, zonality and gradients. Essener Geog. Arbeit. 18, 79–123. Hopley, D. 1989b. The Formation, Use and Management of the Great Barrier Reef. Melbourne, Longman Cheshire. Hopley, D. 1993. Coral reef islands in a period of global sea-level rise. In Saxena, N. (ed.) Recent Advances in Marine Science and Technology, Proc. Pacific Congress on Marine Science and Technology, Honolulu, pp. 453–462. Hopley, D. 1994. Continental shelf reef systems. In Carter, R. W. G. and Woodroffe, C. D. (eds.) Coastal Evolution: Late Quaternary Shoreline Dynamics. Cambridge, Cambridge University Press, pp. 303–340. Hopley, D. 1997a. Coral reef islands: implications of more modest global change predictions. In Saxena, N. (ed.) Recent Advances in Marine Science and Technology, Proc. Pacific Congress on Marine Science and Technology, Honolulu, pp. 249–258. Hopley, D. 1997b. Geology of reef islands of the Great Barrier Reef Australia. Devel. Sedimentol. 54, 835–866.
References
491
Hopley, D. 2006. Coral reef growth on the shelf margin of the Great Barrier Reef with special reference to the Pompey Complex. J. Coast. Res. 22, 150–158. Hopley, D. and Barnes, R. 1985. Structure and development of a windward fringing reef, Orpheus Island, Palm Group, Great Barrier Reef. Proc. 5th Int. Coral Reef Congr., Tahiti 3, 141–146. Hopley, D. and Catt, P. C. 1988. Use of near infra-red aerial photography for monitoring ecological changes to coral reef flats on the Great Barrier Reef. Proc. 6th Int. Coral Reef Symp., Townsville 3, 503–508. Hopley, D. and Choat, H. C. 1990. The effects of mainland land use on adjacent reef systems of the Great Barrier Reef. In Agriculture and the Ecosystem in North Queensland. Townsville, Australian Institute of Agricultural Science, pp. 1–16. Hopley, D. and Harvey, N. 1982. Radiocarbon ages and morphology of reef tops on the Great Barrier Reef between 148 390 S and 208 450 S: indicators of shelf neotectonics. Proc. 4th Int. Coral Reef Symp., Manila 1, 523–530. Hopley, D. and Isdale, P. J. 1977. Coral micro atolls, tropical cyclones and reef flat morphology: a north Queensland example. Search 8, 79–81. Hopley, D. and Murtha, G. G. 1975. The Quaternary Deposits of the Townsville Coastal Plain, Monograph Series 8. Townsville, Department of Geography, James Cook University. Hopley, D. and Partain, B. 1987. The structure and development of fringing reefs off the Great Barrier Reef Province. In C. Baldwin (ed.) Fringing Reef Workshop: Science, Industry and Management. Townsville, Great Barrier Reef Marine Park Authority, pp. 13–33. Hopley, D. and Pichon, M. 1994. Coconut Island Proposed Reclamation, Unpubl. consultancy report. Cairns, Edmiston and Taylor. Hopley, D. and Rasmussen, C. E. 1998. Coconut Island Sand Erosion Study, Unpubl. report. Cairns, Edmiston and Taylor. Hopley, D. and Smithers, S. G. 2003. Queensland. In Bird, E. C. F. and Schwartz, M. (eds.) The World’s Coasts. Available online at http://www.wkap.nl/subjects/ TWCO Hopley, D. and Thom, B. G. 1983. Australian sea levels in the last 15 000 years: a review. In Hopley, D. (ed.) Australian Sea Levels in the Last 15 000 Years: A Review, Monograph Series, Occas. Paper 3. Townsville, Department of Geography, James Cook University, pp. 3–26. Hopley, D. and van Woesik, R. 1988. Turbidity Levels in Nelly Bay, Magnetic Island Nth. Qld. with Reference to Magnetic Quay Proposal. Townsville, McIntyre and Associates. Hopley, D., McLean, R. F., Marshall, J., and Smith, A. S. 1978. Holocene–Pleistocene boundary in a fringing reef: Hayman Island north Queensland. Search 9, 323–325. Hopley, D., Davies, P. J., Harvey, N., and Isdale, P. J. 1982. The geomorphology of Redbill Reef, Central Great Barrier Reef. Proc. 4th Int. Coral Reef Symp., Manila 1, 541–548. Hopley, D., Slocombe, A., Muir, F., and Grant, C. 1983. Nearshore fringing reefs in North Queensland: structure, growth and classification. Coral Reefs 1, 151–160. Hopley, D., Muir, F. J., and Grant, C. 1984. Pleistocene foundations and Holocene growth of Redbill Reef, South Central Great Barrier Reef. Search 15, 288–289. Hopley, D., Parnell, K. E., and Isdale, P. J. 1989. The Great Barrier Reef Marine Park: dimensions and regional patterns. Austral. Geog. Stud. 27, 47–66.
492
References
Hopley, D., van Woesik, R., Hoyal, D. W. D., Rasmussen, C. E., and Steven A. D. L. 1990. Sedimentation Resulting from Road Development Cape Tribulation Area, Technical Memoir 24. Townsville, Great Barrier Reef Marine Park Authority. Hopley, D., Graham, T. L., and Rasmussen, C. E. 1997. Submerged shelf edge reefs, Great Barrier Reef, Australia. In Saxena, N. (ed.) Recent Advances in Marine Science and Technology, Proc. Pacific Congress on Marine Science and Technology, Honolulu, pp. 305–315. Horton, B. P., Edwards, R. J., and Lloyd, J. M. 1999. A foraminiferal-based transfer function: implications for sea-level studies. J. Foramin Res. 29, 117–129. Horton, B. P., Larcombe, P., Woodroffe, S. A., et al. 2003. Contemporary foraminiferal distributions of a mangrove environment, Great Barrier Reef coastline, Australia: implications for sea-level reconstructions. Mar. Geol. 198, 225–243. Hubbard, D. K. 1997. Reefs as dynamic systems. In Birkeland, C. (ed.) Life and Death of Coral Reefs. New York, Chapman and Hall, pp. 43–67. Hubbard, D. K., Gill, I. P., Burke, R. B., and Morelock, J. 1997. Holocene reef backstepping: southeastern Puerto Rico shelf. Proc. 8th Int. Coral Reef Symp., Panama 2, 1779–1784. Hughes, P. J. and Lampert, R. J. 1982. Prehistoric population changes in southern coastal New South Wales. In Bowdler, S. (ed.) Coastal Archaeology in Eastern Australia, Proc. 1980 Valla Conf. on Australian Prehistory, Canberra, pp. 16–28. Hughes, T. P. and Connell, J. H. 1999. Multiple stressors on coral reefs: a long-term perspective. Limnol. Oceanogr. 44, 932–940. Hughes, T. P., Baird, A. H., Dinsdale, E. A., et al. 1999. Patterns of recruitment and abundance of corals along the Great Barrier Reef. Nature 397, 59–63. Hull, K. 2005. Ancient mangroves reveal rapid sea level rise. Austral. Sci. May 2005, 31–33. Idnurm, M. and Cook, P. J. 1980. Paleomagnetism of beach ridges in South Australia and the Milankovitch theory of ice ages. Nature 286, 699–702. Imbrie, J., Hays, J. D., Martinson, D. G., et al. 1984. The orbital theory of Pleistocene climate: support from a revised chronology of the marine 18O record. In Berger, A. L., Imbrie, J., Hays, J. Kukla, G., and Saltzman, B. (eds.) Milankovitch and Climate: Understanding the Response to Astronomical Forcing. Dordrecht, Reidel, pp. 269–305. Intergovernmental Panel on Climate Change. 1990. Climate Change: The IPCC Assessment. Cambridge, Cambridge University Press. Intergovernmental Panel on Climate Change. 2001. Climate Change 2001: The Scientific Basis. Cambridge, Cambridge University Press. International Consortium for Great Barrier Reef Drilling. 2001. New constraints on the origin of the Great Barrier Reef from an international project of deep coring. Geology 29, 483–486. Isdale, P. J. 1984. Fluorescent bands in massive corals record centuries of coastal rainfall. Nature 310, 578–579. Isern, A. R., McKenzie, J. A., and Muller, D. W. 1993. Palaeoceanographic changes ¨ and reef growth off the northeastern Australian margin: stable isotopic data from ODP Leg 133 Sites 811 and 817 and DSDP Leg 2 Site 209. In McKenzie, J. A., Davies, P. J., Palmer-Julson, A. A., and Sarg, J. F. (eds.) Proceedings of the Ocean Drilling Program, vol. 133, Scientific Results, North-East Australian Margin. College Station, TX, Texas A and M University, pp. 263–280. James, M. K. and Mason, L. B. 2005. Synthetic tropical cyclone database. J. Waterway Port, Coast. Ocean Engin. 131, 181–192.
References
493
James, N. P. and Kendall, A. C. 1992. Introduction to carbonate and evaporite facies models. In Walker, R. G. and James, N. P. (eds.) Facies Models: Response to Sea Level Change. St. John’s, Newfoundland, Geological Association of Canada, pp. 265–276. Jardine, F. 1928. The Broad Sound drainage in relation to the Fitzroy River. Repts. GBR Comm. 2, 88–92. Jelgersma, S. 1961. Holocene sea-level changes in the Netherlands. Mededelingen Geol Stichting C-IV(7). Johnson, D. P. 1996. Knowledge summary of terrigenous input to the Central Great Barrier Reef. In Larcombe, P., Woolfe, K., and Purdon, R. (eds.) Great Barrier Reef: Terrigenous Sediment Flux and Human Impacts. Townsville, CRC Reef Research Centre, pp. 86–90. Johnson, D. 2004. The Geology of Australia. Cambridge, Cambridge University Press. Johnson, D. P. and Carter, R. M. 1987. Sedimentary Framework of Mainland Fringing Reef Development, Cape Tribulation Area, Technical Memoir 14. Townsville, Great Barrier Reef Marine Park Authority. Johnson, D. P. and Risk, M. J. 1987. Fringing reef growth on a terrigenous mud foundation, Fantome Is., GBR. Sedimentology 34, 275–287. Johnson, D. P. and Searle, D. E. 1984. Post-glacial seismic stratigraphy, central Great Barrier Reef, Australia. Sedimentology 31, 335–352. Johnson, D. P., Searle, D. E., and Hopley, D. 1982. Positive relief over buried post-glacial channels, Great Barrier Reef Province, Australia. Mar. Geol. 46, 149–159. Johnson, D., Cuff, C., and Rhodes, E. 1984. Holocene reef sequences and geochemistry, Britomart Reef, central Great Barrier Reef, Australia. Sedimentology 31, 515–529. Johnson, D., Belperio, A., and Hopley, D. 1986. A Field Guide to Mixed Terrigenous–Carbonate Sedimentation in the Central Great Barrier Reef Province, Australia, Australasian Sedimentologists’ Group Field Guide Series 3. Sydney, Geological Society of Australia. Johnson, D., Larcombe, P., Carter, B., et al. 1997. Geoscience issues on the Great Barrier Reef: time scales for reef and shelf processes. In Proc. Natl. Conf. The Great Barrier Reef: Science, Use and Management. Townsville, Great Barrier Reef Marine Park Authority, pp. 225–241. Johnston, K. L. 1996. Comparison of sedimentation trends at Low Isles, Northern Great Barrier Reef, 1928–29 and 1991–93. M.Sc. thesis, James Cook University, Townsville. Jones, M. R. 1995. The Torres Reefs, North Queensland, Australia: strong tidal flows a modern control on their growth. Coral Reefs 14, 63–69. Jouannic, C., Hoang, C. T., Hantoro, W. S., and Delinom, R. M. 1988. Uplift rate of coral reef terraces in the area of Kupang, West Timor: preliminary results. Palaeogeog. Palaeoclimatol. Palaeoecol. 68, 259–272. Joyce, K. E., Phinn, S. R., Roelfsema, C. M., Neil, D. T., and Dennison, W. C. 2004. Combining LANDSAT ETM plus and Reef Check classifications for mapping coral reefs: a critical assessment from the southern Great Barrier Reef, Australia. Coral Reefs 23, 21–25. Jukes, J. B. 1847. Narrative of the Surveying Voyage of HMS Fly, commanded by Capt. F. P. Blackwood in Torres Strait, New Guinea and other islands of the eastern archipelago during the years 1842–1846, together with an excursion into the interior of the eastern port of Java, 2 vols. London, T. and W. Boone.
494
References
Jupp, D. L. B., Mayo, K. K., Kuchler, D. A., Heggen, S. J., and Kendall, S. W. 1981a. Remote sensing by LANDSAT as support for management of the Great Barrier Reef. Proc. 2nd Australasian Remote Sensing Conf., Canberra, 9.5.1–9.5.6. Jupp, D. L. B., Mayo, K. K., Kuchler, D. A., Heggen, S. J. and Kendall, S. W. 1981b. The BRIAN method for large area inventory and monitoring. Proc. 2nd Australasian Remote Sensing Conf., Canberra, 6.5.1–6.5.6. Jupp, D. L. B., Mayo, K. K. Kuchler, D. A., et al. 1985. Remote sensing for planning and managing the Great Barrier Reef of Australia. Photogrammetria 40, 21–42. Kan, H., Nakashima, Y., and Hopley, D. 1997. Coral communities during structural development of a fringing reef flat, Hayman Island, the Great Barrier Reef. Proc. 8th Int. Coral Reef Symp., Panama 1, 465–470. Karlson, R. H., Cornell, H. V., and Hughes, T. P. 2004. Coral communities are regionally enriched along an oceanic biodiversity gradient. Nature 429, 867–870. Katz, M. E. and Miller, K. G. 1993. Neogene subsidence along the northeastern Australian margin: benthic foraminiferal evidence. In McKenzie, J. A., Davies, P. J., Palmer-Julson, A. A., and Sarg, J. F. (eds.) Proceedings of the Ocean Drilling Program, vol. 133, Scientific Results, North-East Australian Margin. College Station, TX, Texas A and M University, pp. 75–92. Kayanne, H., Yamano, H., and Randall, R. 2002. Holocene sea-level changes and barrier reef formation on an oceanic island, Palau Islands, Western Pacific. Sed. Geol. 150, 47–60. Kench, P. S. 1998. A currents of removal approach for interpreting carbonate sedimentary processes. Mar. Geol. 145, 197–223. Kench, P. S. and Brander, R. W. 2006. Wave processes on coral reef flats: implications for reef geomorphology using Australian case studies. J. Coast. Res. 22, 209–223. Kench, P. S. and Cowell, P. J. 2002. Variations in sediment production and implications for atoll island stability under rising sea level. Proc. 9th Int. Coral Reef Symp., Bali 2, 1181–1186. Kench, P. S., Nichol, S. L., and McLean, R. F. 2005a. Comment on ‘‘New perspectives for the future of the Maldives’’ by Morner, N. A., et al., Global and Planetans ¨ Change 40 (2004), 177–182. Glob. Planet. Change 47, 67–69. Kench, P. S., McLean, R. F., and Nichol, S. L. 2005b. New model of reef-islands formation: Maldives, Indian Ocean. Geology 33, 145–148. Kench, P. S., McLean, R. F., Brander, R. W., et al. 2006. Geological effects of tsunami on mid-ocean atoll islands: the Maldives before and after the Sumatran tsunami. Geology 34, 177–180. Kennedy, D. M. and Woodroffe, C. D. 2000. Holocene lagoonal sedimentation at the intertidal limits of reef growth, Lord Howe Island, Tasman Sea. Mar. Geol. 169, 287–304. Kennedy, D. M. and Woodroffe, C. D. 2002. Fringing reef growth and morphology: a review. Earth Sci. Rev. 57, 255–277. Kershaw, A. P. 1978. Record of last interglacial–glacial cycle from north-eastern Queensland. Nature 272, 159–161. Kershaw, A. P. and Nanson, G. C. 1993. The last full glacial cycle in the Australian region. Glob. Planet. Change 7, 1–9. King, B. and Wolanski, E. 1992. Coastal dynamics along a rugged coastline. In Prandle, D. (ed.) Dynamics and Exchanges in Estuaries and the Coastal Zone. AGU Coastal & Estuarine Studies 40, 577–598. King, B. and Wolanski, E. 1996. Tidal current variability in the Central Great Barrier Reef. J. Mar. Syst. 9, 187–202.
References
495
King, B., McAllister, F., Wolanski, E., Done, T., and Spagnol, S. 2001. River plume dynamics in the central Great Barrier Reef. In Wolanski, E. (ed.) Oceanographic Processes of Coral Reefs: Physical and Biological Links in the Great Barrier Reef. Boca Raton, FL, CRC Press, pp. 145–159. Kingsford, M. J., Wolanski, E., and Choat, J. H. 1991. Influence of tidally induced fronts and Langmuir circulations on distribution and movements of presettlement fishes around a coral reef. Mar. Biol. 109, 167–180. Kinsey, D. W. 1985. Metabolism, calcification and carbon production. I. Systems level studies. Proc. 5th Int. Coral Reef Congr., Tahiti 4, 505–526. Kinsey, D. W. and Davies, P. J. 1979a. Effects of elevated nitrogen and phosphorus on coral reef growth. Limnol. Oceanogr. 24, 935–940. Kinsey, D. W. and Davies, P. J. 1979b. Inorganic carbon turnover, calcification and growth in coral reefs. In Trudinger, P. and Swaine, D. (eds.) Biogeochemistry of Mineral Forming Elements. Amsterdam, Elsevier, pp. 131–162. Kinsey, D. W. and Hopley, D. 1991. The significance of coral reefs as global carbon sinks: response to Greenhouse. Palaeogeog. Palaeoclimatol. Palaeoecol. 89, 363–377. Kinzie, R. A. and Buddemeier, R. W. 1996. Reefs happen. Glob. Change Biol. 2, 479–494. Kleypas, J. A. 1992. Geological development of fringing reefs of the southern Great Barrier Reef. Ph.D. thesis, James Cook University, Townsville. Kleypas, J. A. 1996. Coral reef development under naturally turbid conditions: fringing reefs near Broad Sound Australia. Coral Reefs 15, 153–167. Kleypas, J. A. and Hopley, D. 1993. Reef development across a broad continental shelf southern Great Barrier Reef Australia. Proc. 7th Int. Coral Reef Symp., Guam 2, 1129–1141. Kleypas, J. A., McManus, J. W., and Menez, L. A. B. 1999a. Environmental limits to coral reef development: where do we draw the line? Am. Zool. 39, 146–159. Kleypas, J. A., Buddemeier, R. W., Archer, D., et al. 1999b. Geochemical consequences of increased atmospheric carbon dioxide on coral reefs. Science 284, 118–120. Kleypas, J. A., Buddemeier, R. W., and Gattuso, J. P. 2001. The future of coral reefs in an age of global change. Int. J. Earth Sci. 90, 426–437. Konishi, K. 1982. Aleyonarian spiculite: limestone of soft corals. Proc. 4th Int. Coral Reef Symp., Manila 1, 643–649. Koop, K., Booth, D., Broadbent, A., et al. 2001. ENCORE: the effect of nutrient enrichment on coral reefs – synthesis of results and conclusions. Mar. Poll. Bull. 42, 91–120. Kuchler, D. A. 1978. Coral cay shoreline movements, historical and seasonal patterns, Green Island, Great Barrier Reef, Australia. Hons. thesis, James Cook University, Townsville. Kuchler, D. A. 1984. Geomorphological separability, LANDSAT MSS and aerial photographic data: Heron Island Reef, Australia. Ph.D. thesis, James Cook University, Townsville. Kutser, T., Dekker, A. G., and Skirving, W. 2003. Modeling spectral discrimination of Great Barrier Reef benthic communities by remote sensing instruments. Limnol. Oceanogr. 48, 497–510. Lambeck, K. 1981a. Flexure of the ocean lithosphere from island uplift bathymetry and geoid height observations: the Society Islands. Geophys. J. Roy. Astron. Soc. 67, 91–114.
496
References
Lambeck, K. 1981b. Lithospheric response to volcanic loading in the southern Cook Islands. Earth Planet. Sci. Lett. 55, 482–496. Lambeck, K. 2004. Sea-level change through the last glacial cycle: geophysical, glaciological and palaeogeographic consequences. C. R. Geosci. 336, 677–689. Lambeck, K. and Chappell, J. 2001. Sea level change through the last glacial cycle. Science 292, 679–686. Lambeck, K. and Nakada, M. 1990. Late Pleistocene and Holocene sea-level change along the Australian coast. Palaeogeog. Palaeoclimatol. Palaeoecol. 89, 143–176. Lambeck, A. and Woolfe, K. J. 2000. Composition and textural variability along the 10 m isobath, Great Barrier Reef: evidence for pervasive northward sediment transport. Austral. J. Earth Sci. 47, 327–335. Lambeck, K., Esat, T. M., and Potter, E. K. 2002. Links between climate and sea levels for the past three million years. Nature 419, 199–206. Langdon, C. 2002. Review of experimental evidence for the effects of CO2 on calcification of reef builders. Proc. 9th Int. Coral Reef Symp., Bali 2, 1091–1098. Larcombe, P. and Carter, R. M. 1998. Sequence architecture during the Holocene transgression: an example from the Great Barrier Reef shelf, Australia. Sed. Geol. 117, 97–121. Larcombe, P. and Carter, R. M. 1999. Discussion: ‘‘Sequence architecture during the Holocene transgression: an example from the Great Barrier Reef shelf, Australia’’ – reply. Sed. Geol. 125, 241–247. Larcombe, P. and Carter, R. M. 2004. Cyclone pumping, sediment partitioning and the development of the Great Barrier Reef shelf system: a review. Quatern. Sci. Rev. 23, 107–135. Larcombe, P. and Woolfe, K. J. 1999a. Terrigenous sediments as influences upon Holocene nearshore coral reefs, central Great Barrier Reef, Australia. Austral. J. Earth Sci. 46, 141–154. Larcombe, P. and Woolfe, K. J. 1999b. Increased sediment supply to the Great Barrier Reef will not increase sedimentation accumulation at most coral reefs. Coral Reefs 18, 163–169. Larcombe, P., Carter, R. M., Dye, J., Gagan, M. K., and Johnson, D. P. 1995a. New evidence for episodic post-glacial sea-level rise, central Great Barrier Reef, Australia. Mar. Geol. 127, 1–44. Larcombe, P., Ridd, P. V., Prytz, A., and Wilson, B. 1995b. Factors controlling suspended sediment on inner-shelf coral reefs, Townsville, Australia. Coral Reefs 14, 163–171. Larcombe, P., Costen, A., and Woolfe, K. J. 2001. The hydrodynamic and sedimentary setting of nearshore coral reefs, central Great Barrier Reef shelf, Australia: Paluma Shoals, a case study. Sedimentology 48, 811–835. Leao, Z. M. A. N. and Ginsburg, R. N. 1997. Living reefs surrounded by siliciclastics sediments: the Abrolhos coastal reefs, Bahia, Brazil. Proc 8th Int. Coral Reef Symp., Panama 2, 1767–1772. Lee, T. T. and Black, K. P. 1979. The energy spectra of surf waves on a coral reef. Proc. 16th Coastal Engineering Conf., Hamburg, 588–608. Levermann, A., Griesel, A., Hofmann, M., Montoya, M., and Rahmstorf, S. 2005. Dynamic sea level changes following changes in the thermohaline circulation. Climate Dynam. 24, 347–354. Lewis, A. 2001. Great Barrier Reef Depth and Elevation Model: GBRDEM. Townsville, CRC Reef Research Centre.
References
497
Lewis, A., Lowe, D., and Jurgen, O. 2003a. Remapping the Great Barrier Reef. Position 4, 46–49. Lewis, A., Slegers, S., Lowe, D., et al. 2003b. Use of spatial analysis and GIS to re-zone the Great Barrier Reef Marine Park. Paper presented at the Coastal GIS Workshop, Wollongong, 7–8 July 2003. Lighty, R. G. 1977. Relict shelf-edge Holocene coral reef: south-east coast of Florida. Proc. 3rd Int. Coral Reef Symp., Miami 2, 215–221. Lighty, R. G., Macintyre, I. G., and Stuckenrath, R. 1978. Submerged early Holocene barrier reef south-east Florida shelf. Nature 276, 59–60. Lighty, R. G., Macintyre, I. G., and Stuckenrath, R. 1982. Acropora palmata reef framework: a reliable indicator of sea-level in the western Atlantic for the past 10 000 years. Coral Reefs 1, 125–130. Limpus, C. J. 1987. A turtle fossil on Raine Island, Great Barrier Reef. Search 18, 254–256. Limpus, C. J., Miller, J. D., Parmenter, C. J., and Limpus, D. J. 2003. The green turtle, Chelonia mydas, population of Raine Island and the Northern Great Barrier Reef: 1843–2001. Mem. Queensland Mus. 49, 349–440. Lloyd, A. R. 1973. Foraminifera of the Great Barrier Reef bores. In Jones, O. A. and Endean, R. (eds.) Biology and Geology of Coral Reefs, vol. 1, Geology. New York, Academic Press, pp. 347–366. Lloyd, A. R. 1977. The basement beneath the Queensland continental shelf. In Jones, O. A. and Endean, R. (eds.) Biology and Geology of Coral Reefs, vol. 4, Geology 2. New York, Academic Press, pp. 261–266. Longuet-Higgins, M. S. and Stewart, R. W. 1964. Radiation stresses in water waves: a physical discussion with applications. Deep Sea Res. 11, 529–562. Lough, J. M. 1994. Climate variation and El-Nino ˜ Southern Oscillation events on the Great Barrier Reef: 1958 to 1987. Coral Reefs 13, 181–195. Lough, J. M. 2001. Climate variability and change on the Great Barrier Reef. In Wolanski, E. (ed.) Oceanographic Processes of Coral Reefs: Physical and Biological Links in the Great Barrier Reef. Boca Raton, FL, CRC Press, pp. 269–300. Lough, J. M. 2004. A strategy to improve the contribution of coral data to highresolution paleoclimatology. Palaeogeog. Palaeoclimatol. Palaeoecol. 204, 115–143. Lough, J. M. and Barnes, D. J. 1997. Several centuries of variation in skeletal extension, density and calcification in massive Porites colonies from the Great Barrier Reef: a proxy for seawater temperature and a background of variability against which to identify unnatural change. J. Exp. Mar. Biol. Ecol. 211, 29–67. Lough, J. M. and Barnes, D. J. 2000. Environmental controls on growth of the massive coral Porites. J. Exp. Mar. Biol. Ecol. 245, 225–243. Loya, Y. 1976. Effects of water turbidity and sedimentation on the community structure of Puerto Rican corals. Bull. Mar. Sci. 26, 450–466. Ludington, C. 1979. Tidal modifications and associated circulation in a platform reef lagoon. Austral. J. Mar. Freshw. Res. 30, 425–430. Ludwig, K. R., Muhs, D. R., Simmons, K. R., Halley, R. B., and Shinn, E. A. 1996. Sea-level records at approximately 80 ka from tectonically stable platforms: Florida and Bermuda. Geology 24, 211–214. Lugo-Fernandez, A., Roberts, H. H., and Suhayda, J. N. 1998. Wave transformations across a Caribbean fringing-barrier coral reef. Continent. Shelf Res. 18, 1099–1124.
498
References
Lugo-Fernandez, A., Roberts, H. H., and Wiseman, W. J. 2004. Currents, water levels, and mass transport over a modern Caribbean coral reef: Tague Reef, St. Croix, U.S.V.I. Continent. Shelf Res. 24, 1989–2009. Lyell, C. 1832. Principles of Geology. London, John Murray. Macdonald, I. A. and Perry, C. T. 2003. Biological degradation of coral framework in a turbid lagoon environment, Discovery Bay, north Jamaica. Coral Reefs 22, 526–535. Macdonald, I. A., Perry, C. T., and Larcombe, P. 2005. Comment on ‘‘Rivers, runoff, and reefs’’ by McLaughlin et al. [Global and Planetary Change 39 (2003) 191–199]. Glob. Planet. Change 45, 333–337. MacGillivray, T. 1852. Narrative of a Voyage of HMS Rattlesnake 1846–1850, 2 vols. London, T. and W. Boone. Macintyre, I. G. 1988. Modern coral reefs of the western Atlantic: new geological perspective. Am. Assoc. Petrol. Geol. Bull. 72, 1360–1369. Macintyre, I. G. and Adey, W. H. 1990. Buck Island Bar, St. Croix, U.S.V.I: a reef that cannot catch up with sea-level. Atoll Res. Bull. 336, 1–7. Macintyre, I. G. and Aronson, R. B. 1997. Field guidebook to the reefs of Belize. Proc. 8th Int. Coral Reef Symp., Panama 1, 203–222. Macintyre, I. G. and Glynn, P. 1976. Evolution of modern Caribbean fringing reef, Galeta Point, Panama. Am. Assoc. Petrol. Geol. Bull. 60, 1054. Macintyre, I. G., Burke, R. B., and Stuckenrath, R. 1977. Thickest recorded Holocene reef section, Isla Pe´rez core hole, Alacran Reef, Mexico. Geology 5, 749–754. Macintyre I. G., Burke, R. B., and Stuckenrath, R. 1982. Core holes in the outer forereef off Carrie Bow Cay, Belize: a key to the Holocene history of the Belizean Barrier Reef Complex. Proc. 4th Int. Coral Reef Symp., Manila 1, 567–574. Macintyre, I. G., Rutzler, K., Norris, J. N., et al. 1991. An early Holocene reef in the western Atlantic: submersible investigations of a deep relict reef off the west coast of Barbados, West Indies. Coral Reefs 10, 167–174. MacNeil, F. S. 1954. The shape of atolls: an inheritance from subaerial erosion forms. Am. J. Sci. 252, 402–427. Mapstone, B. D., Choat, J. L., Cumming, R. L., and Oxley, W. G. 1989. The Fringing Reefs of Magnetic Island: Biota and Sedimentation. Townsville, James Cook University. Maragos, J. F. 1972. A study of the ecology of Hawaiian reef corals. Ph.D. thesis, University of Hawaii, Honolulu. Marine Modelling Unit. 2006. Atlas of Physical Processes in the Great Barrier Reef World Heritage Area. Townsville, School of Engineering, James Cook University. Available online at http://www.mmu.jcu.edu.acc/atlas/ Marshall, J. F. 1977. Marine geology of the Capricorn Channel area. Bureau Min. Resources Geol. Geophys. Austral. Bull. 163. Marshall, J. F. 1983a. The Pleistocene foundations of the Great Barrier Reef. In Baker, J. T., Carter, R. M., Sammarro, P. W., and Stark, K. P. (eds.) Proceedings of the Great Barrier Reef Conference, Townsville, James Cook University, pp. 123–128. Marshall, J. F. 1983b. Lithology and diagenesis of the carbonate foundations of modern reefs in the southern Great Barrier Reef. Bureau Min. Resources J. Austr. Geol. Geophys. 8, 253–265. Marshall, J. F. 1985. Cross-shelf and facies related variations in submarine cementation in the central Great Barrier Reef. Proc. 5th Int. Coral Reef Symp., Tahiti 3, 221–226.
References
499
Marshall, J. F. 1988. Potential effects of oceanic deep waters on the initiation and demise of coral refs. Proc. 6th Int. Coral Reef Symp., Townsville 3, 509–512. Marshall, J. F. and Davies P. J. 1975. High-magnesium calcite ooids from the Great Barrier Reef. J. Sed. Petrol. 45, 285–291. Marshall, J. F. and Davies, P. J. 1978. Skeletal carbonate variation on the continental shelf of eastern Australia. Bureau Min. Resources J. Austr. Geol. Geophys. 3, 85–92. Marshall, J. F. and Davies P. J. 1981. Submarine lithification on windward reef slopes: Capricorn-Bunker Group, southern Great Barrier Reef. J. Sed. Petrol. 51, 953–960. Marshall, J. F. and Davies, P. J. 1982. Internal structure and Holocene evolution of One Tree Reef, southern Great Barrier Reef, Australia. Coral Reefs 1, 21–28. Marshall, J. F. and Davies, P. J. 1984. Last interglacial reef growth beneath modern reefs in the southern Great Barrier Reef. Nature 307, 44–46. Marshall, J. F. and Davies, P. J. 1988. Halimeda bioherms of the northern Great Barrier Reef. Coral Reefs 6, 139–148. Marshall, J. F. and Jacobson, G. 1985. Holocene growth of a mid-Pacific atoll: Tarawa, Kiribati. Coral Reefs 4, 11–17. Marshall, P., Richards, H. C., and Walkom, A. B. 1925. Recent emergence at Holbourne Island, Great Barrier Reef. Repts. GBR. Comm. 1, 29–34. Marshall, S. M. and Orr, A. P. 1931. Sedimentation on Low Isles Reef and its relation to coral growth. Sci. Repts. GBR Exped. 1, 93–133. Marubini, F. and Atkinson, M. J. 1999. Effects of lowered pH and elevated nitrate on coral calcification. Mar. Ecol. Progr. Ser. 188, 117–121. Marubini, F. and Davies, P. S. 1996. Nitrate increases zooxanthellae population density and reduces skeletogenesis in corals. Mar. Biol. 127, 319–329. Marzalak, D. S. 1982. Impact of dredging on a subtropical reef community south-east Florida, USA. Proc. 4th Int. Coral Reef Symp., Manila 1, 147–154. Masse, J. P. and Montaggioni, L. F. 2001. Growth history of shallow-water carbonates: control of accommodation on ecological and depositional processes. Int. J. Earth Sci. 90, 452–469. Massel, S. R. 1992. Wave transformation and dissipation on steep reef slopes. Proc. 11th Australasian Fluid Mechanics Conf., Hobart, 215–232. Massel, S. R. and Brinkman, R. 2001. Wave induced set-up and flow over shoals and coral reefs. I. A simplified bottom geometry case. Oceanologia 43, 373–388. Massel, S. R. and Done, T. J. 1993. Effects of cyclone waves on massive coral assemblages on the Great Barrier Reef: meteorology, hydrodynamics and demography. Coral Reefs 12, 153–166. Massel, S. R. and Gourlay, M. R. 2000. On the modelling of wave breaking and set-up on coral reefs. Coast. Engin. 39, 1–27. Masselink, G. and Hughes, M. G. 2003. Introduction to Coastal Processes and Morphology. London, Arnold. Maxwell, W. G. H. 1968. Atlas of the Great Barrier Reef. Amsterdam, Elsevier. Maxwell, W. G. H. 1969. Radiocarbon ages of sediment: Great Barrier Reef. Sed. Geol. 3, 331–333. Maxwell, W. G. H. 1970. Deltaic patterns in reefs. Deep Sea Res. 17, 1005–1018. Maxwell, W. G. H. 1973a. Geomorphology of eastern Queensland in relation to the Great Barrier Reef. In Jones, O. A. and Endean, R. (eds.) Biology and Geology of Coral Reefs, vol. 1, Geology 1. New York, Academic Press, pp. 233–272.
500
References
Maxwell, W. G. H. 1973b. Sediments of the Great Barrier Reef Province. In Jones, O. A. and Endean, R. (eds.) Biology and Geology of Coral Reefs, vol. 1, Geology 1, New York, Academic Press, pp. 299–345. Maxwell, W. G. H. and Swinchatt, J. P. 1970. Great Barrier Reef: regional variation in a terrigenous–carbonate province. Geol. Soc. America Bull. 81, 691–724. Maxwell, W. G. H., Day, R. W., and Fleming, P. J. G. 1961. Carbonate sedimentation on the Heron Island Reef. J. Sed. Petrol. 31, 215–230. Maxwell, W. G. H., Jell, J. S., and McKellar, R. G. 1964. Differentiation of carbonate sediments on the Heron Island Reef. J. Sed. Petrol. 34, 294–308. McConochie, J. D., Hardy, T. A., and Mason, L. B. 2004. Modelling tropical cyclone over-water wind and pressure fields. Ocean Engin. 31, 1757–1782. McCook, L. J. 1999. Macro-algae, nutrients and phase shifts on coral reefs: scientific issues and management consequences for the Great Barrier Reef. Coral Reefs 18, 357–367. McCook, L. J., Jompa, J., and Diaz-Pulido, G. 2001. Competition between corals and algae on coral reefs: a review of evidence and mechanisms. Coral Reefs 19, 400–417. McCulloch, M. T. and Esat, T. 2000. The coral record of last interglacial sea levels and sea surface temperatures. Chem. Geol. 169, 107–129. McCulloch, M., Pailles, C., Moody, P., and Martin, C. E. 2003a. Tracing the source of sediment and phosphorus into the Great Barrier Reef lagoon. Earth Planet. Sci. Lett. 210, 249–258. McCulloch, M., Fallon, S., Wyndham, T., et al. 2003b. Coral record of increased sediment flux to the inner Great Barrier Reef since European settlement. Nature 421, 727–730. McInnes, K. L., Walsh, K. J. E., and Pittock, A. B. 2000. Impact of Sea Level Rise and Storm Surges on Coastal Resorts: A Report for CSIRO Tourism Research, final report. Aspendale, CSIRO Atmospheric Research. McKenzie, J. A. and Davies, P. J. 1993. Cenozoic evolution of carbonate platforms on the northeastern Australian margin: synthesis of Leg 133 drilling results. In Mckenzie, J. A., Davies, P. J., Palmer-Julson, A. A., and Sarg, J. F. (eds.) Proceedings of the Ocean Drilling Program, vol. 133, Scientific Results, NorthEast Australian Margin. College Station, TX, Texas A and M University, pp. 75–92. McKenzie, J. A., Davies, P. J., Palmer-Julson, A. A., and Sarg, J. F. (eds.) 1993. Proceedings of the Ocean Drilling Program, vol. 133, Scientific Results, North-East Australian Margin. College Station, TX, Texas A and M University. McKergow, L. A., Prosser, I. P., Hughes, A. O., and Brodie, J. 2005. Sources of sediment to the Great Barrier Reef World Heritage Area. Mar. Poll. Bull. 51, 200–211. McLaughlin, C. J., Smith, C. A., Buddemeier, R. W., Bartley, J. D., and Maxwell, B. A. 2003. Rivers, runoff, and reefs. Glob. Planet. Change 39, 191–199. McLean, R. F. 1989. Kiribati and sea level rise, Report to Commonwealth Secretariat Report Group on Climate Change and Sea Level Rise. Sydney, Department of Geography and Oceangraphy, University of New South Wales and Australian Defense Force Academy. McLean, R. F. and Stoddart, D. R. 1978. Reef island sediments of the northern Great Barrier Reef. Phil. Trans. Roy. Soc. Lond. A 291, 101–117.
References
501
McLean, R. F. and Woodroffe, C. D. 1994. Coral atolls. In Carter R. W. G. and Woodroffe C. D. (eds.) Coastal Evolution: Late Quaternary Shoreline Dynamics. Cambridge, Cambridge University Press, pp. 267–302. McLean, R. F., Stoddart, D. R., Hopley, D., and Polach, H. 1978. Sea level change in the Holocene on the northern Great Barrier Reef. Phil. Trans. Roy. Soc. Lond. A 291, 167–186. Mesolella, K. J., Sealy, H. A., and Matthews, R. K. 1970. Facies geometries within Pleistocene reefs on Barbados, West Indies. Am. Assoc. Petrol. Geol. Bull. 54, 1890–1917. Middleton, J. H., Buchwald, V. T., and Huthnance, J. M. 1984. The anomalous tides near Broad Sound. Continent. Shelf Res. 3, 359–381. Milankovitch, M. 1941. Kanon der Erdbestrahlung und seine Anwendung auf da Eiszeitenproblem, Royal Serbian Academy Special Publication 132. Belgrade, Royal Serbian Academy. Miller, D. L. R. and Mackenzie, F. T. 1988. Implications of climate change and associated sea level rise for atolls. Proc. 6th Int. Coral Reef Symp., Townsville 3, 519–522. Milne, G. A., Mitrovica, J. X., and Davis, J. L. 1999. Near-field hydroisostasy: the implementation of a revised sea-level equation. Geophys. J. Int. 139, 464–482. Mitchell, W., Chittleborough, J., Ronai, B., and Lennon, G. W. 2000. Sea level rise in Australia and the Pacific. Sth Pacific Sea Level and Climate Change Newslett. 5, 10–19. Mitrovica, J. X. 2003. Recent controversies in predicting post-glacial sea-level change. Quatern. Sci. Rev. 22, 127–133. Mitrovica, J. X. and Milne, G. A. 2003. On post-glacial sea level. I. General theory. Geophys. J. Int. 154, 253–267. Montaggioni, L. F. 1988. Holocene reef growth history in mid-plate high volcanic islands. Proc. 6th Int. Coral Reef Symp., Townsville 3, 455–460. Montaggioni, L. F. and Venec-Peyre´, M. T. 1993. Shallow-water foraminiferal taphocoenoses at Site 821 : implications for the Pleistocene evolution of the central Great Barrier Reef shelf, north-east Australia. In McKenzie, J. A., Davies, P. J., Palmer-Julson, A. A., and Sarg, J. F. (eds.) Proceedings of the Ocean Drilling Program, vol. 133, Scientific Results, North-East Australian Margin. College Station, TX, Texas A and M University, pp. 365–378. Montaggioni, L. F., Laurenti, A. R., and Faure, G. 1997. Continuous record of reef growth over the past 14 k.y. in the mid Pacific island of Tahiti. Geology 25, 555–558. Moore, J. G. and Fornari, D. J. 1984. Drowned reefs as indicators of the rate of subsidence of the island of Hawaii. J. Geol. 92, 752–759. Moorehouse, F. W. 1933. The recently-formed natural breastwork on Low Isles. Repts. GBR Comm. 4, 35–36. Moorehouse, F. W. 1936. The cyclone of 1934 and its effect on Low Isles with special observations on Porites. Repts. GBR Comm. 4, 37–44. Morelock, J., Boulon, K., and Galler, G. 1979. Sediment stress and coral reefs. In Lopez, J. M. (ed.) Proc. Symp. Energy Industry and Marine Environment in Guayanilla Bay. Mayaguez, Puerto Rico, pp. 46–58. Morner, N. A. 1976. Eustasy and geoid changes. J. Geol. 84, 123–151. ¨ Morner, N. A., Tooley, M., and Possnert, G. 2004. New perspectives for the future of ¨ the Maldives. Glob. Planet. Change 40, 177–182.
502
References
Moss, A., Rayment, G., Reilly, N., and Best, E. 1993. Sediment and Nutrient Exports from Queensland Coastal Catchments: A Desk Study. Brisbane, Department of Environment and Heritage. Moss, P. T. and Kershaw, A. P. 2000. The last glacial cycle from the humid tropics of north-eastern Australia: comparison of a terrestrial and marine record. Palaeogeog. Palaeoclimatol. Palaeoecol. 155, 155–176. Mudelsee, M. 2001. The phase relations among atmospheric CO2 content, temperature and global ice volume over the past 420 ka. Quatern. Sci. Rev. 20, 583–589. Muhs, D. R., Kennedy, G. L., and Rockwell, T. K. 1994. Uranium-series ages of marine terrace corals from the Pacific coast of North America and implications for last-interglacial sea-level history. Quatern. Res. 42, 72–87. Muhs, D. R., Simmons, K. R., and Steinke, B. 2002. Timing and warmth of the Last Interglacial period: new U-series evidence from Hawaii and Bermuda and a new fossil compilation for North America. Quatern. Sci. Rev. 21, 1355–1383. Muir, F. 1986. Islands and reefs surveyed for cyclonic disturbance. In Dutton, I. M. (ed.) The Offshore Effects of Cyclone Winifred. Townsville, Great Barrier Reef Marine Park Authority, pp. 76–78. Muller, R. D., Lim, V. S. L., and Isern, A. R. 2000. Late Tertiary tectonic subsidence ¨ on the northeast Australian passive margin: response to dynamic topography? Mar. Geol. 162, 337–352. Mumby, P. J., Green, E. P., Clark, C. D., and Edwards, A. J. 1998. Digital analysis of multispectral airborne imagery of coral reefs. Coral Reefs 17, 59–69. Mumby, P. J., Green, E. P., Edwards, A. J., and Clark, C. D. 1999. The costeffectiveness of remote sensing for tropical coastal resources assessment and management. J. Environ. Mgmt 55, 157–166. Mumby, P. J., Hedley, J. D., Chisholm, J. R. M., et al. 2004a. The cover of living and dead corals from airborne remote sensing. Coral Reefs 23, 171–183. Mumby, P. J., Skirving, W., Strong, A. E., et al. 2004b. Remote sensing of coral reefs and their physical environment. Mar. Poll. Bull. 48, 219–228. Munk, W. H. and Sargent, M. S. 1954. Adjustment of Bikini Atoll to ocean waves. U.S. Geol. Surv. Prof. Paper 260C, 275–280. Munk, W. H., Ewing, G. C., and Revelle, R. R. 1949. Diffusion in Bikini Lagoon. Trans. Am. Geophys. Union 30, 59–66. Murray, J. W. 1991. Ecology and Palaeoecology of Benthic Foraminifera. Harlow, Longman. Murray-Wallace, C. V., Brooke, B. P., Cann, J. H., Belperio, A. P., and Bourman, R. P. 2001. Whole-rock aminostratigraphy of the Coorong Coastal Plain, South Australia: towards a 1 million year record of sea-level highstands. J. Geol. Soc. 158, 111–124. Mutter, J. C. and Karner, G. D. 1980. The continental margin off northeast Australia. In Henderson, R. A. and Stephenson, P. J. (eds.) The Geology and Geophysics of Northeastern Australia. Townsville, Geological Society of Australia, Queensland Division, pp. 47–69. Nakada, M. and Lambeck, K. 1988. The melting history of the late Pleistocene Antarctic ice sheet. Nature 333, 36–40. Nakada, M. and Lambeck, K. 1989. Late Pleistocene and Holocene sea level change in the Australian region and mantle rheology. Geophys. J. 96, 497–517. Nakamura, T. and van Woesik, R. 2001. Water-flow rates and passive diffusion partially explain differential survival of corals during the 1998 bleaching event. Mar. Ecol. Progr. Ser. 212, 301–304.
References
503
Nakiboglu, S. M., Lambeck, K., and Aharon, P. 1983. Postglacial sea-levels in the Pacific: implications with respect to deglaciation regime and local tectonics. Tectonophysics 91, 335–358. Neef, G., Zhao, J. X., Collerson, K. D., and Zhang, F. S. 2003. Late Quaternary uplift and subsidence of the west coast of Tanna, south Vanuatu, southwest Pacific: U–Th ages of raised coral reefs in the Median Sedimentary Basin. Austral. J. Earth Sci. 50, 39–48. Neil, D. T. 2000. Characteristics and significance of a sub-tropical ‘‘Low Wooded Island’’: Green Island, Moreton Bay, Australia. J. Coast. Res. 16, 287–294. Neil, D. T. and Yu, B. 1996. Fluvial sediment yield to the Great Barrier Reef lagoon: spatial patterns and the effects of land use. In Hunter, H., Eyles, A., and Rayment, G. (eds.) Downstream Effects of Land Use. Brisbane Queensland Department of Natural Resources, pp. 281–286. Neil, D. T., Orpin, A. R., Ridd, E. V., and Yu, B. F. 2002. Sediment yield and impacts from river catchments to the Great Barrier Reef lagoon. Mar. Freshw. Res. 53, 733–752. Nelson, A. W. 1980. Preliminary results of a shallow marine seismic survey: Thursday Island to Townsville. Geol. Surv. Queensland. Rec. 6. Nelson, R. C. 1994. Depth limited design wave heights in very flat regions. Coast. Engin. 23, 43–59. Nelson, R. C. and Lesleighter, E. J. 1985. Breaker height attenuation over platform coral reefs. Proc. Australasian Conf. Coastal and Ocean Engineering, Christchurch, 9–16. Neumann, A. C. 1972. Quaternary sea level history of Bermuda and the Bahamas. Proc. American Quaternary Association 2nd Nat. Conf., Abstr. 41–44. Neumann, A. C. and Macintyre, I. G. 1985. Reef response to sea level rise: keep up catch up or give up. Proc. 5th Int. Coral Reef Congr., Tahiti 3, 105–110. Nicholls, N. 1992. Recent performance of a method for forecasting Australian seasonal tropical cyclone activity. Austral. Meteorol. Mag. 40, 105–110. Nott, J. 1997. Extremely high energy wave deposits inside the Great Barrier Reef, Australia: determining the cause – tsunami or tropical cyclone. Mar. Geol. 141, 193–207. Nott, J. 2000. Records of prehistoric tsunamis from boulder deposits: evidence from Australia. Sci. Tsunami Hazards 18, 3–14. Nott, J. 2003a. The importance of prehistoric data and variability of hazard regimes in natural hazard risk assessment: examples from Australia. Nat. Hazards 30, 43–58. Nott, J. 2003b. Tsunami or storm waves? – Determining the origin of a spectacular field of wave emplaced boulders using numerical storm surge and wave models and hydrodynamic transport equations. J. Coast. Res. 19, 348–356. Nott, J. F. 2003c. Intensity of prehistoric tropical cyclones. J. Geophys. Res. Atmos. 108(D7), 4212–4223. Nott, J. 2004a. Palaeotempestology: the study of and implications of prehistoric tropical cyclones – a review for hazard assessment. Environ. Int. 30, 433–447. Nott, J. 2004b. Washed away: people and buildings during tropical cyclones – are Queensland state and local government policies doing enough? Environ. Plan. Law 21, 227–238. Nott, J. 2006. Tropical cyclones and the evolution of the sedimentary coasts of tropical Australia. J. Coast. Res. 22, 49–62.
504
References
Nott, J. and Hayne, M. 2000. How high was the storm surge from Tropical Cyclone Mahina, North Queensland, 1899? Austral. J. Emerg. Mgmt, Autumn 2000, 11–13. Nott, J. and Hayne, M. 2001. High frequency of ‘‘super-cyclones’’ along the Great Barrier Reef over the past 5000 years. Nature 413, 508–512. Nott, J. F., Thomas, M. F., and Price, D. M. 2001. Alluvial fans, landslides and late Quaternary climatic change in the wet tropics of northeast Queensland. Austral. J. Earth Sci. 48, 875–882. Nunn, P. D. 1993. Role of porolithon algal-ridge growth in the development of the windward coast of Tongatapu Island, Tonga, South Pacific. Earth Surface Proc. Landforms 18, 427–439. Nunn, P. D. 1994. Oceanic Islands. Oxford, Blackwell. Nunn, P. D. 1998. Sea-level changes over the past 1000 years in the Pacific. J. Coast. Res. 14, 23–30. Oberdorfer, J. A. and Buddemeier, R. W. 1986. Coral reef hydrology: field studies of water movement within a barrier reef. Coral Reefs 5, 7–12. Oberdorfer, J. A., Hogan, P. J., and Buddemeier, R. W. 1990. Atoll island hydrogeology: flow and fresh-water occurrence in a tidally dominated system. J. Hydrol. 120, 327–340. Ogston, A. S., Storlazzi, C. D., Field, M. E., and Presto, M. K. 2004. Sediment re-suspension and transport patterns on a fringing reef flat, Molokai, Hawaii. Coral Reefs 23, 559–569. Ohde, S., Greaves, M., Masuzawa, M., et al. 2002. The chronology of Funafuti atoll: revisiting an old friend. Proc. Roy. Soc. Lond. A 458, 2289–2306. O’Keeffe, M. J. 1991. Over and under: geography and archaeology of the Palm Islands and adjacent continental shelf of North Queensland. M.A. thesis, James Cook University, Townsville. Oliver, J. K., King, B. A., Willis, B. L., Babcock, R. C., and Wolanski, E. 1992. Dispersal of coral larvae from a lagoonal reef. II. Comparisons between model predictions and observed concentrations. Continent. Shelf Res. 12, 873–889. Oliver, J. K. and Willis, B. L. 1987. Coral-spawn slicks in the Great Barrier Reef: preliminary observations. Mar. Biol. 94, 521–529. Orme, G. R. 1985. The sedimentological importance of Halimeda in the development of back reef lithofacies, northern Great Barrier Reef, Australia. Proc. 5th Int. Coral Reef Congr., Tahiti 5, 31–37. Orme, G. R. and Flood, P. G. 1980. Sedimentation in the Great Barrier Reef Province, adjacent bays and estuaries. In Henderson, R. A. and Stephenson, P. J. (eds.) The Geology and Geophysics of Northeastern Australia. Townsville, Geological Society of Australia, Queensland Division, pp. 419–434. Orme, G. R. and Salama, M. S. 1988. Form and seismic stratigraphy of Halimeda banks in part of the northern Great Barrier Reef Province. Coral Reefs 6, 131–137. Orme, G. R., Flood, P. G., and Ewart, A. E. 1974. An investigation of the sediments and physiography of Lady Musgrave Reef: a preliminary account. Proc. 2nd Int. Coral Reef Symp., Queensland 2, 371–386. Orme, G. R., Webb, J. P., Kelland, N. C., and Sargent, G. E. G. 1978a. Aspects of the geological history and structure of the northern Great Barrier Reef. Phil. Trans. Roy. Soc. Lond. A 291, 23–35. Orme, G. R., Flood, P. G., and Sargent, G. E. G. 1978b. Sedimentation trends in the lee of outer (ribbon) reefs, Northern Region of the Great Barrier Reef. Phil. Trans. Roy. Soc. Lond. A 291, 85–89.
References
505
Orpin, A. R. and Ridd, P. V. 1996. Sediment distribution and transport mechanisms, Burdekin region, central Great Barrier Reef. In Larcombe, P., Woolfe, K. J., and Purdon, R. G. (eds.) Great Barrier Reef: Terrigenous Sediment Flux and Human Impacts. Townsville, CRC Reef Research Centre, pp. 128–143. Orpin, A. R., Ridd, P. V., and Stewart, L. K. 1999. Assessment of the relative importance of major sediment-transport mechanisms in the central Great Barrier Reef Lagoon. Austral. J. Earth Sci. 46, 883–896. Orpin, A. R., Brunskill, G. J., Zagorskis, I., and Woolfe, K. J. 2004. Patterns of mixed siliciclastic-carbonate sedimentation adjacent to a large dry-tropic river on the central Great Barrier Reef shelf, Australia. Austral. J. Earth Sci. 51, 665–683. Ott, B. 1975. Community patterns on a submerged barrier reef at Barbados, West Indies. Int. Rev. Ges. Hydrobiol. 60, 719–736. Page, M. C. and Dickens, G. R. 2005. Sediment fluxes to Marion Plateau (southern Great Barrier Reef province) over the last 130ka: new constraints on ‘‘transgressive-shedding’’ off northeastern Australia. Mar. Geol. 219, 27–45. Page, M. C., Dickens, G. R., and Dunbar, G. B. 2003. Tropical view of Quaternary sequence stratigraphy: siliciclastic accumulation on slopes east of the Great Barrier Reef since the Last Glacial Maximum. Geology 31, 1013–1016. Paradice, W. E. J. 1925. The pinnacle or mushroom shaped coral growths in connection with the reefs of the outer barrier. Repts. GBR Comm. 1, 52–59. Parnell, K. E. 1987. The hydrodynamics of fringing reef bays in the Great Barrier Reef Marine Park, with emphasis on management. Ph.D. thesis, James Cook University, Townsville. Parnell, K. E. 1988a. The hydrodynamics of fringing reef bays in the Great Barrier Reef Marine Park. Proc. 6th Int. Coral Reef Symp., Townsville 2, 503–508. Parnell, K. E. 1988b. Physical process studies in the Great Barrier Reef Marine Park. Progr. Phys. Geog. 12, 209–237. Parnell, K. E. 1989. Reefs in the Greenhouse: a review. In Welch, R. (ed.) Geography in Action, Proc. 15th New Zealand Geog. Conf., pp. 292–298. Partain, B. R. 1988. Morphology and development of the Cape Tribulation fringing reefs, Great Barrier Reef, Australia. M.Sc. thesis, University of Texas, Austin. Partain, B. R. and Hopley, D. 1989. Morphology and Development of the Cape Tribulation Fringing Reefs, Great Barrier Reef, Australia, Technical Memoir 21. Townsville, Great Barrier Reef Marine Park Authority. Pastorak, R. A. and Bilyard, G. R. 1985. Effects of sewage pollution on coral reef communities. Mar. Ecol. Progr. Ser. 21, 175–189. Peerdeman, F. M. and Davies, P. J. 1993. Sedimentological response of an outer-shelf, upper-slope sequence to rapid changes in Pleistocene eustatic sea level: Hole 820 A, northeastern Australian margin. In McKenzie, J. A., Davies, P. J., PalmerJulson, A. A., and Sarg, J. F. (eds.) Proceedings of the Ocean Drilling Program, vol. 133, Scientific Results, North-East Australian Margin. College Station, TX, Texas A and M University, pp. 303–313. Peltier, W. R. 2002. On eustatic sea level history: Last Glacial Maximum to Holocene. Quatern. Sci. Rev. 21, 377–396. Penck, A. 1896. Das grosse australische Wallriffe. Vortra¨ge Verein z. Verbr. Naturw. Kenttnisse in Wien 36(13), 1–23. Perry, C. T. 2003a. Reef development at Inhaca Island, Mozambique: coral communities and impacts of the 1999/2000 Southern African floods. Ambio 32, 134–139.
506
References
Perry, C. T. 2003b. Coral reefs in a high-latitude, siliciclastic barrier island setting: reef framework and sediment production at Inhaca Island, southern Mozambique. Coral Reefs 22, 485–497. Perry, C. T. and Larcombe, P. 2003. Marginal and non-reef-building coral environments. Coral Reefs 22, 427–432. Perry, C. T., and S. G. Smithers. 2006. Taphonomic signatures of turbid-zone reef development: examples from Paluma Shoals and Lugger Shoal, inshore central Great Barrier Reef, Australia. Palaeogeogr. Palaeoclimatol. Palaeoecol. Petit, J. R., Jouzel, J., Raynaud, D., et al. 1999. Climate and atmospheric history of the past 420 000 years from the Vostok ice core, Antarctica. Nature 399, 429–436. Phinn, S. R., Dekker, A. G., Brando, V. E., and Roelfsema, C. M. 2005. Mapping water quality and substrate cover in optically complex coastal and reef waters: an integrated approach. Mar. Poll. Bull. 51, 459–469. Phipps, C. V. G. and Roberts, H. H. 1988. Seismic characteristics and accretion history of Halimeda bioherms on Kalukalukuang Bank, eastern Java Sea (Indonesia). Coral Reefs 6, 149–159. Phipps, C. V. G., Davies, P. J., and Hopley, D. 1985. The morphology of Halimeda banks behind the Great Barrier Reef east of Cooktown, Queensland. Proc. 5th Int. Coral Reef Congr., Tahiti 5, 27–30. Pickard, G. L. 1986. Effects of wind and tide on upper-layer currents at Davies Reef, Great Barrier Reef, during Mecor (July–August 1984). Austral. J. Mar. Freshw. Res. 37, 545–565. Pickard, G. L., Donguy, J. R., Henin, C., and Rougerie, F. A. 1977. A Review of the Physical Oceanography of the Great Barrier Reef and Western Coral Sea. Townsville, Australian Institute of Marine Science. Pillans, B., Chappell, J., and Naish, T. R. 1998. A review of the Milankovitch climatic beat: template for Plio-Pleistocene sea-level changes and sequence stratigraphy. Sed. Geol. 122, 5–21. Pirazzoli, P. A. 1991. World Atlas of Holocene Sea-Level Changes. Amsterdam, Elsevier. Pirazzoli, P. A. and Montaggioni, L. F. 1988. The 7000yr sea-level curve in French Polynesia: implications for mid-plate volcanic islands. Proc. 6th Int. Coral Reef Symp., Townsville 3, 467–472. Playford, P. E. 1980. Devonian ‘‘Great Barrier Reef’’ of Canning Basin, Western Australia. Am. Assoc. Petrol. Geol. Bull. 64, 814–840. Polach, H. A., McLean, L. F., Caldwell, J. R., and Thom, B. G. 1978. Radiocarbon ages from the northern Great Barrier Reef. Phil. Trans. Roy. Soc. Lond. A 291, 139–158. Potter, E. K., Esat, T. A., Schellmann, G., et al. 2004. Suborbital-period sea-level oscillations during marine isotope substages 5a and 5c. Earth Planet. Sci. Lett. 225, 191–204. Presto, M. K., Ogston, A. S., Storlazzi, C. D., and Field, M. E. 2006. Temporal and spatial variability in the flow and dispersal of suspended-sediment on a fringing reef flat, Molokai, Hawaii. Estuar. Coast. Shelf Sci. 67, 67–81. Pringle, A. W. 1991. Fluvial sediment supply to the north-east Queensland coast, Australia. Austral. Geog. Stud. 29, 114–138. Pringle, A. W. 2000. Evolution of the east Burdekin delta coast, Queensland, Australia 1980–1995. Z. Geomorphol. 44, 273–304. Productivity Commission. 2002. Industries in the Great Barrier Reef Catchment and Measures to Address Declining Water Quality. Brisbane, Productivity Commission.
References
507
Prosser, I. P., Rustomji, P., Young, W. J., Moran, C. J., and Hughes, A. O. 2001. Constructing River Basin Sediment Budgets for the National Land and Water Resources Audit. Canberra, CSIRO. Puotinen, M. L. 2004. Tropical cyclones in the Great Barrier Reef, Australia, 1910–1999: a first step towards characterizing the disturbance regime. Austral. Geog. Stud. 42, 378–392. Puotinen, M. L., Done, T. J., and Skelly, W. C. 1997. An Atlas of Tropical Cyclones in the Great Barrier Reef Region, 1969–1997. Townsville, CRC Reef Research Centre. Purdy, E. G. 1974. Reef configurations cause and effect. In Laporte, L. F. (ed.) Reefs in Time and Space: Selected Examples from the Recent and Ancient, Society of Economic Paleontologists and Mineralogists Special Publication 18. Tulsa, OK, S.E.P.M., pp. 9–76. Purdy, E. G. 1998. Structural termination of the southern end of the Belize Barrier Reef. Coral Reefs 17, 231–234. Purdy, E. G. and Gischler, E. 2005. The transient nature of the empty bucket model of reef sedimentation. Sed. Geol. 175, 35–47. Purdy, E. G., Gischler, E., and Lomando, A. J. 2003. The Belize margin revisited. II. Origin of Holocene antecedent topography. Int. J. Earth Sci. 92, 552–572. Pye, K. 1982a. Morphology and sediments of the Ramsay Bay sand dunes, Hinchinbrook Island, North Queensland. Proc. Roy. Soc. Queensland 93, 31–47. Pye, K. 1982b. Morphological development of coastal dunes in a humid tropical environment, Cape Bedford and Cape Flattery, North Queensland. Geografiska Annaler 64A, 212–227. Pye, K. and Bowman, G. M. 1984. The Holocene marine transgression as a forcing function in episodic dune activity on the eastern Australian coast. In Thom, B. G. (ed.) Coastal Geomorphology in Australia. Sydney, Academic Press, pp. 179–196. Rainford, E. H. 1925. Destruction of the Whitsunday Group fringing reefs. Austral. Mus. Mag. 2, 175–177. Randall, R. H. and Birkeland, C. 1978. Guam’s reefs and beaches. II. Sedimentation studies at Fouha Bay and Ylig Bay. Univ. Guam Mar. Lab. Tech. Rept. 47. Rasmussen, C. E. 1986. An investigation of morphological changes, Low Isles, Northern Great Barrier Reef. Hons. thesis, James Cook University, Townsville. Rasmussen, C. E. 1988. The use of strontium as an indicator of anthropogenically altered environmental parameters. Proc. 6th Int. Coral Reef Symp., Townsville 2, 325–330. Rasmussen, C. E. 1991. Anthropogenic disturbance of environmental signatures retained in massive corals. Ph.D. thesis, James Cook University, Townsville. Rasmussen, C. E. and Hopley, D. 1995. Warraber Island Beach Erosion Review, Unpubl. consultancy report. Cairns, Edmiston and Taylor. Rasmussen, C. E., Cuff, C., and Hopley, D. 1993a. Evidence of anthropogenic disturbance retained in the skeleton of massive corals from Australia’s Great Barrier Reef. Proc. 7th Int. Coral Reef Symp., Guam 1, 201–210. Rasmussen, C. E., Ness, S. E., and Cuff, C. 1993b. A correlation between the El Nino ˜ oscillation in corals of the Great Barrier Reef, Australia. Proc. 7th Int. Coral Reef Symp., Guam 1, 211–220. Rattray, A. 1869. Notes on the geology of the Cape York Peninsula. Geol. Soc. Lond. Quart. J. 25, 297–305. Rees, S. A., Opdyke, B. N., Wilson, P. A., and Fifield, L. K. 2005. Coral reef sedimentation on Rodrigues and the western Indian Ocean and its impact on the carbon cycle. Phil. Trans. Roy. Soc. Lond. A 363, 101–120.
508
References
Reid, R. P., Macintyre, I. G., Browne, K. M., Steneck, R. S., and Miller, T. 1995. Modern marine stromatolites in the Exuma Cays, Bahamas: uncommonly common. Facies 33, 1–18. Reid, R. P., Macintyre, I. G., and Steneck, R. S. 1999. A microbialite/algal fringing reef complex, Highborne Cay, Bahamas. Atoll Res. Bull. 465, 1–18. Revell, C. G. and Goulter, S. W. 1986. South Pacific tropical cyclones and the southern oscillation. Monthly Weather Rev. 113, 1138–1145. Richards, H. C. 1922. Problems of the Great Barrier Reef. Queensland Geog. J. 36–37, 42–45. Richards, H. C. 1923. The Great Barrier Reef of Australia. Proc. 2nd Pan-Pacif. Sci. Congr., Australia 2, 1104–1119. Richards, H. C. and Hill, D. 1942. Great Barrier Reef bores, 1926 and 1937: Descriptions, analyses, and interpretations. Repts. GBR Comm. 5, 122. Risk, M. J. 1992. Musings on monitoring. Reef Encounter 12, 7–9. Roberts, H. H. 1981. Physical processes and sediment flux through reef–lagoon systems. Proc. 17th Coastal Engin. Conf., Sydney, 946–962. Roberts, H. H. and Suhayda, J. N. 1983. Wave–current interactions on a shallow reef (Nicaragua, Central America, Great Corn Island). Coral Reefs 1, 209–214. Roberts, H. H., Murray, S. P., and Suhayda, J. N. 1975. Physical processes in fringing reef systems. J. Mar. Res. 33, 233–260. Roberts, H. H., Aharon, P., and Phipps, C. V. 1988. Morphology and sedimentology of Halimeda bioherms from the eastern Java Sea (Indonesia). Coral Reefs 6, 161–172. Rogers, C. S. 1990. Responses of coral reefs and reef organisms to sedimentation. Mar. Ecol. Progr. Ser. 62, 185–202. Rohling, E. J., Marsh, R., Wells, N. C., Siddall, M., and Edwards, N. R. 2004. Similar meltwater contributions to glacial sea level changes from Antarctic and northern ice sheets. Nature 430, 1016–1021. Rooney, J., Fletcher, C., Grossman, E., Engels, M., and Field, M. 2004. El Nino ˜ influence on Holocene reef accretion in Hawaii. Pacif. Sci. 58, 305–324. Rosen, B. R. 1978. Appendix: Determination of a collection of coral microatoll specimens from the northern Great Barrier Reef. Phil. Trans. Roy. Soc. Lond. B 284, 115–122. Rubin, K. H., Fletcher, C. H., and Sherman, C. 2001. Fossiliferous Lana’i deposits formed by mutiple events rather than a single giant tsunami. Nature 408, 675–681. Russell, R. 1962. Origin of beachrock. Z. Geomorphol. 6, 1–6. Salvat, B. 2002. Agenda 21, International Coral Reef Initiative and the new millenium: progress and prospects for coral reefs. Proc. 9th Int. Coral Reef Symp., Bali 1, 49–55. Sammarco, P. W. and Andrews, J. C. 1988. Localized dispersal and recruitment in Great Barrier Reef corals: the Helix Experiment. Science 239, 1422–1424. Sammarco, P. W. and Andrews, J. C. 1989. The Helix Experiment: differential localized dispersal and recruitment patterns in Great Barrier Reef corals. Limnol. Oceanogr. 34, 896–912. Sammarco, P. W., Andrews, J. C., and Risk, M. J. 1991. Coral-reef geomorphology as a function of seasonal prevailing currents and larval dispersal. Palaeogeog. Palaeoclimatol. Palaeoecol. 88, 1–12. Sammarco, P. W., Risk, M. J., Schwarcz, H. P., and Heikoop, J. M. 1999. Crosscontinental shelf trends in coral delta N-15 on the Great Barrier Reef: further consideration of the reef nutrient paradox. Mar. Ecol Progr. Ser. 180, 131–138.
References
509
Saville-Kent, W. 1893. The Great Barrier Reef of Australia. London, Allen. Sawai, Y., Horton, B. P., and Nagumo. T. 2004. The development of a diatom-based transfer function along the Pacific coast of eastern Hokkaido, northern Japan: an aid in paleoseismic studies of the Kuril subduction zone. Quatern. Sci. Rev. 23, 2467–2483. Scandol, J. P. and James, M. K. 1992. Hydrodynamics and larval dispersal: a population model of Acanthasterplanci on the Great Barrier Reef. Austral. J. Mar. Freshw. Res. 43, 583–596. Schlager, W. 1981. The paradox of drowned reefs and carbonate platforms. Geol. Soc. America Bull. 92, 197–211. Schlanger, S. O. 1963. Subsurface geology of Eniwetok Atoll. U.S. Geol. Surv. Prof. Pap. 260B, 991–1066. Schumm, S. A. 1993. River response to base level change: implications for sequence stratigraphy. J. Geol. 101, 279–294. Scoffin, T. P. 1970. A conglomeratic beachrock in Bimini, Bahamas. J. Sed. Petrol. 40, 756–759. Scoffin, T. P. 1977. Sea-level features on reefs in the northern province of the Great Barrier Reef. Proc. 3rd Int. Coral Reef Symp., Miami 2, 319–324. Scoffin, T. P. 1993. The geological effects of hurricanes on coral reefs and the interpretation of storm deposits. Coral Reefs 12, 203–221. Scoffin, T. P. and Stoddart, D. R. 1978. The nature and significance of microatolls. Phil. Trans. Roy. Soc. Lond. B 284, 99–122. Scoffin, T. P. and Stoddart, D. R. 1983. Beachrock and intertidal cements. In Goudie, A. and Pye, K. (eds.) Chemical Sediments and Geomorphology. London, Academic Press, pp. 401–425. Scoffin, T. P. and Tudhope, A. W. 1985. Sedimentary environments of the Central region of the Great Barrier Reef of Australia. Coral Reefs 4, 81–93. Scoffin, T. P., Stoddart, D. R., McLean, R. F., and Flood, P. G. 1978. The recent development of reefs in the northern province of the Great Barrier Reef. Phil. Trans. Roy. Soc. Lond. B 284, 129–139. Scott, A. J. and Rotondo, G. M. 1983. A model to explain the differences between Pacific plate island-atoll types. Coral Reefs 1, 139–149. Scott, D. B. and Medioli, F. S. 1978. Vertical zonations of marsh foraminifera as accurate indicators of former sea-levels. Nature 272, 528–531. Scott, D. B., Collins, E. S., Duggan, J., et al. 1996. Pacific rim marsh foraminiferal distributions: implications for sea-level studies. J. Coast. Res. 12, 850–861. Searle, D. E. 1983. Late Quaternary regional controls on the development of the Great Barrier Reef: geophysical evidence. Bureau Min. Resources J. Aust. Geol. Geophys. 8, 267–276. Searle, D. E. 1994. Late Quaternary morphology of the Cocos (Keeling) Islands. Atoll Res. Bull. 401, 1–13. Searle, D. E. and Flood, P. G. 1988. Halimeda bioherms of the Swain Reefs, Southern Great Barrier Reef. Proc. 6th Int. Coral Reef Symp., Townsville 3, 139–194. Searle, D. E. and Harvey, N. 1982. Interpretation of inter-reefal seismic data: a case study from Michaelmas Reef, Australia. Mar. Geol. 46, M9–M16. Searle, D. E. and Hegarty, R. A. 1982. Results of a continuous seismic profiling survey in the Princess Charlotte Bay area. Geol. Surv. Queensland Rec. 17. Searle, D. E., Harvey, N., and Hopley, D. 1980. Preliminary results of continuous seismic profiling on the Great Barrier Reef province between 168 100 S and 198 200 S. Geol. Surv. Queensland Rec. 23.
510
References
Searle, D. E., Harvey N., Hopley, D., and Johnson, D. P. 1982. Significance of results of shallow seismic research in the Great Barrier Reef province between 168 100 S and 208 050 S. Proc. 4th Int. Coral Reef Symp., Manila 1, 531–540. Shackleton, N. J. and Opdyke, N. D. 1973. Oxygen isotope and palaeo-magnetic stratigraphy of Equatorial Pacific core V28–238: oxygen isotope temperatures and ice volumes in a 105 year and 106 year scale. Quatern. Res. 3, 39–55. Shackleton, N. J., Berger, A., and Peltier, W. R. 1990. An alternative astronomical calibration of the Lower Pleistocene timescale based on ODP Site 677. Trans. Roy. Soc. Edinb., Earth Sci. 81, 251–261. Shackleton, N. J., Hall, M. A., and Vincent, E. 2000. Phase relationships between millennial-scale events 64 000–24 000 years ago. Paleoceanography 15, 565–569. Shennan, I. 1999. Global meltwater discharge and the deglacial sea-level record from northwest Scotland. J. Quatern. Sci. 14, 715–719. Sheppard, C., Dixon, D. J., Gourlay, M. J., Sheppard, A. and Payet, R. 2005. Coral mortality increases wave energy reaching shores protected by reef flats: examples from the Seychelles. Estuarine Coast. and Shelf, Sci. 64, 223–234. Shinn, E. A., Hudson, J. H., Halley, R. B., et al. 1982. Geology and sediment accumulation rates at Carrie Bow Cay, Belize. In Rutzler, K. and Macintyre, I. G. (eds.) The Atlantic Barrier Reef System at Carrie Bol Cay, Belize, vol. 1, Structure and Communities. Washington, DC, Smithsonian Institution Press, pp. 63–75. Siddall, M., Rohling, E. J., Almogi-Labin, A., et al. 2003. Sea-level fluctuations during the last glacial cycle. Nature 423, 853–858. Siddall, M., Smeed, D. A., Hemleben, C., et al. 2004. Understanding the Red Sea response to sea level. Earth Planet. Sci. Lett. 225, 421–434. Sloss, L. L. 1991. The tectonic factor in sea level change: a countervailing view. J. Geophys. Res. 96, 6609–6617. Smith, A. S. 1978. Magnetic Island and its fringing reefs. In Hopley, D. (ed.) Geographical Studies of the Townsville Area, Monograph Ser. Department of Geography, James Cook University, Townsville, pp. 59–64. Smith, S. V. and Kinsey, D. W. 1976. Calcium carbonate production, coral reef growth and sea level change. Science 194, 937–939. Smith, S. V., Kimmerer, W. J., Laws, E. A., Brock, R. E., and Walsh, T. W. 1981. Kaneohe Bay sewage diversion experiment: perspectives on ecosystem responses to nutritional perturbation. Pacif. Sci. 35, 279–395. Smithers, S. G. 1994. Sediment facies of the Cocos (Keeling) Islands lagoon. Atoll Res. Bull. 407, 1–15. Smithers, S. and Larcombe, P. 2003. Late Holocene initiation and growth of a nearshore turbid-zone coral reef: Paluma Shoals, central Great Barrier Reef, Australia. Coral Reefs 22, 499–505. Smithers, S. G. and Woodroffe, C. D. 2000. Micro-atolls as sea-level indicators on a mid-ocean atoll. Mar. Geol. 168, 61–78. Smithers, S. G. and Woodroffe, C. D. 2001. Coral microatolls and 20th century sea level in the eastern Indian Ocean. Earth Planet. Sci. Lett. 191, 173–184. Smithers, S. G., Woodroffe, C. D., McLean, R. F., and Wallensky, E. 1993. Lagoonal sedimentation in the Cocos (Keeling) Islands, Indian Ocean. Proc. 7th Int. Coral Reef Symp., Guam 1, 273–288. Smithers, S. G., Hopley, D., and Parnell, K. E. 2006. Fringing and nearshore coral reefs of the Great Barrier Reef: episodic Holocene development and future prospects. J. Coast. Res. 22, 175–187.
References
511
Spenceley, A. P. 1980. The geomorphological and zonational development of mangrove swamps in the Townsville area, North Queensland. Ph.D. thesis, James Cook University, Townsville. Spenceley, A. P. 1982. The Geomorphological and Zonational Development of Mangrove Swamps in the Townsville Area, North Queensland, Monograph Series, 11. Townsville, Department of Geography, James Cook University of North Queensland. Spender, M. A. 1930. Island reefs of the Queensland coast. Geog. J. 76, 194–214, 273–297. Stafford-Smith, M. G. 1993. Mortality of the hard coral Leptoria phrygia under persistent sediment influx. Proc. 7th Int. Coral Reef Symp., Guam 1, 289–299. Stanley, G. A. V. 1928. The physiography of the Bowen district and of the northern isles of the Cumberland Group. Repts. GBR Comm. 2, 1–51. Steers, J. A. 1929. The Queensland coast and the Great Barrier Reef. Geog. J. 74, 232–257, 341–370. Steers, J. A. 1937. The coral islands and associated features of the Great Barrier Reefs. Geog. J. 89, 1–28, 119–146. Steers, J. A. 1938. Detailed notes on the islands surveyed and examined by the Geographical Expedition to the Great Barrier Reef in 1936. Repts. GBR Comm. 4, 51–96. Steers, J. A. 1945. Coral reefs and air photography. Geog. J. 106, 232–235. Steers, J. and Stoddart, D. 1977. The origin of fringing reefs, barrier reefs and atolls. In Jones, O. and Endean, R. (eds.) Biology and Geology of Coral Reefs, vol. 4, Geology 2. New York, Academic Press, pp. 21–57. Stein, M., Wasserburg, G. J., Aharon, P., et al. 1993. Tims U-series dating and stable isotopes of the last interglacial event in Papua New Guinea. Geochim. Cosmochim. Acta 57, 2541–2554. Steneck, R. S., Miller, T. E., Reid, R. P., and Macintyre, I. G. 1998. Ecological controls on stromatolite development in a modern reef environment: a test of the ecological refuge paradigm. Carbonates and Evaporites 13, 48–65. Stephenson, P. J., Griffin, T. J., and Sutherland, F. L. 1980. Cainozoic volcanism in northeastern Australia. In Henderson, R. A. and Stephenson, P. J. (eds.) The Geology and Geophysics of Northeastern Australia. Townsville, Geological Society of Australia, Queensland Division, pp. 349–374. Stephenson, P. J., Burch-Johnston, A. T., Stanton, D., and Whitehead, P. W. 1998. Three long lava flows in North Queensland. J. Geophys. Res. 103, 27, 359–27, 370. Stieglitz, T. 2005. Submarine groundwater discharge into the nearshore zone of the Great Barrier Reef, Australia. Mar. Poll. Bull. 51, 51–59. Stieglitz, T. C. and Ridd, P. V. 2000. Submarine groundwater discharge from paleochannels? Wonky Holes on the inner shelf of the Great Barrier Reef, Proc. HYDRO 2006 (26th Nat. and 3rd Int. Symp. Instit. of Engin.), Perth, 1265–1271. Stirling, C. H., Esat, T. M., McCulloch, M. T., and Lambeck, K. 1995. High-precision U-series dating of corals from Western Australia and implications for the timing and duration of the Last Interglacial. Earth Planet. Sci. Lett. 135, 115–130. Stirling, C. H., Esat, T. M., Lambeck, K., and McCulloch, M. T. 1998. Timing and duration of the last interglacial: evidence for a restricted interval of widespread coral reef growth. Earth Planet. Sci. Lett. 160, 745–762. Stoddart, D. R. 1965. British Honduras cays and the low wooded island problem. Trans. Instit. Brit. Geog. 36, 131–147. Stoddart, D. R. 1969. Ecology and morphology of recent coral reefs. Biol. Rev. 44, 433–498.
512
References
Stoddart, D. R. 1978. The Great Barrier Reef and the Great Barrier Reef Expedition 1973. Phil. Trans. Roy. Soc. Lond. A 291, 5–22. Stoddart, D. R. and Fosberg, F. R. 1991. Phytogeography and vegetation of the reef islands of the Northern Great Barrier Reef. Atoll Res. Bull. 349, 1–19. Stoddart, D. R. and Scoffin, T. P. 1979. Micro atolls: review of form, origin and terminology. Atoll Res. Bull. 224, 1–17. Stoddart, D. R. and Steers, J. A. 1977. The nature and origin of coral reef islands. In O. A. Jones and R. Endean (eds.) Biology and Geology of Coral Reefs, vol. 4, Geology 2. New York, Academic Press, pp. 59–105. Stoddart, D. R., McLean, R. F., and Hopley, D. 1978a. Geomorphology of reef islands, northern Great Barrier Reef. Phil. Trans. Roy. Soc. Lond. B 284, 39–61. Stoddart, D. R., McLean, R. F., Scoffin, T. P., and Gibbs, P. E. 1978b. Forty-five years of change on low wooded islands, Great Barrier Reef. Phil. Trans. Roy. Soc. Lond. B 284, 63–80. Stoddart, D. R., McLean, R. F., Scoffin, T. P., Thom, B. G., and Hopley, D. 1978c. Evolution of reefs and islands, northern Great Barrier Reef: synthesis and interpretation. Phil. Trans. Roy. Soc. Lond. B 284, 149–159. Stoddart, D. R., Gibbs, P. E., and Hopley, D. 1981. Natural history of Raine Island, Great Barrier Reef. Atoll Res. Bull. 254, 1–70. Suess, E. 1885. Das Antlitz der Erde. Vienna, Tempsky. Swart, P. K. and Grottoli, A. 2003. Proxy indicators of climate in coral skeletons: a perspective. Coral Reefs 22, 313–315. Sydow, J. C. 1996. Holocene to late Pleistocene stratigraphy of the Mahakam Delta, Kalimantan, Indonesia. Ph.D. thesis, Louisiana State University, Baton Rouge. Symonds, G., Black, K. P., and Young, I. R. 1995. Wave-driven flow over shallow reefs. J. Geophys. Res. Oceans 100, 2639–2648. Symonds, P. A., Davies, P. J., and Parisi, A. 1983. Structure and stratigraphy of the Central Great Barrier Reef. Bureau Min. Resources J. Austral. Geol. Geophys. 8, 277–291. Szabo, B. J., Tracey, J. I., and Goter, E. R. 1985. Ages of subsurface stratigraphic intervals in the Quaternary of Enewetak Atoll, Marshall Islands. Quatern. Res. 23, 54–61. Szabo, B. J., Ludwig, K. R., Muhs, D. R., and Simmons, K. R. 1994. Th-230 ages of corals and duration of the last Interglacial sea-level high stand on Oahu, Hawaii. Science 266, 93–96. Szmant, A. M. 2002. Nutrient enrichment on coral reefs: is it a major cause of coral decline? Estuaries 25, 743–766. Szmant-Froelich, A. M., Johnson, V., Hoehn, T., et al. 1982. The physiological effects of oil-drilling muds on the Caribbean coral, Montastrea annularis. Proc. 4th Int. Coral Reef Symp., Manila 1, 163–168. Tayama, R. 1952. Coral reefs of the South Seas. Bull. Hydrogr. Dept. Tokyo 11, 1–292. Taylor, T. 1924. Movement of sand cays. Queensland Geog. J. 39, 38–39. Teichert, C. and Fairbridge, R. W. 1950. Photo interpretation of coral reefs. Photogramm. Engin. 16, 744–755. Thamrongnawasawat, T. 1996. Development of digital aerial photography for coral reef management. Ph.D. thesis, James Cook University, Townsville. Thamrongnawasawat, T. and Hopley, D. 1995. Digital aerial photography applied to small area management in Thailand. In Bellwood, O., Choat, H., and Saxena, N. (eds.) Recent Advances in Marine Science and Technology, Proc. Pacific Congress on Marine Science and Technology, Townsville, pp. 385–394.
References
513
Thom, B. G. and Chappell, J. 1975. Holocene sea-levels relative to Australia. Search 6, 90–93. Thom, B. G. and Roy, P. S. 1983. Relative sea-levels and coastal sedimentation in southeast Australia in the Holocene. J. Sed. Petrol. 55, 257–264. Thom, B. G., Orme, G. R., and Polach, H. R. 1978. Drilling investigation of Bewick and Stapleton Islands in the Northern Great Barrier Reef. Phil. Trans. Roy. Soc. Lond. A 291, 37–54. Thomas, M. F., Nott, J., and Price, D. M. 2001. Late Quaternary stream sedimentation in the humid tropics: a review with new data from northeast Queensland, Australia. Geomorphology 39, 35–68. Thompson, W. G. and Goldstein, S. L. 2005. Open-system coral ages reveal persistent suborbital sea-level cycles. Science 308, 401–404. Thomson, R. E. and Wolanski, E. 1984. Tidal period upwelling within Raine Island Entrance, Great Barrier Reef. J. Mar. Res. 42, 787–808. Thurber, D. L., Broeker, W. S., Blanchard, R. L., and Potratz, H. A. 1965. Uranium series ages of Pacific atoll coral. Science 149, 55–58. Tiempo. 2002. Global warming and the Third World. Tiempo Newsl. 43, 26. Tomascik, T., Mah, A. J., Nontji, A., and Moosa, M. K. 1997. The Ecology of the Indonesian Seas, 2 vols. Halifax, Nova Scotia, Periplus Editions for Dalhousie University. Tooley, M. J. 1993. Long-term changes in eustatic sea level. In Warrick, R. A., Barrow, E. M., and Wigley, T. M. L. (eds.) Climate and Sea Level Change. Cambridge, Cambridge University Press, pp. 81–107. Toscano, M. A. and Lundberg, J. 1998. Early Holocene sea-level record from submerged fossil reefs on the southeast Florida margin. Geology 26, 255–258. Toscano, M. A. and Lundberg, J. 1999. Submerged Late Pleistocene reefs on the tectonically-stable SE Florida margin: high-precision geochronology, stratigraphy, resolution of substage 5a sea-level elevation, and orbital forcing. Quatern. Sci. Rev. 18, 753–767. Toscano, M. A. and Macintyre, I. G. 2003. Corrected western Atlantic sea level curve for the last 4000 years based on callibrated 14C dates from Acropora palmata framework and intertidal mangrove peat. Coral Reefs 22, 257–270. Tracey, J. I. and Ladd, H. S. 1974. Quaternary history of Enewetak and Bikini Atolls, Marshall Islands. Proc. 2nd Int. Coral Reef Symp., Queensland 2, 537–550. Trenberth, K. E. and Hurrell, J. W. 1994. Decadal atmosphere–ocean variations in the Pacific. Climate Dynam. 9, 303–319. Troedson, A. and Davies P. J. 1997. Sea level controlled sedimentation on a carbonate ramp – southern Great Barrier Reef, Australia. In Saxena, N. (ed.) Recent Advances in Marine Science and Technology, Proc. Pacific Congress on Marine Science and Technology, Honolulu, pp. 297–303. Tudhope, A. W. and Scoffin, T. P. 1984. The effects of Callianassa bioturbation on the preservation of carbonate grains in Davies Reef lagoon, Great Barrier Reef, Australia. J. Sed. Petrol. 54, 1091–1096. Udy, J. W., Dennison, W. C., Lee Long, W., and MacKenzie, L. J. 1999. Responses of sea grass to nutrients in the Great Barrier Reef, Australia. Mar. Ecol. Progr. Ser. 185, 257–271. Umar, M. J., McCook, L. J., and Price, I. R. 1998. Effects of sediment deposition on the seaweed Sargassum on a fringing coral reef. Coral Reefs 17, 169–177. University of Queensland. 2005. Earthquake Maps of Queensland and Australia. Available online at http://www.quakes.uq.edu.au
514
References
van de Plassche, O. 1986. Sea-Level Research: A Manual for the Collection and Evaluation of Data. Norwich, Geo Books. van Woesik, R. 1988. A preliminary examination on the sedimentology, reef growth and hydrology of Green Island. In Baxter, I. (ed.) Green Island Reef. Townsville, Great Barrier Reef Marine Park Authority. van Woesik, R. 1992. Ecology of coral assemblages on continental islands in the southern section of the Great Barrier Reef, Australia. Ph.D. thesis, James Cook University, Townsville. van Woesik, R. 1994. Contemporary disturbances to coral communities of the Great Barrier Reef. J. Coast. Res. (Spec. Issue) 12, 233–252. van Woesik, R. and Done, T. J. 1997. Coral communities and reef growth in the southern Great Barrier Reef. Coral Reefs 16, 103–115. van Woesik, R., Ayling, A. M., and Mapstone, B. 1991. Impact of tropical cyclone ‘‘Ivor’’ on the Great Barrier Reef, Australia. J. Coast. Res. 7, 551–558. van Woesik, R., De Vantier, L. M., and Glazebrook, J. S. 1995. Effects of Cyclone Joy on nearshore coral communities of the Great Barrier Reef. Mar. Ecol. Progr. Ser. 128, 261–270. van Woesik, R., Tomascik, T., and Blake, S. 1999. Coral assemblages and physicochemical characteristics of the Whitsunday Islands: evidence of recent community changes. Mar. Freshw. Res. 50, 427–440. Veeh, H. H. and Veevers, J. J. 1970. Sea level at 175 m off the Great Barrier Reef 13 600 to 17 000 years ago. Nature 226, 536–537. Veron, J. E. N. 1978. Deltaic and dissected reefs of the far northern region. Phil. Trans. Roy. Soc. Lond. B 284, 23–37. Veron, J. E. N. 1987. Checklist of corals from the Daintree Reefs. In Baldwin, C. L. (ed.) Fringing Reef Workshop: Science, Industry and Management. Townsville, Great Barrier Reef Marine Park Authority, pp. 99–103. Veron, J. E. N. 1995. Corals in Space and Time: The Biogeography and Evolution of the Scleractinia. Sydney, University of New South Wales Press. Veron, J. E. N. 2000. Corals of the World, 3 vols. Townsville, Australian Institute of Marine Science. Veron, J. E. N. and Hudson, R. C. L. 1978. Ribbon reefs of the northern region. Phil. Trans. Roy. Soc. Lond. B 284, 3–22. von Arx, W. S. 1954. Circulation systems of Bikini and Rongelap lagoons. U.S. Geol. Surv. Prof. Pap. 260B, 265–273. Wachenfeld, D. R. 1997. Long-term trends in the status of coral reef-flat benthos: the use of historical photographs. In Wachenfeld, D., Oliver, J., and Davis, K. (eds.) State of the Great Barrier Reef World Heritage Area Workshop. Townsville, Great Barrier Reef Marine Park Authority, pp. 134–148. Walcott, R. I. 1972. Past sea levels, eustasy and deformation of the earth. Quatern. Res. 2, 1–14. Walker, G. S. and Brunskill, G. J. 1997. A history of anthropogenic mercury input into the Great Barrier Reef lagoon, Australia. Proc. 8th Int. Coral Reef Symp., Panama 2, 1889–1892. Wallace, C. C. 2002. Journey to the heart of the centre: origins of a high marine faunal diversity in the central Indo-Pacific from the perspective of an acropologist. Proc. 9th Int. Coral Reef Symp., Bali 1, 33–39. Wang, P. X. and Chappell, J. 2001. Foraminifera as Holocene environmental indicators in the South Alligator River, Northern Australia. Quatern. Int. 83–85, 47–62.
References
515
Ware, J. R., Smith, S. V., and Reaka-Kudla, M. L. 1992. Coal reefs: sources or sinks of atmospheric CO2? Coral Reefs 11, 127–130. Watts, K. F., Varga, L. L., and Feary, D. A. 1993. Origins, timing and implications of Miocene to Pleistocene turbidites, debris flows and slump deposits of the Queensland Trough, northeastern Australia. In McKenzie, J. A., Davies, P. J., Palmer-Julson, A. A., and Sarg, J. F. (eds.) Proceedings of the Ocean Drilling Program, vol. 133, Scientific Results, North-East Australian Margin. College Station, TX, Texas A and M University, pp. 379–446. Waycott, M., Longstaff, B. J., and Mellors, J. 2005. Seagrass population dynamics and water quality in the Great Barrier Reef region: a review and future research directions. Mar. Poll. Bull. 51, 343–350. Webster, J. M. and Davies, P. J. 2003. Coral variation in two deep drill cores: significance for the Pleistocene development of the Great Barrier Reef. Sed. Geol. 159, 61–80. Webster, J. M., Davies, P. J., and Konishi, K. 1998. Model of fringing reef development in response to progressive sea level fall over the last 7000 years (Kikai-Jima, Ryukyu Islands, Japan). Coral Reefs 17, 289–308. Webster, J. M., Clague, D. A., Riker-Coleman, K., et al. 2004. Drowning of the 150 m reef off Hawaii: a casualty of global meltwater pulse 1A? Geology 32, 249–252. Wells, S. and Edwards, A. 1989. Gone with the waves. New Scientist, 11 Nov., 29–33. Wheatcraft, S. W. and Buddemeier, R. W. 1981. Atoll island hydrology. Ground Water 19, 311–320. White, N. J., Church, J. A., and Gregory, J. M. 2005. Coastal and global averaged sea level rise for 1950 to 2000. Geophys. Res. Lett. 32. Wilkinson, C. R. 1999. Global and local threats to coral reef functioning and existence: review and predictions. Mar. Freshw. Res. 50, 867–878. Wilkinson, C. R. (ed.) 2004. Status of Coral Reefs of the World: 2002. Townsville, Australian Institute of Marine Science. Williams, D.McB. 2001. Review of the Impacts of Terrestrial Run-Off on the Great Barrier Reef World Heritage Area. Townsville, CRC Reef Research Centre. Wilson, M. E. J. 2005. Development of equatorial delta-front patch reefs during the Neogene, Borneo. J. Sed. Res. 75, 114–133. Wilson, P. R. 1985. Tidal studies in the One-Tree Island Lagoon. Austral. J. Mar. Freshw. Res. 36, 139–156. Wolanski, E. 1983. Water circulation in the central and northern sections of the Great Barrier Reef. In Baker, J. T., Carter, R. M., Sammarco, P. W., and Stark, K. P. (eds.) Proceedings of the Great Barrier Reef Conference. Townsville, James Cook University, pp. 467–472. Wolanski, E. 1986a. Water circulation in a topographically complex environment. In van de Kreeke, J. (ed) Physics of Shallow Estuaries and Bays. Berlin, Springer Verlag, pp. 154–167. Wolanski, E. 1986b. Island wakes and internal tides in stratified shelf waters. Ann. Geophys. 4B, 425–439. Wolanski, E. 1986c. Observations of wind-driven surface gravity-waves offshore from the Great Barrier Reef. Coral Reefs 4, 213–219. Wolanski, E. 1993. Facts and numerical artifacts in modelling the dispersal of crown-of-thorns starfish larvae in the Great Barrier Reef. Austr. J. Mar. Freshw. Res. 44, 427–436. Wolanski, E. 1994. Physical Oceanographic Processes of the Great Barrier Reef. Boca Raton, FL, CRC Press.
516
References
Wolanski, E. (ed.) 2001. Oceanographic Processes of Coral Reefs: Physical and Biological Links in the Great Barrier Reef. Boca Raton, FL, CRC Press. Wolanski, E. and Bennett, A. F. 1983. Continental-shelf waves and their influence on the circulation around the Great Barrier Reef. Austral. J. Mar. Freshw. Res. 34, 23–47. Wolanski, E. and De’ath, G. 2005. Predicting the impact of present and future human land-use on the Great Barrier Reef. Estuar. Coast. Shelf Sci. 64, 504–508. Wolanski, E. and Hamner, W. M. 1988. Topographically controlled fronts in the ocean and their biological influence. Science 241, 177–181. Wolanski, E. and Jones, M. 1980. Water circulation around Britomart Reef, Great Barrier Reef, during July 1979. Austral. J. Mar. Freshw. Res. 31, 415–430. Wolanski, E. and Jones, M. 1981. Physical properties of Great Barrier Reef lagoon waters near Townsville: effects of Burdekin River floods. Austral. J. Mar. Freshw. Res. 32, 305–319. Wolanski, E. and King, B. 1990. Flushing of Bowden Reef Lagoon, Great Barrier Reef. Estuar. Coast. Shelf Sci. 31, 789–804. Wolanski, E. and Pickard, G. L. 1983. Upwelling by internal waves and Kelvin waves at the continental shelf break on the Great Barrier Reef, Austr. J. Mar. Freshw. Res. 34, 65–80. Wolanski, E. and Pickard, G. L. 1985. Long-term observations of currents on the Central Great Barrier Reef continental shelf. Coral Reefs 4, 47–57. Wolanski, E. and Ridd, P. V. 1990. Mixing and trapping in Australian tropical coastal waters. In Cheng, R. T. (ed.) Residual Currents and Long-Term Transport. New York, Springer Verlag, pp. 165–183. Wolanski, E. and Ruddick, B. 1981. Water circulation and shelf waves in the Northern Great Barrier Reef Lagoon. Austral. J. Mar. Freshw. Res. 32, 721–740. Wolanski, E. and van Senden, D. 1983. Mixing of Burdekin River flood waters in the Great Barrier Reef. Austral. J. Mar. Freshw. Res. 34, 49–63. Wolanski, E. and Spagnol, S. 2000. Sticky waters in the Great Barrier Reef. Estuar. Coast. Shelf Sci. 50, 27–32. Wolanski, E. and Thomson, R. E. 1984. Wind-driven circulation on the northern Great Barrier Reef continental shelf in summer. Estuar. Coast. Shelf Sci. 18, 271–289. Wolanski, E., Imberger, J., and Heron, M. L. 1984. Island wakes in shallow coastal waters. J. Geophys. Res. 89C6, 10553–10569. Wolanski, E. J., Drew, E., Abel, K. M., and O’Brien, J. 1988. Tidal jets, nutrient upwelling and their influence on the productivity of the alga Halimeda in the ribbon reefs, Great Barrier Reef. Estuar. Coast. Shelf Sci. 26, 169–201. Wolanski, E., Burrage, D., and King, B. 1989. Trapping and dispersion of coral eggs around Bowden Reef, Great Barrier Reef, following mass coral spawning. Continent. Shelf Res. 9, 479–496. Wolanski, E., Fabricius, K., Spagnol, S., and Brinkman, R. 2005. Fine sediment budget on an inner shelf coral fringed island, Great Barrier Reef, Australia. Estuar. Coast. Shelf Sci. 65, 153–158. Wood, R. 1999. Reef Evolution. Oxford, Oxford University Press. Woodroffe, C. D. 1993. Morphology and evolution of reef islands in the Maldives. Proc 7th Int. Coral Reef Symp., Guam 7, 1217–1226. Woodroffe, C. D. (ed.) 1994. Ecology and geomorphology of the Cocos (Keeling) Islands. Atoll Res. Bull. 399–414. Woodroffe, C. D. 2002a. Coasts: Form, Process and Evolution. Cambridge, Cambridge University Press.
References
517
Woodroffe, C. D. 2002b. Reef island sedimentation on Indo-Pacific atolls and platform reef. Proc. 9th Int. Coral Reef Symp., Bali 2, 1187–1192. Woodroffe, C. D. 2005. Late Quaternary sea-level highstands in the central and eastern Indian Ocean: a review. Glob. Planet. Change 49, 121–138. Woodroffe, C. D. and McLean, R. F. 1990. Microatolls and recent sea level change on coral atolls. Nature 344, 531–534. Woodroffe, C., Thom, B., and Chappell, J. 1985. Development of widespread mangrove swamps in mid-Holocene times in northern Australia. Nature 317, 711–713. Woodroffe, C. D., Veeh, H. H., Falkland, A. C., McLean, R. F., and Wallensky, E. 1991. Last interglacial reef and subsidence of Cocos (Keeling) Islands, Indian Ocean. Mar. Geol. 96, 137–143. Woodroffe, C. D., McLean, R. F., and Wallensky, E. 1994. Geomorphology of the Cocos (Keeling) Islands. Atoll Res. Bull. 402, 1–33. Woodroffe, C. D., Kennedy, D. M., Hopley, D., Rasmussen, C. E., and Smithers, S. G. 2000. Holocene reef growth in Torres Strait. Mar. Geol. 170, 331–346. Woodroffe, C. D., Kennedy, D. M., Jones, B. G., and Phipps, C. V. G. 2004. Geomorphology and Late Quaternary development of Middleton and Elizabeth Reefs. Coral Reefs 23, 249–262. Woodroffe, C. D., Dickson, M. E., Brooke, B. P., and Kennedy, D. M. 2005. Episodes of reef growth at Lord Howe Island, the southernmost reef in the southwest Pacific. Glob. Planet. Change 49, 222–237. Woodroffe, C. D., Kennedy, D. H., Brooke, B. P., and Dickson, M. E. 2006. Geomorphological evolution of Lord Howe Island and carbonate production at the latitudinal limit of coral growth. J. Coast. Res. 22, 188–201. Woodroffe, S. A. and Horton, B. P. 2005. Holocene sea-level changes in the IndoPacific. J. Asian Earth Sci. 25, 29–43. Woodroffe, S. A., Horton, B. P., Larcombe, P., and Whittaker, J. E. 2005. Intertidal mangrove foraminifera from the central Great Barrier Reef shelf, Australia: implications for sea-level reconstruction. J. Foramin. Res. 35, 259–270. Woolfe, K. J. and Larcombe, P. 1998. Terrigenous sediment accumulation as a regional control on the distribution of reef carbonates. In Camoin, G. and Davies, P. (eds.) Reefs and Carbonate Platforms in the Pacific and Indian Oceans. Oxford, Blackwell Science, pp. 295–310. Woolfe, K. J., Larcombe, P., Orpin, A. R., et al. 1998a. Controls upon inner-shelf sedimentation, Cape York Peninsula, in the Region of 128 S. Austral. J. Earth Sci. 45, 611–621. Woolfe, K. J., Larcombe, P., Naish, T., and Purdon, R. G. 1998b. Lowstand rivers need not incise the shelf: an example from the Great Barrier Reef, Australia, with implications for sequence stratigraphy models. Geology 26, 75–78. Woolfe, K. J., Larcombe, P., Orpin, A. R., and Purdon, R. G. 2000. Spatial variability in fluvial style and likely responses to sea-level change, Herbert River, Queensland. Austral. J. Earth Sci. 47, 689–694. Wyrwoll, K. H., Zhu, Z. R., Collins, L. B., and Hatcher, B. G. 2006. Origin of blue hole structures in coral reefs: Houtman Abrolhos, Western Australia. J. Coast. Res. 22, 202–208. Yabe, H. 1942. Problems of the coral reefs. Tokyo Imp. Univ. Geol. Palaeo. Inst. Rept. 39, 1–6. Yamano, H., Miyajima, T., and Koike, I. 2000. The importance of foraminifera for the formation and maintenance of a coral sand cay, Green Island, Australia. Coral Reefs 19, 51–58.
518
References
Yamano, H., Kayanne, H., Matsuda, F., and Tsuji, Y. 2002. Lagoonal facies ages and sedimentation in three atolls in the Pacific. Mar. Geol. 185, 233–247. Yentsch, C. S., Yentsch, C. M., Cullen, J. J., et al. 2002. Sunlight and water transparency: cornerstones in coral research. J. Exp. Mar. Biol. Ecol. 268, 171–183. Yokoyama, Y. and Nguyen, H. V. 1980. Direct and non-destructive dating of marine sediments, manganese nodules and corals by high resolution gamma-ray spectrometry. In Goldberg, E. D., Horibe, Y., and Saruhashi, K. (eds.) Isotope Marine Chemistry. Tokyo, Uchida Rokakuho, pp. 259–289. Yokoyama, Y., Esat, T. M., and Lambeck, K. 2001a. Last glacial sea-level change deduced from uplifted coral terraces of Huon Peninsula, Papua New Guinea. Quatern. Int. 5, 275–283. Yokoyama, Y., De Deckker, P., Lambeck, K., Johnston, P., and Fifield, L. K. 2001b. Sea-level at the Last Glacial Maximum: evidence from northwestern Australia to constrain ice volumes for oxygen isotope stage 2. Palaeogeog. Palaeoclimatol. Palaeoecol. 165, 281–297. Young, I. R. 1988. A parametric hurricane wave prediction model. J. Waterway Port, Coast. Ocean Engin. 114, 637–652. Young, I. R. 1989. Wave transformation over coral reefs. J. Geophys. Res. 94, 9779–9789. Young, I. R. and Sobey, R. J. 1980. A predictive model of tropical cyclone windwaves. Proc. 7th Australasian Conf. Hydraulics and Fluid Mechanics, Brisbane, 480–483. Young, I. R., Black, K. P., and Heron, M. L. 1994. Circulation in the Ribbon Reef Region of the Great Barrier Reef. Continent. Shelf Res. 14, 117. Zann, L. P. 1996. State of the Marine Environment Report for Australia: Technical Summary. Canberra, Department of Environment, Sport and Territories. Zwartz, D. 1995. The recent history of the Antarctic ice sheet: constraints from sea-level change. Ph.D. thesis, Australian National University, Canberra.
Geographic index
Abrolhos Reef (see also Houtman–Abrolhos Islands), 400 Africa, 136 Agincourt Reef, 121 Fig. 4.12, 141 Fig. 5.1, 142 Table 5.1, 286 Airlie Beach, 196 Aitutaki, 408 Alacran Reef, 400, 402, 405, 407 Aladdin’s Cave, 51 Ambergris Cay, 400 Anchor Cay, 34 Antarctic, 74, 85 Antarctica, 18 Antigua, 402 Arc Reef, 421 Arlington Reef/Cay, 316, 322, 323, 417 Fig. 12.3 Ashmore Reef/Cay/Banks, 271, 322, 323 Asia, 136 Atherton Tableland, 2 Fig. 1.1, 19, 424, 445 Atui, 408 Ayr, 93 Fig. 4.1, 97 Table 4.1 Bacchi Cay, 323 Baird Island, 325 Balding Bay, 63 Fig. 3.6, 66, 67 Bali, 15 Bantem Bay, 432 Barbados, 5, 43, 52, 53, 59, 59 Fig. 3.5, 60, 296, 437 Barnett Patches/Shoals, 79, 150 Fig. 5.7 Barron River, 21 Fig. 2.2, 23, 177, 440, 441 Barrow Point, 93 Fig. 4.1, 94 Bathurst Bay, 122 Table 4.2, 124 Beanley Islands, 334 Fig. 10.6, 342 Beaver Cay, 318, 330, 354 Beesley Cay, 323 Belize, 3, 353, 400, 401 Fig. 11.11, 402, 408 Bell Cay, 323, 345 Bench Point, 107 Beverley Island Group, 210 Bewick Reef/Island, 77, 144, 144 Fig. 5.3, 145, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 238–239, 238 Fig. 8.3, 250 Table 8.2, 254, 268 Fig. 8.15,
320, 325, 332, 338, 339 Fig. 10.9, 340, 341 Fig. 10.11, 342, 343, 369, 370, 381 Big Bank Shoals, 185, 187 Bikini Atoll, 6, 13, 395 Binstead Cay, 322, 325, 330, 342 Bird Cay, 325 Black Rock, 2 Fig. 1.1, 34 Bligh Trough, 20 Fig. 2.1, 27, 30 Boot Reef, 2 Fig. 1.1, 38, 271 Boulder Reef, 2 Fig. 1.1, 35, 36 Fig. 2.11, 38, 39, 40, 59 Fig. 3.5, 60, 75, 79, 108, 234 Table 8.1, 237 Fig. 8.2, 239, 249, 250 Table 8.2, 267, 268 Fig. 8.15, 372, 373, 381 Bowden Reef, 107 Bowen, 23, 54, 63 Fig. 3.6, 77, 93 Fig. 4.1, 97 Table 4.1, 118, 122 Table 4.2, 168, 381 Bowl Reef, 54, 60, 80, 179, 180, 236 Fig. 8.1, 273 Table 9.1, 289, 292, 292 Fig. 9.14, 294 Fig. 9.16, 295 Fig. 9.17, 373, 375, 381, 418 Fig. 12.4, 421 Fig. 12.7 Bowling Green Bay, 179, 424 Fig. 12.8 Bramble Cay, 34, 170, 451 Brazil, 400 Breaksea Spit, 2 Fig. 1.1, 8, 24 Brisbane, 345 Britomart Reef, 80, 82, 84, 108, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 240, 249, 250 Table 8.2, 251, 253, 258, 266, 268, 268 Fig. 8.15, 369, 371 Broad Sound, 82, 83, 100, 102, 122 Table 4.2, 123, 132, 134, 178, 217 Broomfield Reef, 173 Bundaberg, 82, 87, 93 Fig. 4.1, 97 Table 4.1, 122 Table 4.2 Bunker–Capricorn Reef/Island Group, 32, 83, 159, 162, 169 Fig. 6.1, 172, 173, 236 Fig. 8.1, 245, 246 Fig. 8.7, 247, 269, 271, 272, 309, 310, 314, 315, 320, 332, 334 Fig. 10.7, 337, 345, 355, 358, 361, 370, 390, 414, 418, 423, 448, 451, 453, 465 Burdekin Dam, 127 Burdekin River, 19, 21 Fig. 2.2, 23, 25 Fig. 2.5, 37, 127, 129, 131, 176, 178 Fig. 6.3, 179, 236 Fig. 8.1, 272, 414, 415, 418 Fig. 12.4, 421, 440, 443 Burnett River, 127
519
520
Geographic index
Bushy Island (see also Bushy-Redbill Island, Redbill Reef), 237 Fig. 8.2, 319, 345, 362, 363, 427, 445, 453, 453 Table 13.3, 455, 459 Bushy–Redbill Island, 323 Bylund Cay, 323 Cairns, 1, 7, 9, 10, 18, 19, 23, 30, 32, 34, 37, 55, 77, 87, 93 Fig. 4.1, 97 Table 4.1, 111, 112 Fig. 4.9, 113, 118, 122 Table 4.2, 123, 124, 128, 159, 168, 171, 173, 176, 177, 178 Fig. 6.3, 180, 200, 225, 275, 285, 286, 314, 338, 344, 345, 347, 354, 355, 365, 369, 416, 421, 423, 440, 442, 444, 451, 452, 465 Cairns Reef, 79 Camp Island, 54, 81 Campeche Bank, 404 Canning Basin, 414, 415 Fig. 12.2 Cape Bedford, 24, 184 Fig. 6.5 Cape Capricorn, 24, 93 Fig. 4.1, 97 Table 4.1 Cape Clinton, 2 Fig. 1.1, 23, 24, 169 Fig. 6.1, 422 Cape Conway, 196 Cape Ferguson, 87 Cape Flattery, 24, 25 Fig. 2.5 Cape Gloucester, 196, 422 Cape Grafton, 23 Cape Grenville, 120, 121 Fig. 4.12, 197, 271 Cape Hillsborough, 2 Fig. 1.1, 19 Cape Kimberley, 197 Cape Melville, 55, 77, 86, 197, 370 Cape Tribulation, 2 Fig. 1.1, 23, 23 Fig. 2.4, 75, 122 Table 4.2, 174, 195 Fig. 7.2, 197, 198 Fig. 7.3, 203, 207, 212, 214, 220, 376, 381, 426, 433, 433 Table 13.1, 434 Fig. 13.1, 435 Fig. 13.2, 436 Fig. 13.3, 13.4, 437, 438, 438 Table 13.2, 440, 447, 449 Cape Upstart, 174, 196, 422 Cape Weymouth, 18, 200 Cape York, 93 Fig. 4.1, 97 Table 4.1, 118, 125, 159 Cape York Peninsula, 26, 32, 96, 130, 197, 200 Capricorn Basin, 20 Fig. 2.1, 34 Capricorn Channel, 2 Fig. 1.1, 32, 55, 100, 169 Fig. 6.1, 171, 179, 236 Fig. 8.1, 274 Fig. 9.1, 296, 414, 416, 417, 421, 423 Capricorn Coast, 111, 112 Fig. 4.9, 113 Captain Billy Landing, 197 Cardwell, 122 Table 4.2, 128, 423 Caribbean, 117, 136, 384, 400–403 Carmila, 199 Carter Reef, 77, 78, 273 Table 9.1, 280, 283, 284 Cato Trough, 418 Fig. 12.4 Cay Cay, 354 Centipede Reef, 27 Central America, 136 Central Queensland Shelf, 54 Channel Reef, 79 Chapman Cays, 322, 325 Clack Island, 325 Clack Reef, 160, 177 Clairview, 122 Table 4.2 Clare, 127 Cleveland Bay, 132, 133, 182, 222
Cockatoo Reef, 81, 142, 143 Fig. 5.2, 144, 236 Fig. 8.1, 273 Table 9.1, 297 Fig. 9.18, 300, 300 Fig. 9.21, 301, 304 Fig. 9.24, 369, 376, 383, 448 Cockburn Reef, 184 Fig. 6.5 Cockermouth Reef/Island, 54, 75, 81, 204, 214, 368, 376 Cockle Bay, 69, 204 Cocoa Creek, 69 Table 3.2 Coconut Island, 249, 453, 454 Fig. 13.9, 456, 462 Fig. 13.11 Cocos–Keeling Islands, 393–395, 394 Fig. 11.9, 404 Table 11.2, 407 Coen, 93 Fig. 4.1, 97 Table 4.1 Coil Reef, 421 Cook Islands, 395 Cooktown, 93 Fig. 4.1, 97 Table 4.1, 122 Table 4.2, 124, 159, 185, 187, 218, 369, 370, 372, 381 Coolgaree Bay, 203, 207 Coombe Island, 323 Coquet Island, 325, 340, 342 Coral Sea, 19, 27, 32, 35, 37, 96, 99, 111, 118, 425 Coral Sea Basin, 20 Fig. 2.1, 27, 30 Corbett Reef, 177, 393 Cowley Beach, 182 Cullal Cullal Reef, 197 Cumberland Island Group, 2 Fig. 1.1, 26, 82, 160, 199, 370, 371, 424, 426 Cungulla, 54 Curacoa Island, 228, 350 Curtis Island, 2 Fig. 1.1, 21 Fig. 2.2, 24 Daintree, 93 Fig. 4.1, 96 Daintree River, 174, 177, 441 Darley Reef, 81, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 243–244, 243 Fig. 8.5, 249, 250 Table 8.2, 251, 253, 268 Fig. 8.15, 381 Darnley Island, 2 Fig. 1.1, 34, 184 Fig. 6.5 Darnley Valley, 179 Dauar Island, 34 Davies Reef, 108, 110, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 242, 243 Fig. 8.5, 250 Table 8.2, 253, 254, 268, 268 Fig. 8.15, 372, 418 Fig. 12.4 Day Reef, 236 Fig. 8.1, 282, 412 Daydream Island, 438 Digby Island, 51, 54, 204, 210, 368, 369, 372 Dip Reef, 418 Fig. 12.4 Double Island, 200 Douglas Island, 351 Dunk Island, 79, 177, 203, 207, 214 Eagle Cay, 323, 331 East Fairfax Island (see also Fairfax Islands/ Reef), 336 East Hope Reef/Island (see also Hope Reef, West Hope Reef/Island), 79, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 238 Fig. 8.3, 239, 250 Table 8.2, 251, 267, 323, 332 East Hoskyn Island (see also Hoskyn Reef/Island, West Hoskyn Island), 323, 336 Eastern Plateau, 19, 20 Fig. 2.1, 27, 236 Fig. 8.1
Geographic index Ecuador, 89 Eddy Reef, 177 Edgecombe Bay, 442 Elizabeth Reef, 391, 392, 392 Table 11.1 Elliot River, 54 Ellis Island, 317, 322, 335, 354 Ellison Reef, 79, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 240, 250 Table 8.2, 376, 381 Emmagen Reef, 79, 203 Endeavour River, 69 Enewetak Atoll, 6, 13, 34, 185, 393, 395, 404 Table 11.2 Erskine Island, 323, 332, 334 Fig. 10.7, 355, 357 Fig. 10.14, 362, 453, 455, 456, 459 Eulalie Reef, 285, 285 Fig. 9.9 Euston Reef, 417 Fig. 12.3 Fairey Reef, 150 Fig. 5.7 Fairfax Reef/Islands (see also East Fairfax Island), 81, 83, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 247, 248 Fig. 8.8, 249, 250 Table 8.2, 268 Fig. 8.15, 269, 314, 320, 337, 345, 370, 372, 383 Family Island Group, 200 Fantome Island, 80, 203, 219, 222, 376 Faraday Reef, 179, 418 Fig. 12.4 Fife Island, 323 Fin Reef, 417 Fig. 12.3 Fisher Reef/Island, 59 Fig. 3.5, 77, 78, 325, 442 Fitzroy Island, 93 Fig. 4.1, 97 Table 4.1, 200 Fitzroy Reef, 81, 83, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 247, 250 Table 8.2, 253, 254, 268, 268 Fig. 8.15, 370, 371, 372, 383 Fitzroy River, 19, 21 Fig. 2.2, 22, 24, 37, 127, 178, 179, 414, 440, 443 Flinders Island, 78 Flinders Island Group, 314 Flinders Passage, 174 Flinders Reef, 29 Flora Pass, 178, 181 Fig. 6.4, 182, 416 Florida, 437 Florida Keys, 437 Fly River, 128, 170 Flynn Reef, 417 Fig. 12.3 Forbes Island, 26, 184 Fig. 6.5 Four Mile Beach, 196 Fraser Island, 2 Fig. 1.1, 18, 21 Fig. 2.2, 24, 125 Frigate Cay, 185, 323 Funafuti Atoll, 6, 9, 183, 393, 404 Table 11.2 Gable Reef, 14 Fig. 1.4, 81, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 244–245, 245 Fig. 8.6, 250 Table 8.2, 253, 254, 262–265, 263 Fig. 8.13, 264 Fig. 8.14, 268, 370, 383 Galeta Point, 402 Gannet Cay, 323, 458 Geoffrey Bay, 204, 223 Geranium Passage, 178 Gilbert Islands, 395 Gladstone, 93 Fig. 4.1, 97 Table 4.1, 122 Table 4.2, 465
521
Glasshouse Mountains, 19 Gloucester Island, 199 Glovers Reef, 400, 402 Goble Reefs, 244 Goold Island, 79, 218 Grafton Passage, 124, 181 Fig. 6.4, 416, 417 Fig. 12.3 Grand Cayman, 407 Great Australian Bight, 412 Great Detached Reef, 183, 271, 276, 278, 371 Great North East Channel, 93 Fig. 4.1, 94 Great Palm Island, 80 Table 3.3, 195 Fig. 7.2, 203 Table 7.2, 205, 207, 214 Green Island, Moreton Bay, 345 Green Reef/Island, 2 Fig. 1.1, 7, 102, 140, 141 Fig. 5.1, 142 Table 5.1, 171, 184 Fig. 6.5, 323, 332, 345, 355, 358, 361, 362, 417 Fig. 12.3, 427, 444, 445, 446 Fig. 13.6, 451, 452, 453, 455, 456, 458, 459 Greenland, 74 Grub Reef, 80, 177, 179, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 241–256, 241 Fig. 8.4, 250 Table 8.2, 256 Fig. 8.9, 266, 267, 381, 418 Fig. 12.4 Guam, 5, 437 Gulf of Carpentaria, 84, 96, 199, 432, 450 Gulf of Papua, 34, 35, 37, 55, 94, 170, 279 Halifax Basin, 20 Fig. 2.1, 32, 236 Fig. 8.1, 275 Halifax Bay, 197, 199, 213, 222, 236 Fig. 8.1, 449, 450 Hammond Island, 59 Fig. 3.5, 78, 203, 207, 218, 376 Hampton Island, 78, 325 Hanauma Bay, 396, 397 Fig. 11.10 Hannah Island, 340 Fig. 10.10, 342 Hastings Reef, 417 Fig. 12.3 Hawaii, 135, 216, 396, 397 Fig. 11.10, 407, 409, 437 Hayman Island, 59 Fig. 3.5, 60, 81, 82, 93 Fig. 4.1, 97 Table 4.1, 204, 207, 209, 214, 217, 218, 368, 369, 372, 373, 423, 424 Hazard Bay, 80 Hedge Reef, 269 Hedley Reef, 79 Helix Reef, 173, 174, 418 Fig. 12.4 Herald Island, 68, 80, 82 Herbert River, 21 Fig. 2.2, 22, 128, 174, 175 Fig. 6.2, 176, 178, 178 Fig. 6.3 Heron Island, 2 Fig. 1.1, 9, 12, 14, 32, 34, 38, 48, 49, 93 Fig. 4.1, 97 Table 4.1, 230, 314, 315, 319, 323, 332, 335 Fig. 10.8, 358, 370, 451, 453, 455, 456, 458, 459, 465 Hicks Reef, 236 Fig. 8.1, 282, 412 High Island, 224, 442 High Peak Island, 81, 204, 376 Hinchinbrook Channel, 24, 26 Fig. 2.6 Hinchinbrook Island, 2 Fig. 1.1, 21 Fig. 2.2, 24, 82, 177, 199, 311 Holmes Reef/Cay, 29, 329 Fig. 10.3 Hope Reef (see also East Hope Reef/Island, West Hope Reef/Island), 417 Fig. 12.3 Hopkinson Reef, 418 Fig. 12.4
522
Geographic index
Horn Island, 315 Hoskyn Reef/Islands (see also East Hoskyn Island, West Hoskyn Island), 150 Fig. 5.7, 320, 337, 345 Houghton Island, 78, 325, 342 Houtman–Abrolhos Islands (see also Abrolhos Reef), 391, 392, 392 Table 11.1, 404 Table 11.2, 407 Howick Group, 144 Howick Island/Reef, 26, 160, 167, 169 Fig. 6.1, 325, 370 Hull River, 127 Fig. 4.14, 177 Huon Peninsula, 43, 46 Fig. 3.1, 50 Fig. 3.2, 51, 52, 59 Fig. 3.5, 60, 63 Fig. 3.6 Hydeaway Bay, 23, 81, 196, 204, 223 Hydrographers Passage, 286, 287 Fig. 9.10 Indian Ocean, 125 Indonesia, 3, 183, 432 Ingham, 34, 93 Fig. 4.1, 96, 97 Table 4.1 Ingram Island, 325, 332, 334 Fig. 10.6, 342, 343 Innaminka Shoal, 169 Fig. 6.1 Innisfail, 23, 70, 82, 93 Fig. 4.1, 97 Table 4.1, 286, 388 Iris Point, 63 Fig. 3.6, 75, 80, 195 Fig. 7.2, 198 Fig. 7.3, 200, 203, 211, 214, 219, 226, 227, 228, 229 Isbell Shoal, 173 Isla Perez Reef, 404 Jamaica, 437 Jenny Louise Shoal, 285, 285 Fig. 9.9 John Brewer Reef, 418 Fig. 12.4 Johnstone River, 178 Joist Reef, 285, 285 Fig. 9.9 Joseph Bonaparte Gulf, 50 Fig. 3.2, 53 Kalukalukuang Bank, 185, 190 Kaneohe Bay, 135, 437 Kapingamarangi Atoll, 353 Karamea Bank, 169 Fig. 6.1 Kay Islet, 323, 331 Keeper Reef, 80, 179, 180, 418 Fig. 12.4 Kennedy River, 177 Keppel Island Group, 199, 209 Kimberley Coast, 417 King Island, 78 King Reef, 196, 207 Kurramine Beach, 196 Lady Elliot Island, 86, 93 Fig. 4.1, 97 Table 4.1, 114, 115, 312, 319, 323, 335, 336, 337, 349, 452 Lady Musgrave Reef/Island, 265, 319, 323, 335, 336 Lark Pass Reef, 78, 273 Table 9.1, 283 Layoak Island, 359 Fig. 10.15 Leggatt Island, 78, 325 Lifou, 412 Lighthouse Reef, 400, 402 Lihou Reef, 29 Linden Bank, 417 Fig. 12.3 Line Islands, 395 Linnet Reef, 317, 318
Little Pioneer Bay, 80 Liverpool Creek, 177 Lizard Island, 2 Fig. 1.1, 8, 11 Fig. 1.3, 26, 159, 167, 171, 183, 184 Fig. 6.5, 187, 188, 236 Fig. 8.1, 314, 315, 317, 423, 458 Lockhart River, 93 Fig. 4.1, 97 Table 4.1, 122 Table 4.2, 124 Lodestone Reef, 418 Fig. 12.4 Long Reef, 78 Lord Howe Island, 391, 392, 392 Table 11.1 Louisiade Archipelago, 422 Low Isles, 1, 2 Fig. 1.1, 7, 10, 10 Fig. 1.2, 11, 12, 63 Fig. 3.6, 93 Fig. 4.1, 97 Table 4.1, 128, 177, 312, 314, 315, 325, 343, 348, 350, 440, 441, 441 Fig. 13.5, 453, 458 Low Wooded Island, 59 Fig. 3.5, 75, 77, 78, 320, 325, 338, 340 Lowrie Island, 325 Lowry Pass, 177 Loyalty Islands, 412 Lucinda, 93 Fig. 4.1, 97 Table 4.1 Lugger Bay, 79, 222 Lugger Shoals, 203, 207, 220 Fig. 7.12, 221, 222, 229 Lynches Reef, 150 Fig. 5.7 MacGillivray Reef, 318 Table 10.1, 331 Mackay, 22, 32, 87, 89, 93 Fig. 4.1, 96, 97 Table 4.1, 111, 112 Fig. 4.9, 122 Table 4.2, 345, 369, 370, 442 Mackay Cay, 322, 323 MacLennan Cay, 323, 457 Fig. 13.10, 459 Madagascar, 3 Magnetic Island, 63 Fig. 3.6, 66, 67, 81, 82, 204, 219, 222, 223, 224, 314, 433, 438 Magnetic Passage, 418 Fig. 12.4 Magpie Reef, 269 Magra Island, 323 Mahakam Delta, 185 Makassar Strait, 185, 190 Makatea Island, 412 Maldives, 125, 463 Mare´, 412 Maria Creek, 177 Marion Plateau, 19, 20 Fig. 2.1, 27, 29, 41, 275, 299 Marion Reef, 29 Martin Reef, 317, 318 Masig–Kodall (see also Yorke Island), 320, 337, 338, 345, 358, 359 Fig. 10.15 Masthead Island, 324, 331, 351, 358, 362, 453, 455, 457 Fig. 13.10, 459, 465 Mauritius, 399, 400, 407 Mayotte Island, 399, 400, 407 Mer Island, 34 Mexico, 400 Michaelmas Reef/Cay, 2 Fig. 1.1, 9, 34, 38, 48, 49, 79, 324, 417 Fig. 12.3, 451, 451 Fig. 13.8, 452, 465 Middle Cay, 141 Fig. 5.1 Middle Island, 81, 82 Middle Percy Island, 204 Middle Reef, 222, 229, 435, 449 Middleton Reef, 391, 392, 392 Table 11.1 Midway Atoll, 13, 34
Geographic index Milln Reef, 417 Fig. 12.3 Milman Island, 332 Missionary Bay, 24, 69 Moa Island, 203, 207, 368 Molar Reef, 81, 236 Fig. 8.1, 297 Fig. 9.18, 300, 300 Fig. 9.21, 448 Mombasa, Kenya, 225 Mooloolaba, 111 Moonlight Bay, 196 Moreton Bay, 345 Morgan River, 69 Morris Cay, 324 Table 10.2 Moss Reef, 79, 236 Fig. 8.1, 240, 285 Fig. 9.9, 286–289, 288 Fig. 9.11, 292, 292 Fig. 9.14, 293 Fig. 9.15, 295 Fig. 9.17, 307, 376, 381 Mossman, 199 Mount Bartle Fre`re, 19, 21 Fig. 2.2 Mount Bellenden Ker, 19, 21 Fig. 2.2 Mount Dalrymple, 19, 21 Fig. 2.2 Mount Elliott, 19, 23 Fig. 2.4 Mount Inkerman, 54 Mowbray River, 196, 197, 442 Mozambique, 222 Mulgrave River, 178 Murdoch Island, 319, 337, 345 Murray Islands, 2 Fig. 1.1, 22, 22 Fig. 2.3, 32, 184 Fig. 6.5 Murray River, 69 Mururoa Atoll, 6, 13, 34, 393, 404 Table 11.2 Mustard Patches, 53, 285, 285 Fig. 9.9 Mutchero Inlet, 70 Myall Reef, 79, 203, 207 Myrmidon Reef, 2 Fig. 1.1, 42, 53, 54, 60, 80, 99, 102, 105, 106, 111, 179, 180, 236 Fig. 8.1, 273 Table 9.1, 285 Fig. 9.9, 286, 289, 290 Fig. 9.12, 291 Fig. 9.13, 292, 293 Fig. 9.15, 295 Fig. 9.17, 296, 368, 381, 390, 415, 450 Naracoote, South Australia, 48 Needle Reef, 418 Fig. 12.4 Nelly Bay, 224 New Caledonia, 3, 225, 393, 396–398, 397 Fig. 11.10, 407 New South Wales, 86, 87, 196 New Zealand, 73, 99 Newton Island, 325, 342 Nicaragua, 185 Nicholas Reef, 377 Fig. 11.3 Night Island, 319, 337, 345 Ningaloo Reef, 391, 392, 392 Table 11.1, 404 Table 11.2 Noah Head, 197 Noble Island, 78 Table 3.3 Nonsuch Bay, 402 Norman Reef, 377 Fig. 11.3 Normanby River, 69, 127, 177 North Direction Island, 26 North Keppel Island, 315 North Pickersgill Cay, 322, 354 North Reef, 453, 459 North Stradbroke Island, 111, 112 Fig. 4.9
523
North West Island, 324, 331, 362 Northumberland Island Group, 2 Fig. 1.1, 26, 160, 199, 371, 422, 424, 426 Northwest Reef/Island, 173, 459 Nymph Island, 78 Table 3.3, 325, 342 Okinawa, 15 Ollera Shoals, 229 Olympic Reef, 297 Fig. 9.18, 307 One Tree Reef/Island, 81, 83, 108, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 245, 247, 248 Fig. 8.8, 249, 250 Table 8.2, 253, 258, 260, 265, 266, 267, 268, 268 Fig. 8.15, 315, 319, 324, 336, 337, 349, 368, 370, 371, 375, 383, 387, 394, 396, 456 Onyx Reef, 417 Fig. 12.3 Opal Reef, 79 Orpheus Island, 63 Fig. 3.6, 109, 110, 124, 196, 200, 200 Fig. 7.4, 203, 207, 219, 222, 223, 224, 226, 315 Osprey Embayment, 20 Fig. 2.1, 27 Ouvea, 412 Oyster Reef, 141 Fig. 5.1, 142 Table 5.1, 415 Fig. 12.2 Palau, 396, 398–399, 405, 407, 416, 419 Fig. 12.5 Palfrey Island, 171 Palm Island Group, 2 Fig. 1.1, 26, 77, 82, 178, 424, 426 Palm Passage, 178, 418 Fig. 12.4 Paluma Shoals, 80, 193, 195, 195 Fig. 7.2, 197, 204, 207, 209, 220 Fig. 7.12, 221, 222, 229, 435, 449 Panama, 402 Pandora Cay, 324 Pandora Reef, 26, 318, 327, 330 Fig. 10.4 Pandora Trough, 20 Fig. 2.1, 27, 30 Papua New Guinea, 1, 3, 18, 43, 46 Fig. 3.1, 167, 170, 177, 368, 416, 422 Pelican Cay, 324 Penrith Island, 81, 83, 86, 204, 214, 215 Fig. 7.11, 368, 371, 376 Percy Island/Isles, 81, 376 Petherbridge Island, 325 Petricola Shoal, 188 Pickard Island, 322 Pickersgill Cay, 316, 317, 318, 350, 362 Pine Inlet, 97 Table 4.1 Pine Island, 93 Fig. 4.1 Pioneer Bay, 63 Fig. 3.6, 80, 203, 207, 208 Fig. 7.8, 219, 222, 223, 224, 226 Pioneer Bay, Whitsunday Coast, 196 Pioneer River, 21 Fig. 2.2, 22 Piper Island, 325 Pipon Island, 325 Pith Reef, 418 Fig. 12.4 Pixie Reef, 173, 354 Pompey Reefs/Complex, 2 Fig. 1.1, 40, 41, 53, 77, 83, 153, 159, 167, 168, 169 Fig. 6.1, 171, 173, 180, 233, 236 Fig. 8.1, 269, 271, 272, 274, 274 Fig. 9.1, 275, 276, 285, 296–309, 297 Fig. 9.18, 298 Fig. 9.19, 299 Fig. 9.20, 302 Fig. 9.22, 303 Fig. 9.23, 306 Fig. 9.25, 308 Fig. 9.26, 310, 345, 369, 370, 383, 413, 414, 415, 416, 417, 419, 423, 448
524
Geographic index
Port Douglas, 93 Fig. 4.1, 96, 97 Table 4.1 Portlock Reef, 2 Fig. 1.1, 38, 271 Poruma (see also Coconut Island), 317, 358 Potter Reef, 79, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 240, 249, 250 Table 8.2, 254, 376, 381 Price Cay, 324 Princess Charlotte Bay, 24, 32, 86, 105, 118, 159, 167, 171, 177, 184 Fig. 6.5, 189, 269, 317 Puerto Rico, 400, 402, 403, 437 Queensland Plateau, 19, 20 Fig. 2.1, 27, 29, 37 Queensland Trough, 20 Fig. 2.1, 27, 30, 168 Quoin Island, 26, 184 Fig. 6.5 RAAF Shoal, 285, 285 Fig. 9.9 Raine Island, 183, 236 Fig. 8.1, 272, 273 Table 9.1, 276–278, 277 Fig. 9.2, 9.3, 284, 369, 371, 373, 375, 376, 381, 384, 387, 451, 452, 453, 455, 464 Raine Island Entrance, 189, 271 Raine Reef/Island, 78, 312, 314, 324, 335, 345, 349, 352, 353 Rattlesnake Island, 80, 82, 204, 207, 314 Rattray Island, 105, 107 Red Sea, 42, 437, 459 Redbill Reef (see also Bushy Island, Bushy–Redbill Island), 26, 75, 81, 83, 86, 171, 200, 228, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 245, 246 Fig. 8.7, 249, 250 Table 8.2, 253, 258, 259 Fig. 8.11, 266, 268, 269, 368, 371, 372, 383, 453 Reef 17–065, 79, 178, 234 Table 8.1, 236 Fig. 8.1, 240, 250 Table 8.2, 376, 381 Reef 17–066, 178 Reef 20–374, 236 Fig. 8.1, 297 Fig. 9.18, 300, 448 Reef 20–389, 236 Fig. 8.1, 297 Fig. 9.18, 300, 448 Reunion Island, 399, 407 Rib Reef, 175 Fig. 6.2 Ribbon 3 Reef, 189 Ribbon 4 Reef, 189 Ribbon 5 Reef, 35, 35 Fig. 2.10, 36 Fig. 2.11, 38–39, 40, 53, 54, 77, 79, 236 Fig. 8.1, 272, 273 Table 9.1, 281, 281 Fig. 9.6, 282 Fig. 9.7, 283, 284, 289, 309, 371, 372, 381, 404 Table 11.2, 412 Riptide Cay, 236 Fig. 8.1 Rockhampton, 23, 178 Rodrigues Island, 394 Fig. 11.9, 399 Rollingstone, 54 Rosslyn Bay, 87 Russell River, 178 Rykers Reef, 79, 203 Sabai Island, 368, 369 Sabari Island, 420 Fig. 12.6, 422 Sand Bank Number 6, 322 Sand Bank Number 7, 322, 344 Sand Bank Number 8, 324, 344 Sand Cay, 322, 325 Sandfly Creek, 69 Sandpiper Cay, 322 Saunders Beach, 69 Saunders Island, 324
Saxon Reef, 417 Fig. 12.3 Scawfell Island, 81, 195 Fig. 7.2, 202, 204, 210, 376 Seychelles, 462 Shelburne Bay, 2 Fig. 1.1, 24, 142 Sherrard Reef/Island, 317, 325 Shoalwater Bay, 193, 229 Showers Reef, 449 Fig. 13.7 Sinclair Island, 325 Sir Charles Hardy Islands, 2 Fig. 1.1, 26, 184 Fig. 6.5 Slashers Reef, 418 Fig. 12.4 Small Detached Reef, 271 Society Islands, 395 South America, 136 South Direction Island, 26 South Johnstone, 93 Fig. 4.1, 97 Table 4.1 South Myall Reef, 79, 195 Fig. 7.2 South Pickersgill, 322 St. Croix, 400 Stainer Reef/Island, 78, 324, 332 Stanley Reef, 81, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 244, 250 Table 8.2, 251, 253, 255, 257 Fig. 8.10, 268 Fig. 8.15, 368, 369, 381 Stapleton Reef/Island/Cay, 234 Table 8.1, 235, 236 Fig. 8.1, 237 Fig. 8.2, 250 Table 8.2, 251, 252, 319, 324, 331, 332, 333 Fig. 10.5, 361, 373 Stephens Island, 34 Stewart River, 177 Stingaree Reef/Bay, 203, 207 Stone Island, 81, 311 Sudbury Cay, 322, 324, 330, 459 Surfers Paradise, 122 Table 4.2 Swain Reefs, 2 Fig. 1.1, 77, 100, 120, 121 Fig. 4.12, 153, 159, 160, 161, 169 Fig. 6.1, 171, 185, 187, 188, 190, 236 Fig. 8.1, 271, 272, 309, 331, 344, 345, 352, 362, 363, 414, 417, 423, 448, 451, 453, 458 Tahiti, 15, 57, 59 Fig. 3.5, 60, 254, 396, 397 Fig. 11.10, 398, 404, 407, 409 Tarawa Atoll, 393, 395–396, 404 Table 11.2 Tasman Sea, 18, 391 Taylor Reef, 79, 82, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 240, 249, 250 Table 8.2, 255, 267, 376 Tern Reef, 26 Thailand, 139 Thetford Reef, 79, 141 Fig. 5.1, 142 Table 5.1, 417 Fig. 12.3 Thimble Reef, 418 Fig. 12.4 Thomas Island, 324 Thread Reef, 418 Fig. 12.4 Three Isles, 78, 325, 326, 340, 343, 348, 350, 453, 453 Table 13.3 Thursday Island, 93 Fig. 4.1, 97 Table 4.1, 455 Tongue Reef, 377 Fig. 11.3 Torilla Plains, 82 Torres Strait, 1, 2 Fig. 1.1, 16, 20 Fig. 2.1, 26, 32, 34, 51, 54, 55, 57 Fig. 3.4, 77, 89, 92, 94, 100, 140, 170, 173, 174, 192, 205, 207, 209, 210, 211, 218, 231, 258, 260, 275, 276, 278 Fig. 9.4, 279, 301, 312, 317, 319, 320, 336, 337, 344, 345, 349, 358, 369, 370, 380, 381, 425, 453, 458, 461, 463
Geographic index Townsville, 22, 26, 30, 32, 34, 42, 71, 85, 86, 87, 93 Fig. 4.1, 97 Table 4.1, 100, 111, 112 Fig. 4.9, 121 Fig. 4.12, 122 Table 4.2, 123, 132, 133, 159, 173, 174, 182, 197, 213, 218, 222, 229, 258, 286, 314, 345, 370, 381, 383, 421, 435, 438, 449, 450 Townsville Trough, 20 Fig. 2.1, 27, 30 Tregrosse Reef, 29 Trinity Opening, 124, 177, 416 Tropic of Capricorn, 314 Tryon Island/Reef, 150 Fig. 5.7, 324, 362, 453, 455, 459 Tully River, 168, 177 Turneffe Islands, 400, 402 Turtle Reefs/Islands, 322, 324, 326, 340, 342, 348 Twin Reef/Cays, 318, 331 Two Isles, 326 Tydeman Cay, 345 Undine Cay, 322, 354, 459 Upolu Reef/Cay, 141 Fig. 5.1, 142 Table 5.1, 324, 332, 417 Fig. 12.3 Upstart Bay, 179 Valla Beach, 67 Vanuatu, 43 Victoria, 412, 413 Fig. 12.1 Viper Reef, 80, 173, 236 Fig. 8.1, 273 Table 9.1, 289, 292, 292 Fig. 9.14, 293 Fig. 9.15, 295 Fig. 9.17, 373, 381, 390 Vlasoff Reef, 142 Table 5.1 Vostok, 45, 46 Fig. 3.1 Waier Island, 22 Fig. 2.3, 34 Warraber Reef/Island, 78, 234 Table 8.1, 235, 236 Fig. 8.1, 237 Fig. 8.2, 250 Table 8.2, 258–260, 261, 266, 268 Fig. 8.15, 269, 348, 349, 376, 416, 420 Fig. 12.6, 453, 455, 456, 461, 463 Warrior Islet, 345 Warrior Reefs, 94, 184 Fig. 6.5 Water Park Point, 196 Waterwitch Cay, 322, 353 Watson Island, 322, 326, 338, 342
525
West Hope Island/Reef (see also East Hope Reef/ Island, Hope Reef), 238 Fig. 8.3, 326, 338, 340 West Hoskyn Island (see also Hoskyn Reef/Islands, East Hoskyn Island), 324 West Java, 432 Wheeler Reef/Cay, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 241 Fig. 8.4, 242, 249, 250 Table 8.2, 251, 252, 252 Table 8.3, 253, 254, 258, 260, 261 Fig. 8.12, 262, 266, 267, 269, 318, 322, 330, 345, 347, 354, 355, 356 Fig. 10.13, 362, 364, 373, 375, 381, 418 Fig. 12.4, 453, 453 Table 13.3, 455, 458 Whitsunday Island Group, 2 Fig. 1.1, 26, 82, 105, 107, 119, 123, 142, 159, 160, 168, 196, 199, 207, 218, 227, 425, 442, 443 Whitsunday Passage, 2 Fig. 1.1, 23, 438 Williamson Reef, 79, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 239, 250 Table 8.2, 251, 252, 270, 373, 381 Willis Reef/Island, 29, 93 Fig. 4.1, 97 Table 4.1 Wilson Island, 324, 453, 459 Wreck Reef/Island, 2 Fig. 1.1, 34, 38, 81, 83, 234 Table 8.1, 236 Fig. 8.1, 237 Fig. 8.2, 245–246, 246 Fig. 8.7, 249, 250 Table 8.2, 268 Fig. 8.15, 269, 324, 332, 383, 459 Yam Island, 57 Fig. 3.4, 59 Fig. 3.5, 78, 203, 207, 218, 368, 376 Yamacutta Reef, 150 Fig. 5.7 Yankee Reef, 179, 421 Yellowstone National Park, 447 Yeppoon, 87, 122 Table 4.2, 196 Ylig Bay, 437 Yonge Reef, 77, 78, 236 Fig. 8.1, 273 Table 9.1, 280 Fig. 9.5, 283, 283 Fig. 9.8, 284, 292, 369, 371, 376, 387 Yorke Island (see also Masig–Kodall), 320, 337, 345 Yosemite National Park, 447 Yule Detached Reef, 271 Yule Entrance, 236 Fig. 8.1, 275 Yule Point, 79 Table 3.3, 196, 197, 207, 221, 223, 224, 442
Subject index
Aboriginal/indigenous peoples, 312, 428, 429 Acanthaster plancii (crown of thorns), 135 Accommodation space, 39, 40, 48, 191, 210, 211, 214, 307, 402 Aeolianite (eolianite, dune calcarenite), 54 Table 3.1, 68, 205, 214, 391 Agassiz, Alexander, 8 Algae, blooms, 131, 444 coralline, 38, 42, 183, 196, 309, 370, 386, 387, 388, 396, 398, 399, 400 macro, 443, 444 Caulerpa sp., 185 Lithophyllum sp., 398 Pencillus sp., 185 Udotea sp., 185 Algal pavement, 113, 227, 228, 279, 338, 464 Algal ridges/rims, 64, 83, 258 microbialite reefs, 401–402, 406 Fig. 11.12, 409 Algal turf, 227, 338 Andrews, E. C., 8, 9 Antecedent karst control hypothesis, 13, 183, 305 Anthropogenic impacts, 3, 23, 134–135, 136, 192, 219, 226, 230, 232, 314, 428, 429, 432, 433–435, 434 Fig. 13.1, 435 Fig. 13.2, 438 Table 13.2, 439, 445, 459, 468 see also Cays, use; Fertilizers; Land-use, deforestation; Nutrients, enhanced; Reefs, stresses and disturbance; Reefs, use and management; Sediments, impacts on corals and reefs, yield to GBR; Tourism Aragonite saturation state, 42, 444 Atolls, 1, 3, 6, 396–399, 400, 402 Australian Institute of Marine Science (AIMS), 242, 315 Australian National Tidal Facility, 87 Bacteria, 443 Banks, Joseph, 6 Barnacles, 66 Basset edges, 226, 227, 330, 341 Fig. 10.11B see also Platforms, promenades, cemented
Beachrock/conglomerate, 8, 54 Table 3.1, 63 Fig. 3.6E, 67–68, 77, 82, 133, 228, 322 Table 10.2, 331, 332, 335 Fig. 10.8, 336, 338, 342, 343, 348, 349, 353, 354, 355, 358, 359 Fig. 10.15, 360, 393, 464 Beach ridges/barriers, 24, 25 Fig. 2.5A, 123, 174, 182, 196, 348, 413, 414, 416, 422, 428 see also Cheniers Beˆche-de-mer, 314, 452 Beechey, Frederick William, 6 Bernoulli effect, 104, 105, 189, 190 Bioerosion, 72, 444, 446 Bioturbation, 109, 224 Blackwood, Capt. F. P., 8 Bleaching, see Climate change (Greenhouse effect) Blue holes, 41, 300, 300 Fig. 9.21, 305, 392, 414, 448 Boulder beaches/tracts, 54 Table 3.1, 72, 77, 197, 200, 200 Fig. 7.4, 202, 214, 226, 227, 228 British Association for the Advancement of Science, 10 Bryozoa/bryozoan sediments, 29, 39, 170, 171, 187, 295 Bureau of Mineral Resources Canberra (now Geoscience Australia), 242, 244, 258 Calcrete, 62, 370 Callianassa sp., 109, 185 Cays (reef islands), 10, 249, 311, 468 cementation, see Basset edges; Beachrock/ conglomerate; Platforms, promenades, cemented classifications, 160, 315–343, 318 Table 10.1, 320, 327 Fig. 10.2, 364 Fig. 10.16 climate controls, 317, 353, 355, 357 Fig. 10.14, 358, 453–455 distribution, 312, 344 Fig. 10.12 dynamics/erosion, 117 Fig. 4.11, 314, 315, 316–317, 330, 343, 351, 352–358, 356 Fig. 10.13, 359 Fig. 10.15, 360–365, 445, 446 Fig. 13.6, 451, 452–455, 453 Table 13.3, 456, 457 Fig. 13.10A, 458, 463 effects of man-made structures, 446 Fig. 13.6, 454 Fig. 13.9B, 456–458
526
Subject index formation, 116, 346, 455 numbers/size, 148, 160, 161 Table 5.8, 322 Table 10.2 terraces, 332, 342, 343, 348–349, 351 types, multiple, 318 Table 10.1, 320, 337–338, 345 mangrove, 318 Table 10.1, 320, 337, 345 low wooded island, 160, 161 Table 5.8, 162, 226, 228, 235, 238, 249, 318 Table 10.1, 320, 322 Table 10.2, 327 Fig. 10.2, 338–343, 339 Fig. 10.9, 340 Fig. 10.10, 344 Fig. 10.12, 345, 347, 381, 427, 453, 458 unvegetated sand and shingle, 160, 161 Table 5.8, 240, 241 Fig. 8.4B, 242, 317–331, 318 Table 10.1, 320, 322 Table 10.2, 329 Fig. 10.3, 330 Fig. 10.4, 343–345, 344 Fig. 10.12, 350, 354–355, 356 Fig. 10.13, 453, 458, 459 vegetated sand, 160, 161 Table 5.8, 235, 239, 245, 247, 277 Fig. 9.2, 318 Table 10.1, 320, 322 Table 10.2, 331–335, 333 Fig. 10.5, 334 Fig. 10.6, 10.7, 344 Fig. 10.12, 345, 349–350, 355–358, 451 Fig. 13.8, 453, 457 Fig. 13.10B, 458, 459 vegetated shingle, 160, 161 Table 5.8, 247, 248 Fig. 8.8B, 318 Table 10.1, 320, 336–337, 349–350 vegetated, mixed, 335–336 use, 312–314, 450–452 see also Chapter 10 Chamisso, Albert, V., 6 Cheniers, 24, 71, 83, 174 Chlorophyll, 444 Climate, present, 93 Fig. 4.1, 94–99, 95 Fig. 4.2, 97 Table 4.1 Climate change (Greenhouse effect), 3, 44, 87, 92, 135–136, 220, 231, 438, 459–461 bleaching, 228, 450, 459, 460, 462 CO2 levels, 460 island effects, 459, 461, 463–467 ocean pH, acidity, 444, 460, 463 rainfall, 460, 461 sea level, 5, 88 Fig. 3.10, 446, 460, 461, 464 sea surface temperature, 459 Coastal plains, 19, 22–23, 414 Coastal ranges, 19 Coffee rock, 68, 196 Continental drift/plate tectonics, 18–19, 27, 28 Fig. 2.7, 29, 43, 392–393, 408 Continental islands, 18, 22, 26–27, 199–202 Continental shelf, 18, 27, 30–34, 31 Fig. 2.8, 146 Fig. 5.4, 147 Fig. 5.5, 274, 274 Fig. 9.1, 275, 279, 285, 290 Fig. 9.12, 291 Fig. 9.13, 309 Cook, James, 1, 6, 314 Corals, ahermatypic, 295 Lobophyton sp., 228 Sinularia sp., 228, 426 Corals, hermatypic, 42, 459 calcification/skeletal growth, 42, 444, 446, 460, 462, 465, 466 Table 13.4 community dynamics, 64
527
depth range, 42, 289 environmental influences, 42–43 larval dispersal/refuges/recruitment, 107–108, 373, 416, 425, 432, 438, 450, 468 sea-level indicators, 61–62, 64 tolerance to sediments, 432–439, 438 Acropora sp. 63 Fig. 3.6B, 216, 217, 223, 227, 228, 279, 386, 388, 392, 398 Acropora cervicornis, 400, 402, 407 Acropora dana, 400 Acropora palmata, 62, 400, 401 Fig. 11.11B, 409 Acropora robusta, 400, 409 Dendrophyllia sp., 187 Endophyllia sp., 289 Favia sp., 187, 223, 226, 386 Galaxea fascicularis sp., 221 Goniastrea sp., 212, 221, 223, 227, 228, 229, 400 Goniopora sp., 386 Leptoseris sp., 289 Montipora sp., 223 Pachyseris sp., 289 Pocillopora sp., 386 Porites sp., 130, 187, 212, 216, 217–218, 221, 223, 227, 228, 229, 386, 398 Platygyra sp., 386 Symphyllia sp., 386 Corals, morphotypes, 63, 279, 387 Crocodiles, 416 Currents mean circulation, 44 topographic modification, 103, 105–108 within reefs, 108–109 within reef framework, 109–110 see also Eddies; Tides Cyanobacteria, 131, 188 Daly, R. A., 8 Darwin, Charles, 6, 8, 9 Dates and dating Accelerator Mass Spectrometer Dating (AMS), 68, 349 ESR, 40 radio carbon, 13, 15, 40, 43, 60, 61, 67, 69, 74, 75, 76 Fig. 3.8, 77, 78 Table 3.3, 80, 124, 180–182, 188, 197, 203 Table 7.2, 204, 209, 210, 228, 233, 234 Table 8.1, 235, 250 Table 8.2, 251, 252 Table 8.3, 255–258, 262, 273 Table 9.1, 276, 283, 289, 305, 342, 347–350, 371, 381–383, 382 Fig. 11.6, 393, 395 palaeomagnetic, 40 strontium, 404 Table 11.2 uranium-series, 40, 51, 52, 371, 391, 395, 404 Table 11.2 David, T. Edgeworth, 6, 9 Davis, W. M., 7 Deltas, deltaic deposits, 22–23, 174, 177, 197, 207, 214, 220, 223, 415, 416, 436 Fig. 13.4 Denham, Capt. H. M., 8 Diagenesis, 62, 187, 262, 301, 368, 371
528
Subject index
Drilling programmes, 9, 13, 14, 14 Fig. 1.4, 15, 34, 35 Fig. 2.10, 36 Fig. 2.11, 38–40, 48, 209, 233, 234 Table 8.1, 236 Fig. 8.1, 256 Fig. 8.9, 257 Fig. 8.10, 259 Fig. 8.11, 261 Fig. 8.12, 263 Fig. 8.13, 272, 277 Fig. 9.3, 283 Fig. 9.8, 286, 289–290, 293 Fig. 9.15, 294 Fig. 9.16, 297 Fig. 9.18, 301, 304 Fig. 9.24, 367 Dugong, 412, 416 Dune calcarenite, see Aeolianite Dunes, 24, 25 Fig. 2.5B, 48, 129, 171, 174, 414, 428 East Australian Current (EAC), 100–102, 103 Ecology, 4, 468 Eddies, 105–107, 106 Fig. 4.7, 109 see also Currents; Tides El Nino-Southern Oscillation (ENSO), 87–89, ˜ 95 Fig. 4.2, 99, 135, 136, 216, 409, 429, 460, 461 Estuaries, estuarine sediments, 174, 177, 180, 416, 421, 423 Euphotic zone, 47, 57 Eutrophication, see Nutrients, enhanced Fairbridge, Rhodes W., 12, 314, 316 Fan deposits, 413 Fertilizers, agricultural, 444, 445 Flinders, Matthew, 8, 314 Floods, flood plumes/runoff, 3, 23, 92, 102, 125–128, 127 Fig. 4.14, 131, 139, 167, 199, 200, 213, 218, 230, 385, 386, 433, 436 Fig. 13.3, 440, 441, 443, 444, 446 Foraminifera, 39, 45, 71, 170, 171, 174, 182, 187, 227, 309, 349, 362, 386, 445 Amphistegina sp., 227 Baculogypsina sp., 227 Calcarina sp., 227 Freycinet, L-C de S., 6 Fringing reefs, 18, 23, 60, 103, 105, 106, 140, 148, 149, 152 Table 5.3, 5.4, 154 Table 5.6, 155 Table 5.7, 159, 160, 161 Table 5.8, 194 Fig. 7.1, 201 Fig. 7.5, 205 Fig. 7.6, 208 Fig. 7.8, 215 Fig. 7.11, 269, 396–399, 426, 438, 442 distribution, 193 Table 7.1 drilling data, 203 Table 7.2, 204 foundations, 203 Table 7.2, 204, 368, 369 incipient, 142, 148, 149, 152 Table 5.3, 152 Table 5.4, 154 Table 5.6, 155 Table 5.7, 159, 192, 196, 199, 218 lateral growth, 211–222, 212 Fig. 7.9, 226, 230, 231, 428, 448 low sea level, 416, 419–421 nearshore shoals, 193, 207, 209, 220 Fig. 7.12, 229, 449 vertical accretion, 211, 216, 219, 222, 225, 227–228, 231, 385, 386, 428 zonation, 222 see also Chapter 7 Gaimard, J. P., 6 Gardiner, J. S., 8
Geographical Information Systems (GIS), 4, 138, 140–145, 141 Fig. 5.1, 142 Table 5.1, 144 Fig. 5.3, 148, 162–163, 363 Geomorphology, 4–5, 431, 467–468 Gold mining, 444 Grainstones, 38 Great Barrier Reef age, 34–37, 38–41 dimensions, 142 Table 5.1, 146, 146 Table 5.2, 147–160, 152 Table 5.4, 153 Table 5.5 Great Barrier Reef Committee, 1, 8, 9, 10, 314 Great Barrier Reef Marine Park (GBRMP), 1, 2 Fig. 1.1, 140, 147–151, 162, 192, 447 Great Barrier Reef Marine Park Authority (GBRMPA), 3, 139, 146, 162, 199, 244, 432, 447, 448, 450 gazetteer, 139, 140–145, 141 Fig. 5.1, 143 Fig. 5.2, 151, 312, 343, 447 zoning, 142, 162–163, 163 Table 5.9, 164 Table 5.10, 199, 447 see also Representative Areas Programme (RAP) Great Barrier Reef World Heritage Area (GBRWHA), 312 Groundwater discharge to continental shelf (‘‘Wonky holes’’), 128–129, 131, 176, 444 Ghyben-Herzberg lens, 109–110, 110 Fig. 4.8, 465 layered-aquifer model, 465–467 Guano 276, 277 Fig. 9.2, 312, 336, 451, 452–455, 465 see also Phosphatic cay sandstone Halimeda (Halimeda bioherms), 39, 103, 105, 145, 151, 153, 166, 170, 171, 173, 174, 184 Fig. 6.5, 186 Fig. 6.6, 6.7, 262, 276–278, 284, 289, 371, 386, 395, 422, 423, 425, 442 Halosteric effects, 90 Hedley, Charles, 9, 314 Heinrich events, 52 High islands, see Continental islands Holmes, Arthur, 7 Holocene ‘‘high energy window’’, 64, 213, 218, 219, 226, 227, 362, 381, 425, 427–428 Holothurians, 187 Hydrographic survey, 12 Hydro-isostasy, 44, 47, 75, 83, 84 Fig. 3.9, 85, 269, 275, 343, 345, 347, 348, 351, 370, 380, 383, 410, 426, 445 International Consortium for Great Barrier Reef Drilling, 38–40, 239, 272 International Society for Reef Studies (ISRS), 15 Intertropical Convergence Zone (ITCZ), 96 Island research stations, 12, 315 Isostasy, 47, 58, 83–86 see also Hydro-isostasy James Cook University (of North Queensland), 14 Fig. 1.4, 236 Fig. 8.1, 242, 244, 258 Jukes, J. B., 8, 314
Subject index Karst (karstified reef), 41, 108, 110, 179, 398, 409, 414, 417 see also Antecedent karst control hypothesis; Pleistocene foundations King, Capt. P. P., 8 Land-use, deforestation, 439, 440, 442, 443, 444 Larvae, dispersal, 107–108, 425 recruitment, 432, 438 refuges, 373 Limestone islands (emerged reef, Makatea), 55, 180, 189, 409, 411, 412, 416, 419 Fig. 12.5, 420 Fig. 12.6A, B, 421, 422, 423, 429 Lithoclasts, 38, 40 Lithification (submarine), 390, 396 Lyell, Charles, 6 MacGillivray, J., 314 Madden-Julian circulation, 96 Mainland coastline, 23–24 Main Dividing Ranges, 19 Mangroves swamps, 24, 26 Fig. 2.6, 69 Table 3.2, 71, 174, 180–182, 196, 224, 411, 414, 416, 417 Fig. 12.3, 418, 418 Fig. 12.4, 421–422, 423, 424 Fig. 12.8, 425, 428, 429 on low wooded islands, 144–145, 238, 249, 337, 340–342, 340 Fig. 10.10, 342 peats and muds, 63 Fig. 3.6F, 68–71, 82, 83, 180, 181 Fig. 6.4, 173, 182, 209, 221, 348 Avicennia sp., 196, 224 Rhizophora sp., 196, 224, 337 Marchant, E. C., 10 Marginal plateaux, 27–29 Marshall, Sheina, 11 Maxwell, W. G. H., 12–13, 276, 305 Meltwater discharge pulses, 58, 59–60, 72 Microatolls, 4, 64–66, 227, 229 fossil, 63 Fig. 3.6C, 66, 75, 77, 82, 86, 211, 223, 227, 269, 342, 343, 347, 348, 381, 393 Microbialite reefs, see Algal ridges/rims Microfossils, 71 Milankovitch’s astronomical theory, 45 Moating, 64, 65, 227, 228–229, 230, 231, 232, 363 see also Shingle ramparts Mollusks, 170, 171, 180, 182, 187, 309, 349 Monitoring, 3, 5, 438, 467 Moorehouse, F. W., 314 Mudflats (saline flats), 24, 174 Navigation aids (light towers), 8, 276, 312, 329 Fig. 10.3, 451, 452, 456, 458 Notches, drowned, 53, 54 Table 3.1, 68, 412 Nutrients, 18, 28 Fig. 2.7, 35, 40, 42, 129, 133, 419, 428, 429 effects on corals, 443, 444, 446, 447 enhanced/eutrophication, 134, 296, 437, 442–444, 462, 465 see also Fertilizers; Land-use, deforestation
529
nutrient upwelling, 102, 104–105, 188–190, 296, 422, 429, 442 yield to GBR, 130–131, 130 Table 4.3, 131, 422, 443 Ocean Drilling Program (ODP), 15, 27, 32, 35, 74, 275 Oceanographic regions, 93 Fig. 4.1 Ooids, 171 Orr, A. P., 11, 314 Oxygen isotope analysis, 38, 39, 45, 48, 50, 51, 52, 74 see also Paleoenvironmental analysis Oysters, 63 Fig. 3.6D, 66–67, 70, 82, 86 Pacific Decadal Oscillation (PDO), 87, 89 Packstones, 38 Paleo-drainage channels, 70, 129, 174, 175–180, 175 Fig. 6.2, 178 Fig. 6.3, 414, 417, 421, 425 incision into shelf, 175, 176, 178, 179 Paleoenvironmental analysis, 16, 45, 135, 212, 217–218, 384, 409, 417 Fig. 12.3, 418 Fig. 12.4, 429, 444, 445 Paleosols, 39, 48, 62, 242, 368, 372, 398 Phosphatic cay sandstone, 276, 278, 335, 336, 349, 353, 465 see also Guano Photosynthesis, Photosynthetically Active Radiation (PAR), 42, 44, 432, 444 Pisonia grandis, 465 Plankton, phytoplankton, 103, 105, 133, 189, 443, 444 Plate tectonics, see Continental drift Platforms, promenades, cemented, 338, 343, 348 Pleistocene clays, 70, 170, 172, 187, 202, 214, 219, 220 Pleistocene foundations, 149 Fig. 5.6, 152, 170, 173, 180, 258, 269, 284, 288, 343, 389 Fig. 11.8, 411, 462 constituents and age, 205, 207, 209, 210, 214, 283–284, 370–372, 398, 404 Table 11.2 depth, 60, 234 Table 8.1, 250 Table 8.2, 252, 252 Table 8.3, 273 Table 9.1, 289, 301, 368–372, 369 Fig. 11.1, 374 Fig. 11.2, 383, 391, 393, 395, 398, 399, 400, 402, 408 morphology and influence on modern reefs, 233, 251, 254, 269–270, 387 Pollen, pollen analysis, 71, 212, 228 Purdy, E. G., 13 Quoy, J. R., 6 Rainfall and runoff, 96, 126 Fig. 4.13 see also Floods, flood plumes/runoff Reef islands, see Cays Reefs chronology/evolution, 265–270, 268 Fig. 8.15, 306 Fig. 9.25, 361 Table 10.3, 363–365 classification, 148, 149 Fig. 5.6, 150 Fig. 5.7, 152 Table 5.3, 247–249, 250 Table 8.2, 252 Table 8.3 distribution of reef types, 151–160, 156 Fig. 5.8, 157 Fig. 5.9, 158 Fig. 5.10 environmental controls and limits, 3, 42–43, 199, 212, 296, 391
530
Subject index
Reefs (cont.) facies/internal structure/processes, 109–110, 202–207, 255–260, 265–270, 386–390 Reefs, growth strategies catch-up (‘‘Katch’’ up), 193, 233, 383, 388, 394, 395, 398, 402, 407, 411, 425, 426 keep-up, 57, 60, 62, 254, 269, 295, 383, 384, 388, 395, 398, 399, 400, 401, 402, 407 turn-off/dead/give-up, 38, 40, 57, 216, 282, 282 Fig. 9.7, 283, 289, 293, 296, 384, 403 turn-on/take-off, 37, 49, 193, 222, 373, 390, 398, 402, 408, 418, 419, 423, 425, 461 Reefs, stress and disturbance, 3, 135–137, 216, 217, 222, 230 see also Anthropogenic impacts Reefs, structure and growth rates, 39, 210, 284, 289, 301, 375–380, 378 Fig. 11.4, 383–385, 392, 392 Table 11.1, 395, 396, 398, 399, 402, 409, 425, 427 detrital, 209, 210, 235, 249, 254, 258, 261, 262, 265, 268 Fig. 8.15, 284, 361 Table 10.3, 373, 376–380, 378 Fig. 11.4, 386, 387, 388, 390, 395, 399, 410, 447, 448, 449 Fig. 13.7 framework, 209, 210, 235, 239, 240, 262, 266, 284, 292, 295 Fig. 9.17, 373, 374 Fig. 11.2, 375, 376–380, 377 Fig. 11.3, 378 Fig. 11.4, 379 Fig. 11.5, 385 Fig. 11.7, 387, 389 Fig. 11.8, 390, 396, 399, 447, 448 Reefs, types size as influence, 235, 249, 254, 266, 270, 346–347, 348 see also Pleistocene foundations, morphology and influence on modern reefs types in evolutionary order submerged, 53, 57, 61, 142, 150, 152 Table 5.3, 5.4, 154 Table 5.6, 155 Table 5.7, 159, 161 Table 5.8, 163, 179, 183, 185, 199, 271, 272, 274, 275, 285–296, 285 Fig. 9.9, 287 Fig. 9.10, 288 Fig. 9.11, 298–299, 299 Fig. 9.20, 305, 360, 370, 403, 412, 413, 415, 443, 448, 450, 468 patch, 109, 150 Fig. 5.7, 152 Table 5.3, 5.4, 154 Table 5.6, 155 Table 5.7, 159, 161 Table 5.8, 163, 240, 247, 360, 427, 447 crescentic, 150 Fig. 5.7, 152 Table 5.3, 5.4, 154 Table 5.6, 155 Table 5.7, 159, 161 Table 5.8, 235, 239, 240, 242, 244, 247, 249, 252–255, 266, 360, 427, 447 lagoonal, 150 Fig. 5.7, 152 Table 5.3, 5.4, 154 Table 5.6, 155 Table 5.7, 159, 161 Table 5.8, 240, 243 Fig. 8.5A, 242, 244, 245, 246 Fig. 8.7A, 247, 248, 248 Fig. 8.8, 249, 251, 253–254, 255, 258, 265–268, 360, 447, 448, 449 Fig. 13.7 planar, 150 Fig. 5.7, 152 Table 5.3, 5.4, 154 Table 5.6, 155 Table 5.7, 159, 160–162, 161 Table 5.8, 163, 235, 239, 241 Fig. 8.4B, 242, 245, 246 Fig. 8.7B, 247, 249, 251–252, 253, 254, 258–260, 317, 338, 343, 345, 346,
351, 352, 358, 360, 361 Table 10.3, 363, 364, 449 other types deltaic, 272, 274, 275–276, 278–279, 278 Fig. 9.4, 299, 302 Fig. 9.22, 303 Fig. 9.23, 305–307, 412, 448 detached, 271, 272, 276–278, 349, 415 resorbed, 267, 305 ribbon, 64, 104, 142, 148, 152 Table 5.3, 5.4, 154 Table 5.6, 155 Table 5.7, 159, 161 Table 5.8, 163, 167, 183, 189, 251, 271, 272, 274, 275, 279, 280 Fig. 9.5, 281 Fig. 9.6, 286, 307, 390, 412, 413 Fig. 12.1, 415 Fig. 12.2, 416, 422, 425, 448 see also Fringing reefs Reefs, use and management, 138, 165, 312–314, 438, 447–459, 467, 468 see also GBRMPA Reefs, zonation, geomorphological features, aligned coral zone, 284, 381 boulders, 8, 124, 125 channels, 94, 104, 105, 174, 189, 244, 275, 278, 279, 298 Fig. 9.19, 299, 301, 305 double fronts, 240, 242, 244, 245 Fig. 8.6, 260–265, 264 Fig. 8.14 hardline edge, 240, 242, 243, 247, 255, 260 lagoonal patch reefs, 48, 240, 242, 244, 254, 255, 258, 260, 266, 368, 380, 387, 392 Table 11.1, 448, 449 Fig. 13.7 lagoonal ridge reefs, 301, 305, 307 lagoons, 241, 244, 265, 301, 305, 308 Fig. 9.26, 346, 392, 392 Table 11.1, 407, 448 reef flats and reef flat age, 249–253, 254, 380 reef slope, 228, 281 Fig. 9.6, 282 Fig. 9.7, 290 Fig. 9.12, 291 Fig. 9.13, 421 Fig. 12.7 reef terraces, 52, 53 sand slopes/aprons, 62, 266, 280, 284, 387, 388 spur-and-groove systems, 113, 279, 281 zonation, 113, 226–228, 229, 279, 284, 338, 380, 388, 415 Remote sensing, 102, 138–139 aerial photography, 12, 138, 139, 145, 183, 192, 199, 314, 343, 452, 455 Laser Airborne Depth Sounder (LADS), 145 Remotely Operated Vehicle (ROV), 73, 166, 272, 282, 285, 286–288, 412 Satellite imagery, 89, 90, 138–139, 140, 145, 147, 189, 192, 199, 343 Representative Areas Programme (RAP), 145, 162, 447–448 see also Great Barrier Reef Marine Park Authority, zoning Rhodoliths (see also Algae, coralline), 34–37, 38, 40, 227 Richards, Prof. H. C., 9, 314 Royal Geographical Society of Australasia, 9 Royal Society, London,
Subject index 1896–98 Funafuti Coral Reef Boring Expedition, 6, 9, 183 1928–29 GBR Expedition, 1, 7, 10 Fig. 1.2, 11 Fig. 1.3, 128, 138, 452 1973 Royal Society–Universities of Queensland Expedition to the northern GBR – 11, 11 Fig. 1.3, 13, 14, 228, 235, 251, 315, 317, 348, 452 Royal Society of New South Wales, 9 Salinity, 42 Saville-Kent, I. W., 314 Sclerochronology, 212 see also Paleoenvironmental analysis Sea-level change eustatic, 45, 47 GBR response, 53–55 high-frequency oscillations, 55, 74, 87 rates of change, 55–57 relative, 43, 44, 395, 399, 400, 402, 405, 408, 439 see also Hydro-isostasy; Isostasy Sea-level curves, 72, 73 Fig. 3.7, 77, 84 Fig. 3.9, 85, 406 Fig. 11.12 Sea-level indicators, 61–72, 63 Fig. 3.6 Sea levels, past Cainozoic/Quaternary, 6, 29, 30, 37, 44–49, 46 Fig. 3.1, 49–58, 272, 282 last glacial maximum (LGM), 44, 52–53, 54 Table 3.1, 272, 309, 412–416 last interglacial, 40, 51, 54 Table 3.1, 57 Fig. 3.4, 166, 269, 278, 305, 371, 408 late Holocene, 86–87, 214, 226, 228, 230, 362, 426, 428, 445, 463 low stands, 44, 271 mid-Holocene, 13, 23 Fig. 2.4, 24, 58, 66, 75–86, 211, 214, 223, 228, 229, 231, 258, 347, 351, 380 modern changes, 87–90, 88 Fig. 3.10, 102 post-glacial/Holocene transgression, 58–61, 59 Fig. 3.5, 72–75, 166, 209, 295, 296, 298–299, 372–375, 411, 416–426, 417 Fig. 12.3, 418 Fig. 12.4, 419 Fig. 12.5, 421 Fig. 12.7, 432 stadial/interstadials, 42–43, 50 Fig. 3.2, 52, 260, 274, 275, 309, 408, 421 see also Hydro-isostasy Sea Surface Temperatures (SSTs), 98–99 Seabirds, 314, 381, 412, 450, 451–452, 451 Fig. 13.8, 465 Seagrass, 133, 249, 362, 428, 429, 445, 446 Fig. 13.6 Sediment, 18, 23, 29, 37, 432–442 bedload, 129, 131, 440 Halimeda production and sediments, 188 impact on corals and reefs, 5, 133, 134, 231, 432, 435, 437–438 movement/dynamics, 24, 103, 106, 109, 116, 117, 118, 119, 120–123, 168, 224, 353, 427, 440, 453 Table 13.3, 461, 463, 465 offshelf flux, 57, 296, 423, 430
531
rates of settlement, 180, 182, 188, 433 Table 13.1, 437–438, 438 Table 13.2, 440–442, 441 Fig. 13.5, 453 Table 13.3 reef production (carbonate), 5, 167, 199, 217, 218, 230, 266, 267–269, 350–351, 424, 462, 462 Fig. 13.11, 463, 464, 466 Table 13.4 relict, 170, 171, 172 resuspension, 132–133, 213, 218, 224, 225, 428, 433, 439, 440, 441 shelf zones, 94, 167–168, 169 Fig. 6.1 suspended, 129, 131, 132, 432, 433 Table 13.1, 440 yield to GBR, 19, 127 Fig. 4.14, 130, 168, 170, 213, 218, 223, 225, 415, 419, 422, 424, 432, 433 Table 13.1, 435 Fig. 13.2A, 436 Fig. 13.3, 439–442 Sediment compartments, 24 Sediment traps, 11, 433, 435 Fig. 13.2, 437–438, 438 Table 13.2, 440, 453 Sediment wedge/prism (inshore), 131–132, 168, 174–175, 180, 182, 213, 213 Fig. 7.10, 219, 221, 428, 433, 439 Seismic survey, 13, 14, 27, 48, 49, 53, 170, 171–172, 173, 175, 175 Fig. 6.2, 176, 177, 179, 185, 187, 233, 242, 244–245, 247, 250 Table 8.2, 253, 261, 262, 272, 276, 284, 286, 288 Fig. 9.11, 297 Fig. 9.18, 299 Fig. 9.20, 301, 303 Fig. 9.23, 412, 413 Shingle ramparts, 64, 68, 72, 226, 227, 228–229, 230, 232, 335, 336, 338–340, 339 Fig. 10.9, 340, 341 Fig. 10.11, 342, 349–350, 442 see also Basset edges/promenades Shore platforms, 72, 412 Shorelines, submerged, 54 Table 3.1, 73, 412 Siboga Expedition, 183 Silcrete, 68 Solution unconformity, 14, 39, 48 see also Pleistocene foundations Spender, Michael, 7, 10, 314, 315 Spiculite (Alcyonarian), 426 Spits, island, 202, 207, 453, 454 Fig. 13.9A, 456 Sponges, 295 Stanley, G. A. V., 314 Stanley, Capt. Owen, 8 Steers, J. Alfred, 7, 10, 14, 314, 436 Fig. 13.3, 450, 451, 452 Stoddart, D. R., 11 Fig. 1.3, 14 Storm deposits, 64, 336 see also Boulder beaches/tracts; Shingle ramparts Submersible, manned, 35 Fig. 2.10, 185, 272, 281, 281 Fig. 9.6, 289, 412 Taphonomy, 217, 221 Tectonic activity earthquakes, 32–34, 33 Fig. 2.9 see also Continental drift/plate tectonics subsidence, 30, 32, 43, 48, 85, 191, 275, 292, 370, 372, 383, 425, 427 uplift, 19, 22, 43, 44, 63 Fig. 3.6A, 82, 83, 225, 266, 420 Fig. 12.6 Teichert, Kurt, 12, 314 Thermocline, 102
532
Subject index
Thermosteric effects, 90 Tides and tidal currents, 92, 99–100, 100 Fig. 4.4 currents, 94, 118, 132, 168, 170, 173, 179–180, 223–224, 275–276, 279, 286, 296, 298, 305, 423, 433, 435, 437–438, 440, 445 effects on waves, 114 internal tides, 102 jets, 103, 104–105, 104 Fig. 4.6, 189, 422, 442 range, 61–62, 64, 68, 75, 100, 101 Fig. 4.5, 217, 275, 279, 363 velocities, 103, 104 see also Eddies Tourism and the GBR, 314, 450, 451–452 Tridacna sp. 278, 336, 349, 350 Tropical cyclones, 4, 94, 96, 98 Fig. 4.3, 118–124, 221, 226, 354, 460, 462 cyclones Althea (1971), 123, 327, 355 David (1976), 456 Emily (1972), 456, 458–459, 465 Ingrid (1984), 124, 343 Ivor (1990), 119, 280–281, 379 Mahina (1899), 124 Manu (1986), 436 Fig. 13.3, 438 Sadie (1994), 128 Simon (1980), 456–458 Winifred (1986), 119, 127 Fig. 4.14, 167, 443 cyclonic surges, 118, 119, 120–124, 122 Table 4.2, 460, 462, 465 effects on reefs and islands, 119, 229, 230–231, 280, 336, 343, 345, 351, 358, 361–362, 379, 388, 407, 438, 442, 456, 458, 465 past frequency, 218, 337
Tsunami, 94, 124–125 Tubeworms (Serpulids), 66, 67 Turbidites, 38 Turbidity, turbid environments, 39, 57, 132, 133–135, 140, 197, 199, 210, 217, 218, 221–222, 224, 225, 229, 230, 385, 415, 419, 432–435, 437, 442, 449, 450, 467 Turtles, 133, 314, 381, 412, 416, 451, 452 Upwelling, 104, 105, 190, 450 Vermetid gastropods, 386 Volcanic activity, 19–22, 32 Volcanic islands, 22 Fig. 2.3, 392, 396 Waves fringing reef interaction, 117, 224 gauges, recorders, 111, 112 Fig. 4.9 interaction with reefs, 113–117, 117 Fig. 4.11, 224, 225, 307, 354, 398, 409, 435, 463–464 low frequency, 102–103 modeling, 111, 113, 116, 120–123, 121 Fig. 4.12 refraction and diffraction, 116, 117 Fig. 4.11, 161, 346, 352, 353 setup on reefs, 114 Fig. 4.10, 115–116, 123 shoaling/breaking, 113, 114–115 wind waves and swell, heights and periods, 111–113, 112 Fig. 4.9 Yonge, C. A., 10, 14 Younger Dryas, 51, 59, 72 Yule, Capt. I., 8 Zooxanthellae, 432, 444, 459