peony

gt Ne wth” NS Via sesame

goes Ju

_ Tel. (03) 836 1800

useum Victoria

PAPERS AND PROCEEDINGS OF

Tine Royall Society of Tasmania

Edited by M. R. Banks and S. J. Smith and Published by the Society Hobart, Tasmania

August 1988

Volume 122, Part I

Papers presented at the Macquarie Island Symposium Hobart

May 11-15, 1987

I. C. Carter, Acting Government Printer, Tasmania

ISSN — 0080-4703

COUNCIL AND OFFICE BEARERS

Patron

Her Majesty the Queen

President

His Excellency the Governor of Tasmania Sir James Plimsoll succeeded by General Sir Phillip Bennett

Vice-Presidents

Dr P.G. Quilty Mr C.A.H. Payne Council Mr I.D. Bennison Dr J.S. Rogers eae Cameron Mr R.D. Shepherd _ Mr G. Davis Mr C.B. Tassell Dr. P.W. McCartney Mr M.J. Walsh Prof. O.M. Roe ae

Honorary Secretary Mr D.R. Gregg

Honorary Treasurer , Mr T.A.H. Knight

Honorary Auditor Mr R.J. Dwyer

Honorary Editor Dr M.R. Banks

Honorary Assistant Editor Dr S.J. Smith

ill PAPERS AND PROCEEDINGS OF THE ROYAL SOCIETY OF TASMANIA Volume 122(1)

CONTENTS

Introduction BRYDEN, M.M. — Macquarie Island — a wonder spot of the world

Session One — Management

DAVIS, B.W. — Heritage conservation in antarctic and subantarctic jurisdictions: the case of Macquarie Island

Session Two — History and Archaeology TOWNROW, K. — Sealing sites on Macquarie Island: an archaeological survey

Session Three — Earth Sciences

WILLIAMSON, P.E. — Origin, structural and tectonic history of the Macquarie Island region

DUNCAN, R.A. & VARNE, R. — The age and distribution of the igneous rocks of Macquarie Island

JONES, T.D. & McCUE, K.F. — The seismicity and tectonics of the Macquarie Ridge

SELKIRK, P.M., ADAMSON, D.A. & SEPPELT, R.D. — Terrace types and vegetation dynamics on Macquarie Island

ADAMSON, D.A., SELKIRK, P.M. & COLHOUN, E.A. — Landforms of aeolian, tectonic and marine origin in the Bauer Bay-Sandy Bay region of subantarctic Macquarie Island

SELKIRK, D.R., SELKIRK, P.M., BERGSTROM, D.M. & ADAMSON, D.A. — ~ Ridge top peats and paleolake deposits on Macquarie Island

Session Four — Meteorology

STRETEN, N.A. — The climate of Macquarie Island and its role in atmospheric monitoring

ADAMSON, D.A., WHETTON, P. & SELKIRK, P.M. — An analysis of air temperature records for Macquarie Island: decadal warming, ENSO cooling and southern hemisphere circulation patterns

FRANCEY, R.J., PEARMAN, G.I. & BEARDSMORE, D.J. — A southern view of the global atmosphere

MALLIS, M. — A quantitative investigation of airborne sea salt over Macquarie Island

Session Five — Biological Sciences

BROTHERS, N.P. & COPSON, G.R., — Macquarie Island flora and fauna management — interpreting progress and predictions for the future

COPSON, G.R. — The status of the black-browed and grey-headed albatrosses on Macquarie Island

WOEHLER, E.J. & JOHNSTONE, G.W. — Banding studies of giant petrels, macronectes spp, at Macquarie Island

Session Six — Biological Sciences

BRYDEN, M.M. — Southern elephant seals as subjects for physiological research

HINDELL, M.A. & BURTON, H.R. — The history of the elephant seal industry at Macquarie Island and an estimate of the pre-sealing numbers

SHAUGHNESSY, P.D., SHAUGHNESSY, G.L. & FLETCHER, L. — Recovery of the fur seal population at Macquarie Island

Page

113 121

129 137

143

153 159

177

ROUNSEVELL, D. — Periodic irruptions of itinerant leopard seals within the Australasian sector of the Southern Ocean, 1976-86

MORGAN, I.R. — Viruses in Macquarie Island birds

KLEMM, M. & HALLAM, N. — Conceptacle development, gamete maturation and embryology of durvillaea antarctica from Macquarie Island

Session Seven — Biological Sciences

SIMPSON, R.D. — Developments in the studies of subantarctic shores, with particular reference to Macquarie Island

DAWSON, E.W. — The offshore fauna of Macquarie Island: history and biography — results from New Zealand and United States research cruises

WILLIAMS, R. — The nearshore fishes of Macquarie Island

PETERSON, J.A. & SCOTT, J.J. — Interrelationships between wind exposure, vegetation distribution and pollen fallout between Bauer Bay and Sandy Bay, Macquarie Island

SCOTT, J.J. — Rabbit distribution history and related land disturbance, Macquarie Island Session Eight — Physics COLE, K.D. — Importance of Macquarie Island for upper atmosphere physics

ESPOSITO, C. — A comparison of map systems with Macquarie Island monthly median ionospheric data

ESSEX, E. — Macquarie Island and the AUSSAT radio beacon experiment Session Nine — Physics

MALLIS, M. — Total ozone fluctuations observed at Macquarie Island: 1981-82

GRANT. I.F., BURNS, G.B. & COLE, K.D. — PiB and Pi2 geomagnetic pulsations at auroral break-up

JONGENS, S. & ESSEX, E.A. — A low-cost ionospheric doppler and propagation delay monitor

189 193

199

211

219 233

247 255

267

273 283

289 303

313

MACQUARIE ISLAND SYMPOSIUM HOBART, MAY 1987

Wireless Hill

LEGEND Plateau edge Lake Mountains

Walking tracks ~-7 7

Huts H

Hasselborough Bay Kilometres AMAR. The Isthmus Gadget Gully °

Langdon Point Nuggetts Point

Bauer Bay Mawson Point

12785ha

4 Zi PACIFIC Yat

Pb MACQUARIE

PACIFIC

N

Net

{// Hl Lusitania Bay Precarious Point

Hurd Point

Papers and Proceedings of the Royal Society of Tasmania, Volume 122(1), 1988 !

MACQUARIE ISLAND — A WONDER SPOT OF THE WORLD

by M.M. Bryden

BRYDEN, M.M.,

1988 (viii): Macquarie Island — a wonder spot of the world. Pap. Proc. R. Soc. Tasm., 122(1): 1-7.

Papers presented at the Macquarie Island Symposium, Hobart, May 1987. ISSN 0080-4703. Formerly University of Queensland, now Department of Veterinary Anatomy, University of Sydney, New South Wales, Australia 2006.

The geographical position, and geological and cultural history of Macquarie Island render it an important site for continued observations in upper atmosphere physics, meteorology, geology and many aspects of biology, especially those concerned with the Southern Ocean ecosystem and the control of introduced species. (Ed.)

Key Words: Macquarie Island, Southern Ocean, population studies, island flora and fauna, meteorology, aurora, mid-ocean

ridge.

INTRODUCTION AND HISTORICAL COMMENTS

The title of this address reflects the sentiment of Sir Douglas Mawson, who, ina paper read to the South Australian branch of the Royal Geographical Society of Australasia in 1919, stated:

“This little island is one of the wonder spots of the world. It is the great focus of the seal and bird life in the Australasian sub-Antarctic regions, and is consequently of far greater significance and importance in the economy of that great area than its small-dimensions suggest.”

This contrasts sharply with earlier statements (see Cumpston 1968, pp.60, 66) by Captain Douglass, master of the ship Mariner that visited Macquarie Island in 1822:

“As to the island, it is the most wretched place of involuntary and slavish exilium that can possibly be conceived; nothing could warrant any civi- lized creature living on such a spot, were it not the certainty of industry being handsomely rewarded...”,

and the Hobart Town Gazette of 4 February 1826, commenting on the advisability of ese a penal settlement there:

“The remote and stormy region in which Macquarie Island is placed, is a strong reason against the adoption of that sterile place as a penal settlement.”

Macquarie Island, at latitude 54°30’S and longitude 158°57’E, lies just to the north of the Antarctic Convergence, 1466 km (833 nautical miles) southeast of Hobart, and 1294km (735

nautical miles) from the antarctic continent. It is an exposed fragment of the crust of the Macquarie Ridge Complex which runs south from New Zealand, towards the Balleny Islands. The island is 34 km long and up to 5 km wide (120 km?), its long axis running roughly northeast, and comprises a long narrow plateau, 250-350 m above sea level, bounded on all sides by steep slopes (20°-40°) or cliffs rising either directly from the coast or froma raised beach terrace. Cumpston (1968) saw it as

“standing like a signpost pointing the way to the frozen lands and seas of the antarctic region ... the last vestige of land to be found in the vast expanse of the Southern Ocean to the south of Australia.”

Since the earliest days of visits and temporary occupation, Macquarie Island has been attached as a dependency to Tasmania. It is within the Municipality of Esperance. In 1890 an endeavour was made by the New Zealand Government to annex it, but this was unsuccessful (Mawson 1922). Subsequently, representations were made to trans- fer jurisdiction over the island to the Common- wealth Government, but that was never effected.

Early scientists to visit the island commented particularly on the abundance of wildlife, and its scientific interest. Prior to the Australasian Ant- arctic Expedition 1911-14, the only important contributions to the natural history of Macquarie Island were those of Scott (1883) and Hamilton (1895). Sir Douglas Mawson devoted the greater part of his paper “Macquarie Island and Its Future”, read to the Royal Society of Tasmania (Mawson 1922), to a discussion of the animal life of the island. It is clear that he saw that as scientifically the most interesting aspect of Macquarie:

2 M.M. Bryden

“

. the island is quadruply unique in the Australasian seas. Firstly, for the abundance of its life; secondly, for the variety of species frequenting its shores ...; thirdly, for the fact that it is the ... only possible breeding place for such life in those seas; finally, for the fortunate circumstance that up to the present man has not completely wrecked nature’s handiwork, though certain spheres of life formerly abundant are now extinct, and others so greatly reduced that they are in danger of complete extermination.”

This gives me some comfort in delivering the present address; although much has changed in the 65 years since Mawson’s address, I perceive biology at Macquarie Island as being most important still. However, the fact that I accentuate biology in this address reflects my knowledge and special interests, and not necessarily the relative importance of the various scientific disciplines. 1 have had lengthy discussions with a number of people in other scientific areas, and their opinions and advice have been most valuable to me in composing this address.

One might expect that Mawson would have emphasised the geology of Macquarie Island more than he did, instead of its biological significance. However, it is as a result of concepts developed in recent years, especially the concepts of plate tec- tonics, that the great geological importance of Macquarie Island has been realised.

The Australian National Antarctic Research Expeditions personnel have maintained a scientific station on Macquarie Island continuously since 1948. Research projects have concentrated in the disciplines of biology, geology, upper atmosphere physics, cosmic ray physics, and meteorology,

although other disciplines have been represented from time to time.

CONTRIBUTIONS OF RESEARCH AT MACQUARIE ISLAND

As I see it, the contributions to these different scientific disciplines of research at Macquarie Island fall into three major categories: (1) those involving collection of data at Macquarie Island, to be added to those gathered elsewhere, thereby providing a broader picture; (2) intrinsic studies, largely descriptive, that reflect the geological history and natural history of the island itself; (3) those that exploit the unique position or features of the

island, and potentially provide new scientific insights.

Broad Picture

I place in this category the fields of seis- mology, cosmic ray physics, and certain aspects of geology, meteorology and biology. Because the island is so far from any other land mass, meteoro- logical, seismological and cosmic ray information collected there provides a body of data in what would otherwise be a large void. This contribution is becoming less important in certain respects now, with access to advanced technology, but, nonethe- less, local observations do provide detailed in- formation not obtainable by other means.

For historical reasons, the island has pre- sented an opportunity to examine several general questions in biology: for example, the extensive depletion and subsequent recovery of populations of animals (royal and king penguins), colonisation or recolonisation (fur seals), and the effects of introduced species (rats, mice, cats, rabbits, wekas) on the ecology of the island. Further, the results of control or eradication of introduced species on the fauna and flora of the island could provide valuable insights. Ecological investigations of animal populations at Macquarie Island and antarctic stations have provided further information on which to base ecological concepts (see Carrick 1964, Carrick & Ingham 1967).

Descriptive Studies

The structure and ecology of Macquarie Island have been examined from many points of view. There is a substantial body of knowledge relating to its climate, geology, lakes and streams, flora, fauna and littoral zone; some oceanographic investigations have been conducted in the region. (For a comprehensive list of publications, see Selkirk et al. 1986.) A very brief summary is presented here, much of the information having been drawn from Jenkin’s excellent descriptions (Jenkin 1975).

The dominant influence on the climate of the island, associated with its location, is the surface temperature of the surrounding ocean. This, in turn, is influenced strongly by ocean currents. Asa result of this and of the position of the island, to the north of the Antarctic Convergence, the climate is extremely oceanic with a uniform temperature regime. Precipitation occurs in some form on most days of the year (in excess of 300 days). This resultant excess of precipitation over evaporation and consistently high humidity result in a moist, drought-free environment.

The rocks of Macquarie Island form an almost unaltered ophiolite complex of mid-Tertiary age, upthrust in association with tectonic activity along the margin of the Indian-Australian plate. Evidence of apparent glacial activity exists, but there is debate about whether, in fact, the island was glaciated during the last glacial maximum. It has been suggested that supposedly glacial features can be explained in terms of structural control, by faulting and rapid uplift.

The soils in areas of maximum exposure, as on the plateaux, are mainly gravelly loams. Else- where acid peats predominate, differentiated into various types according to drainage. The organic matter generally masks the effects of different parent materials.

Drainage of water from the plateaux is either directly via streams, or into the numerous lakes, some of which in turn are drained by streams, others by subterranean seepage. Periods of heavy rain usually result in increased erosion of the steep coastal slopes. Although it has been assumed that grazing by rabbits has denuded the slopes, causing extensive erosion, there is recent evidence that rabbit activity has had minimal effect on overall rates of erosion, even though it appears significant in localised areas (Griffin 1980, Selkirk e¢ al. 1983).

The chemistry of the island lake waters is influenced by the surrounding sea: salinity of lake waters depends on distance from the west coast, that is, in the direction of the prevailing westerly winds.

The flora comprises vascular species, mosses, liverworts and lichens. Broadly, grassland occurs on the coastal slopes, herbfield, fen and bog on the beach terraces and relatively flat, sheltered upland areas, and fellfield on the plateaux. With increasing altitude or exposure, the vegetation in all forma- tions shows a progressive reduction in size and vigour.

Because the island always has been isolated from other land masses, the establishment of its non-marine flora must have been by long-distance transoceanic dispersal. Numerous exotic pollen and spore types occur in Holocene lake deposits.

The fauna of Macquarie Island has received considerable attention. There are many publica- tions describing the species of invertebrates, including arthropods (more than 100 species), annelids, platyhelminths and a land mollusc on the island; littoral molluscs, echinoderms, brachiopods and chordates (ascidians); and marine bacteria and other micro-organisms, echinoderms, molluscs, nemerteans and arthropods (crustaceans). The vertebrates are spectacular, and large in numbers

Maquarie Island — awonder spot 3

but not in variety. In the seas surrounding the island, several species of fishes and certain cetaceans have been identified. Macquarie Island is a breeding ground for very large populations of seabirds and seals. A few species of non-marine birds have been recorded. The vertebrate wildlife was exploited for more than acentury from the time of discovery of the island in 1810, and this inevitably modified the original biota. The specific identity of the original fur seal, which was exterminated within a few years after 1810, is not known. An endemic rail and a parakeet became extinct, and rats, mice, wekas, cats, rabbits, a slug and a mite became established. Predation by cats probably has affected bird populations, particularly burrow-nesting species; burrowing and grazing by rabbits have modified vegetation to some extent and contribute locally to erosion.

Studies Exploiting the Unique Position or Features of the Island

Meteorology

The Australasian Antarctic Expedition in 1911-14 established the value of Macquarie Island as a meteorological station. F.M.B. Fisher, Minister in Charge of the Meteorological Department in New Zealand in 1913, stated, among other things, that he attached great importance and much value to the daily messages from Macquarie Island. A committee of the Australasian Association for the Advancement of Science stressed the importance of maintaining the wireless station at Macquarie Island in that

“(1) in the matter of meteorological forecasts and warnings to shipping, the station has already proved its value, (2) important general results are to be obtained by the scientific study of the weather conditions of the far south, as those conditions are related to those in the temperate regions.” (For sources see Cumpston 1968.)

These points are still valid as I understand it. When the antarctic station of Mawson was estab- lished, it was necessary to close one of the other stations. A major reason for retaining the station at Macquarie Island rather than Heard Island was that the former is more valuable as a meteorological station in relation to Australasian meteorological studies: Heard Island contributes only indirectly.

Overall, the benefits of Macquarie Island as a weather station for short-term forecasting in southern Australia and New Zealand have dimin- ished in importance as improved records from

4 M.M. Bryden

satellites have become available. However, at the same time the importance of Macquarie Island asa ground-truthing station has increased. It is a most valuable platform from which to monitor long- term changes or trends in the climatic record.

Upper Atmosphere Physics

Camps were established at the isthmus and Hurd Point, near the northern and southern ends of the island respectively, during the International Geophysical Year, 1956. Simultaneous observa- tions enabled accurate estimates to be made of the distance of the aurora, obtained by triangulation, a measurement that is surprisingly difficult to obtain.

Macquarie Island lies towards the equator from the southern auroral zone, forming a conju- gate pair with Kotzebue in Alaska. It has been Suggested that this conjugate pair is well positioned to investigate the characteristics of auroral sub- storms on the global scale. These regions are particularly suitable for conjugate investigations, because they lie within the closed magnetic field line region; in addition, post-breakup pulsating auroras, associated -with strong absorption in- creases, occur most frequently equator-wards of the auroral zone (Hajkowicz 1969).

Investigations of the visible aurora are made difficult at Macquarie Island because of the pre- valence of cloud cover. The interrelationship of the auroral phenomenon and other events in the upper atmosphere at Macquarie Island are somewhat similar to that in parts of New Zealand and southern Australia.

Geology

The Macquarie Ridge runs south from New Zealand, Macquarie Island itself being an exposed portion of the ridge. The entire ridge formed as a result of interactions along the congruent margins of the Indian—-Australian and Pacific plates. The island is a rare example of uplifted oceanic crust, detailed examination of which provides insight into how the oceanic crust was formed. Such insight is difficult, if not impossible, to obtain from studies of the ocean floor itself. Macquarie Island is, therefore, a critical area for the study of the Oceanic crust and ocean tectonics.

Zoology

Macquarie Island has been recognised as being important zoologically for a long time, as we have seen already. Because of its position, the only speck of land in the vast expanse of ocean to the south of Australia and New Zealand between latitude 52°S and the Antarctic Circle, it is an

important place to examine critical aspects of the ecosystem of the Southern Ocean.

Some descriptions and species lists resulted from the activities of members of the Australasian Antarctic Expedition 1911-14 and the British, Australian and New Zealand Research Expedition 1919-31, and in the early days of the Australian National Antarctic Research Expeditions (see references in Selkirk ef a/. 1986). A major upturn in biological research at Macquarie Island occurred in the 1950’s and 1960’s, when long-term investiga- tions of elephant seals and royal penguins, super- vised by Dr R. Carrick of the CSIRO Division of Wildlife Research, were undertaken. A series of papers by Carrick and others was published in CSIRO Wildlife Research, volume 7, in 1962. As an adjunct to those projects several more restricted but related investigations involving anatomy, physiology and ethology of seals and penguins were undertaken (see Bryden 1964 er seq., Ling 1965 et seq., Shaughnessy 1970a, 1974, 1975, Smith 1974), and studies of certain other species, par- ticularly birds, were made (see for example Mackenzie 1968, Purchase 1980, Shaughnessy 1970b, c, 1971, Simpson 1965a, b).

In the 1970’s the emphasis of zoology, by then supervised within the Antarctic Division, shifted to the antarctic continent, where extensive studies were centred at Davis. The major thrust was in limnology at that time. Monitoring of popula- tions continued at Macquarie Island, and an intensive investigation of the control of the annual cycle of elephant seals was conducted (see Griffiths etal. 1979, Griffiths & Bryden 1981, Griffiths 1984 et seq.). At the end of that decade the emphasis shifted again, when marine biological investigations began. This was an important development and was seen as a valuable contribution by Australia to the BIOMASS Programme (Biological Investiga- tions of Marine Antarctic Systems and Stocks).

In the last three or four years the importance of Macquarie Island zoologically, mainly as a result of its location, has been re-emphasised. The knowledge now that certain animal populations are undergoing dynamic change in the Southern Ocean has pointed up the need for major zoological studies at Macquarie Island to supplement the marine biology programmes. There is solid evidence that the elephant seal population at Macquarie Island has declined by 50% in the last 30-odd years (Hindell & Burton 1987), and some preliminary investigations of factors associated with that decline have been undertaken (Little e¢ al. 1987).

Throughout the 1970’s and early 1980's other aspects of biology continued. These included

— making collections of biota for selected studies of flora and fauna, in particular algae and aquatic invertebrates;

— continuation of long-term population studies of seabirds, particularly the wandering albatross, the light-mantled sooty albatross and the giant petrels; more recently, the black-browed albatross and the grey-headed albatross have been included;

— control programmes: initial surveys and experi- mental control of rabbits and cats were conducted as early as the 1960’s, and control procedures have been employed since then; control of rodents and wekas has been attempted more recently;

— monitoring seabirds: this represented an involve- ment in an International Survey of Antarctic Seabirds; photographic censuses of all the major penguin colonies on the island were carried out;

— making acensus of fur seals: general study of the fur seals at Macquarie Island was conducted as early as the 1950’s (Csordas & Ingham 1965); recently it has been recognised that there are three species of fur seal at the island, the New Zealand fur seal which is a visitor only, the antarctic fur seal and the subantarctic fur seal;

— recording and tagging leopard seals;

— studies of diet of the four species of penguin, gentoo, royal, king and rockhopper.

The Commission for the Conservation of Antarctic Marine Living Resources, now estab- lished in Hobart, has recognised for several years that high-level consumers, such as seals, penguins and other seabirds, may be valuable “biological indicators” of the condition of the Southern Ocean ecosystem. The long-term censuses and population studies that have been conducted over several years will continue to be most valuable in this regard.

THE FUTURE

There is little doubt that Macquarie Island continues to be an important meteorological station and presents opportunities to conduct unique investigations of the aurora and other aspects of the upper atmosphere. Exciting geological work is currently being conducted at the island. All these aspects will be addressed elsewhere in this volume.

The recognition by Mawson and some before him that the fauna of Macquarie Island is of principal importance is as true today as it was early this century. However, our perception of what constitutes its importance could hardly be more different from that of Mawson. In this regard it is interesting to reflect on the future management of

Macquarie Island — a wonder spot 5

the fauna of Macquarie Island put forward by Mawson (1919):

(1) ensure that only redundant bull elephant seals and decrepit animals are killed;

(2) _ kill any sea leopards that come ashore;

(3) reduce the number of skua gulls;

(4) investigate the possibility of a penguin egg industry;

(5) reduce, and if possible exterminate, the wild cats;

(6) breed a few sheep in a restricted area on North Head;

(7) introduce the fur seal and possibly other fur- bearing animals.

The potential value of seals and seabirds as biological indicators of the status of the ecosystem of the Southern Ocean has been mentioned above; censuses of these species should continue, in par- ticular the penguins and elephant seals. Detailed ecological investigations of these species may shed valuable light on the factors responsible for fluc- tuations in population size. Technology is being developed, and in some cases is available now, to examine their range, short-term and long-term movements, energetics, diving and foraging behaviour. It will be possible to compare these factors among age and sex classes.

The fur seals at Macquarie Island present an excellent opportunity to examine ecology and behaviour in an apparently developing population that may be unique. The interrelationships among New Zealand, antarctic and subantarctic fur seals, and possible interbreeding between subantarctic and antarctic species, potentially are a fertile field for ecological study.

Hitherto, physiological investigations of the fauna at Macquarie Island have been few. The opportunity for such studies provided by the island fauna is not unique on a world scale, but it does present the only practicable opportunity for such work by Australians in Australia. Recent investiga- tions have been aimed at examining the role of the pineal gland in southern seals (Griffiths & Bryden 1981, Bryden et al. 1986, Griffiths et al. 1986), and mechanisms of body temperature control in new- born seals (Little & Bryden, work in progress). It appears that changes in day length, which are marked in polar regions, are monitored by a mechanism involving the pineal gland, and are particularly important in controlling at least some annual cycles in the southern seals. It is likely that light, upon which most phytoplankton are depen- dent, has a significant influence on the marine

6 M.M. Bryden

fauna of the Southern Ocean at many trophic levels. The physiology of energy transfer within species is potentially extremely important to our understanding of the mechanisms of shifts in the Southern Ocean system.

CONCLUDING REMARKS

Insummary, Macquarie Island is as fascinat- ing today as it was 70 and more years ago when scientists first visited it. The animal life is of particular interest, as it was when Sir Douglas Mawson visited the island earlier this century; the strategic importance of the island in the Southern

Ocean has a different and even greater significance’

in this regard than was foreseen earlier. The declaration of the island as a fauna and flora reserve and, more recently, placing it under the control of the National Parks and Wildlife Service, Tasmania, have been vital. The introduction of measures to control feral species and ensure the continued well-being of natural populations have contributed significantly to improving the value of the island’s populations of animals as monitors of the Southern Ocean ecosystem.

As a result of investigations in a number of scientific disciplines, we have an appreciation of the geological history of the island and its flora and fauna.

Macquarie Island continues to be an impor- tant meteorological station. Largely as a result of its geographic position, useful data in a number of scientific spheres continue to be recorded there.

In the 1980’s we are reaping the benefits of wise planning and execution of scientific pro- grammes at Macquarie Island by the Australian National Antarctic Research Expeditions over the past 39 years.

REFERENCES

BRYDEN, M.M., 1964: Insulating capacity of the sub- cutaneous fat of the southern elephant seal. Nature, 203: 1299-1300.

BRYDEN, M.M., GRIFFITHS, D.J.. KENNAWAY, D.J. & LEDINGHAM, J., 1986: The pineal gland is very large and active in newborn antarc- tic seals. Experientia, 42: 564-566.

CARRICK, R., 1964: Southern seals as subjects for ecological research. In Carrick, R., Holdgate, M.W. & Prevost, J. (Eds): BIOLOGIE ANT- ARCTIQUE. PREMIER SYMPOSIUM OR- GANISE PAR LE S.C.A.R. PARIS 1962. Herman, Paris: 421-432.

CARRICK, R. & INGHAM, S.E., 1967: Antarctic sea-birds as subjects for ecological research. JARE SCIENTIFIC REPORTS. SPECIAL ISSUE NO.I1. PROCEEDINGS OF THE SYMPOSIUM ON PACIFIC-ANTARCTIC SCIENCES, TOKYO: 151-184.

CSORDAS, S.E. & INGHAM, S.C., 1965: The New Zealand fur seal, Arctocephalus forsteri (Lesson), at Macquarie Island, 1949-1964. C.S.1.R.0O. Wildl. Res., 10: 83-99.

CUMPSTON, J.S., 1968: Macquarie Island. ANARE Sci. Rep., Ser. A(1), Narrative. Publication no.93. Antarctic Division, Department of Ex- ternal Affairs, Australia.

GRIFFIN, B.J., 1980: Erosion and rabbits on Mac- quarie Island: some comments. Pap. Proc. R. Soc. Tasm., 114: 81-83.

GRIFFITHS, D.J., 1984: The annual cycle of the epi- didymis of the elephant seal (Mirounga leonina) at Macquarie Island. J. Zool. (Lond.), 203: 181-191.

GRIFFITHS, D.J. & BRYDEN, M.M., 1981: The annual cycle of the pineal gland of the elephant seal (Mirounga leonina). In Matthews, C.D. & Seamark, R.J. (Eds): PINEAL FUNCTION. Elsevier/ North Holland, Amsterdam: 57-66.

GRIFFITHS, D.J., BRYDEN, M.M. & KENNAWAY, D.J., 1986: A fluctuation in plasma melatonin level inthe Weddell seal during constant natural light. J. Pineal. Res., 3: 127-134.

GRIFFITHS, D.J., SEAMARK, R.F. & BRYDEN, |

M.M., 1979: Summer and winter cycles in plasma melatonin levels in the elephant seal

(Mirounga leonina). Aust. J. Biol. Sci., 32:

581-586. HAJKOWICZ, L.A., 1969: The role of post-breakup pulsating aurora in conjugate point absorption

increases. J. Atmos. Terr. Phys., 31: 1365-1370. |

HAMILTON, A., 1895: Notes on a visit to Macquarie Island. Trans. Proc. N.Z. Inst. 1894, 27: 559-579,

HINDELL, M.A. & BURTON, H.R., 1987: Past and present status of the southern elephant seal (Mirounga leonina Linn.) at Macquarie Island. J. Zool. (Lond.): in press.

JENKIN, J.F., 1975: Macquarie Island, Subantarctica. In Rosswall, T. & Heal, O.W. (Eds): STRUC- TURE AND FUNCTION OF TUNDRA ECO- SYSTEMS. Ecological Bulletins, Stockholm, 20.

LING, J.K., 1965: Functional significance of sweat glands and sebaceous glands in seals. Nature, 208: 560-562.

LITTLE, G.J.. BRYDEN, M.M. & BARNES, A., 1987: Growth from birth to 20 days of the elephant seal (Mirounga leonina) at Macquarie Island. Aust. J. Zool., 35: in press.

MACKENZIE, D., 1968: The birds and seals of the Bishop and Clerk Islets, Macquarie Island. Emu, 67: 241-245.

MAWSON, D., 1919: Macquarie Island. A sanctuary for Australasian subantarctic fauna. Proc. R. Geogr. Soc. Aust., South. Aust. Branch 1918- 1919: 71-85.

MAWSON, D., 1922: Macquarie Island and its future. Pap. Proc. R. Soc. Tasm. 1922: 40-54. PURCHASE, D., 1980: Giant petrel records from Fiji.

Notornis, 27: 406-506.

SCOTT, J.H., 1883: Macquarie Island. Trans. Proc. N.Z. Inst. 1882, 15: 484-493.

SELKIRK, P.M., COSTIN, A.B., SEPPELT, R.D. & SCOTT, J.J., 1983: Rabbits, vegetation and erosion of Macquarie island. Proc. Linn. Soc. N.S.W., 106: 337-346.

SELKIRK, D.R., SELKIRK, P.M. & SEPPELT, R.D., 1986: An annotated bibliography of Macquarie Island. ANARE Res. Notes, 38.

SHAUGHNESSY, P.D., 1970a: Ontogeny of haemo- globin in the Royal penguin Eudyptes chryso- lophus schlegeli. J. Embryol. Exp. Morph., 24: 425-428.

SHAUGHNESSY, P.D., 1970b: Genetics of plumage phase dimorphism of the southern giant petrel Macronectes giganteus. Heredity, 25: 501-506.

Macquarie Island — a wonder spot 7

SHAUGHNESSY, P.D., 1970c: Serum proteins of two sibling species of giant petrel (Macronetes spp.). Comp. Biochem. Physiol., 33: 721-723.

SHAUGHNESSY, P.D., 1971: Frequency of the white phase of the southern giant petrel Macronetes giganteus (Gmelin). Aust. J. Zool., 19: 77-83.

SHAUGHNESSY, P.D., 1974: Anelectrophoretic study of blood and milk proteins of the southern elephant seal, Mirounga leonina. J. Mammal., 55: 796-808.

SHAUGHNESSY, P.D., 1975: Variation in facial colour of the royal penguin. Emu, 75: 147-152. SIMPSON, K.G., 1965a: The dispersal of regurgitated

pumice gizzard-stones by the southern skua at Macquarie Island. Emu, 65: 119-124. SIMPSON, K.G., 1965b: First record of a grey plover at Macquarie Island. Emu, 65: 77-78. SMITH, G.T., 1974: An analysis of the function of some displays of the royal penguin. Emu, 74: 27-34.

(accepted 13 August 1987}

Papers and Proceedings of the Royal Society of Tasmania, Volume 122(1), 1988 9

HERITAGE CONSERVATION IN ANTARCTIC AND SUBANTARCTIC JURISDICTIONS: THE CASE OF MACQUARIE ISLAND

by Bruce W. Davis

DAVIS, B.W., 1988 (viii): Heritage conservation in antarctic and subantarctic jurisdictions: the case of Macquarie Island. Pap. Proc. R. Soc. Tasm., 122(1): 9-14. Papers presented at the Macquarie Island Symposium, Hobart, May 1987. ISSN 0080-4703. Formerly University of Tasmania, now Public Policy and Administration, School of Social Inquiry, Murdoch University, Perth, W.A., Australia 6150.

Macquarie Island, in the subantarctic, is in a position of great importance for programmes of scientific research and nature conservation, but its historic value in the area of heritage conservation should also be recognised. A management plan should be formulated for protection of the features of the island in all these aspects. Key Words: Macquarie Island, conservation, Antarctic Treaty.

INTRODUCTION

Most Australians know little of Macquarie Island but, if they have any vision of the place, probably assume it to be part of Antarctica, teeming with wildlife. In this symposium it is important not only to place on record current knowledge about Macquarie Island, but also to explain its contextual relationship to international programmes of scientific investigation and nature conservation. It is equally essential to recognise that Antarctica and the Southern Oceans are becoming a focus of international rivalries and concern; thus subantarctic dependencies such as Macquarie Island take on a new Strategic signi- ficance. This paper is intended to set the scene for more detailed expositions of characteristics of the island, but the political implications of its existence as part of Australia must also be recognised.

TERRAIN AND WILDLIFE

Other commentators will explain in their papers details of the island’s topography, climate, vegetation and wildlife, but here some basic facts are outlined.

Macquarie Island was discovered in 1810 by Frederick Hasselborough in the Sydney sailing brig Perseverance (Cumpston 1968). The island is located at latitude 54°37’S and longitude 158°54’E, in the Southern Ocean, approximately 1500 km southeast of Tasmania. When Van Diemen’s Land was proclaimed in 1825, Macquarie Island was included in its charter and has subsequently been considered a subantarctic dependency of Esperance municipality, Tasmania.

The island is approximately 12 343 hectares in extent, some 34 km long (north-south) and up to 5 km wide, with a rocky shoreline backed by steep

hills and cliffs, which form an undulating plateau 200-300 maz.s.|., with amaximum altitude of 433 m at Mt Hamilton. There are several adjacent sea- stacks and offshore islets. Macquarie Island is of volcanic origin, a horst block subsequently eroded by glacial action and possessing several lakes, but with no permanent ice. The island has a cold temperate climate with strong westerly winds, heavy cloud and a rainfall of approximately 900 mm per year.

There are no trees, but the island is extensively vegetated with tussock grasslands, herbs and sedges, and areas of peatbog. Numerous varieties of birds, penguins, elephant seals and fur seals abound; indeed, Macquarie Island is one of the great wildlife preserves of the world. Unfortunately, there are also some feral species, such as mice, rats, cats and rabbits; a mere remnant of diverse species introduced by man at various stages since 1810. Early interest in the island was entirely commercial, based on exploitation of penguin and seal resources for fur and oil. Macquarie Island was declared a sanctuary in 1933 and the only human habitation now extant is an Australian National Antarctic Research Expedition (ANARE) base in the north of the island, intended for meteorological and scientific use.

PROTECTION

Scientists began visiting Macquarie Island from 1833 onwards, and during the 1911-14 Australasian Antarctic Expedition (AAE), Sir Douglas Mawson established the first major scien- tific research station on the island. The efforts of various scientists, but principally Mawson, brought commercial exploitation of wildlife to a close by 1919 and in 1933 the Tasmanian Government declared the island to be a Wildlife Sanctuary

10. ~~ Bruce W. Davis

under the Animals and Birds Protection Act 1928. It was subsequently gazetted a State Reserve under the Tasmanian National Parks and Wildlife Act 1970, which came into effect in 1971-72, and in 1978 overall protection was upgraded to that of a Nature Reserve. All wildlife and natural features are now conserved and entry to the reserve is by permit. Alteration of regulations or any attempt at revocation of part of the reserve would require ihe assent of both Houses of the Tasmanian Parliament, so security of protection is high (TNPWS Report 1980). A detailed management plan is currently in preparation.

BIOSPHERE RESERVE STATUS

Macquarie Island is one of the twelve bio- sphere reserves within Australian jurisdiction, hav- ing been granted this status by the United Nations Educational, Scientific and Cultural Organisation (UNESCO) in October 1977, as part of its Man and Biosphere (MAB) programme (Rounsevell & Copson 1985). Biosphere reserves are areas which, because of their characteristic plants and animals, and the way they are used by people, provide scientific reference points to compare human impacts on the environment with the effects of natural forces. A biosphere reserve therefore con- tains both relatively pristine areas and zones which have been modified by human activity, scientific monitoring being used to assess the differences. There is usually a protected core of relatively undisturbed terrain and one or more “buffer zones” in which experimental work is carried out (Davis & Drake 1983).

How are biosphere reserves different from other protected areas? The International Union for the Conservation of Nature (IUCN) recognises ten different types of protected areas, each with par- ticular characteristics (IUCN 1978):

Category I Scientific reserve/strict nature reserve Category II National Park

Category III Natural monument/ national landmark Nature conservation

reserve/ wildlife sanctuary

Category IV

Category V Protected landscape Category VI Resource reserve Category VII Natural biotic

area/ anthropological reserves Category VIII Multiple-use management area Category IX Biosphere reserve Category X World Heritage site (natural).

It is fairly obvious that Macquarie Island could qualify for any or all of Categories I, II, IV or IX status, but overall the reserve possesses qualities mainly evident in Categories I and IX, beinga strict nature reserve, with limited areas assigned for scientific monitoring and investigation.

Australia’s system of biosphere reserves has an important international value in setting reference points for assessing air quality, pollution and radioactivity levels, meteorological effects, the impact of introduced species and environmental damage caused by human activities. There are, however, many deficiencies and omissions in Australian monitoring programmes (Davis 1986a), and it cannot be argued that a systematic and thematic investigation plan exists for Macquarie Island as yet. A recent report on the conservation of subantarctic islands, prepared by an IUCN/ SCAR (Scientific Committee on Antarctic Re- search) workshop (SCAR/IUCN 1986) does provide guidelines which should be followed in the future. Details of the\principal recommendations are given as Appendix A of this paper.

ANTARCTIC TREATY PROVISIONS

The Antarctic Convergence lies some 200- 250 km south of Macquarie Island; thus the nature reserve does not form part of Antarctica or auto- matically come under provisions of the Antarctic Treaty of 1959. In practice, however, there is a sensible and necessary connection. Many of the scientific programmes carried out on Macquarie Island are operated by personnel with antarctic interests, projects and experience; moreover, the data collected link in the comparative material from other subantarctic dependencies and the Antarctic Continent itself (Antarctic Division 1987). This is not to suggest that a// research has antarctic implications, but the nexus is considerable and strengthening.

The three principles upon which the Antarctic Treaty is based, namely demilitarisation, inter- national scientific co-operation and environmental protection, are as pertinent to Macquarie Island as to the frozen continent further south. These pre- cepts are further reinforced by Australian statutes regulating research in Antarctica and subantarctic islands such as Heard, McDonald (Keage 1981) and Macquarie. The relevant legislation is as follows:

~ the Wildlife Protection (Restriction of Exports and Imports) Act 1982;

— the Whale Protection Act 1980;

- the Environmental Protection (Impact of Proposals) Act 1974;

- the Quarantine Act 1908.

In addition, two statutes exist which give effect to agreements under the Antarctic Treaty system:

- the Antarctic Treaty (Environment Protection) Act 1980;

- the Antarctic Marine Living Resources Con- servation Act 1981.

As we shall discover later, some aspects of the Australian Heritage Commission Act 1975 are also relevant.

CAMLR

The Convention on the Conservation of Antarctic Marine Living Resources (CAMLR) was signed in Canberra on 21 May 1980 and came into force on 7 April 1982. The fifteen original con- tracting parties included all thirteen full members of the Antarctic Treaty, together with the Federal Republic of Germany and the German Democratic Republic. The convention is widely acknowledged by member states to be part of the Antarctic Treaty system and operates south of 60° latitude and the Antarctic Convergence. Its primary aim is to secure international co-operation in marine research and ensure the conservation of particular fish species and the ecosystems upon which they depend (Commission CAMLR 1984). In recent times the Australian Government has expressed concern about slow implementation of the Convention, particularly ecosystems management; nonetheless, some achievements have been recorded and more are in prospect (Commonwealth of Australia 1985).

In theory, CAMLR does not apply to the waters surrounding Macquarie Island, but in prac- tice there is a linkage through marine research programmes in the Southern Ocean. Australia is permitted to declare a 200-mile economic zone around Macquarie Island, under the provisions of UNCLOS III (the Law of the Sea), but has not yet done so. However, a 200-mile“fishing zone does apply, although it is difficult to envisage how this could be effectively monitored or enforced. In brief, our knowledge of marine resources adjacent to Macquarie Island is scant and needs to be improved in the future.

Heritage conservation I]

HERITAGE CONSERVATION

Thus far, the paper has focussed on aspects of nature conservation and scientific research on Macquarie Island, but this neglects another impor- tant dimension, namely heritage conservation. Here we are concerned with aspects such as historic shipwrecks, industrial archaeology and the built environment, as much as natural features. The Australian Heritage Commission, a federal statu- tory authority established in 1975, has jurisdiction in this field. The Commission is responsible for advising the Federal Government about policy relating to conservation of the natural, built and cultural environment, and also compiles an inven- tory of significant places, known as the Register of the National Estate. The latter is defined (Australian Heritage Commission 1985) as “... those places being components of the natural environment of Australia or the cultural environment that have aesthetic, historic, scientific or social significance or other special value for future generations as well as the present community”.

Except in special circumstances, listing of a place only highlights its significance and does not preclude any private owner, state or local government from exercising judgement about use of the place; however, the Act does bind the Commonwealth in a number of ways.

Section 30 of the Australian Heritage Com- mission Act 1975 requires all Federal Ministers to conserve National Estate values. Thus, if a Com- monwealth action will impact upon listed places, ministers must examine the situation, consult the Commission and take all prudent and feasible measures which will conserve the essential heritage characteristics of the area. Since Macquarie Island is listed on the Register of the National Estate, Federal agencies and employees are enjoined to minimise human impacts upon biota, natural features, marine or industrial relics or historic buildings. The work of documenting and listing specific sites and artefacts on Macquarie Island is far from complete, but was described in other contributions to this symposium.

STRATEGIC IMPLICATIONS

It is tempting to assume that Macquarie Island will remain a windswept, hostile but fascina- ting wildlife refuge, with little human occupation or impact, but the island must be viewed against the broader spectrum of international rivalries and global politics. Although Antarctica is frequently depicted-as.a zone of peace, there are rival territorial

12 Bruce W. Davis

claims and underlying political tensions; more nations are becoming interested in the region and wish to have some say in its management (Beck 1986, Hall 1986). The question is whether the Treaty is robust or flexible enough to meet new demands in the years ahead. Similarly, the CAMLR convention is under test and the Law of the Sea has induced some developing nations to seek a share in the spoils, through “common heritage” arguments (Davis 1986b). In brief, we must be cognisant with the reality that Macquarie Island could have an importance beyond its natural beauty; that in formulating management plans for the island, state, national and international factors must be

considered and the political implications of actions .

carefully weighed.

The stark grandeur of Macquarie Island, once witnessed, is not easily forgotten. May it continue to be haven of peaceful scientific research, aesthetic inspiration and wildlife conservation.

REFERENCES

ANTARCTIC DIVISION, 1987: APPROVED ANARE SCIENCE PROGRAMME 1986-87. Depart- ment of Science, Canberra.

AUSTRALIAN HERITAGE COMMISSION, 1985: AUSTRALIA'S NATIONAL ESTATE: THE ROLE OF THE COMMONWEALTH. Aus- tralian Government Publishing Service, Can- berra.

BECK, P.J., 1986: Challenges to the Antarctic Treaty regime: The external political challenge. In Millar, T.B. (Ed.): AUSTRALIA, BRITAIN AND ANTARCTICA. Australian Studies Centre, Institute of Commonwealth Studies, University of London: 60-79.

COMMISSION CAMLR, 1984: BASIC DOCUMENTS OF THE CAMLR CONVENTION. CAMLR Secretariat, Hobart.

COMMONWEALTH OF AUSTRALIA, 1983: THE ANTARCTIC TREATY SYSTEM. Australian Government Publishing Service, Canberra.

COMMONWEALTH OF AUSTRALIA, SENATE STANDING COMMITTEE ON NATURAL

RESOURCES, 1985: THE NATURAL RESOURCES OF THE AUSTRALIAN ANTARCTIC TERRITORY. AGPS, Canberra.

CUMPSTON, J.S., 1968: Macquarie Island. ANARE Sci. Rep., Ser. A(1). Antarctic Division, Department of External Affairs, Melbourne.

DAVIS, B.W., 1986a: The contribution of the social sciences to Australian antarctic research. ASAC FORUM, PAPERS AND SUBMISSIONS, Vol.2. Department of Science, Canberra, October 1986.

DAVIS, B.W., 1986b: Antarctica: Background to the present situation. /n Millar, T.B. (Ed.): AUS- TRALIA, BRITAIN AND ANTARCTICA. Australian Studies Centre, Institute of Com- monwealth Studies, University of London: 4-18.

DAVIS, B.W. & DRAKE, G.A., 1983: AUSTRALIA’S BIOSPHERE RESERVES: CONSERVING ECOLOGICAL DIVERSITY. Australian Na- tional Commission for UNESCO, Australian Government Publishing Service, Canberra.

HALL, R., 1986: Antarctica in World Politics, Unpub- lished M.A. thesis, University of Tasmania.

IUCN, 1978: CATEGORIES, OBJECTIVES AND CRITERIA FOR PROTECTED AREAS. Morges, Switzerland.

JOHNSTONE, G.W., 1985: Threats to birds on sub- antarctic islands. CPB (International Council for Bird Preservation) Tech. Bull., No.3.

KEAGE, P.L., 1981: THECONSERVATION STATUS OF HEARDISLAND AND THE McDONALD ISLANDS. Environmental Studies, Occasional Paper 13. Centre for Environmental Studies, University of Tasmania.

ROUNSEVELL, D. & COPSON, G., 1985: Southern ocean sanctuary. UNESCO Review, No.10, March 1985, Canberra.

SCAR/IUCN, 1986: The biological basis for conserva- tion of subantarctic islands. REPORT OF A JOINT SCAR/IUCN WORKSHOP, PAIM- PONT, FRANCE, SEPTEMBER 1986.

TASMANIAN NATIONAL PARKS AND WILDLIFE SERVICE, 1980: MACQUARIE ISLAND NATURE RESERVE. Hobart.

(accepted 17 February 1988)

Heritage conservation 13

APPENDIX A

CONSERVATION MEASURES FOR SUBANTARCTIC ISLANDS IUCN/ SCAR RECOMMENDATIONS, 1986

Recommendations

The workshop made the following unanimous recommendations:

1. Recommend

that the severity of the impact of introduced plants and animals on these sensitive island ecosystems be urgently assessed, that appropriate control mea- sures be instituted as soon as possible to minimise damage and that the ecosystems be monitored to assess recovery;

Encourage national authorities to conserve, under strictly controlled conditions, a sample of those introduced species whose genetic resources are regarded as potentially useful, if this does not conflict with the conservation of endangered indigenous species.

2. Recommend

that all islands be protected from any new accidental introductions by Man and that all necessary inspection and quarantine procedures to ensure this be brought into use as soon as possible;

Encourage national authorities to consider very carefully any voluntary or planned introductions of flora, fauna or soil to the islands, to subject such plans to rigorous and prior Environmental Impact Assess- ment, to monitor the effect of the introduction and, whenever practicable, to destroy or remove the introduction once its original purpose is complete.

3. Recommend

that particular consideration be given, when con- sidering fishery controls, to the dependence of these terrestrial ecosystems on nutrient input from the Southern Ocean; that marine buffer zones should be instituted (where practicable under extant legislation) to provide some limited protection of food resources;

Encourage appropriate national authorites to develop agree- ments for protection of these food chains under the Convention for the Conservation of Antarctic Marine Living Resources (CCAMLR),

Recognise that mortality in some bird and seal species during fishing operations or by the ingestion of plastic

debris gives cause for serious concern and could be minimised by appropriate national action.

4. Recommend

that each national authority develops and imple- ments conservation policies and plans, devised specifically for each island or island group, and incorporating a full consideration of the control of human impact on the natural ecosystem;

Encourage the implementation of scientific research and monitoring programmes necessary for the conser- vation management of the island ecosystems;

Recognise that although achievement of the objectives of conservation plans will be subject mainly to self- assessment, the use of independent observers appointed by each national authority is likely to contribute to greater success.

5. Recommend that island ecosystems as well as specific sites are accorded special legal protection to ensure their integrity is maintained;

Encourage national authorities to identify a fully representative series of strictly protected areas of restricted access, as well as sites of special scientific interest, and to consider which areas might be proposed for inter- national designation, as World Heritage Sites or Biosphere Reserves, for example.

6. Recommend

that organisation and control of stations, logistics . and scientific programmes be exercised to ensure minimal impact on the natural ecosystem, and that station development be constrained within a designated area; ;

Encourage national authorities to follow the SCAR guidelines on pollution control published for the Antarctic Treaty area.

14 Bruce W. Davis

7. Recommend

that historical sites and artefacts of shipwreck, ©

sealing, whaling or other human activities are mapped, documented and conserved as far as possible.

Encourage

the involvement of specialist advisors on archae- ology and history in this;

Recognise that this subject merits detailed treatment outside the remit of this workshop.

8. Recommend

that national authorities give immediate considera- tion to ensuring adequate education for all their island and ship personnel in the conservation objectives for each island in all their island groups;

Encourage both scientists and administrators to provide public education on the significance and value of the subantarctic islands; encourage responsible and controlled tourism;

Recognise that SCAR and especially IUCN could make an important contribution to global awareness of subantarctic conservation.

9. Recommend

that IUCN convene a meeting of island manage- ment authorities to discuss implementation of these recommendations;

Encourage national authorities, operating agencies and scien- tists to promote free and full exchange of all information and data, especially on those aspects which concern conservation and environmental protection of these unique islands.

Papers and Proceedings of the Royal Society of Tasmania, Volume 122(1), 1988 15

SEALING SITES ON MACQUARIE ISLAND: AN ARCHAEOLOGICAL SURVEY

by K. Townrow

(with five text-figures and three plates)

TOWNROW, K., 1988 (viii): Sealing sites on Macquarie Island: an archaeological survey. Pap. Proc. Roy. Soc.

Tasm. 122(1): 15-25, 3 pl. Paper presented at the Macquarie Island Symposium, Hobart, May 1987. ISSN 0080-4703. Department of Lands, Parks and Wildlife, Hobart, Tasmania, Australia,

A brief history of Macquarie Island is outlined with particular note of the different types of sites and sealing equipment used. Their spatial relationships and the evidence for technological change are viewed in the light of this historical evidence. The results of a small excavation challenge the historically derived view that living quarters were continuously repaired and/or re-used. Problems in the management of these resources are outlined.

Key Words: Macquarie Island, sealing, early Australasian industry, early technology, archaeology.

INTRODUCTION

Macquarie Island is a subantarctic island situated in the Southern Ocean at latitude54°30’S and longitude 158°57’E. It lies approximately 1500 km southsoutheast of Tasmania and 640 km west of the Auckland Islands. The island is approximately 34km long. Its width varies from approximately 250 mat the Isthmus to a maximum of 5km. The island is dominated by a central plateau area which is, on average, 150 m high. The coastal fringe offers no good all-weather anchor- ages. The fauna exploited by the sealers is concen- trated on this coastal fringe and includes fur and elephant seals and four species of penguins. The vegetation consists of low shrubs and tussock with no trees. The climate can generally be described as wet, windy, cold-temperate, foggy and humid with a small temperature range.

The cultural remains on the island fall into three categories: shipwreck material, Australasian Antarctic Expedition sites and sealing sites. The sealing sites include living quarters and production sites. Prior to the survey described here, limited work had been conducted on the island by Davies (1982). The current survey was conducted during the five-week Australian National Antarctic Research Expedition (ANARE) summer season in 1986-87. Its main objectives were to survey all the cultural remains on the island, assess their condition and the archaeological potential of the island, and make management recommendations. The survey was funded by a grant from the National Estate, through the Australian Heritage Commission. It was administered by the Department of Land, Parks and Wildlife.

MACQUARIE ISLAND

Ww

Isthmus

Nuggets

NORTH

Aurora Point

W_ wreckage

180°

Lusitania Bay 150 New\Zealand

Caroline Cove

Hurd Point ‘ 60°

FIG.1 — Macquarie Island, showing sources of cultural remains,

16 K. Townrow

The history of Macquarie Island is anexample of the industries of the new colonies, and their technology. Emphasis from the outset was on exploitation, with maximum gain for minimum input, rather than management for maintenance of the resource. This was especially so in the earliest stages. The fur seal population was rapidly devas- tated and the activity became non-viable; but there was a “fall back”. Elephant seal oil production quickly took over and lasted much longer. Penguins were exploited for a short period of time later. Initially the production of oil was by tryworks. This type of technology was crude and inefficient but productive. The more efficient digester tech- nology was only introduced about 80 years later, under the management of one man, Joseph Hatch, who may have been thinking about a more long- term industry, rather than the “hit and run” style of his predecessors.

HISTORICAL BACKGROUND

Macquarie Island was discovered in 1810 by Captain Hasselborough of the Perseverance while seeking fur seals (this, and other references in this section, from Cumpston (1968)). A gang was landed in July 1810 to assess the potential of the fur seal population and start harvesting. They con- sidered the potential to be good and tried to keep secret the location of the island, but they failed and by December 1810 a further three gangs had been landed. One of these started rendering down elephant seals, in addition to hunting fur seals. At this stage, gangs appear to have comprised about twelve men. The earliest record of a hut was of one built from the sails and skins salvaged off the wreck of the Campbell Macquarie in 1812, but it was soon accidently burned down. The visit of the Mary and Sally in 1813, however, indicates that there were huts at the Isthmus, as a gang of eleven men was landed. Tryworks had certainly been established, as men were sent ashore to put them in order (Cumpston 1968, p.30). This party also established living quarters and a tryworks at Sandy Bay.

It appears that the island was then worked seasonally by two to three gangs until the late 1820’s. Underwood and MacQueen were the main operators. Huts and tryworks were located at the Isthmus, Sandy Bay, Lusitania Bay and Caroline Cove. It is not clear when the Hurd Point tryworks were first established, but graves dating from 1821- 26, described by Cook in 1851, would suggest a starting point. Bellingshausen (a Russian explorer) mentioned separate gangs, totalling 40 men, working the island during his visit in 1820. By that

time, the only industry appears to have been elephant seal oil production.

Fur seals were slaughtered to the point of Virtual extinction within five years. Records of the number of pelts taken are contradictory, but it certainly exceeded 150000. Both historically and archaeologically, very little evidence has been found to substantiate the location or locations of this industry on the island.

Underwood ceased operations in 1828 and McQueen in 1829. Bellingshausen’s description of the headman’s hut at Sandy Bay (Cumpston 1968, p.45) provides an idea of the type of dwellings being used on the island at this time. It was

“20 feet long by 10 feet broad, inside it was lined with skins of seals, the outside was covered with a kind of grass which grows on the island. At one end there was a small hearth, and a lamp was always kept alight. Beside the hearth was a bedstead. Provisions were stored at the other end of the hut.”

Windows are described as bladders stretched over holes in the wall. Carpenters had been employed in various gangs, and it is assumed that timber brought down on ships (and perhaps some shipwreck material) was used to form, at least, a framework for the hut. From 1830 to 1874, there are only records for three trips to Macquarie Island. John Cook’s gang of 14 men worked the North End in 1851. His diary provides an insight into. the scale of operations in former times. He mentions huts and old casks at a place that appears to have been Unity Bay, and describes the South End works as consisting of houses, tryworks,.two cooper’s shops and seven graves. Of North Head he states (Cumpston 1968, p.78) that

“Our flat seems to have been where the most business has been done. Over one dozen old houses and a great quantity of bones and ... hoops (are) buried here; there are Tri-pots set all about in different places; there are 2 graves on the flat ...”.

Between 1860 and 1873 there are no shipping records of visits to Macquarie Island at all. There was a brief period of greater interest in elephant seal oil production between 1874 and 1884 by Nichol and Tucker, followed by Cormack, Elder and Co. Activity appears to have been concentrated initially between the Isthmus and Sandy Bay, with repairs and additions to huts. Existing trypots were re-used with relocation of some to Nuggets Creek. Cormack, Elder and Co. gradually extended operations over the ten-year period to include

Lusitania Bay and Hurd Point. Gangs were generally smaller than in early years, averaging eight men.

Intermittent working continued on the island until 1889 when Joseph Hatch began to dominate operations. Hatch was a New Zealand entrepeneur who became involved in the trade initially to supplement his pharmaceutical business. He had gangs of about eight men operating on the island for the next 30 years. During this period, the transition was made from a low-technology, small capital operation to a more intensive and expensive industry. Headquarters were initially at Lusitania Bay. Here, elephant seals were rendered down and the exploitation of the king penguin colony began. Digesters were introduced at Lusitania in Novem- ber 1889. Coal, timber and firewood then became regular items on the list of stores taken to the island. Depots in other parts of the island were also used. For example, the Gratitude collected oil from North End, South End and Lusitania Bay in February 1891. z

In early 1892, Hatch’s headquarters were transferred to the Nuggets. A gradual transition from the exploitation of king penguins to royal penguins also seems to have begun. This was possibly because of the problems of fermentation of the king penguin oil, since it had not been possible to reduce the blood content in it, and/or too great a reduction in king penguin numbers. Digesters and boilers were also set up at the Isthmus, Hurd Point and Hasselborough Bay. The industry then settled into a two-part operation conducted in two seasons: the royal penguin season operated from the Nuggets during November and December and the elephant seal season operated from North Head from mid-August until the end of October. Hatch’s confidence in being able to maintain a continuing operation on Macquarie Island seems to be reflected in the introduction of digester plants. He also secured a sole occupational licence in 1902.

However, opposition to his operation began during the 1910’s. Headed by Mawson, who had led the 1911-14 Australasian Antarctic Expedition, a campaign to declare the island a nature reserve and to condemn the royal penguin industry, in particular, gained momentum. After a brief respite in this campaign, during World War I, it seems to have become somewhat inflammatory. Claims, often by people who had never been to the island, became increasingly exaggerated. Despite continual public denials by Hatch of these exaggerated claims, he was finally forced, through the cancella- tion of his licence, to cease operations in 1920. Even

Archaeological study of sealing sites 17

without this cancellation, he might not have been able to continue, due to increasing financial difficulties.

ARCHAEOLOGICAL EVIDENCE

An archaeological survey of the island was undertaken in late 1986-early 1987. The sealing activities have left three main categories of remains; shipwreck material, living quarters and two types of production sites. Although all categories are interrelated, they are not necessarily all represented at every site. Living sites are distinct from produc- tion sites, in that they contain either tryworks or steam digester plants. There would appear to have been little or no intermingling of the functional use of the huts.

Shipwreck Material

Without diving it was only possible to observe the material which has gathered along the coastlines in the form of flotsam and jetsam. This is mixed with some of the refuse of the Southern Ocean. The material is concentrated on the east coast between North Head and Sandy Bay and right along the west coast. The east coast scatter is undoubtedly related to the concentration of known shipwrecks in this area. The scatter on the west coast is probably attributable to its windward location, in the path of the general drift of the Southern Ocean. The last obvious remains of the Gratitude, which was beached at the Nuggets, were carried out to sea in 1975-76 (G. Copson, pers. comm.).

Living Quarters

Quarters can be found at the Isthmus, the Nuggets, Lusitania Bay and Hurd Point. There are possibly remains of living quarters at Sandy Bay but they have deteriorated to the point of being nebulous.

Three huts were identified at the Isthmus. All are located under the eastern lee of Razorback Hill. Associated with these huts was a large scatter of artefacts, including parts of an iron stove, a clay tobacco pipe bowl, barrel staves and assorted timber.

The Nuggets living quarters consist of a large and small hut. The larger is more clearly defined by a row of large beach stones on its seaward side. The layout of floorboards at this hut suggests that the smaller, sleeping area was at the northeast end of the hut with a larger, communal/store area at the other. Adjacent to the western corner of the hut is a

18 K. Townrow

=| 2 planking = «© a=

ISTHMUS LIVING QUARTERS

NORTH rs 1 » ? 5 RAZORBACK HILL 10 metres ‘ Was =e wy in a ~ wallow ' 3 Ds glass 4 cr NA iat ° = N Noah to corrugated iron planking =~ \ 0 = ky \ ~ oy fas L a a °o metal ring Joven fragment co\y AAS * corrugated iron ¢ 2 metal pulley Ik I TRENCH a ~ 7 shoe leather ~ Hasselborough Bay a ay, i rinane sty MES a Rae lO Statio® Se 1 Ay — . ° EM ° al Artefact Corral’ 7, p? S o Transmitter 8

put THE ISTHMUS

26/12/86 SCALE 1:7 500

é x

FIG.2 — Isthmus living quarters, showing location of artefacts.

Archaeological study of sealing sites 19

PLATE 1 Hurd Point tryworks, an example of the problems of archaeological survey during the penguin breeding season.

small outhouse, which may have acted as an extra store or pantry area. The small living quarters are defined by some corner posts. Associated with the huts are parts of a stove, a standing mooring post, bricks, a mallet head and bottle glass fragments.

Both the Hurd Point and Lusitania Bay living quarters were almost totally obscured when fieldwork was being conducted, because both sites are situated within large royal or king penguin colonies (plate 1). However, one living hut was positively identified at each site. The Hurd Point hut lies on the western edge of the colony, perched half-way up the beach bank. As with the Isthmus huts, it had stilts on the seaward side. The Lusitania hut was much smaller than the others and, at this stage, could only be identified by four corner posts, an anvil, barrel base and timber scatter.

Production Sites

These can be divided into two types: trypot works and digester works. Evidence of tryworks is present at Hurd Point, Caroline Cove, Lusitania Bay and the Isthmus. Of these sites, Hurd Point is the most intact example, with a cluster of six trypots and the scattered remains of three others in the general vicinity. The cluster of trypots is arranged so that one feeds into two. Two such

groups are positioned north-south down the beach bank. Remains of a building which housed these works are present. At the other sites, tryworks are represented by two trypots at Caroline Cove, one of which is three-quarters buried; one at Lusitania Bay with associated timber; and one at the Isthmus, almost totally buried.

Digester works can be found at the Isthmus, the Nuggets, Hasselborough Bay, Lusitania Bay and Hurd Point. Both Hurd Point and Hassel- borough Bay have only one small digester and one boiler. Remains of the sheds which housed them are evident at both locations. Lusitania Bay has one small and one large digester (plate 2) and a boiler, which were all originally housed in a large shed. The Nuggets has a group of two small and two large digesters fed by three boilers within a shed. Two discarded boilers lie to the southwest of the site. The Isthmus works consist of two digesters, one boiler, and a discarded boiler to the east of the site. Both these sites also have the foundations of the sheds which housed them. Artefacts associated with this type of site include tools, such as vices and anvils, barrels, coal and settling tanks. The Nuggets and Lusitania Bay sites both appear to have piped in water to the boilers. At the Nuggets, there are two stationary pumps which would have been used for this purpose.

20K. Townrow

PLATE 2 Lusitania Bay, steam digester plant.

TUSSOCK LINE

ay ee x

living quarters

trypot °

trypot ‘ fragment \ oa

is J aye

FIG.3 — Hurd Point production site, showing location of artefacts.

EXCAVATION

A Im by 6m trial trench was excavated through one of the huts at the Isthmus (plate 3). The excavation was conducted in an attempt to determine the effectiveness of this form of investi- gation on the island. The specific site was chosen because the hut was known to have been used by Hatch’s men in the 1910’s (archival photographs), thereby providing an end-date to occupation at this site. Hence it could be more reliably assessed whether the hut represented the re-use of an earlier construction. The trench was orientated to run east-west, covering what was thought to be the side of the structure. Standard archaeological procedure was followed, although the soil was not sieved for artefacts. ’

The trench established that considerable quantities of material are being obscured by vegetation. The main location of this material was just beneath the regrowth which has been the principal agent in its preservation. Unfortunately the trench did not expose the side of the hut but appears to have uncovered a “cold porch” or workshop area where tools and the horse harness could have been stored. The horse was undoubtedly one of the Clydesdales used by Hatch’s men to pull the blubber sled, during the later phase of his operations. Only the floor joists and posts with some wall planking, complete with tar (bitumen insulation) paper, remained of the hut. It, therefore, seems likely that the floorboards were scavenged at a later date and perhaps used to repair other buildings or as fuel. The hut appears to have been stilted to the east (this is confirmed by archival photographs). It was not possible to determine whether the hut had been refurbished (the trench proved to be too small), although it may not have represented an original 1820’s hut, since a penguin bone refuse pile was found under the floor joists which may pre-date the hut. The total depth of occupation at the western end was I.1 m. It was generally found that the state of preservation of the artefacts was good. Leather, iron, fabric and wood artefacts were in reasonable condition, although, due to the acidity of the soil, bone artefacts were not as well preserved. The finds included a cart- horse harness, various iron tools, guttering, wall insulation, glassware, domestic pottery, elephant seal teeth, two spent cartridges from different calibre firearms and parts of a cupboard or tea-chest.

Archaeological study of sealing sites 2]

PLATE 3 — Trial trench, Isthmus living quarters, after removal of vegetation.

DISCUSSION

From this evidence the following comments can be made:

(1) Where living quarters and production sites are found at the one location, the living quarters are situated 25-50 m upwind of the production sites, unless confined by topographical restraints (for example see fig. 3). The reason for this seems to be purely one of relative comfort. The smell of rancid blubber from the boiling process was notorious.

(2) Living quarters appear to have a rather limited life. No evidence was found of the early tussock-with-wood frame huts. Remains of huts were identified at the Isthmus, some of them possibly from the dozen huts described in 1851, but none could be positively established as such. The historical records mention constant repairs to huts, but the limited excavation undertaken suggests that the same huts were not continuously repaired.

22. K. Townrow

BB v.04 leather [+] tron [-] chaik pottery (_] otass

EAST EAST

FIG.4 — Isthmus test trench — horizontal layout; numbers refer to different levels of excavation.

23

Archaeological study of sealing sites

*‘SUOIJIAS [DI1]49A — YIUAA Isa] snulyIs] — C DJA

aqjow 4

sy901 Jews YM pues Aai6 pue moljeA 2 (J

souog daays yyIM pues uMOsG-pal 9 [<j os" sauoq deays ym pues Aaib ¢ Fy << ae

0

pues umoiq-pal y fe) eam

yead Aafejo pas pue Aoib ¢ (4 sayoyed Aejo ym pues Aaib Z (J

yead umojg-pal |} ES} HINos NOILOSS LSVA

HLYON

asjow 4

NOILOSS LSAM

HLNOS

sajew g

LSAM

NOILOSS HLNOS

sojew 9

ISAM

NOILOAS HLYON

isva

24 ~~ K. Townrow

Instead, when a hut was beyond repair, material was salvaged and re-used in building a new hut ina new location. However, additional excavations would need to be carried out to confirm this or highlight differences between sites.

(3) The tryworks reflect the longer industrial phase and have consequently suffered the most from re-use and/or relocation of the pots from their original situations. The loss of context of any artefact is directly related to its portability. The trypots, being relatively portable, compared to boilers and digesters, have been more subject to this problem.

(4) The trypots show no uniformity of size or design. They have little, if any, decoration. This reflects the low level of technology required for the boiling-down process.

(5) This simple technology was in use during the peak period of production between 1810 and 1830 (Hindell & Burton, this volume). The obvious reason contributing to the relative success of this technology, compared to the “higher technology” of steam digesters, was greater availability of the resource, but other factors may have contributed, such as higher labour input per gang and, in total, stimulus of rivalry between gangs or the stimulus of higher profitability.

(6) The use of digesters in the production of seal and penguin oils would appear to be unique to Macquarie Island. It is likely that they were adapted from similar plants used on whaling boats to render down blubber. In the whaling industry, similar types of digesters, of Norwegian design, were in use by 1889 (Tonenessen & Johnsen 1982). At the time of their introduction on Macquarie Island, at Lusitania Bay in 1889, therefore, they may have been “state of the art”,

(7) The digesters appear to represent two phases of development; the smaller and single sets repre- senting the trial usage and the larger representing the more cost efficient locations for this type of machinery. Size may also be related to the resource being used. The larger digesters are present at the Nuggets and Lusitania Bay, indicating that a larger size of digester was viable where large numbers of easily obtainable penguins were available, but was not necessary or viable when rendering down elephant seal blubber.

(8) The digesters and boilers represent an increas- ing availability of large manufactured machinery during the late nineteenth and early twentieth centuries. They are standardised in design and manufacture, made from hot-rivetted sheet iron. There are distinct sizes. The digester works, there- fore, may be said to reflect the rapid changes in

industrial technology in the Australasian colonies during this time.

(9) The sites represent entrepreneurial activity rather than a Government-initiated or subsidised industry. As such, the remains tend to reflect their functional and essentially temporary nature rather than any degree of permanence. Equipment was left where it stood because it was not a profitable proposition to remove it.

(10) The sites may also be a reflection on general work and living conditions in Australia and New Zealand during the last century. Conditions on Macquarie Island were always harsh, yet men were prepared to work there, often for relatively long periods. This would suggest that either there was not a sufficient range of job opportunities at home or that conditions were not substantially different or both.

FACTORS AFFECTING THE SURVIVAL OF REMAINS

Factors affecting the rate of attrition of this resource are weather and sea, flora and fauna, and human intervention.

The climate of Macquarie Island is cold, wet and saline. This has resulted in the deterioration of ironwork, the collapse and scattering of huts and the rotting of wood. Storms add to this, with high seas tending to redistribute remains along the coast.

The fauna can be a problem. The survival of remains within penguin rookeries is limited. Seals wallow, and tend to relocate or obscure remains or expose them to the weather. Although the flora, particularly tussock-grass, tends to totally obscure remains, excavation has shown that it may also act as a protective covering, thereby keeping the remains in a damp but relatively stable environ- ment.

Human intervention has possibly been the greatest drain on the resource. Initially, any available salvageable material seems to have been re-used where at all possible. Although this is also part of the archaeological record, it tends to introduce interpretative problems. In later years, the evidence that has retained its context appears to have suffered from souveniring, and from the lack © of a policy on what was to be done with the artefacts and the sites. The policy of leaving artefacts in situ, unless under immediate threat, was introduced in 1982 and has helped to reduce the amount of preventable damage caused by human intervention.

CONCLUSION

The survey conducted on Macquarie Island during the 1986-87 summer season indicates its potential for increasing understanding of early Australasian industry and technology. The industry and its remains may also reflect some societal attitudes of that time, in regard to the use of primary resources. It is important that the work done in the study is followed up, through additional excavation and by survey, during the non-breeding season, of remains within penguin colonies. Archaeological evidence, by providing additional information and throwing new light on present records, represents the best remaining source of material for understanding the early human activity on Macquarie Island.

Archaeological study of sealing sites 25

REFERENCES

CUMPSTON, J.S., 1968: Macquarie Island. ANARE Sci. Rep., Ser. A(1). Antarctic Division, Department of External Affairs, Australia.

DAVIES, M., 1982: Macquarie Island: Archaeological Report. National Parks and Wildlife Service Report, unpubl.

HINDELL, M. & BURTON H., 1988: The history of the elephant seal industry at Macquarie Island and an estimate of the pre-sealing numbers. Pap. Proc. R. Soc. Tasm. 122(1): 159-176.

TONNESSON, J.N. & JOHNSEN, A.L., 1982: THE HISTORY OF MODERN WHALING. Aus- tralian National University Press, Canberra.

(accepted 18 September 1987)

Papers and Proceedings of the Royal Society of Tasmania, Volume 122(1), 1988 27

ORIGIN, STRUCTURAL AND TECTONIC HISTORY OF THE MACQUARIE ISLAND REGION

by P.E. Williamson

(with eleven text-figures)

WILLIAMSON, P.E., 1988 (viii): Origin, structural and tectonic history of the Macquarie Island region. Pap. Proc. R. Soc. Tasm., 122(1): 27-43. Papers presented at the Macquarie Island Symposium, Hobart, May 1987. ISSN 0080-4703. Bureau of Mineral Resources, Canberra, A.C.T., Australia.

Macquarie Island, in the Southern Ocean, was formed by oceanic crust uplift due to transpressive forces between the Indian/ Australian and Pacific oceanic plates, in a transpressional regime which has persisted over the last 10 Ma. The amount of uplift is affected by regional isostatic compensation for crustal thickening; accompanying effects are tilting of rocks and rotation of the southern segment of the island. Gabbro and serpentinite, in the north, and basalts, in the south, all of which were formed in the primary oceanic crust, are now exposed. Consequently, magnetic properties of igneous rocks on the island correlate with similar features on the Indian plate which is on both sides of it.

In conflict with evidence from younger palacontological and potassium-argon (K-Ar) dating, which may reflect later episodes, this suggests that the original oceanic crust composing the island was formed at the Indian-Antarctic accreting

mid-oceanic ridge around the time of anomaly 7 (27 Ma B.P.). Key Words: oceanic crust uplift, magnetic anomalies, sea-floor history, Macquarie Island.

INTRODUCTION

The tectonic and physiographic expression of Macquarie Island and the surrounding region (fig. 1) is the result of oblique compression (transpression) between the Indian/ Australian and Pacific oceanic plates near to their common boundary. This boundary, south of New Zealand, lies within the Macquarie Ridge Complex, an arcuate system of ridges and troughs, extending from the Macquarie triple junction between the Indian, Pacific and Antarctic plates near 62°S, 160°E (Sykes 1970) to the Alpine Fault of the South Island of New Zealand. The major physio- graphic element of the complex is an arcuate ridge (the Macquarie Ridge) with a trench (the Macquarie Trench) to the east in the central area and a trench to the west in the northern area (Puysegur) and the southern area (Hjort).

Geological and geophysical studies of the Macquarie Ridge began with the New Zealand programme to investigate the marine geology of the New Zealand subantarctic sea floor (Summer- hayes 1969). Reconnaissance geophysical investi- gations in the region were later carried out by Lamont-Doherty Geological Observatory vessels Vema and Conrad, and were expanded between 1968 and 1971, using the USNS Eltanin. Data were also collected during acruise by the Japanese vessel Umitaka-Maru and during aeromagnetic studies

by the US Navy Oceanographic Office. In the Macquarie Island region, marine magnetic and bathymetric data were collected, primarily by the author, from the Australian National Antarctic Research Expedition (ANARE) relief vessel Nella Dan between 1970 and 1973.

Magnetic lineations have been used to resolve the sea-floor spreading history between Australia and Antarctica (Weissel 1972, Weissel & Hayes 1971, 1972, Weissel er al. 1977) and in the Pacific plate south of New Zealand (Christoffel & Falconer 1972). The broad crustal structure and tectonic history of the Macquarie Ridge Complex and the. Macquarie Island region has been discussed by Hayes & Talwani (1972), Woodward (1973), Williamson & Johnson (1974), Williamson (1974, 1975), and Weissel et al. (1977), utilising a pre- dominantly marine geophysical data base, and by Ruff & Cazenave (1985), using SEASAT data.

Analyses of sea-floor spreading magnetic lineations and gravity from. Macquarie Island imply that the origins of the rocks of Macquarie Island were mainly at the Indian/ Australian- Antarctic spreading centre, probably around 27 Ma B.P. — anomaly 7 time (Williamson 1974, 1978, Williamson er al. 1981). Subsequent rotation of island segments, probably during uplift of Macquarie Island, has been demonstrated by Williamson (1978) and Varne & Rubenach (1972).

28 P.E. Williamson

7

<P; : VAUST: 2. o Wan netay SN 4 So Camp “ig MACQUARIE IS ieee eet

os, Bin, § ty, } 2.

ba 3 n/a “rp,

This paper represents a review, and, for the palaeomagnetic component, a reassessment of the geophysical and geological studies relating to the

broad structure and origins of the Macquarie -

Island region, as a basis for future study.

MORPHOLOGY Macquarie Ridge Complex

The morphology of the major segments of the Macquarie Ridge Complex (figs 2 and 3) has been described earlier (Brodie & Dawson 1965, Summerhayes 1969, Hayes & Talwani 1972). Except for a number of breaks (sometimes pronounced), the ridge lies at a depth less than 2000 m between latitudes of 47°S and 57°S. South of 57°S it becomes less regular and branches into two segments. The Macquarie Trench lies east of the Macquarie Ridge between about 51°S and 56°S. It is a continuous deep over a distance of 550 km. It reaches a maximum depth of 6000 m and is typically deeper than 5200 m.

The Emerald Basin, east of the Macquarie Trench, is characterised by numerous large sea- mounts and pronounced depressions, especially in the vicinity of 55°S-59°S (Hayes & Talwani 1972). The basin lying to the west of the Macquarie Ridge is characterised by relatively low relief, generally less than 400m, and variable sediment thickness (Houtz et al. 1971, Hayes & Conolly 1972).

FIG. | — Locality map. Isobaths at 2, 4 and 6 km.

5Q0

Macquarie Island Region

The bathymetry of the Macquarie Island region (Summerhayes 1969) shows gentle gradients to the west of the island which continue down tothe level of the ocean basin (fig.3). Much steeper gradients occur to the east of the island and extend into the Macquarie Trench. The major bathymetric trends in the area parallel the Macquarie Ridge, with only minor irregularities along these trends. The Macquarie Ridge is consistently shallow and rises above sea level in the cases of Macquarie Island, Bishop and Clerk Islands to the south, and Judge and Clerk Islands to the north.

Macquarie Island itself trends northnorth- easterly, paralleling the Macquarie Ridge in the region. The island is characterised by a narrow coastal strip with steep slopes up to an undulating plateau at a maximum elevation of about 400 m. The island plateau is cut by a major valley between Bauer Bay and Sandy Bay. ¢

GEOLOGY Macquarie Ridge Complex

Dredge hauls along the Macquarie Ridge (Summerhayes 1969, Watkins & Gunn 1970) obtained in situ tholeiitic basalt, harzburgite, peridotite, sodic olivine gabbro and troctolite, similar to such rocks observed on Macquarie Island. The Macquarie Ridge Complex south of 51°S appears from these data to be composed of rocks similar to normal oceanic crust.

The Macquarie Ridge is virtually devoid of sediment cover, and sedimentation in adjacent areas is extremely variable (Houtz & Markl 1972, Houtz et al. 1971, Hayes & Talwani 1972). The sediments in the Macquarie Trench are generally variable in thickness (from zero to | km) but are locally undisturbed and horizontally layered (Hayes & Talwani 1972), with thick accumulations of sediments occurring in small enclosed basins (Houtz ef al. 1971).

Tectonic history 29

ieeciaan 3

0 200km st we omental

FATHOMS

[ea] 0- 1000 a 1000 - 2000 2000 - 2500 ie) 2500 - 3000 VZZ,

YZ > 3000

152°

) EMERALD

BASIN

FIG. 2 — Generalised bathymetry of the Macquarie Ridge Complex showing seismicity (after Jones & McCue, this volume) and vectors of relative plate motion of the Indian/ Australian and Pacific plates in cm/yr (after Minster & Jordan 1978); A-A’ is the location of the gravity profile shown in figure 6.

30 ~—sO*&P.E.. Williamson

eee 159°E CE 7 . 54°S

L// II

erk Islands

MACQUARIE ISLAND //- "+

au // ! OM AM y | s } h = / S}|§ If l\

FIG. 3 — Bathymetry of the Macquarie Island region contoured in metres (after Summerhayes 1967).

Macquarie Island

Early work carried out on Macquarie Island by Mawson (1943) has largely been superseded by more detailed work (Varne ef al. 1969, Varne & Rubenach 1972, Quilty et a/. 1973, Griffin & Varne 1980, Griffin 1982, Christodoulou er al. 1984). Varne & Rubenach (1972) concluded that the island is formed of fault-bounded blocks. The northern part of the island is described as composed mainly of serpentinised peridotites, gabbros, and dolerite dyke swarms (fig. 5), and the southern part of extrusive rocks (fig.4) which occur both as basaltic lava and as basaltic volcanic breccias. Parallel dolerite dykes cut the lavas, volcanic breccias and associated sediments in all parts of the island, but there is no continuous gradation from areas of extrusive rocks cut by dykes into the areas of dyke swarms. Globigerina ooze occurs as interstitial materials in the pillow lavas, and poorly preserved coccoliths have been found at North Head.

The detailed structure and stratigraphy of the extrusive rocks of the island are complex and poorly understood. The pillow forms of some lavas imply that these have been tilted or near-vertical attitudes, but the steepest dips which could be reliably measured on lava flows with interbedded mudstone are about 45°. These flows were pre- sumably horizontal or near-horizontal at the time of deposition. Many high and low angle faults cut the extrusive rocks.

Dykes cut the interbedded lava flows, breccias and sediments as a series of parallel sheets and do not intersect one another. Acute angles between the planes of the dykes and the bedding planes of the extrusive rocks are almost always large, no matter how steep the inclinations of the bedding planes, implying that the dykes were originally steeply inclined intrusions into relatively flat-lying extru- sive rocks, and that the tilting occurred sub- sequently. In addition, the strikes of the tilted lava flows, breccias, and sediments are parallel or near- parallel to the strikes of the dykes that cut them. Seemingly, therefore, the axes about which the tilting took place must have been approximately horizontal and parallel to the dyke-bedding plane intersections (Varne & Rubenach 1972).

In the northern part of the island, coarse- grained intrusive rocks representing deeper levels in normal oceanic crust are exposed. The serpen- tinised peridotites are usually massive rocks. The only layering interpreted as compositional occurs at Eagle point, where adjacent layers differ in their proportions of serpentinised olivine and ortho-

Tectonic history 31

Toutcher

Lusitania xy] serpentinised peridotite and gabbro

+] extrusive volcanics and =<2] associated sediments

[-) dyke swarms

— fault

vitere ahlaey ares ast artes 7

<7. act

—- strikes on lavas

-= strikes on dykes

gradational or uncertain

FIG. 4 — Generalised geological map of Macquarie Island with superficial deposits omitted (after Varne & Rubenach 1972).

32. ~P-.E. Williamson

pyroxene. The style and regularity of this layering are similar to that found in cumulate rocks, and the layers are continuous over the length of the outcrop. A layered gabbro complex disturbed by faulting and dyke swarm emplacement occurs in the north- west section of the island, and large outcrops showing spectacular layering are exposed.

Varne & Rubenach (1972) argued that the lack of schistosity or penetrative cleavage in the metamorphosed basalts and dolerites in the north of the island, and the survival of igneous téxtures, indicate that the metamorphism was not accom- panied by deformation, and was similar in nature to burial metamorphism.

Petrographic analyses of rocks from Mac- quarie Island by Varne & Rubenach (1972) show the majority of the dolerites and basalts are similar to normal ocean crust. On the basis of trace element studies, the rocks from Macquarie Island also clearly correspond to the ocean-floor basalts, although some variation in the trace element content in the rocks occurs (Varne & Rubenach 1972, Griffin & Varne 1980).

Although Macquarie Island geologically «

forms part of the crest of the Macquarie Ridge, the rocks of the island were not, in general, formed during tectonic evolution and emergence of the ridge within the last 10 Ma. This is supported by the observation that some of the freshest and pre- sumably youngest volcanic rocks on Macquarie Island are associated with sediments that appear to be of Miocene age (11-25 Ma) (Quilty e¢ al. 1973).

CRUSTAL STRUCTURE AND TECTONIC HISTORY

During the last 10 Ma and up to the present time, the dominant motion between the Indian/ Australian and Pacific plates south of New Zealand has been interpreted as mainly transcurrent, with compression particularly at the Hjort and Puysegur Trenches, where varying degrees of nascent sub- duction of the Indian plate are indicated from earthquake studies by Banghar & Sykes (1969) and from studies of relative plate motion by Weissel (1972), Hayes & Talwani (1972), Hayes er al. (1972) and Weissel es al. (1977).

Seismicity Studies of the seismicity of the Macquarie Ridge Complex have been made by Cook (1966)

and Barazangi & Dorman (1969). Seismic epicentres (Barazangi & Dorman 1969, Jones & McCue, this

volume) show zones of intense activity at the southern end of New Zealand, between about 49°S and 53°S, and at about 56°S. First motion studies have been made on events from the Macquarie Ridge by Bangar & Sykes (1969), Sykes (1967, 1970), Johnson & Molnar (1972), Ruff er al. (1982) and Jones & McCue (this volume). Results of Johnson & Molnar (1972) indicate that under- thrusting of the Tasman Sea exists along the Macquarie Ridge Complex south to about 52°S. Dextral strike-slip motion is interpreted ona fault plane approximately parallel to the Macquarie Ridge at 56.2°S while, for an event at 57.5°S, formal faulting is interpreted. Jones & McCue (this volume) show dextral strike slip motion for the Macquarie Island region.

Johnson & Molnar (1972) concluded that the pole of rotation of the Australian-Pacific plate is located east of and close to the ridge, at about 56-57°S. Ruff eral. (1982), however, have generated focal mechanism solutions, in basic agreement with the globally derived RM2 rotation pole of Minster & Jordan (1978), that place the pole at 60.5°S and imply transpression of varying degrees for the whole Macquarie Ridge complex. That compres- sion along the plate boundary is a recent activity is supported by the plate tectonic reconstruction of Molnar er al. (1975) and the studies of Walcott (1978) and Scholz er al. (1973). These studies indicate that, prior to 10 Ma B.P., the Macquarie Ridge was characterised by strike-slip and spreading motions.

Gravity

The gross crustal structure associated with the central Macquarie Ridge Complex has been investigated using gravity data. Marine gravity data over this area have been inverted to show a progressive thickening of the Indian/ Australian oceanic crustal layer from approximately 7 to l4km, over a 200km distance as the ridge is approached from the west (Williamson & Johnson 1974). A root zone, some 30 km wide and approxi- mately 4km thick, is indicated not symmetrically beneath the Macquarie Ridge itself but displaced towards the trench. Another root zone, approxi- mately 50km wide and 3km thick, within the Emerald Basin to the east of an unthickened segment of oceanic crust immediately west of the Macquarie Trench, appears to be associated witha local bathymetric high, possibly caused by volcanic activity (Hayes & Talwani 1972) or the presence of an ancient spreading ridge associated with the opening of the Emerald Basin.

33

Tectonic history

[888] serpentinised peridotite

d

[#4] layered gabbro

gabbro

North Hea

= te» ° aq N v >4 “ br ertseersi7 S vr. Ss > > D fO8 Pf abe Ss » ° & 7 S iS

a e

FE LESesetete Sttett srt yy:

r+ eee4

®

x Sete, >

ak +++, 2s

S

ec hen ° $

3S '9

= = =

other symbols as in figure 4

54° 30'S

o8 °

Eagle Point

43° YS Nuggets Point

: Island Lake

Langdon Point

348

Mt Blair

Douglas Point

v

0

36°

37 Fr,

v toe e cde, ater g<ty

Mawson Point 0

Sandy Bay

158° 58'E

54° 33'S

23/09/110

s omitted (after

Sit

ial depo.

ic

ie Island with superfi

quarie

al map of the northern part of Mac

Varne & Rubenach 1972).

Geologic

a=

FIG.

34 P.E. Williamson

DISTANCE (km) A 0 200 400 600

Bouguer anomaly

(Macquarie Ridge, using densities shown below) 1000

= = Bak \ 3 B a, : \ Bowers /1°%e \ Ridge ; le Macquarie \ eh | F 25 1 ‘Puerto Rico B 5 & — € 3 15

fea] Volumetric shortening/thickening of Indian Plate near plate boundary

25

FIG. 6 — (A) Crustal structure (lower surfaces) over the central Macquarie Ridge Complex, arrived at by inversion of Bouguer gravity anomaly data, and bathymetry (upper surfaces); demonstrating incipient subduction compared to progressively more developed subduction for the Bowers Ridge and Puerto Rico Trench region. MR is Macquarie Ridge; MT is Macquarie Trench; EBH is Emerald Basin High. Densities used for the inversion are also shown. Location of profile is shown in figure 2. (After Williamson 1975.) (B) Stippled area shows the crustal thickening relative to the original Indian/Australian oceanic plate during the forma- tion of the Macquarie Ridge, which could accom- modate crustal shortening of 65 km.

Asymmetry similar to that observed for the central Macquarie Ridge (excluding the Emerald Basin feature) is typical of island arc systems (fig. 6). The Tonga~Kermadec area has been inves- tigated by Talwani er al. (1961), the mature Puerto Rico Trench region, by Talwani, Sutton & Worzel (1959), the nascent Bowers Ridge Complex by Kienle (1971), and the Japan region by Tomoda (1973). All the above regions exhibit asymmetry in the free air and Bouguer gravity anomalies, and in their inferred crustal structure. Also, in island arc regions, negative isostatic anomalies are associated with the trench and positive isostatic anomalies with the arc or ridge (Heiskanen & Vening Meinesz

1958, Tomoda 1971). Similar relationships are observed for the Macquarie Ridge Complex (Hayes & Talwani 1972). For well-developed island arc systems, such as that associated with the Puerto Rico Trench (Talwani, Sutton & Worzel 1959), where extensive loading by island arc volcanics has occurred, this asymmetry is not as marked as it is in the case of partially developed systems, stich as in the western part of the Bowers Ridge. In the latter area, crustal thickening increases towards the trench, with the root zone occurring beneath the trench. This is most similar to that observed for the central region of the Macquarie Complex. A similar, though less pronounced, crustal structure occurs in the Mendocino Escarpment, where a root zone is present and is displaced from the bathy- metric high towards a buried trough feature masked by thick sediments (Talwani, Worzel & Landisman 1959). This structure would appear to be the result of secondary intra-crustal compressive forces and to represent the oppposite case to that of the “leaky transform fault” which reflects secondary intra- crustal extensional forces.

The various cases cited can be used to show progressive stages in the development from an incipient to a mature subduction zone (fig. 6). The asymmetry of gravity signature and inferred crustal structure of the central Macquarie Ridge Complex (along with the shallow seismicity) correspond to those of an incipient subduction system and support the predicted compressional component of relative plate motion between the Indian and Pacific plates in the region, as shown in figure 2. It is further impled that the Indian/Australian-Pacific plate boundary occurs at the Macquarie Trench where incipient westward dipping subduction of the Pacific plate is now occurring. For the northern and southern ends of the complex, gravity anomalies and inferred crustal structures (Wood- ward 1973) both show thickening towards the ridge. In the case of the Puysegur Trench, however, a more pronounced asymmetric root zone is defined. This, together with the occurrence of a considerably more pronounced component of underthrusting of the ocean plate in the area (Johnson & Molnar 1972), is consistent with a better developed subduction zone in that region.

Loss of around 65 km of oceanic crust of the Indian plate, due to crustal shortening, and thickening to around 150% (fig. 6) is implied by gravity results in the central Macquarie Ridge Complex north of Macquarie Island. The area is north of the zone where Weissel er al. (1977) interpreted crustal loss from sea-floor spreading magnetic anomaly patterns, which they considered

to be due to possible subduction or entrapment by the Pacific plate. Investigation of crustal loss in the area of the gravity profile was not possible from magnetic studies due to lack of data (J. Weissel, pers. comm.). The oceanic crust in the Macquarie Island region has probably been further thickened and shortened, compared with the location of the gravity profile, to accommodate the additional uplift of the ridge to above sea level. Crustal thickening at Macquarie Island and close to the boundary between the Indian/ Australian and Pacific plates in general, is the result of trans- pressional forces, and is thus likely to be associated with structural features, including “flower struc- tures”, that usually occur within transpressional regimes (M. Etheridge, pers. comm.).

Uplift of some rocks of Macquarie Island by approximately 10km during crustal thickening is indicated, since harzburgites representative of the lower ocean crust formed at depths of 5 km (Griffin & Varne 1980) have been uplifted to outcrop on Macquarie Island from water depths of around 5km. The high dips of lavas on the island and the rotation of the south of the island relative to the central island block, by 55° clockwise around a vertical axis, which has been demonstrated from palaeomagnetic results (Williamson 1978), are also apparently associated with the transpression along the ridge crest (fig. 7). Rotation of rocks on the island is also supported from geological argument (Varne & Rubenach 1972).

The Bouguer gravity anomaly associated with Macquarie Island shows only a Bouguer gravity gradient across the island (Williamson & Rubenach 1972). This is inconsistent with local isostatic compensation of the island but consistent with deepening of the Moho towards the plate boundary at the Macquarie Trench east of Macquarie Island, as indicated by inversion of the gravity profile (fig. 6).

The characteristics of the Bouguer gravity anomalies and the inferred crustal structure over the central Macquarie Ridge Complex (fig. 6) are dissimilar to those over accreting mid-ocean ridges (e.g., for the Mid-Atlantic Ridge and the East Pacific Rise (Talwani er a/. 1965), for the Pacific- Antarctic Ridge (Tomoda 1971), for the North Atlantic and Rykjanes Ridges (Talwani er al. 1971)). These typically show gravity lows which are symmetrical about the ridge axis.

SEASAT Data

SEASAT data, along with earthquake seismicity data collected over the Macquarie Ridge Complex

Tectonic history 35

(Ruff & Cazenave 1985), show geoid anomalies over the Hjort and Puysegur Trenches which are interpreted as being characteristic of subduction zones (fig. 8). They display the characteristic geoid signature of mature subducting trenches. This includes a narrow geoid minimum of 5-20m amplitude and 100-400 km width directly at the trench axis. From the trench to the landward side there is a geoid high, coinciding generally with the island arc. Superimposed on this short wavelength variation is a broad positive anomaly in the geoid, with a maximum in the vicinity of the trench.

The Hjort Trench region is interpreted (Ruff & Cazenave 1985) as an oblique subduction zone

r) normal palaeomagnetic 60° directions X

159°00'

O reversed directions

Block 1 Le eee, ee 28° Block 2 55° Assumed rotation of Block 3 aN 0 4km beamed

FIG. 7 — Rotation of the southern block (3) of Macquarie Island relative to the central block (2) during crustal thickening and uplift of the island, inferred from requirements for statistical com- patability of apparent palaeomagnetic poles from the central and southern areas of the island. (After Williamson 1978.)

36 ~~ P.E. Williamson

A PUYSEGUR TRENCH Profile 125 * 103 122 5m

LES eee

| 160°E

200

t ot ie {i 500 km

180°E 170°E

FIG. 8 — SEASAT geoid anomalies over the Macquarie Ridge Complex (A and B)and locations of profiles (C) (after Ruff & Cazenave 1985). Anomalies over the Puysegur (A) and Hort Trenches (B; profiles 43-353) are interpreted as indicating nascent subduction. Anomalies over the Macquarie Trench (fig. B; profiles

239-136) are non-diagnostic consistent with incipient subduction.

similar to the Aleutian-Komandorski region on the basis of its geoid anomaly, while the geoid anomaly of the Puysegur Trench region is considered to be consistent with a nascent subduction zone environ- ment. The geoid anomalies over the central Mac- quarie Ridge are non-diagnostic, but they are clearly not consistent with more than incipient subduction in that region; in particular, the characteristic narrow geoid minimum directly over the trench for mature subduction zones is absent.

ORIGIN OF MACQUARIE ISLAND ROCKS

Regional Pattern of Sea-Floor Spreading Magnetic Anomalies

The formation of the eastern Indian/ Australian oceanic plate south of Australia, to which the Macquarie Island rocks belong, has been investigated from sea-floor spreading magnetic anomalies (Le Pichon & Heitzler 1968, Weissel & Hayes 1971, Weissel ef al. 1977) tied to the geomagnetic time-scale of Heitzler er al. (1968). The magnetic anomaly lineations on the Indian/ Australian plate south of the Resolution Ridge at 48°S strike eastnortheast and are interpreted as resulting from formation of Cainozoic oceanic crust at the Indian-Antarctic Ridge (fig. 9). Anomalies to anomaly 21 are identified. Anomaly lineations are offset in detail (fig. 10) by fracture zones striking northnorthwest. To the west of the Macquarie Ridge Complex north of 48°S (fig. 9), the magnetic anomalies trend in a northnorth- westerly direction and have been correlated with specific sea-floor spreading anomalies associated with the opening of the Tasman Sea (Ringis 1972, Hayes & Ringis 1973, Weissel et al. 1977).

An interpretation of the magnetic anomalies on the Pacific plate to the southeast of the Macquarie Ridge Complex shows the maximum age of oceanic crust in the region is about 80 Ma, anomaly 32(78 Ma) occurring close to the Campbell Plateau (Christoffel & Falconer 1972). A sequence down to anomaly 25 occurs south of this. The next reliable identification, however, is of anomaly 5, which occurs close to the Pacific-Antarctic spread- ing ridge. The results imply that distinct periods of sea-floor spreading have occurred within this region, as many of the sea-floor spreading anoma- lies between 5 and 25 do not appear to be repre- sented in the area (Christoffel & Falconer 1972, Adams & Christoffel 1965).

To the east of the Macquarie Ridge, at about 51.5°S and 162.5° E, anomaly lineations are poorly

Tectonic history 37

150° 155° 160° 165°

TASMAN SEA

apparent crustal loss from Indian/Australian plate (Weissel et al. 1977)

\

y

S 3 ~ Ry aS) S & is) ea)

55° SOUTHWEST

PACIFIC OCEAN

60°

FIG. 9 — Generalised trends of marine magnetic lineations surrounding the Macquarie Ridge Complex. For the trenches 2500 fathom contours are shown.

defined but appear to trend approximately parallel to the ridge (Hayes & Talwani 1972). However, the tentative correlations may reflect structure in the sea floor rather than the normal polarity reversals. The direction of magnetic lineations has also not been established in the Emerald Basin which lies to the south of the above area. The orientation of tracks and the apparent wavelengths displayed in that area, suggest that, if lineations are present, they are probably oriented in an approximately northnortheasterly direction.

Sea-Floor Spreading Magnetic Anomalies in the Macquarie Island Region

The history of oceanic crust around Macquarie Island was investigated, using magnetic data collected along profiles parallel to the Macquarie Ridge from the Nella Dan (Williamson et al. 1981). Anomaly 7 (27 Ma) is identified, on Nella Dan data, to the west of Macquarie Island (fig. 10), and anomalies 8 and 9 north of the island.

38 P.E. Williamson

These anomalies have been identified with greater confidence due to their existence as part of a longer sequence of anomalies, which extend from anomaly 7 to anomaly 21, and are observed in the Nella Dan data between Melbourne and Macquarie Island. The magnetic lineations parallel sea-floor spreading anomalies on the Indian/ Australia plate west of

158°E ~ 159°E

the island and terminate east of Macquarie Island at the Macquarie Trench, which is interpreted to be the Indian/Australian-Pacific plate boundary; this implies that Macquarie Island occurs on the Indian/Australian plate. Offsets in the magnetic lineations suggest the presence of two minor trans- form faults to the west of the island. These are expressed as a deflection to the north in the magnetic lineations interpreted in the broader study of Weissel er al. (1977). To the west of the island, anomalies 8 and 9, observed in the Eltanin 44 data, occur about 80km south of the same anomalies on the Nella Dan tracks east of a separating transform fault. A second transform fault, with a small offset, appears to trend in a northnorthwesterly direction close to the west coast of Macquarie Island. Here, anomaly 9 is interpreted to occur about 20 km to the south of the same lineations further west. Anomaly 9 is ten- tatively correlated over the ridge north of Macquarie Island. The shape of the anomaly is distorted, however, probably as a result of the magnetic anomaly associated with the ridge.

Sea-Floor Spreading Magnetic Anomaly Data on Macquarie Island

A magnetic anomaly lineation is interpreted as crossing Macquarie Island (fig. 10). The preferred identification is that it is anomaly 7 (27 Ma), since the broad features of the upwardly continued magnetic anomaly profile (to 1 km) along the island can be correlated with anomaly 7, interpreted on the adjacent marine magnetic profiles. In the onland data, the character and anomaly widths of the magnetic high and the associated low to the south are also similar to those of anomaly 7 and the associated magnetic low in the adjacent marine

FIG. 10 — Marine magnetic anomaly profiles in the vicinity of Macquarie Island to the west of the plate boundary at the Macquarie Trench. Identified anomalies are from the Heirtzler et al. (1968) sequence. Inferred fracture zones are interpreted Srom offsets of the anomaly sequences. The lower part of the figure shows marine magnetic anomaly profiles close to Macquarie Island. Also shown is the main north-south profile along the island, upwardly continued to a level of | km, which is similar to the depths to sea floor on the marine profiles. The tentative correlation of the anomaly on either side and across the island corresponds to anomaly 7 (27 Ma B.P.).

data. This implies the presence of remanent sea- floor spreading anomalies over, and generated by, the island lavas.

Magnetic properties of the Macquarie Island rocks, confirming the ability of the basalt lavas on the island to contribute to sea-floor spreading magnetic anomalies, have been reported by Williamson (1974, 1978), Butler & Banerjee (1973) and Butler er al. (1975, 1976).

Modelling of the main north-south onland magnetic profile from Macquarie Island (William-

Macquarie Island Symposium

2000 nT

D = 0° (180°)

Acknowledgement

Tectonic history

D = 55° (235°)

Dr Eric Guiler not only made the initial suggestion for the Symposium but then helped to bring the Symposium to

fruition as a member of the Organising Committee.

Ms Wendy Cuskelly was the Secretary of the Organising Committee and also helped in many other ways to make the

Symposium a success.

P. G. Quilty D.E. Rounsevell

Honorary Organisers

39

40 ~P.E. Williamson

Palaeomagnetic Data

Palaeomagnetic apparent pole data from Macquarie Island (Williamson 1978) have been reviewed and recalculated (paper in prep.) in relation to the revised polar wander path for the Australian plate (Idnurm 1985). No unique apparent palaeomagnetic pole for Macquarie Island is now considered possible from present data, due to the inability to resolve rotations around a vertical axis in relation to any simple structural model generated from other data. The possible range of poles in not inconsistent with inferences from marine magnetic data but cannot be said to uniquely support them.

DISCUSSION

Geophysical results can be summarised as follows:

(1) The delineation of a remanent sea-floor spreading magnetic anomaly lineation across Macquarie Island, with a trend similar to those on the Indian plate to the west, suggests that the rocks of the island were formed at the Indian/ Australian—Antarctic spreading ridge.

(2) The coincidence of the trends of Indian/ Australian plate sea-floor spreading lineations with the strike of dykes in the central region of the island is consistent with the model of dyke injection at a mid-ocean ridge, and further argues for the forma- tion of the island rocks at the Indian/ Australian- Antarctic spreading ridge.

(3) The interpretation of crustal structure from gravity data implies that the eastern extent of the Indian/ Australian plate is at the Macquarie Trench, east of Macquarie Island, which again implies that the rocks of Macquarie Island were formed as part of the Indian/Australian oceanic plate.

(4) The preferred interpretation of the remanent magnetic lineation crossing Macquarie Island is anomaly 7 (27 Ma). This interpretation is an exten- sion of interpretations of remanent magnetic anomalies over the oceanic crust for the Indian/ Australian plate west of Macquarie Island. The interpreted age is considered to be correct within approximately +3 Ma, which corresponds to the anomaly being misinterpreted anomaly 6 or 8. Due to the structural complexity of the region such a misinterpretation may be possible. Detachment of oceanic crustal blocks at the ridge could contribute to this so that the age is more likely to be of anomaly 8 than 6, given that the motion of the Indian/ Australian plate relative to the Pacific plate is dextral. That is, the age is more likely to be underestimated than overestimated. An age for

Macquarie Island oceanic crust of around 10 Ma (anomaly 5) age, as suggested from recent K-Ar dating, would require either (a) that there is a spreading ridge jump of around 500km to the north and also that the sea-floor spreading results published for the eastern Indian plate are seriously in error; or (b) that Macquarie Island crust was locally emplaced at the plate boundary; sea-floor spreading remanent anomaly trends would then parallel the Macquarie Ridge, which does not appear to be the case.

The geophysical data, with the exception of the palaecomagnetic data, are broad scale and do not relate to specific rock samples, and certainly those data argue that the rocks of Macquarie Island in general were formed at the Indian/ Antarctic spreading ridge at around 27 Ma B.P.

A discrepancy exists, however, between geo- physical and some palaeontological dating and all K-Ar age estimates. Palaeontological data asso- ciated with rocks from North Head (“some of the freshest volcanic rocks”) on the island have given ages of middle to early Miocene (Quilty er al. 1973). Recent K-Ar data reported by Varne (pers. comm.), however, give young ages, normally within the last 10 Ma, for rocks of the island. Those ages essentially correspond to the normally accepted time of uplift of the ridge and island estimated from geophysical data. K-Ar ages could not be obtained for the rocks used in the palaeomagnetic study because their preservation was not considered sufficiently good for that purpose.

The apparent discrepancy between the two sets of ages could reflect two factors:

(1) _ essentially different data sets relating, on the one hand, to original oceanic crust and later intrusive and extrusive activity (such as seamount activity) associated with the already formed oceanic crust or, on the other, to crustal compression during formation of the Macquarie Ridge; this could result in the fresher samples being more recent, which may be particularly relevant to samples from North Head which are unusually “fresh” samples; or

(2) inaccuracies in the relatively small number of palaeontological age determinations, due to poor preservation of specimens and misidentification: loss of residual argon, due to tectonic activity during the last 10 Ma, giving young and incorrect K-Ar dates, although some of these are *’Ar/*’Ar dates, which are not as susceptible to Ar loss or alteration; dating simply reflecting time of uplift; or fortuitous correlation of unrelated geophysical anomalies.

Some igneous episodes after the formation of the original oceanic crust are to be expected, however, particularly given the level of tectonic activity of the region and the nearness to the plate margin. This could suggest that factor (1) above is the dominant cause of the discrepancy. Nonetheless, it is important that investigations be carried out to resolve the discrepancy, and to this end more extensive palaeontological studies on sediments between lava layers from all areas of the island

could be employed.

CONCLUSION

Macquarie Island is a rare example of uplifted oceanic crust. As such it is a geological enigma and acritical area for the study of ocean tectonics. The structure and physiographic expression of the island and the surrounding region are the result of a tectonic history of oblique compression (trans- pression) between the Indian/Australian and Pacific oceanic plates at their common boundary within the Macquarie Ridge Complex south of New Zealand. Intra-plate transpressive forces in this region resulted in the uplifted of oceanic crust of the Indian/Australian plate to form the Macquarie Ridge and Macquarie Island, and in incipient subduction of the Pacific oceanic plate at the Macquarie Trench. This uplift has been effected by thickening and shortening of the oceanic crust of the Indian/ Australian plate, the amount of uplift being reduced by regional isostatic compen- sation of the increased crustal thickness. The transpressional regime has persisted during the last 10 Ma; prior to that time, strike-slip and exten- sional regimes occurred along the Macquarie Ridge Complex.

The uplift of Macquarie Island has been accompanied by tilting of rocks and rotation of the southern segment of the island, relative to the central segment, by about 55° in a clockwise direction around a vertical axis. In the north, gabbro and serpentinite formed at depths of up to 5km in the primary oceanic crust, are currently exposed above sea level. Structural tilting of the island lavas has not been sufficient to reverse the primary in situ directions of remanent magneti- sation. Consequently, long wavelength remanent magnetic sea-floor spreading anomalies associated with areas of outcropping lavas in the centre and south of Macquarie Island correlate, with the same

Tectonic history 41

polarity, with similar features in marine magnetic data on the Indian plate on both sides of the island. Measured magnetic properties of the island’s igneous rocks confirm that their remanent magne- tisations could contribute to sea-floor spreading magnetic anomalies, with the dominant contribu- tion coming from the lava sequences and lesser contributions from gabbros and dolerites. The serpentinite samples measured do not exhibit stable NRM.

Sea-floor spreading magnetic anomaly 7 is tentatively correlated across Macquarie Island, indicating that oceanic crust, of which the island is composed, was formed at the Indian-Antarctic accreting mid-oceanic ridge around anomaly 7 time (27 Ma B.P.).

This is in conflict with evidence from younger palaeontological and K-Ar dates from the island (from Miocene to Recent) which may reflect later metamorphic or igneous episodes after the estab- lishment of the original oceanic crust. The youngest K-Ar date, for instance, is on a metamorphic dyke from the sheeted dyke complex and reflects the timing of the closure of the metamorphic minerals as they cooled (R. Varne, pers. comm.).

To date, broad tectonic studies have presented insights into the mode of formation of Macquarie Island, the origins of its dominant rock types and the nature and result of the large-scale structural movements which affected the region. The detailed structural configurations involved are, however, poorly understood. Future work could usefully investigate the inferred presence of transpressional

«structures on the island, associated with the crustai

thickening and uplift of deep crustal rocks, assuming sufficient data reflecting stress and strain directions can be obtained. In addition, more extensive palaeontological studies could help to resolve discrepancies in the ages of the island rocks. as interpreted from existing palaeontological,

K-Ar and geophysical data.

ACKNOWLEDGEMENTS

The work described here was carried out under the auspices of ANARE with assistance from many ANARE personnel. Constructive reviews by C.J. Pigram, B.D. Johnson and R. Varne have helped bring the text to its final form.

42 P.E. Williamson

REFERENCES

ADAMS, R.D. & CHRISTOFFEL, D.A., 1965: Total magnetic field surveys between New Zealand and the Ross Sea. J. Geophys. Res., 70: 805-814.

BANGHAR, A.R. & SYKES, L.R., 1969: Focal mechanisms of earthquakes in the Indian Ocean and adjacent regions. J. Geophys. Res., 74(2): 632-649. :

BARAZANGI, M.W. & DORMAN, J., 1969: World seismicity maps, compiled from ESSA Coast and Geodetic Survey epicenter data, 1961-1967. Bull. Seism. Soc. Am., 59(1): 369-380.

BRODIE, J.W. & DAWSON, E.W., 1965: Morphology of North Macquarie Ridge. Nature, 207: 844-845.

BUTLER, R.F. & BANERJEE, S.K., 1973: Magnetic properties of exposed oceanic crust on Macquarie Island. Nature, 244(138): 115-118.

BUTLER, R.F., BANERJEE, S.K. & STOUT, J.H., 1975: Magnetic properties of ocean pillow basalts from Macquarie Island. Nature, 257: 302-303.

BUTLER, R.F., BANERJEE, S.K. & STOUT, J.H., 1976: Magnetic properties of oceanic pillow basalts: evidence from Macquarie Island. Geophys. J. R. Astro. Soc., 47: 179-196.

CHRISTODOULOU, C., GRIFFIN, B.J. & FODEN, J., 1984: The geology of Macquarie Island. ANARE Res. Notes, 21.

CHRISTOFFEL, D.A. & FALCONER, R.K.H., 1972: Marine magnetic measurements in the southwest Pacific Ocean and the identification of new tectonic features. Jn Hayes, D.E. (Ed.):; ANT- ARCTIC OCEANOLOGY II: THE AUSTRA- LIAN-NEW ZEALAND SECTOR. Antarctic Research Series, 19. American Geophysical Union, Washington, D.C.: 197-209.

COOK,R.J.S., 1966: Some seismological features of the North Macquarie Ridge. Nature, 221: 953-954.

FALCONER, R.K.H., 1972: The Indian-Antarctic- Pacific triple junction. Earth Planet. Sci. Lett., 17: 151-158.

GRIFFIN, B.J., 1982: Igneous and metamorphic petro- logy of lavas and dykes of the Macquarie Island ophiolite complex. Unpublished Ph.D. thesis, University of Tasmania.

GRIFFIN, B.J. & VARNE, R., 1980: The Macquarie Island ophiolite complex: mid-Tertiary oceanic lithosphere from a major ocean basin. Chem. Geol., 30: 285-308.

HAYES, D.E. & CONOLLY, J.R., 1972: Morphology of the southeast Indian Ocean. /n Hayes, D.E. (Ed.): ANTA RCTIC OCEA NOLOGY II: THE AUSTRALIAN-NEW ZEALAND SECTOR. Antarctic Research Series, 19. American Geo- physical Union, Washington, D.C.: 125-146.

HAYES, D.E. & RINGIS, J., 1973: Sea-floor spreading in the Tasman Sea. Nature, 243: 454-458.

HAYES, D.E. & TALWANI, M., 1972: Geophysical investigation of the Macquarie Ridge Complex.

In Hayes, D.E. (Ed.): ANTARCTIC OCEAN- OLOGY Il: THE AUSTRALIA N-NEW ZEA~ LAND SECTOR. Antarctic Research Series, 19. American Geophysical Union, Washington, D.C.: 211-234.

HAYES, D.E. TALWANI, M. & CHRISTOFFEL, D.E., 1972: The Macquarie Ridge. PROCEED- INGS OF THE SCAR/IUGS SYMPOSIUM OF ANTARCTIC GEOLOGY AND SOLID EARTH GEOPHYSICS, OSLO, 1970. Spon- sored by the Scientific Committee on Antarctic Research. Universitets Forlaget, Oslo.

HEIRTZLER, J.R., DICKSON, G.O., HERRON, E.M., PITMAN, W.C. III] & LE PICHON, xX., 1968: Marine magnetic anomalies, geomagnetic field reversals and motions of the ocean floor and continents. J. Geophys. Res., 73(6): 2119-2136.

HEISKANEN, W.A. & VENING MEINESZ, F.A., 1958: THE EARTH AND ITS GRAVITY FIELD. McGraw Hill Book Co., Inc.

HOUTZ, R.E. & MARKL, R.G., 1972: Seismic profiler data between Antarctica and Australia. /n Hayes, D.E. (Ed.): ANTARCTIC OCEANOLOGY II: THE AUSTRALIAN-NEW ZEALAND SEC- TOR. Antarctic Research Series, 19. American Geophysical Union, Washington, D.C.: 147-164.

HOUTZ, R., EWING, J. & EMBLY, R., 1971: Profiler data from the Macquarie Ridge area. /n Reid, J.L. (Ed.): ANTARCTIC OCEANOLOGY |. Antarctic Research Series, 15: 239-245.

IDNURM, M., 1985: Late Mesozoic and Cenozoic palaeomagnetism of Australia — I. A redeter- mined apparent polar wander path. Geophys. J. R. Astr. Soc., 83: 399-418,

JOHNSON, T. & MOLNAR, P., 1972: Focal mechan- isms and plate tectonics of the south west Pacific. J. Geophys. Res., 77(26): 5000-5032.

JONES, T. & McCUE, K., 1988: The seismicity and tectonics of the Macquarie Ridge. Pap. Proc. R. Soc. Tasm., 122(1): 51-57.

KIENLE, J., 1971: Gravity and magnetic measurements over Bowers Ridge and Shirshoy Ridge, Bering Sea. J. Geophys. Res., 76(29): 7138-7153.

LE PICHON, X. & HEIRTZLER, J.R., 1968: Magnetic anomalies in the Indian Ocean and sea-floor spreading. J. Geophys. Res., 73: 2101-2117.

MAWSON, D., 1943: Australian Antarctic Expedition Sci. Rep., 1911-1914. Ser. A(V). Government Printing Office, Sydney, Australia.

MINSTER, J.B. & JORDAN, T.H., 1978: Present-day plate motions. J. Geophys. Res., 83: 5311-5354,

MOLNAR, P., ATWATER, T., MAMMERICKX, J. & SMITH, S.M., 1975: Magnetic anomalies, bathymetry, and tectonic evolution of the South Pacific since the Late Cretaceous. Geophys. J. R. Astron. Soc., 49: 383-420.

QUILTY, P.G., RUBENACH, M. & WILCOXON, J.A., 1973: Miocene ooze from Macquarie Island. Search, 4(5): 163-164.

RINGIS, J., 1972: The structure and history of the Tasman Sea and the southeast Australian margin. Unpublished Ph.D. thesis, University of New South Wales.

RUFF, L. & CAZENAVE, A., 1985: SEASAT geoid anomalies and the Macquarie Ridge Complex. Phys. Earth Planet. Interiors, 38: 59-69.

RUFF, L., GIVEN, G. & SANDERS, C., 1982: The tectonics of the Macquarie Ridge, New Zealand: new evidence of strike-slip motion from the earthquake of May 25, 1981, Mw=7.7. EOS, 63(18): 384.

SCHOLZ, C.H., RYNN, J.M.W., WEED, R.W. & FROHLICH, C., 1973: Detailed seismicity of the Alpine Fault Zone and Fiordland Region, New Zealand. Geol. Soc. Amer. Bull., 84: 3297-3316.

SUMMERHAYES, C.P., 1969: Marine geology of the New Zealand Subantarctic sea-floor. N.Z. Dep. Sci. Ind. Res. Bull., 190.

SYKES, L., 1967: Mechanism of earthquakes and nature of faulting on the mid-ocean ridges. J. Geophys. Res., 72(8): 2131-2153. %

SYKES, L.R., 1970: Seismicity of the Indian Ocean and a possible nascent island between Ceylon and Australia. J. Geophys. Res., 75(26): 5041-5055.

TALWANI, M., HEEZEN, B.C. & WORZEL, J.L., 1961: Gravity anomalies, physiography and crustal structure of the Mid-Atlantic Ridge. Trav. Sci. Bur. Centr. Seism. UGGI, Int. Ser. A, fasc. 22: 81-111.

TALWANI, M. LE PICHON, X. & EWING, M., 1965: Crustal structure of mid-ocean ridges, 2. Computed model from gravity and refraction data. J. Geophys. Res., 70: 341-352.

TALWANI, M., SUTTON, G.H. & WORZEL, J.R., 1959: Acrustal section across the Puerto Trench. J. Geophys. Res., 64: 1545-1555.

TALWANI, M., WINDISCH, C. & LANGSETH, M.G. Jr., 1971: Reykjanes ridge crest. A detailed geophysical study. J. Geophys. Res., 76(2): 437-517.

TALWANI, M., WORZEL, J.L. & LANDISMAN, M., 1959: Rapid gravity computations for two- dimensional bodies with application to the Mendocino submarine fracture zone. J. Geophys. Res., 64: 49-59.

TOMODA, Y., 1971: Gravity anomalies in the Pacific Ocean. Paper presented at 12th Pacific Science Congress, Canberra, Australia.

TOMODA, Y., 1973: FREE AIR AND BOUGUER ANOMALIES IN AND AROUND JAPAN. University of Tokyo Press.

VARNE, R., GEE, R.D. & QUILTY, P.G., 1969: Macquarie Island and the cause of oceanic linear magnetic anomalies. Science, 166(3902): 230-232.

VARNE, R. & RUBENACH, M.J., 1972: Macquarie Island and its relationship to oceanic crust. /n Hayes, D.E. (Ed.): ANTARCTIC OCEAN- OLOGY Il: THE AUSTRALIA N-NEW ZEA- LAND SECTOR. Antarctic Research Series,

Tectonic history 43

19. American Geophysical Union, Washington, D.C.: 251-266.

WALCOTT, R.I., 1978: Present tectonics and Late Cenozoic evolution of New Zealand. Geophys. J. R. Astron. Soc., 52: 137-164.

WATKINS, N.D. & GUNN, B.M., 1970: Petrology, geochemistry and magnetic properties of some rocks dredged from the Macquarie Ridge. N.Z. J. Geol. Geophys., 14: 153-163.

WEISSEL, J.K., 1972: Analysis of marine magnetic data from south of Australia. Unpublished Ph.D. thesis, University of New South Wales.

WEISSEL, J.K. & HAYES, D.E., 1971: Asymmetric spreading south of Australia. Nature, 231(5304): 517-521.

WEISSEL, J.K. & HAYES, D.E., 1972: Magnetic anomalies in the southeast Indian Ocean. Jn Hayes, D.E. (Ed.): ANTARCTIC. OCEAN- OLOGY II: THE AUSTRALIAN-NEW ZEA- LAND SECTOR. Antarctic Research Series, 19. American Geophysical Union, Washington, D.C.: 165-198.

WEISSEL, J.K., HAYES, D.E. & HERRON, E.M., 1977: Plate tectonics synthesis: the displace- ments between Australia, New Zealand, and Antarctica since the Late Cretaceous. Mar. Geol., 25: 231-277.

WILLIAMSON, P., 1974: The structural evolution of the Macquarie Island region and its relation to ocean crust. Unpublished Ph.D. thesis, Univer- sity of New South Wales.

WILLIAMSON, P., 1975: Structure of the Macquarie Ridge Complex in relation to the formation and polarity of new and fossil subduction zones. Bull. ASEG., 6: 54-55.

WILLIAMSON, P., 1978: The paleomagnetism of out- cropping oceanic crust on Macquarie Island. J. Geol. Soc. Aust., 25: 387-394.

WILLIAMSON, P. & JOHNSON, B.D., 1974: Crustal structure of the central region of the Macquarie Ridge Complex from gravity studies. Mar. Geophys. Res., 11(2): 127-132.

WILLIAMSON, P. & RUBENACH, M.J., 1972: Pre- - liminary report on geophysical studies on Macquarie Island. /n Hayes, D.E. (Ed.): ANTARCTIC OCEANOLOGY II: THE AUS- TRALIAN-NEW ZEALAND SECTOR. Ant- arctic Research Series, 19. American Geophysi- cal Union, Washington, D.C.: 234-249.

WILLIAMSON, P., HAWKINS, L.V. & LONG, B., 1981: An examination of the possible occur- rences of seafloor spreading magnetic anomalies on Macquarie Island. Mar. .Geophys. Res., 5: 139-155.

WOODWARD, D.J., 1973: Gravity and magnetic anomalies over the Macquarie Ridge. /n Fraser, R. (Compiler): OCEANOGRAPHY OF THE SOUTH PACIFIC.N.Z. Nat. Comm. UNESCO, 1972. Wellington.

(accepted 13 January 1988)

Papers and Proceedings of the Royal Society of Tasmania, Volume 122(1), 1988 45

THE AGE AND DISTRIBUTION OF THE IGNEOUS ROCKS OF MACQUARIE ISLAND

by R. A. Duncan and R. Varne

(with one table and one text-figure)

DUNCAN, R.A. & VARNE, R., 1988 (viii): The age and distribution of the igneous rocks of Macquarie Island. Pap. Proc. R. Soc. Tasm., 122(1): 45-50. Papers presented at the Macquarie Island Symposium, Hobart, May 1987. ISSN 0080-4703. College of Oceanography, Oregon State University, Corvallis, Oregon, U.S.A. 973331; and Department of Geology, University of Tasmania, Hobart, Tasmania, Australia 7001.

Basalts associated with the formation of the Macquarie Ridge have been radiometrically dated at between 9.7 Ma and 11.5 Ma, giving a Middle Miocene age in approximate agreement with the age indicated by fossils in intercalated sedimentary rocks but younger than suggested by the nearby sea-floor spreading anomaly pattern. (Ed.) Key Words: igneous rocks, radiometric analysis, tectonic uplift, Macquarie Island.

INTRODUCTION

Macquarie Island is unique among islands of the major ocean basins in being composed of Oceanic crust and marks the tectonic emergence above sea level of part of the Macquarie Ridge (Varne et al. 1969, Varne & Rubenach 1972). The latter is a narrow arcuate welt of ridges and trenches that runs southward from New Zealand to join the Indian-Pacific mid-ocean ridge system.

The first geological map of the island was produced by L. R. Blake, a surveyor with the 1911-1914 Australasian Antarctic Expedition led by Sir Douglas Mawson. Blake was killed in the First World War, and his map was published by Mawson in his account (the first) of the geological history of the island, also based on Blake’s work (Mawson 1943). This assigned the rocks of the island to an older basic group of lavas, in places intensely folded, intruded by a group of peridotites and gabbros, with both groups, following a period of erosion which exposed the intrusive rocks, overlain unconformably by a younger basic group of pillow lavas and extrusive fragmental rocks. Mawson also tentatively suggested a Cretaceous age for the formation of the rocks of the island, by analogy with the timing of tectonic and magmatic activity in New Zealand.

More recent geological mapping, including that of Varne et al. (1969), Varne & Rubenach (1972), Christodoulou et al. (1984), Crohn (1986) and Lees (1987), generally confirms the lithological distributions recorded in Blake’s map, but Mawson’s stratigraphic column has been ques- tioned. Varne et al. (1969), who first showed that the island is formed from uplifted oceanic crust,

pointed out that the older basic group described by Mawson (1943) is made up of dolerite dyke swarms and not of extrusive rocks. They also argued that the extrusive rocks of Mawson’s younger basic group might be intruded by the dyke swarms, the gabbros and the serpentinised peridotites, and could therefore be the oldest rocks exposed on the island. Foraminifera from silicified oozes associated with pillow lavas were considered to be of Late Miocene or Pliocene age.

Varne & Rubenach (1972) presented new maps and geological data showing that Macquarie Island seems to be composed of a number of fault- bounded blocks, probably on all scales, derived from different layers of the oceanic lithosphere. This interpretation implies that all of the igneous rocks of the island could have been formed at much the same time, in a spreading zone in an oceanic environment, passibly the Indian-Antarctic spread- ing ridge. Poorly-preserved coccoliths recovered - from oozes associated with North Head pillow lavas were identified by Quilty ez a/. (1973) as being of Early or Middle Miocene age.

Geochemical and geophysical data have supported the identification of Macquarie Island as uplifted oceanic crust. The basalts and dolerites are compositionally very like ocean-floor basalts from the Mid-Atlantic Ridge (Griffin & Varne 1980), and pillow lavas have magnetic properties which correspond well with those of ocean-floor basalts (Banerjee et al. 1974, Butler et al. 1976, Levi et al. 1978). A magnetic profile along the northern part of the island (Williamson 1974) shows a broad anomaly which has been correlated with similar anomalies in marine magnetic profiles around the island, believed by Williamson e¢ al. (1981) to be

46 R.A. Duncan and R. Varne

part of sea-floor spreading anomaly 7. Williamson (this volume) therefore suggests that the rocks of the island formed at the Indian-Antarctic spreading ridge at anomaly 7 time, about 27 Ma, considerably earlier than the palaeontological data suggest.

In this article, we briefly describe the geology of Macquarie Island, present a revised geological map, and provide the first radiometric determina- tions of the ages of Macquarie Island rocks.

THE GEOLOGY OF MACQUARIE ISLAND

The island is about 35 km long but less than 6 km wide, elongated northnorthwest along the axis of the Macquarie Ridge. It is topped by an elevated and undulating lake-studded summit plateau, best developed in the northern part of the island, which stands about 300 m above sea level and is cut off seaward by steep cliffs. A marked coastal platform with relict sea stacks occurs in places at the base of the cliffs on both sides of the island, with its greatest development at Handspike Point, where it is a kilometre wide and rises to an elevation of about 15 m above sea level (Mawson 1943).

Marine, glacial and other subaerial erosional processes may have all played a part in shaping Macquarie Island from the ridge as it rose. Varne et al. (1969) identified raised beach deposits at 200 m on the northern summit plateau of the island, but Colhoun & Goede (1974) and Loffler & Sullivan (1980) proposed that many landforms from this part of the island are of glacial origin. Some terraces in the northern part of the island may be relicts of periglacial activity, formed at a time when cooler climatic conditions favoured the operation of solifluction processes (Loffler et al. 1983). Others are associated with raised beach deposits that were apparently lifted to their present position by neotectonic processes (Varne et al. 1969, Selkirk et al. 1983, Ledingham & Peterson 1984). These raised beaches are not yet dated, but deposits resting upon them are young (Selkirk er a/. 1983). It could be that the summit plateau was largely cut by marine erosion, either before renewed compression along the Macquarie Ridge uplifted it to its present position or when sea levels were higher, in the past.

The landscape has also been modified by recent faulting. Ledingham & Peterson (1984) described recent lineaments and fault lines: some apparently displace young topographical features and are associated with topographic depressions and lakes. These young lineaments and fault lines follow generally northerly trends, subparallel to the east and west coastlines of the island and to the

general trend of the Macquarie Ridge. The Brothers fault line (fig. 1) and its northern continuation are the most prominent, but less distinct, apparently older, linear features trend northwest and southwest (Ledingham & Peterson 1984, plates 6 and 7).

The northern part of Macquarie Island is formed mainly of intrusive igneous rocks, whereas the southern part is formed mainly of extrusive volcanic rocks and associated sedimentary rocks (fig. 1). The contact zone between the mainly intrusive rocks of the northern part of the island and the mainly extrusive rocks south of Bauer Bay and Sandy Bay follows a topographic depression, and corresponds in part with a northwest-trending fault zone that is marked by mylonites and flaser gabbros(Varne et al. 1969) and invaded by gabbroic veins, and therefore unlikely to be due to near- surface neotectonic movements. Other northwest- striking faults in the northern part of the island also appear to be older than the generally north- trending fault lines recorded by Ledingham & Peterson (1984).

The extrusive volcanic rocks are basaltic. Pillow lavas are common, and volcanic breccias hyaloclastites, and massive lavas are also present (Mawson 1943, Varne & Rubenach 1972, Christo- doulou et al. 1984, Crohn 1986). Thin lenses of volcaniclastic sediments, ranging from mudston® to conglomerate, tend to be associated with volcanic sequences which contain massive lavas, as at Mawson Point. Calcareous oozes occur betwee? pillows, and are widely distributed although not abundant. Volcanic activity, therefore, occurred if a submarine setting, but the water depths at the time of deposition are unknown and may have ranged from deep to relatively shallow. Little i also known as yet of the volcanic successions 0! Stratigraphy. An angular discordancy may occu!

FIG. | — Interpretative geological map of Macquarie Island with superficial deposits omitte® compiled from the mapping of Varne et al. (1969 Varne & Rubenach (1972), Griffin (1982), Christ” doulouet al. (1984), Crohn (1986) and Lees (1987) The structural lineaments are from Ledingham Peterson (1984).

[2] serpentinised peridotite dolerite dyke swarms gabbro [,] layered gabbro extrusive volcanics and associated sedimentary rocks —— /—— major/minor faults —— /—-— major/minor structural lineaments

s+ Qradational or uncertain contacts —— / -- strikes on lavas/dykes

AURORA POINT SOUCEK BAY

v

39°00 E

NORTH HEAD

BAY

MH ify, i BS BROTHERS

App

N N

\ | Pyrami

PRECARIOUS POINT

CAPE STAR G \,30 YY \

CAROLINE COVE x25

SOUTH WEST POINT

A) SADDLE POINT

Wie he aMount\ Ainsworth

iy

ANS

5

HURD POINT

Igneous rocks 47

just east of Mawson Point (Varne & Rubenach 1972), although Lees (1987) has suggested that the variations in strike in the Mawson Point-Bauer Bay area are due to recent fault movement.

Two strike directions are commonly found in the volcanic and sedimentary rocks. South of Mawson Point and near Green Gorge, the domi- nant strike trends are westerly, but elsewhere they lie in the northwest quadrant. Dips range from shallow to steep, and are variable even in areas of constant strike. Narrow basaltic dykes cut the extrusive rocks. South of Brothers Point, dykes form as much as 40% of the coastal outcrop, but are generally less abundant elsewhere. The volcanic rocks may have been first tilted around axes that were near-horizontal and parallel to dyke bedding- plane intersections, causing variations in dip, and later rotated about vertical axes, causing variations in strike (Varne & Rubenach 1972, Williamson 1978).

Sheeted dyke complexes, composed of dip- ping series of parallel to subparallel dolerite dykes, occur mainly in the north of the island, associated with gabbros and peridotites, but are also exposed on the coasts north of Cape Toutcher and Lusitania Bay (fig. 1). Screens of gabbro, serpentinised peridotite, brecciated dolerite, and very rarely, pillow lava, have all been found within the dyke complexes, but no screens are identifiable over large areas.

Varne et al. (1969) reported that in the north of the island, the dykes are distributed around a near-horizontal axis trending 300°, with most dipping at about 50° to the west. They commented . that the reversals in dyke dips were taken by Blake (see Mawson 1943) to be due to folding of originally-horizontal lava sheets. No systematic study has yet been made of the orientations of the dykes, of their thicknesses, or of the facings of their chilled edges. At Double Point, a transition may occur between extrusive rocks cut by many basaltic dykes into a sheeted dolerite dyke complex.

A recently-revised interpretative geological map of Macquarie Island, with superficial deposits omitted, is presented as figure 1. This map has been compiled from the mapping of Varne et al. (1969), Rubenach & Varne (1972), Griffin (1982), Christo- doulou et al. (1984), Crohn (1986) and Lees (1987). In general, there is little outcrop inland on Macquarie Island, and the map is therefore more reliable along the coast. The maps of Crohn (1986) and Lees (1987), which are mainly of the northern part of the island, indicate areas where outcrop is relatively good.

South of Sandy Bay and Bauer Bay, this new

48 R.A. Duncan and R. Varne

version differs little from that of Varne & Rubenach (1972), other than in the addition of the structural lineaments discovered by Ledingham & Peterson (1984). Compared with earlier maps of the northern part of the island, the main difference lies in the incorporation of a north-trending fault, northwest of Sandy Bay, mapped by Lees (1987) in its southern part. Recent east-side-up movement has occurred along this fault, which runs near-parallel to the young structural lineaments of Ledingham & Peterson (1984). In the map, this fault is sketched as continuing northwards to Hasselborough Bay, along the eastern contact of the massive gabbro, but has not yet been located by field mapping. The northnorthwest-trending faults cutting the east coast between Nuggets Point and Buckles Bay may splay off the main north-trending offshore struc- tural lineament drawn by Ledingham & Peterson (1984) and may also be young features.

PETROLOGY AND GEOCHEMISTRY OF THE IGNEOUS ROCKS OF THE ISLAND

Gabbros and peridotites are restricted to the northern third of the island. The peridotites are Serpentinised; most were originally harzburgitic assemblages of olivine, orthopyroxene and spinel (Varne & Rubenach 1972) although dunite has been found as loose blocks (Mawson 1943) and may outcrop northwest of Sandy Bay (Lees, per- sonal communication 1987). Layered sequences of gabbroic rock are exposed at Handspike Point and along Half Moon Bay (Varne & Rubenach 1972; Crohn 1986). At Handspike Point (fig. 1), olivine gabbro is common, although troctolite and dunite are the dominant rock types (Christodoulou et al. 1984). Several massive gabbro bodies are exposed along the north and east coasts, and gabbro probably underlies much of the plateau (fig. 1), although exposures there are poor.

Mawson (1943), Varne & Rubenach (1972), Cameron et al. (1981), Griffin & Varne (1980), and Griffin (1982) have described the petrography and chemical compositions of some basalts, dolerites and gabbros. Basalts and dolerites range from coarsely porphyritic to aphyric, and from almost wholly crystalline to almost wholly glassy. Many of the extrusive rocks have been altered and meta- morphosed, although fresh glass is also widespread. Where alteration is slight, crystalline groundmasses are mainly composed of clinopyroxene, plagioclase, Opaques, and recrystallised glass, with pseudo- morphs after olivine and rare amphibole in rela- tively alkaline varieties (Griffin & Varne 1980).

On North Head, Mawson Point, and in the Pyramid Peak area (fig. 1), the rocks are relatively fresh, and at worst have suffered only the smettite- carbonate alteration that is characteristic of ocean- floor weathering. Rocks from these three locations have been radiometrically dated. Elsewhere basalts were metamorphosed to higher grades, and Griffin (1982) described a section from the south of the island where the volcanic succession is about 1.4km thick, and passes from zeolite facies assemblages through to lower greenschist facies assemblages. Intense veining and alteration is associated with fault zones (Crohn 1986). The dolerites are compositionally similar to the basalts (Griffin & Varne 1980), but tend to be more altered, displaying a widespread replacement of primary mafic minerals by actinolite. However, plagioclase survives little-changed in the dolerites, as do their doleritic textures (Varne & Rubenach 1972, Banerjee et al. 1974, Griffin 1982). Locally the metamorphic grade in the dolerite dykes lies within the amphibole facies.

THE RADIOMETRIC AGES OF SOME MACQUARIE ISLAND ROCKS

Seven basaltic rocks were selected from the University of Tasmania Macquarie Island collection for radiometric analyses. Results are summarised in table 1; additional analytical details are available on request. All selected samples were well crystal- lised and compact (rare glass and vesicles) and showed the least low-temperature alteration in the collection.

K-Ar ages were determined for four whole rock samples and one hornblende mineral concen- trate from a fifth sample. The hornblende is a fresh, primary igneous phase and its age (11.50.3 Ma) is the time of crystallisation of its host basalt. The measured ages for the whole rock samples range from 3.6 to 9.6 Ma. Based on the petrographic evidence for low temperature formation of K-rich zeolites and clays (samples 47959, 47977 and 60635) and higher temperature metamorphism (sample 38248) these ages are minimum estimates of the true crystallisation ages. Replicate analyses of one of the samples (47977) show that the ages are reproducible and may relate to the age of alteration, from 5.8 to 9.6 Ma. The youngest measured age (3.6 Ma) comes from the stratigraphically deepest sample (38248) and may reflect the addition of K to the rock during chemical exchange with seawater.

The “°Ar/*’Ar experiments were undertaken to establish more reliable crystallisation and metamorphic ages for the basaltic rocks. Three samples (60635, 47977 and 38248) were investigated

Igneous rocks 49

TABLE 1 K-Ar and “’Ar/” Ar radiometric ages of basaltic rocks from Macquarie Island Sampl Titans K-Ar ages** “Ar/*’ Ar ages** %K Radiogenic “Ar %Radiogenic Age -+lo Total Plateau (x10°''mol/g) Ar (Ma) Fusion Age Age (recombined)

38248 (Nuggets 0.434 0.02748 5.9 3.60.3 6,940.7 6.50.5 Point) — dyke mie0 (Pyramid 0.389 0.80696 36.0 11.5-+0.3

eak) — dyke-hornblende Separate 47959 (Pyramid 0.464 0.54797 7.0 6.7£1.0

€ak) — lava flow 47977 (Aerial 0.999 1.01698 7.6 5.80.8 9.20.4 9.70.3

Ove) — lava 0.99950 7.6 5.7£0.8

Ow 60635 (Mawson 0.506 0.85452 6.7 9.640.9 13.4+0.3 11.5+0.3 Point) — lava flow 60685 (Hurd 7.2£0.6***

Oint) — dyke 60820 (Souchek 6.7+0.3***

ay) — Java flow

a LE ROE aN NE RN ENO OARS Vo noe ol onlay EO SOLE er ee Leen Cee ee

) ; 5 ‘ University of Tasmania collection numbers.

Ages calculated using the following decay and abundance constants: A, = 0.581x10°° yr‘;

\,= - -1, - -4. i tag > 4.962x10° yr; “°K/K = 1.167x10~ yr! mol/mol.

* . not recombined.

Using incremental heating methods (e.g. Walker & '¢Dougall 1982) which reveal internally consistent : Plateau” ages from the argon composition released 'N increasing temperature steps. Total fusion ages ere calculated for these samples by recombining © individual step compositions; another two Samples (60685 and 60820) were analysed in a Single (total fusion) experiment. Samples 60635 ‘nd 47977 are from near the top of the section and °xhibit zeolite facies metamorphism to cold sea- ater alteration only. These show well-defined i ateau ages at 11.5+0.3 and 9.7+0.3 Ma, respec- Bel which we interpret as original crystallisation Rese The older plateau age is identical to the K-Ar an of sample 47956 (11.5 Ma) which dates the dest volcanic activity at Macquarie Island. The unger age dates the eruption of more alkalic °Ws which comprise North Head. Samples 38248, 685 and 60820 are from the dyke swarm and mater Pillow lava section and their ages reflect the aad of greenschist metamorphism conditions. The 8 range, 6.5 to 7.2 Ma, records the time of closure these rocks to diffusion of argon. Griffin & Varne (1980) suggested that the

coastal traverse from North Head, via Hassel- borough Bay and Handspike Point to Eagle Point might represent a tilted section through ocean crust, with the extrusive rocks of North Head at the top, overlying dolerite dyke swarms and massive and layered gabbros, with serpentinised peridotites at the base. However, if volcanism at both Pyramid Peak and at North Head occurred at the Indian- Antarctic spreading ridge, this would imply a very low spreading rate of the order of 1 mm/yr.

A more reasonable interpretation is that the younger volcanism at North Head may have occurred away from the spreading axis. Off-axis seamount volcanism is observed in the Pacific Ocean (Batiza & Vanko 1984) and leads to the growth of submarine volcanoes, commonly located on or near transform faults, and ridge deflections. This would be compatible with the alkaline nature of the volcanism. Alternatively, the young north- trending lineaments and faults cutting the isthmus between North Head and Hasselborough Bay may mark the site of recent transcurrent movement along the Macquarie Ridge, throwing two dia- chronous sections of ocean crust against one

50 R.A. Duncan and R. Varne

another. The crustal section including North Head may terminate at the zone of major faulting between Bauer Bay and Sandy Bay. This could be in part the trace of an old fracture zone separating two segments of ocean crust.

SUMMARY

Macquarie Island is made up of fault-bounded blocks of oceanic lithosphere, planated by marine erosion as they were lifted and tilted during tectonic activity associated with the formation of the Macquarie Ridge. Radiometric ages for basalts range from 11.5 Mato 9.7 Ma, broadly compatible with the Middle Miocene ages of fossils in inter- calated sedimentary rocks, but younger than the age inferred from the sea-floor spreading anomaly pattern west of the Macquarie Ridge.

REFERENCES

BANERJEE, S.K., BUTLER, R.F. & STOUT, J.H., 1974: Magnetic properties and mineralogy of exposed oceanic crust on Macquarie Island. Zeitschr. fur Geophys., 40: 537-548.

BATIZA, R & VANKO, D., 1984: Petrology of young Pacific seamounts. J. Geophys. Res., 89: 11235-11260.

BUTLER, R.F., BANERJEE, S.K. & STOUT, J.H., 1976: Magnetic properties of oceanic pillow basalts: evidence from Macquarie Island. Geophys. J. R. Astr. Soc., 47: 179-196.

CAMERON, W.E., NISBET, E.G. & DIETRICH, V.J., 1981: Petrographic dissimilarities between ophiolitic and ocean-floor basalts. Proc. Internat. Ophiolite Symp. Nicosia, 1979: 182-192.

CHRISTODOULOU, C., GRIFFIN, B.J. & FODEN, J., 1984: The geology of Macquarie Island. ANARE Res. Notes, 21: 1-15.

COCKER, J.D., GRIFFIN, B.J. & MUEHLENBACHS, K., 1982: Oxygen and carbon isotope evidence for seawater-hydrothermal alteration of the Macquarie Island ophiolite. Earth Plan. Sci. Lett., 61: 112-122.

COLHOUN, E.A. & GOEDE, A., 1974: A reconnais- sance survey of the glaciation of Macquarie Island. Pap. Proc. R. Soc. Tasm., 108: 1-19.

CROHN, P.W., 1986: Geology and geomorphology of Macquarie Island with special emphasis on heavy metal trace element distribution. ANA RE Res. Notes, 39: 1-28.

GRIFFIN, B.J. & VARNE, R., 1980: The Macquarie Island ophiolite complex: mid-Tertiary oceanic lithosphere from a major ocean basin. Chem. Geol., 30: 285-308.

GRIFFIN, B.J., 1982: Igneous and Metamorphic Petrology of Lavas and Dykes of the Macquarie Island Ophiolite Complex. Unpublished Ph.D. thesis, University of Tasmania.

LEDINGHAM, R. & PETERSON, J.A., 1984: Raised beach deposits and the distribution of structural lineaments on Macquarie Island. Pap. Proc. R.

pe 506 Fam ue 223-235,

eR : REPORT ON GEOL OBSERVATIONS, MACQUARIE ISBNS DECEMBER 1986 TO JANUARY 1987 Ant- arctic Division, Kingston, Tasmania.

LEVI, S., BANERJEE, S.K., BESKE-DIEHL, S. & MOSKOWITZ, B., 1978: Limitations of ophio- lite complexes as models for the magnetic layer

of the oceanic lithosphere. G. oa aA phere. Geophys. Res. Lett.,

Teco & SUEELVAN, M.E., 1980: The extent of former glaciati i 5 Ssearehsti(Ts8)246-249) oa LOFFLER, E., SULLIVAN, M.E. & GILLISON. A.N tea verielacial landforms on Macquarie nd, Subantarctic. Zei EN aaa Zeitschr. fur Geomorph. MAWSON, D., 1943: Macquarie Island, its geography and geology. Australasian Antarctic Expedition a at Ser. A(5). Government Printing Office, Se eS aa Nee M.& WILCOXON, J.A : Miocene o ie Island, es teas om Macquarie Island. SELKIRK, D.R., SELKIRK, P.M. & GRIFFIN, K 1983: Palynological evidence for Holocene environmental change and uplift on Wireless ut pepeauarie island. Proc. Linn. Soc. N.S.W., SELKIRK, D.R., SELKIRK, P.M, & SEPPELT, R.D 1986: An annotated bibliography of Mac tafic Island. ANARE Res. Notes, 38: 1-133 : VARNE, R., GEE, R.D. & QUILTY, P.G., 1969: Macquarie Island and the cause of oceanic linear magnetic anomalies. Science, 166: 230-232 VARNE, R. & RUBENACH, M.J., 1972: Geolog of Macquarie Island and its relationship to ent crust. In Hayes, D.E. (Ed.): ANTARCTIC GEOLOGY Il. THE AUSTRALIAN-NEW ZEALAND SECTOR. Antarct. Res. Ser., 19: WALKER, D.A. & McDOUGALL, L., 1982: “Ar/22Ar and K-Ar dating of altered glassy volcanic rocks: the Dabi Volcanics, P.N.G. Geochim Costs chim. Acta, 46: 2818-2190, WILLIAMSON, P., 1978: The Palaeomagnetism of out- cropping oceanic crust on Macquarie Island. J- geol. Soc. Aust., 21(7): 387-394 WILLIAMSON, P., HAWKINS, L.V. & LONG. B. 1981: An examination of the possible occur: rence of seafloor spreading magnetic anomalies : ss MU Island. Mar. Geophys. Res., 5: WILLIAMSON, R., 1988: Origin, structural and tectonic history of the Macquarie Island region. Pap. Proc. R. Soc. Tasm., 12X(1):27-43..

(accepted 12 April 1988)

Papers and Proceedings of the Royal Society of Tasmania, Volume 122(1), 1988 51 THE SEISMICITY AND TECTONICS OF THE MACQUARIE RIDGE by T.D. Jones and K.F. McCue (with two tables and four text-figures)

JONES, T.D. & McCUE, K.F., 1988 (viii): The seismicity and tectonics of the Macquarie Ridge. Pap. Proc. R. Soc. Tasm., 122(1): 51-57. Papers presented at the Macquarie Island Symposium, Hobart, May 1987. ISSN 0080-4703. Australian Seismological Centre, Bureau of Mineral Resources, G.P.O. Box 378, Canberra, A.C.T., Australia 2601.

Nunee The Macquarie Ridge extends over 1200 km from southern New Zealand to a triple junction of the Pacific, oe tralian and Antarctic plates near 61°30’S, 161°00’E. Large earthquakes occur frequently on the ridge, with an age return period of one year for an event of magnitude 6.2 or more and 10 years for one of magnitude 7.2 or more.

a pasts of 16 focal mechanism solutions indicates that the direction of the principal stress is consistently horizontal, and Hite the northern and central ridge segments it strikes east-west. In the southern ridge segment, its azimuth is poorly

eH but appears to rotate towards the northeast with increasing southerly latitude. North of 51°S, the characteristic isplacement is along reverse dip-slip faults, whilst south of 51°S, fault motion is predominantly by right-lateral

‘trike-slip; throughout, the strike direction is parallel to the ridge axis. *y Words: Macquarie Ridge, earthquakes, strike-slip, seismicity.

INTRODUCTION

The Macquarie Ridge marks the southern- boundary between the Australian and Pacific spheric plates. It extends south from New ae to the triple junction at about 61°30’S, 00°E, where these two plates abut against the Baeerclic plate (Falconer 1972). The ridge is a Mplex topographic structure (fig. 1, from Hayes - alwani 1972) in juxtaposition with a trench ‘tem which cuts through it at about 50°S and cae neatly subdividing it into northern, central rid Southern segments. At its northern end, the 8€ Is separated from the continental (?) Campbell oe to the east by the Solander Trough, with uysegur Trench extending from 46°S to 49°S baie west. The central ridge, which is narrow and ed in the east by the Macquarie Trench, rises etitke Sea level at 54°30’S, at Macquarie Island, ti Can be traced from 51°S to 56°S. The southern cane Segment is a broad feature, sloping gently the aS to the Emerald basin and separated from ta Ustralian-Antarctic Rise in the west by the ae Trench which stretches from 56°S to 60°S the yes & Talwani 1972). Cross sections showing varying morphology through the three ridge Ments are shown in figure 2. Associated with the Macquarie Ridge system eILOW, belt of shallow seismicity of moderately activity (Gutenberg & Richter 1954). Epi- ¢s of the early earthquakes were located from Talian and New Zealand data, but since 1950 a

Most lith

Seg

Centr Ust

seismograph has been operated at Macquarie Island by the Australian Bureau of Mineral Resources (BMR), and the antarctic network too has slowly been extended. This improvement in the regional seismograph network has resulted in more accurate epicentral determinations and a database sufficient to compute focal mechanisms from first motion data.

SEISMICITY

We have used hypocentres from the Inter- national Seismological Centre (ISC) supplemented with data from the BMR datafile, the DSIR in Wellington, and the United States Geological Survey and its predecessors. They span the years 1960, which was the year that heralded the intro- duction of the Worldwide Standardised Seismo- graphic Network (WWSSN) and the consequent dramatic improvement in precision of hypocentral parameters, up to the end of 1984. Hypocentres of earthquakes 1918 to 1959 are from Gutenberg & Richter (1954) and those from 1960 are from the International Seismological Summary (ISS).

The historical record is very scanty, with human habitation of Macquarie Island limited to occasional visits by sealers or shipwrecked mariners prior to 1948 when the Australian National Ant- arctic Research Expedition established a base there. Probably the earliest report is of a series of violent earthquakes commencing on 31 October 1815, with aftershock activity continuing to April

52. T.D. Jones and K.F. McCue

FATHOMS : ee 160° - 164° 168° [4] 0- 1000 DY > 3000 _ ;

1000 - 2000 0M 5-6. YAS. TEALAND

2000 - 2500 46°

2500 — 3000

FIG. 1 — Topograph y and seismicity of the Macquarie Ridge. Earthquake epicentres are re J dots, the size indicating the magnitude. Presented by th

1816 (Sydney Gazette, 22 June 1816). The main shock is reported to have overthrown rocks and Sliven the ground a wave-like motion for some Seconds, several men being thrown off their legs and one considerably hurt. Another sealer claimed ht witnessed the fall of several mountains and the rocking of others, which seemed to have separated from the summit to the base. The precise epicentre and magnitude cannot of course be determined, ut the duration of the aftershock sequence and the act that such a long aftershock sequence was felt at all implies that the main shock had to be of Magnitude 6 or above and must have occurred very Close to Macquarie Island.

It is interesting to compare this with a more recent report. Extensive ground deformation was Observed around the shores of Macquarie Island following the 7 February 1980 earthquake of Magnitude 6.5. There were extensive rock falls and Mudslides and widespread ground fractures up to ie m long and 0.5 m deep, the largest wide enough

walk in (Ledingham 1981). Earthquakes were “ported felt through to mid-October. Allowing for € difference in time and writing style between 5; €s€ two reports, the effects seem to be fairly similar, so the earthquakes were probably of a Imilar Size and similarly located near the island. i Figure | shows epicentres for events of é 4gnitude 5.0 or more. With few exceptions the pidechtres of the post-1964 events form a single ere band only a few tens of kilometres wide, 'ch is within the locatable error. The focal sabulls are poorly constrained but, from the Ppearance of the seismograms, the foci appear to = very shallow and probably limited to the crust, ie a (according to Williamson & Johnson 1974) &a ss than 15 km thick under the central ridge ed Common seismological practice was to Sites a depth of 33 km, later 10 km, to earthquakes ‘a Obviously shallow but unconstrained depths, ine of a lack of close seismograph stations.

Sexton (1981) and Davies (1983) tabulated Ma mes for local earthquakes recorded at i" eat Island and found a significant number dist €S-P range of 0-3 sec, corresponding to focal ees of 0 to about 20 km and indicating that at

Some of the earthquakes are very shallow. er In the northern and central segments, earth- tides largely occur within the ridge rather than

. "the trench but this pattern is not so evident in Southern segment, where the relatively few fentres occur beneath or near the Hjort Trench. off There is a low level of background seismicity activi Tidge, commensurate with normal intraplate Ity

Seismicity and tectonics of the Macquarie Ridge 53

A pe ee: Sea level 167°E my 48°S = Campbell = 1000 Plateau o ee, Puysegur - 2000 Trench = © 3000

B Macquarie

Island 156° 161°E

ag Sea level 54.5°S i= i=] = 1000 & Macquarie = 2000 Trench a poenanel eo S 3000

(C;

° 162° > qe Sea level 57.8°S E ° -, = 1000 oe = 2000 = os ao a © 3000

FIG. 2 — Topographic cross sections across the north, central and south segments of the Macquarie Ridge.

The greatest activity currently occurs near the southern end of the northern ridge segment. In marked contrast, immediately to the south, where the trench cuts the ridge, there has been an almost complete absence of seismicity in the past quarter- century.

The central ridge is marked by seismicity which is evenly distributed along its 500 km length, in contrast with the southern segment, where recent earthquakes are sparse but relatively large in © magnitude. The lack of smaller events is not just a symptom of the poor seismographic coverage, since there were many small events in this period located farther southeast of the Hjort Trench, beyond the triple junction. :

Epicentres of the 12 largest earthquakes (table 1) which have occurred along the length of the ridge since 1918 are marked by the large dots in figure 1. Some of the epicentres are more than 100 km off the ridge axis, but this distance is within the uncertainty in the epicentres and the earth- quakes probably occurred along the ridge axis, though where on the ridge is impossible to determine.

54 T.D. Jones and K.F. McCue

o

ML<7, 1962 — 1984 Ms 27, 1920 - 1984

Cumulative number of earthquakes per year

S

4) Magnitude

FIG. 3 — Recurrence relation for Macquarie Ridge earthquakes.

MAGNITUDE RECURRENCE RELATIONS

A plot of the cumulative frequency of earthquakes versus magnitude, for the ridge seg- ment bounded by 47°-61°S and 155°-165°E, is shown in figure 3. The data cover the years 1962-84 for magnitudes less than Ms 7.0 and 1920-84 for magnitudes greater than or equal to Ms 7.0 (table 1). The linear portion of the plot yields a “b” value (slope) of 0.97 (£0.02). On average, the once-a-year earthquake has a magnitude of Ms 6.2, the 10-year earthquake a magnitude of Ms 7.2 and the extra- polated 100-year earthquake a magnitude of Ms 8.2. The largest earthquakes this century (table 1) occurred in 1924 and 1943 and were of magnitude Ms 7.8.

Asignificant feature of the recurrence relation is the relatively small number of moderate earth- quakes located for the period 1962-84. We interpret this flattening of the recurrence curve to mean that some earthquakes in that period up to a Richter magnitude of 5.5 were not located; hypocentres appear in the bulletins for only half of the expected number of earthquakes of magnitude 5. This situation probably continues to the present day, due to the scarcity of sensitive seismographs in the southwest Pacific and Antarctica.

FAULT PLANE SOLUTIONS

In table 2, details are given of the 16 earth- quakes for which fault plane solutions have been determined. The greatest and least principal stress axes are assumed to act at angles of 45° to the nodal planes. In addition to those already published, the

+ 150

0 200km pS eect

5 a *C1 15 fe— 14 i? < 4 °. a os | u/ - “4 ©} Fault plane solution with 16 ~ Principal stress direction

Relative plate motion Dip slip ——™ Strike slip L 24026

As 4

FIG. 4 — Fault plane solution coincide with those in table 2. Ss. The numbers

authors have drawn up an additional sj

solutions from the ISC bulletins ARE that have occurred since the inception Of the WWSSN. All events lie on the Macquarie Ridge except for event 16 which is located on the era fault between the Australian and Antarctic plates near the triple junction. a

The five solutions (fig. 4, events I-

northern Macquarie Rideetene, ae ae ia the Puysegur Bank earthquake of 12 October 1979 which was felt widely through the southern island of New Zealand. All five are east-west thrust mechanisms, as is that for 3] January 1985, off the southwest coast of New Zealand USC Bulletin January 1985); the mean azimuth for the eet axis is 257° and the mean plunge 27°. In ee of the five cases, one of the nodal planes dips at a shallow angle in an approximately easterly direction, and in all five cases the complementary nodal plane dips steeply in an approximately westerly direction

: The Six solutions (events 6-1 1) for the central ridge section have remarkably similar directions for the pressure axis, with a mean value azimuth of

Seismicity and tectonics of the Macquarie Ridge 55

TABLE 1 Major earthquakes along the Macquarie Ridge Date Time Latitude Longitude Depth Mag.* y Tithe deisel ei °s oF km

1924 067 = 26" 20137, 56 157.5 33 7.8 1924 O7ne 4 04055 49.5 159 50 7.5 1926 10) 103) 19) 38 49 161 50 7.5 1936 02522) ees | 49.5 164 33 7.2 1943 09 #06 #03 41 55.1 158.5 0 7.8 1960 1213) 07, 36 52.09 161.04 0 7A 1964 OOF 12) 222) 07, 48.95 164.46 33 (7.0) 1965 Og 02). 183 19 55.90 157.70 33 (7.1) 1970 (on lem 10,946 58.86 157.60 3 7.2 1979 1 AR AD Bs 46.54 165.91 33 7.2 1981 05h 25) 05,8 25 48.55 164.70 0 7.4 1982 OF AD 8 51.15 160.64 10 7.0

* Magnitudes are surface-wave Ms. (7.0) implies Ms computed from body-wave magnitude MB.

TABLE 2 Fault plane solutions for Macquarie Ridge earthquakes* No. Date Time Lat. Long. Depth Mag. Principal Stress Axes Source** y md hm p axis t axis pl/az pl/az 1 1979 10 12 10 25 46.54 165.91 2k) Th 34/251 54/049 4 2 1981 05 25 05 25 48.55 164.70 0 7.4 27/241 62/080 l 3 1964 09 12 22 07 48.95 164.46 33 6.1MB 17/274 73/004 2 4 1964 11 08 02 43 48.98 164.20 33 5.6MB 28/236 62/056 3 5 1967 09 20 09 39 49.57 163.93 34 5.8MB 27/281 63/101 6 6 1982 07 07 +10 43° SI.15 160.64 1034720 08/101 07/193 I 4 1984 05 23 05 16 51.88 161.08 10 5.8 05/100 02/192 l 9 1969 06 17 23 58 52.53 159.70 33 6.6 27/098 02/006 l 10 1966597057225) 19) 20) o277n 100i, 35" 0,0MB 09/078 81/258 8 1 19795 0423) 21 54.5203, 159184 10 6.0 05/089 - 20/181 l 1D 1977? (07, Qi ll, 03153182 158184) 33h OeT, 00/084 89/347 4 3 ED We Poe Tih SA 1s) TM) 26/271 35/161 5 14 1965 08 02 13 19 55.90 157.70 33 6.2MB 26/051 16/313 8] 15 196305) 12).09) 43) 57.06 15978 335. 2014 35/116 39/240 6 16 1970 06 Il 16 46 58.86 157.60 33 7.2 07/201 43/107 1 : 1983) (04°27) Ol) 39° 62/50 155:30) = 1063519 07/105 01/015 7 M

fr agnitude is Ms (surface-wave scale) or MB (body-wave scale). Time is Universal Time, hypocentres are +4 ™ the ISC or ISS. cyene mechanism sources are (1) Jones & McCue, this paper; (2) Sykes (1967); (3) Denham (1977); eters etal. (1985); (5) Dziewonski et al. (1983); (6) Johnson & Molnar (1972); (7) ISC; (8) Banghar & 69)

56 T.D. Jones and K.F. McCue

92° and a mean plunge of 9°. Solutions 6, 7, 8 and 10 are almost pure strike-slip, and solutions 9 and 11 are pure east-west thrust mechanisms.

Earthquake 12 occurred some 100 km east of the trench crossover which separates the central and southern ridge sections, and its mechanism is dominantly strike-slip, similar to the strike-slip mechanisms of earthquakes on the central ridge.

Earthquakes 13 and 15 seemingly occurred beneath the Hjort Trench, the axis of which rotates increasingly counterclockwise in the south. The mechanism for each contains a strong component of strike-slip motion, with one of the nodal planes striking parallel to the trench axis; the direction of the pressure axis for the more southerly event (15) is accordingly northnortheast compared to the northeast direction of the pressure axis for event 13.

Earthquake 14 occurred in the eastern part of the broad southern ridge and its fault plane solution is a combination of thrust faulting at 116°, and of strike-slip faulting. However, while the direction of thrust faulting is approximately east-west, in agreement with other thrust mechanisms along the ridge, the strike-slip component in the mechanism implies a sense of movement opposite to that of the other strike-slip mechanism, i.e. left-lateral slip along the ridge axis rather than right-lateral slip, if slip occurs along the nodal planes parallel to the ridge axis.

The focal mechanism for the earthquake of 27 April 1983 (event 16), which occurred on the Balleny Fracture Zone, resembles closely the strike-slip mechanisms of the central ridge, with a pressure axis almost east-west.

DISCUSSION

At least along the northern and central Macquarie Ridge the foci are under the ridge rather than the parallel trench. The epicentres form a single, rather narrow band until well south of Macquarie Island, where earthquakes are appar- ently less frequent than in the north, although the location capability decreases with distance south along the ridge.

There is some suggestion that the trench- ridge crossover points mark the places where the largest earthquakes have occurred this century, yet over the last 25 years there has been a relative lack of small and moderate earthquakes. They may be sites where either the accumulating stress is lower than in the neighbouring ridge segments because it has been released by previous large earthquakes, or where the stress level is higher but so also is the

crustal strength.

Fr Our ability to decide between the two choices is limited by the uncertainty in position of the large earthquakes earlier in the century, but the question may be clarified by an analysis of slippage rates along the ridge, which can be calculated from estimated fault dimensions of the largest recent earthquakes.

More frequent seismicity accounts for stress release along the northern and, in particular, the central Macquarie Ridge segments. The level of seismicity is lower along the southern ridge segment than the northern segments, possibly due to its position relative to the pole of rotation of the Pacific plate.

___ The lack of moderate or deep seismici indicates that no well-developed siteduistics tooo exists. Perhaps the onset of subduction has only recently begun, analogous to the plate boundary along the northwest coast of the United States where there is also an absence of subcrustal seis- micity. In the Fiordland region of New Zealand there is active subduction of the Australian plate (Christoffel & van der Linden 1972): studies by the authors with P.E. Williamson, suggest that in the central Macquarie Ridge the Australian plate has been foreshortened and thickened by approximately 65 km in a process preceding subduction.

__ The frequency of large earthquakes is quite high, with average return periods of | and 10 years for earthquakes of magnitude 6.2 and 7 2 respec- tively. The slope of the recurrence relation op" is 0.97 (£0.02). The decrease in “b” with decreasing magnitude Suggests that some earthquakes, even up to magnitude 5.5, are not being located

We interpret the fault plane solutions in the following way. In the northern Macquarie Ridge segment, where the trench is to the west of the ridge, the nodal plane which dips shallowly east could represent a fault plane on which subduction has recently begun in a similar sense to that off the southwest coast of New Zealand.

In the central ridge segment 5 have both nodal planes striking niente wee and either one could be the fault plane. For the strike-slip solutions, the nodal plane parallelin the ridge is probably the fault plane, in which caret sense vf moun is right-lateral.

n the southern ridge sepm t planes also lie along the ridge eau Pas: inferred mechanisms are again strike-slip ina right- lateral sense. Hence, there is a Consistent regime 0 right-lateral strike-slip motion, albeit with some element of east-west thrusting, along the ridge from about latitude 51°S to the triple junction

If we choose the nodal plane parallel to the

Seismicity and tectonics of the Macquarie Ridge 57

transform fault as the fault plane for event 16, then the motion is left-lateral strike-slip, which is con- Sistent with the motion on the transforms along the South-Indian and Pacific Ocean spreading centres. The only inconsistent mechanism is that for €vent 14 and the solution has not been re-examined. his earthquake occurred in May 1963, before the Completion of the World Standard seismograph network, and it is possible that few data were available to determine its mechanism.

CONCLUSIONS

From the fault plane solutions we infer that Movement along the ridge is consistent with an tauclockwise rotation of the Pacific plate, relative ‘ the Australian plate, about a pole located east of at central or south Macquarie Ridge and within are 1000 km of the ridge, as suggested by other i Ors (CI in fig. 4 from Christoffel & van der

den 1972, Minster & Jordan 1978, Johnson & Olnar 1972). Such a rotation would account for © Compressive regime producing subduction in lordland, New Zealand, possible incipient sub- ae beneath the northern Macquarie Ridge, a the combination of right-lateral strike-slip and West thrusting observed in the central Mac- TENG Ridge. It is also consistent with the observed ight-lateral strike-slip regime in the south Bere atic Ridge section, which continues south-

St to the triple junction.

ACKNOWLEDGEMENTS

Peewee wish to thank Natasha Kozin and Brian

con ey who drafted the figures, James Lewis who

; Piled much of the data, and Leonie Jones who fred many useful suggestions.

POSTSCRIPT

as On 3 September 1987, at 0640 UT, a major

ap quake occurred beneath the Hjort Trench at

Sealey 59°12’S, 158° 18’E. It was a shallow

Tan quake and the surface wave magnitude was

will nd 7,3, The fault plane solution for this event

int 8lve additional control on the tectonic regime € southern ridge area.

(accepted 5 January 1988)

REFERENCES

BANGHAR, A.R. & SYKES, L.R., 1969: Focal mech- anisms of earthquakes in the Indian Ocean and adjacent regions. J. Geophys. Res., 74(2): 632-649.

CHRISTOFFEL, D.A. & VAN DER LINDEN, W.J.M., 1972: Macquarie Ridge-New Zealand Alpine Fault transitions. Jn Hayes, D.E. (Ed.): ANT- ARCTIC OCEANOLOGY II: THE AUSTRA- LIAN-NEW ZEALAND SECTOR. Antarctic Research Series, 19: 235-242.

DAVIES, P.M., 1983: Macquarie Island Geophysical Observatory Annual Report. Bur. Min. Resour. Rec. 83(2): 39 pp.

DENHAM, D., 1977: SUMMARY OF EARTHQUAKE FOCAL MECHANISMS FOR THE WEST- ERN PACIFIC-INDONESIAN REGION, 1929-1973. Report SE-3, World Data Centre A for Solid Earth Geophysics: 110 pp.

DZIEWONSKI, A., FRIEDMAN, A., GIARDINI, D. & WOODHOUSE, J.H., 1983: Global seismicity of 1982: centroid-moment tensor solutions for 308 earthquakes. Phys. Earth Planet. Inter., 33: 76-90.

FALCONER, R.K.H., 1972: The Indian-Antarctic- Pacific triple junction. Earth Planet. Sci. Lett., 17: 151-158.

GIARDINI, D., DZIEWONSKI, A. & WOODHOUSE, J., 1985: Centroid-moment tensor solutions for 113 large earthquakes in 1977-1980. Phys. Earth Planet. Inter., 40> 259-272.

GUTENBERG, B. & RICHTER, C.F., 1954. SEIS- MICITY OF THE EARTH. Princeton Uni- versity Press: 310 pp.

HAYES, D.E. & TALWANI, M., 1972: Geophysical investigations of the Macquarie Ridge Complex. In Hayes, D.E. (Ed.): ANTARCTIC OCEAN- OLOGY II: THE AUSTRALIAN-NEW ZEA- LAND SECTOR. Antarctic Research Series 19: 211-234.

INTERNATIONAL SEISMOLOGICAL CENTRE, 1961-84: Bulletin of the International Seismo- logical Centre, 1(1)-21(12).

JOHNSON, T. & MOLNAR, P., 1972: Focal mech- anisms and plate tectonics of the southwest Pacific. J. Geophys. Res., 77(26): 5000-5032.

LEDINGHAM, R., 1981: Macquarie shakes again. Aurora, October 1981: 6-8.

MINSTER, J.B. & JORDAN, T.H., 1978: Present day plate motions. J. Geophys. Res., 10: 5331-5334.

SEXTON, M.J., 1981: Macquarie Island Geophysical Observatory Annual Report. Bur. Min. Resour. Rec., 81(17): 23 pp.

SYKES, L.R., 1967: Mechanisms of earthquakes and nature of faulting on the mid-oceanic ridges. J. Geophys. Res., 72(8): 2131-2153.

WILLIAMSON, P. & JOHNSON, B.D., 1974: Crustal structure of the central region of the Macquarie Ridge complex from gravity studies. Mar. Geophys. Res., 2: 127-132.

Papers and Proceedings of the Royal Society of Tasmania, Volume 122(1), 1988 DY

TERRACE TYPES AND VEGETATION DYNAMICS ON MACQUARIE ISLAND

by P.M. Selkirk, D.A. Adamson and R.D. Seppelt

(with two text-figures and three plates)

SELKIRK, P.M., ADAMSON, D.A. & SEPPELT, R.D., 1988 (viii): Terrace types and vegetation dynamics on Macquarie Island. Pap. Proc. R. Soc. Tasm., 122(1): 59-64, 3 pl. Papers presented at the Macquarie Island Symposium, Hobart, May 1987. ISSN 0080-4703. School of Biological Sciences, Macquarie University, New South Wales, Australia 2109; and Antarctic Division, Channel Highway, Kingston,

Tasmania, Australia 7150.

thn Feldmark vegetation occurs extensively on the plateau of Macquarie Island above about 200 m. Alternating ae vegetation and gravel are frequently associated with terraced terrain. Detailed study of 38 terraced sites with mereuuward aspects shows vegetation-terrace interaction to be active under present climatic conditions. Wind, Sture, hillslope, slope stability and vegetation all affect terrace form. ey Words: feldmark vegetation, terrace formation, vegetation dynamics, Macquarie Island.

INTRODUCTION

Macquarie Island is a small subantarctic ‘nd in the Southern Ocean, southeast of jenni the path of strong west to north- wee winds. Feldmark (or fellfield or fjaeld- : ) vegetation occurs extensively on the island Plateau above about 200 m altitude. Amongst the ria Species recorded in the feldmark, bryo- oe rae the cushion-forming vascular plant, om e 4 selago, are dominant. Vegetation cover is remely variable. Large expanses of gravel-

Tog ground are common (Selkirk & Seppelt

islan

Vegetation disposition ranges from regularly ating stripes of vegetation and gravel to sat vegetation patches amongst gravel. Regu- Sg es es be associated with “steps and slopes frrain, “Leeward terraces "on east-facing Orin eee which vegetation forms risers and gravel aviss rads of the stairs, have been described by Slopes ( 955). “Windward terraces” on west-facing acral less regular, with gravel risers and ne won treads. Taylor interpreted both terrace Prevaili, responses by vegetation to the present (1983) ng strong westerly winds. Léffler er al. lerasee se Loffler (1983) regarded windward relict ar wind formed, but leeward terraces as ormer Mluction landforms developed during a terraces More severe climate. They described eeward Ay slopes with neither windward nor aspects as intermediate in form. Struc stton & Gill (1965) described in detail the Ure of terraces on windward slopes on North

altern Trepy

Veg,

Mountain and the involvement of wind in their . formation, maintenance and gradual upslope progression.

This paper concentrates on terraces on hills with other than windward aspects, following observation of a wide variety of terrace forms and dimensions: wind, moisture, hillslope, slope stabi- lity and vegetation all affect terrace form. Studies were also made of dynamic features of terraces which imply that they are not merely inherited landforms from a glacial period, but that develop- ment, stabilisation and destruction are all con- temporary processes. Although terrace initiation is difficult to observe, it is believed that this could develop on unterraced hillslopes in the present environment, as, for example, following landslip.

CLIMATIC FACTORS AND FROST SORTING

Although climatic records for Macquarie Island indicate that average daily minimum, mean and maximum temperatures are all above 0°C, 51 days per year, on average, have air temperatures below zero, and 99 days per year have surface ground temperatures below zero (LOffler 1983). Ground temperatures at depths greater than 100 mm approximate air temperatures, so soil frost formation is restricted to the surface 100 mm or so (L6ffler er al. 1983).

As meteorological data on the island are recorded at 6m a.s.l., and temperatures are approximately I°C lower for each 100 m increase

60 P.M. Selkirk, D.A. Adamson and R.D. Seppelt

in altitude, occurrence of ground frost at plateau sites is likely to be significantly more frequent. Loffler (1983) estimated there would be approxi- mately 170 days with frost at 200 m a.s.1. Precipi- tation averages about 900mm per annum at the isthmus weather station, and 42% more than this at the northern end of the plateau (Mallis, this volume) so moisture for needle ice formation on the plateau is abundant. : On bare ground on the plateau, stone sorting resulting from frost action is frequently seen. Loffler et al. (1983) described small stone polygons on level ground, and stone stripes on ground sloping at between 2° and 20°. We have also observed frost-mediated general downslope move- ment of gravel, sorted so that larger stones are towards the edges and front of active chutes.

VEGETATION AND TERRACE OBSERVATIONS

Vegetation disposition in feldmark is clearly associated with hillslope, varying with increasing steepness.

No Slope

On more or less level ground, vegetation may consist of irregularly placed moss polsters and Azorella cushions (e.g. Boot Hill) or elongated stripes of moss, aligned at an angle of about 20° to the north of prevailing wind direction (e.g. on Windy Ridge: an unofficial name for a ridge | km southeast of Mt Fletcher). In some areas with slopes of less than 2° these moss stripes may be arranged in bands of alternating vegetation and bare gravel, associated with very slight terracing of the ground (e.g. Windy Ridge) (plate 1). This type of vegetation stripe has not previously been des- cribed from Macquarie Island.

Gentle Slopes

On gently sloping ground (about 0-15°) hill- sides are markedly terraced approximately along the contours, with vegetated terrace risers and gravel-covered terrace treads (e.g. at Ratsak Junction: an unofficial name for the track junction on the southeast flank of Mt Martin) (plate 2 and fig. 1A). Treads may slope slightly (1-2°) either outwards or inwards, and usually slope slightly (0-S°) along their length (fig.2). These are the leeward terraces of Taylor (1955), Léffler er al. (1983) and LOoffler (1983).

no lag, fines erode, vegetation undercut

gravel lag at front of tread

vegetation skirt gravel lag over entire tread

FIG. | — Notional sections through terraces approximately along contours. (A) Risers are vegetated, treads have gravel lag. (B) Vegetation on risers can be undercut at base when fines are eroded from a surface lacking gravel lag. (C) Vegetation extending as a“‘skirt” from the riser above can help stabilise the back of a tread.

Terrace vegetation dynamics yield several minor variations in the details of plant and gravel boundaries.

At the back of a tread, near the base of the vegetation of the riser above, stirring by needle ice action, followed by water washing down the riser, can gradually erode fine material away (fig. 1B). Over time, this will steepen the angle of the riser. and, as vegetation slowly grows into gravel at the front of the tread above, the riser will move upslope. The hill remains terraced but individual terraces are not fixed in their form or positions 09 the slope. In other cases, the back of the tread may be well stabilised by vegetation extending onto the tread from the riser above, as described below (fig. 1C).

Vegetation on the risers is predominantly Azorella, with Agrostis magellanica, Festucé contracta, Acaena minor, Ranunculus biternatus and a variety of bryophytes and lichens. Rhaco- mitrium crispulum and Ditrichum strictum some times form a “skirt” along the base of the riset

Terrace types and vegetation dynamics 61

PLATE 1 Bands of moss stripes alternate with bare gravel on very slightly terraced ground on Windy Ridge, approximately | km southeast of Mt Fletcher. Geology hammer gives scale.

PLATE 2 Hillslope terraced approximately along contours with vegetated risers and gravel covered treads, southeast flank of Mt Martin.

62 P.M. Selkirk, D.A. Adamson and R.D. Seppelt

30 ae safe a cor - @ a o . oe oon o a 2 el aheoge _ oe 2 20 . ae z= . as; COL . ee . poten . 4 . en at ° ° ee % 104- oe . = . aed .

i) 5 10 15 20 25 Slope along terrace tread (degrees)

FIG. 2 — Plot of slope along terrace tread against general hillslope for 38 sites. The change from subhorizontal terrace treads to sloping treads occurs at hillslopes of approximately 11°. Re- gression equation: hillslope = 10.9 + 0.71 treadslope.

extending a variable distance onto the back (or inner edge) of the gravel tread. This “skirt” helps stabilise the terraces and minimise their upslope migration by protecting the back of the tread and foot of the riser from erosion.

Frost sorting of surface gravel on treads is common, with larger stones sorting to the front, or outer edge of the tread. Similar sorting occurs on the chutes which lead from one tread to the tread below. Only a few scattered polsters of Ditrichum Strictum, Andreaea spp. and Rhacomitrium crispulum occur on treads on which frost sorting appears active. On such treads, a high proportion of fines is visible at the surface amongst the gravel. Oia treads where only a very small proportion of fines is visible, frost sorting of gravel appears less active, and more moss polsters, small areas of Rhacomitrium crispulum turf, small Azorella selago clumps and runners of Ranunculus biter- natus and Acaena minor occur.

In a few places, completely vegetated terraces occur (e.g. southwest of Mt Martin; west of the largest of the three lakes south of Mt Blair). These completely vegetated terraces do not appear to have been described previously on Macquarie Island. The terrace tread surface becomes com- pletely vegetated, presumably having stabilised after fines have been washed out, frost heaving has decreased, and additional supply of debris from above has been exhausted.

Measured across their surfaces at right angles to contours, treads and risers are wider on gently sloping hillsides than on slightly steeper slopes.

This is often visible on a single hillside, terraced approximately parallel to its contours, where wide terraces occur low on the hill and narrower terraces higher up.

Where terrace treads are not quite parallel to contours but slope slightly along their length, the tread may itself be terraced with alternating vege- tation and gravel stripes. This gives the appearance of intersecting terraces (e.g. approximately | km south of Mt Martin). Rarely, on much steeper slopes, completely unvegetated transverse terracing of gravel treads was observed (e.g. on the southeast flank of Mt Hamilton).

Moderately Steep Slopes

On more steeply sloping sites, both terraces and associated vegetation and gravel stripes are oriented at an increasingly steep angle to the contours. Figure 2 shows that this change from subhorizontal or gently sloping terraces occurs at a hillslope of approximately 11°. Ata particular site, this threshold seems to be affected by a number of factors, including the size of gravel fragments and proportion of gravel to fines in the tread material, presence of bedrock at or near the surface to which the terrace materials become pinned, available moisture, wind exposure, altitude and hence extent of needle ice formation.

As frost action moves gravel downslope along terrace treads, larger stones are sorted to the outer edges of treads (e.g. southern shoulder of Mt Hamilton) and to the edges of chutes.

It is common to see gravel overriding vege- tation at the top of risers (contrary to the observa- tions of Loffler er a/. 1983), to see spills of gravel across vegetation onto lower levels (plate 3), and to see chutes of gravel connecting one tread to a lower one (e.g. northeast of Mt Hamilton). Surface water flows downslope from tread to tread along such chutes, washing fine material downwards. The orientation of frost stripes on the treads, where they occur, indicates the direction of gravel move- ment and erosion in general.

Extremely Steep Slopes

On slopes on which gravel movement is active, vegetation occurs in patches elongated downslope, alternating with gravel stripes. Terrac- ing is apparent at some such sites (e.g. high on the south face of Mt Hamilton).

PLATE 3 Gravel commonly overrides vegetation at top of risers, spilling across vegetation onto lower levels. Notebook (150 x 90 mm) gives scale.

Other Sites

Although the terraces so far discussed have vegetated risers, terraces are not necessarily depen- dent on vegetation banking for either their forma- tion or maintenance. Terracing in completely unvegetated gravel is very occasionally seen (as _ described at end of “Gentle Slopes” section above). At several sites (e.g. 750 m northeast of Mt Law; at the foot of the southern side of Mt Tulloch) terraces occur in short grassland. Gravel treads are well dotted with Ditrichum strictum polsters; the change of slope from tread to riser occurs in gravel; the upper third of the riser is devoid of vegetation, while the lower two-thirds supports short grassland vegetation (principally Luzula crinita, Agrostis magellanica, Festuca contracta, Acaena minor).

Terrace types and vegetation dynamics 63

This type of terrace has not previously been described on Macquarie Island.

INTERPRETATION OF TERRACES

Inthe Macquarie Island plateau environment there are suitable temperature and moisture condi- tions for needle ice formation in the top 100 mm of soil. When a mixture of fines and gravel is present, frost action causes sorting and, even on gently sloping sites, downslope movement of gravel and removal of fines from the surface leaves a surface lag of gravel. Although strong westerly to north- westerly winds are common on the island, on non- windward terraced slopes most movement of fines is downslope, washed by water from the almost constant precipitation.

Ata few sites unvegetated terraces occur, but more commonly vegetation clothes the risers of terraces, stabilising them. Sorting of stones and downslope movement of gravel occur only on the treads. Gravel particles follow a zigzag path downa terraced slope, via terrace treads and intercon- necting gravel chutes. Terracing lengthens the path and reduces the angle of slope down which particles travel during erosion of hillslopes.

Figure | and observations of the instability of the surface material imply that the mantle of mineral particles on the hillslopes is mobile and moving downslope, even if the form of the terraces may move upslope. Bedrock surfaces bear no relation to the surface profile: they may be smooth beneath strongly developed terraces. Terracing is a feature of the mobile mantle.

The nearly contour-parallel terraces resemble some forms of unsorted steps described from periglacial environments. Although Washburn (1979) described non-sorted steps as elongated downslope, Embleton & King (1975) described some forms parallel with slope contours, with vegetated risers and bare treads. Steps merge into stripes running downslope as slope increases. Embleton & King (1975) also commented that neither steps nor stripes need necessarily be associated with permafrost.

Kirkpatrick (1984) described a change from non-sorted stone steps to non-sorted stone stripes at a general slope angle of approximately 8° in some alpine areas of Tasmania. There, risers are well vegetated with herbs and shrubs to 0.5 m tall. Our observations of terraces subparallel to contours on low hillslopes and angled terraces on hillslopes steeper than approximately 11° seem analagous to Kirkpatrick’s.

64 P.M. Selkirk, D.A. Adamson and R.D. Seppelt

Observations show that terrace-vegetation interaction on Macquarie Island is active at present. Although terracing on hillslopes is likely to have been present for a considerable length of time, it is not necessary to postulate that terraces are relict features or that they required a more severe former climate for their formation.

ACKNOWLEDGEMENTS

Permission from the Macquarie Island Advi- sory Committee to conduct fieldwork on the island and logistic support from the Australian Antarctic Division are gratefully acknowledged. Financial assistance was provided by the Australian Research Grants Scheme and a Macquarie University Research Grant.

REFERENCES

ASHTON, D.H. & GILL, A.M., 1965: Pattern and process in a Macquarie Island feldmark. Proc. R. Soc. Vict., 79: 235-245.

EMBLETON, C. & KING, C.A.M., 1975: PERI- GLACIAL GEOMORPHOLOGY. John Wiley and Sons, New York. :

KIRKPATRICK, J.B., 1984: Tasmanian high mountain vegetation Il — Rocky Hill and Pyramid Mountain. Pap. Proc. R. Soc. Tasm., 118: 5-20.

LOFFLER, E., 1983: Macquarie Island — eine vom Wind gepragte Naturlandschaft in der Sub- Antarktis. Polarforschung, 53(1): 59-74.

LOFFLER, E., SULLIVAN, M.E. & GILLISON, A.N., 1983: Periglacial landforms on Macquarie Island, Subantarctic. Zeitschr. Geomorph. N. F., 27(2): 223-236.

MALLIS, M., 1988: A quantitative investigation into scavenging of airborne sea salt over Macquarie Island. Pap. Proc. R. Soc. Tasm., 122(1): 121-128.

SELKIRK, P.M. & SEPPELT, R.D., 1984: Fellfield on Macquarie Island. Tasm. Nat., 78: 24-26.

TAYLOR, B.W., 1955; The flora, vegetation and soils of Macquarie Island. ANARE Rep., Ser. B(Il), Botany.

WASHBURN, A.L., 1979: GEOCRYOLOGY. Edward Arnold, London.

(accepted 2 October 1987)

Papers and Proceedings of the Royal Society of Tasmania, Volume 122(1), 1988 65

LANDFORMS OF AEOLIAN, TECTONIC AND MARINE ORIGIN IN THE BAUER BAY-SANDY BAY REGION OF SUBANTARCTIC MACQUARIE ISLAND

by D.A. Adamson, P.M. Selkirk and E.A. Colhoun (with seven text-figures and twelve plates)

ADAMSON, D.A., SELKIRK, P.M. & COLHOUN, E.A., 1988 (viii): Landforms of aeolian, tectonic and marine origin in the Bauer Bay-Sandy Bay region of subantarctic Macquarie Island. Pap. Proc. R. Soc. Tasm., 122(1): 65-82, 12 pl. Papers presented at the Macquarie Island Symposium, Hobart, May 1987. ISSN 0080-4703. School of Biological Sciences and Quaternary Research Group, Macquarie University, New South Wales, Australia 2109; and Department of Geography, University of Newcastle, New South Wales, Australia 2308.

Aeolian landforms on Macquarie Island, in the Southern Ocean, occur above 100 m on the plateau between Bauer Bay and Sandy Bay. Anextensive sand sheet, at least 6000 to 7000 years old, is wind and water eroded to produce blow-outs with bare troughs and vegetated elevated margins. Sand moved eastwards before the prevailing winds. Wind-polished bedrock on the western edge of the plateau is a relict feature. Wind-polishing of bedrock occurs today near sea level at the head of Bauer Bay. The area is tectonically active. No evidence of former glacial activity was observed. Topographic features in this area, previously regarded as glacial, are attributed to faulting. Tectonic uplift and sea-level changes are invoked to explain the formation and age of the broad and gently sloping raised marine terrace around much of the western and northern coasts. This explanation also sets constraints on the ages

of the sand sheet and the cobble beach at about 100m on the plateau east of Bauer Bay. Key Words: tectonic activity, fault movement, aeolian landforms, marine landforms, Macquarie Island.

INTRODUCTION

Macquarie Island (54°37’S, 158°53’E) is the highest point on the Macquarie Ridge, part of the tectonically active boundary between the Pacific and Australian plates. The plates are moving northwards at different rates with some conver- gence. The island is in a zone of frequent earth- quakes (Jones & McCue, this volume), crustal shortening (Williamson, this volume), block fault- ing (Duncan & Varne, this volume), and relatively rapid uplift (Colhoun & Goede 1973, Selkirk et al. 1983). Macquarie Island is seamed by scores of active normal faults, many represented by youthful scarps.

At 54°S, Macquarie Island is in the path of high latitude cyclones. At the meteorological station near sea level, annual average wind speed is 8.3 ms ', annual precipitation about 900 mm, annual mean temperature 4.8°C and mean temperature for the coolest month (July) 3.3°C. Most of the island is between 100 and 300 m elevation. Snow can fall at any time but lies only briefly.

Hamilton (1926) reported that the only sand area on the island occurred on the east coast at its northern end, immediately south of the junction

between the isthmus and the main island. Taylor (1955) noted the abundant sand at the head of Bauer Bay. Other areas of sand accumulation occur, particularly on the plateau on the western side of the island. The present paper describes and puts age constraints on an extensive plateau sand sheet east of Bauer Bay.

There is a long-standing and still active controversy over glaciation on the island during the last global glacial period, about I5 to 25 thousand years ago, when world sea level was some 100 to 140m lower than today. Crohn (1986) maintains that glaciation affected most or all of the present plateau surface, whereas Ledingham & Peterson (1984) consider that many features pre- viously ascribed to glaciation are better explained by tectonic activity, particularly as beach cobbles- occur at high altitudes in supposedly glaciated areas. The present paper supports and develops Ledingham and Peterson’s view that faulting is responsible for landforms previously interpreted as glacial.

A raised marine terrace (Blake in Mawson 1943, Colhoun & Goede 1973) extends along: the northern two-thirds of the west coast of the island — a gently sloping platform rising to about 15m

66 D.A. Adamson, P.M. Selkirk and E. A. Colhoun

a.s.l., up to 1km wide and mantled by peat and mire. It extends from the shoreline (tidal range about | m; Bye, pers. comm.) to the base of the steep slopes or cliffs which bound the plateau. Elsewhere along the coast, except in the far south- east near Hurd Point, steep slopes and cliffs rise to the high plateau directly from the ocean or froma narrow rocky shore.

This paper describes landforms in the area near Bauer Bay and Sandy Bay, where there is interplay of past and present aeolian, tectonic, fluvial and marine processes. An extensive sand- sheet, containing terrestrial peats of mid to early Holocene age and now being dissected by water and wind erosion, occurs at over 100 maz.s.]. Near sea level, present-day aeolian processes are also active. Recent faulting has affected drainage, particularly on the eastern side of the region. Finally we propose a model for the formation and age of the broad marine-cut platform on the western coast by relating it to global sea-level changes, uplift rates and dated deposits.

AEOLIAN LANDFORMS Sand Sheet

Between Bauer Bay and Sandy Bay, a low col, at about 100 to 150 m altitude, separates higher plateau surfaces to the north from those to the south (fig. 1). On the northern side of Stony Creek is an extensive thin sand sheet (dated at least 6000 to 7000 years old from interbedded peats) which mantles the plateau surface at between 100 and 200 m altitude (figs2 and 3). The sand extends across the col and around the flanks of Hill 236 to spill over into the head of the Finch Creek valley (fig. 4). (Note that unnamed hills are referred to by their spot heights in metres on the 1:50,000 Macquarie Island map, first edition, 1971, NMP/68/025, Division of National Mapping, Canberra, Australia.) The full extent of this sand to the east of Bauer Bay has not been traced on the ground, but it occurs north of a line from Stony Creek, across the col between Hill 210 and Hill 236 to Finch Creek. The sand deposit is dissected by wind and water erosion to form blow-outs, gullies and bare surfaces flanked by low sandy scarps (fig. 2). Layers of peat and iron-cemented sand are exposed in the scarps and on bare surfaces. Cobbles of an undated beach (Beach 5 of Ledingham & Peterson 1984) occur on the largest bare-eroded surface.

Plates 1, 2 and 3 (see figs | to 5 for all plate locations) show the general topography of the

sandy area west of the col. Plate | shows the view northnorthwest across Stony Creek valley to the sand-mantled plateau and hill slope where blow- outs and gullies are visible. Plate 2 shows the view westnorthwest down the incised valley of Stony Creek with its obvious valley-in-valley form. On the northern 10° to 20° slopes of the valley, linear blow-outs (see arrow) occur with their long axes oriented roughly parallel to the contours. The 25° to 40° southern slopes are deeply gullied and free of sand. Plate shows the view from the site of the dated peat deposit (fig.2) on the sand-mantled plateau, across the hidden, deep valley of Stony Creek to the gullies on its southern side. The general concordance of plateau surfaces north and south of Stony Creek is obvious. The distribution of sand on the plateau south of Stony Creek has not been investigated, but a cobble and sandy beach has been recorded there (Ledingham & Peterson 1984, Beach 7).

Blow-outs

Blow-outs have not been described previously from Macquarie Island. They are trough-shaped elongated hollows formed by wind eroding a pre- existing sand deposit, the sand blown from the trough accumulating downwind and to the side of the trough in a blow-out dune (Whittow 1984). Figures 2 and 3 show that the long axes of the blow-outs are parallel to the dominant and strong west to northwest winds. The slight northwards rotation of the blow-outs on the lower slopes, near Stony Creek (fig. 2, plate 2 arrow), compared with those on the plateau above, matches precisely the orientation of the valley and is accounted for by winds channelled up the valley. Small blow-outs of the eastern edge of figure 2 show slight but definite divergence due to wind flow around a ridge.

Figure 3 shows several blow-outs on a 10° south-facing slope, mantled by sand and dissected by drainage lines. Cross-sections (AF, GL) of the blow-out in the centre of the diagram show the convex shape of the sandy body (BE, HK), inset into which is a trough V-shaped in plan view with an east-pointing apex. Its sandy floor (CD, IJ) !§ flanked to north and south by steep edges. At the eastern end of the blow-out (east of IJ) the sandy trough changes to a sandy tongue roughly coin cident with the dune surface. Plate4 shows the sandy trough in the foreground, widening west wards. In Plate 5, a person stands where the deeply inset trough, sloping westwards behind the person: changes into an eastward-sloping sandy tongué roughly level with the surface of the convex blow~ out dune.

Origins of landforms 67

uare ke/ / / /

—— Scarpof fault/inferred fault

-—--Track =] Swamp °210 Spot height asl (m)

—— Stream | Lake

...? Approx. extent of sand —+ Photo location, direction on plateau, northern limit & Plate number not determined

Fig. | — The Bauer Bay-Sandy Bay region of Macquarie Island showing the location of larger scale maps (figs 2, 4 and 5). Information for the maps was derived from: 1:50,000 map Australia, Macquarie Island, first edition, November 1971, Division of National Mapping, Canberra; from 1976 air photos; and from field work. The drainage pattern differs substantially from that shown on the published 1:50,000 map.

68 D.A. Adamson, P.M. Selkirk and E. A. Colhoun

~, 410 * Sand aoa i Hummocks AY act ler

~—— Stream Bare surface © Section &C date

Swamp +m Scarp(sandy) (V) Wind-polished rock = Oo Beach cobbles sme:

2 Blow-out O° (uppermost lateau edge

—* Photo location, direction & Plate numbef

Fig. 2— Part of the sand sheet on the plateau north ofStony Creek. Location of figure 3 is shown. The date4 terrestrial peat (Beta 20167: 6060 + 70 B. P.) occurred as a basal band 30 mm thick, 2.78 m below the surfacé of the sand sheet and overlying angular cobbles and weathered bedrock.

Origins of landforms 69

Aspect of hillslope

Blow-out -— Outer margin (B,E,H,K) ——-- Drainage line +r Scarp(C,D,I,J) —* Photo location, direction & Plate number

Sandy floor

Fig. 3 — Detailed plan and sections of blow-outs on the sand sheet at about 100m altitude on the plateau north of Stony Creek.

70 D.A. Adamson, P.M. Selkirk and E. A. Colhoun

Starting from the wide western end, the sandy floor of the blow-out slopes very gently up towards the east, but as the V narrows the slope steepens slightly to reach a broad crest (close to the position of IJ in fig. 3 and the person in plate 5). To the east of this crest the slope falls gently towards the tip of the bare sandy tongue (plate 5) at the eastern end of the blow-out dune. The dominant and severe westerly winds accelerating into the funnel of the V-shaped trough presumably pick up sand from the gently rising western two-thirds of the trough and drop it to the east of the crest on the gentle downslope. The rising western (windward) side of the longitudinal trough is a wind erosion zone but the eastern (lee) slope is a depositional zone. So long as the sand supply is adequate, and while water erosion does not carry too much sand downhill, the blow-out trough and dune move slowly downwind towards the east, the trough progressively eating into, and blowing forward, its own dune.

The blow-outs are oriented approximately across the hillslope and, because of their convex transverse profile, pond or divert drainage (fig. 3, B and H). Ponding on the upslope side of blow-outs results in seepage of water into their troughs. Between transects AF and GL, a drainage line has broken through the blow-out dune, so that the trough westwards has become a minor creek. The floors of blow-outs also intercept subsurface water moving downslope along relatively impervious surfaces of weathered bedrock, buried peat or layers of iron-cemented sand. Consequently the floors and sides of troughs are often saturated with water, as shown in plate 4 and in plate 5 beyond the person. Water flow and seepage probably contri- bute to the development of blow-outs by transport- ing sand from the steeper sides and slopes of the trough to the gentler slopes of the floor. Borowka (1980) found that wind around 15 to 20 ms"', even for brief periods, is effective at moving sand from wet, cold (2° -6°C) surfaces, and that changes in air viscosity and density at low temperatures increase the effectiveness of cold air in moving sand. Conditions described by Borowka are similar to those on Macquarie Island. Jungerius et al. (1981) also noted that rain had little effect on wind erosion in blow-outs in Holland, possibly because strong winds desiccate the surface of wet sand quickly so that the effect of precipitation is short-lived.

The flanks of the blow-out are well vegetated. Only the incised trough (CD, IJ) and the narrow sandy tongue east of IJ are bare or sparsely vegetated (plates4 and 5). Sparse colonisation is visible in the background in plate 4, at the broad

PLATE! :

Stony Creek valley looking northnorthwest to the sand-mantled plateau and hill slope (much of which is shown in fig. 2). Part of the marine terrace north of Bauer Bay is visible on the coast. The beach marked by rounded cobbles (Beach 5 of Ledingham & Peterson 1984) is shown by the large horizontal arrowhead; the eastern edge of the blow-outs mapped in figures 2 and 3, by the small horizontal arrow; the location of the dated peat (Beta 20167: 6060 ' 70 B.P.) near the base of the sand sheet, by the small vertical arrowhead; the wind-polished bedrock (fig. 2 and plates 6,7 and 8), by the large vertical arrow. Locations of all plates are shown on the relevant figures.

PLATE 2 Stony Creek valley looking westnorthwest. Linear blow-outs occur on the northern (right-hand) slope, one at low level is marked by the vertical arrow. The arrowhead marks the crest of the steep southern slope, also marked on plate 3.

PLATE 3

Looking southwards across the deeply incised valley of Stony Creek, most of which is hidden. The arrowhead shows the same location marked in plate 2. The surface and base of the sand sheet is marked by two horizontal white arrows. Dated peat (Beta 20167: 6060 ‘70 B.P.) was excavated at this site from the base of the sand sheet. Beneath the sand sheet is a strew of angular cobbles and weathered bedrock.

western part of the trough where the sand supply has been depleted. Plate 5 shows a narrow rim of Acaena magellanica adjacent to the bare sand, its stolons growing onto the sandy surface. Luzuld crinita and Pleurophyllum hookeri occur with Acaena magellanica to form a complete cover 0 vegetation over the rest of the blow-out dune. The vegetated eastern end of the blow-out (the depos!- tion zone) traps sand and builds up its convex surface as fast as sand is blown from the west (the erosional zone), plant growth roughly keeping pace with sand supply. The eastern advancing tip of the dune is about 500 mm above the adjacent surface: Build-up and advance of the blow-out dune wil! cease when sand supply declines. In the wet environment of the island, plant growth is rapid, $9 that no leeward slip-slope of bare sand develops, 45 might occur in drier, warmer environments. Once the sand grains are blown from the trough of thé

Origins of landforms 71

%

nf Witmb:

72. D.A. Adamson, P.M. Selkirk and E. A. Colhoun

erosional zone, they are trapped where they fall by the vegetation in the accumulation zone. At the leeward (eastern) end of the blow-out, instead of a slip-slope of bare sand, there is a broad