australian journal of earth sciencespeople.rses.anu.edu.au/lambeck_k/pdf/108.pdfpost-palaeozoic...

This article was downloaded by:[Australian National University]On: 18 December 2007Access Details: [subscription number 768221878]Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Australian Journal of Earth SciencesAn International Geoscience Journal of theGeological Society of AustraliaPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t716100753

The post-Palaeozoic uplift history of south-easternAustraliaKurt Lambeck a; Randell Stephenson aa Research School of Earth Sciences, Australian National University, Canberra,ACT, Australia

Online Publication Date: 01 June 1986To cite this Article: Lambeck, Kurt and Stephenson, Randell (1986) 'Thepost-Palaeozoic uplift history of south-eastern Australia', Australian Journal of EarthSciences, 33:2, 253 - 270

To link to this article: DOI: 10.1080/08120098608729363URL: http://dx.doi.org/10.1080/08120098608729363

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article maybe used for research, teaching and private study purposes. Any substantial or systematic reproduction,re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expresslyforbidden.

The publisher does not give any warranty express or implied or make any representation that the contents will becomplete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should beindependently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with orarising out of the use of this material.

http://www.informaworld.com/smpp/title~content=t716100753http://dx.doi.org/10.1080/08120098608729363http://www.informaworld.com/terms-and-conditions-of-access.pdf

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7

Australian Journal of Earth Sciences (1986) 33, 253-270

The post-Palaeozoic uplift history of south-eastern Australia

Kurt Lambeck and Randell Stephenson

Research School of Earth Sciences, Australian National University, GPO Box 4, Canberra, ACT 2600,Australia.

Vertical movements within mountainous terrain may be indicative of either active tectonism orof passive isostatic rebound of the crust to the erosion of the highlands. During the active orogenicphase, the landscape evolution is controlled largely by the tectonic process but once this ceasesor is reduced in intensity, the erosional unloading and concomitant rebound becomes the dominantlandscape-forming process. It is this latter phase that is examined here. It is argued that the south-eastern highlands are a residual of the Palaeozoic Lachlan Fold Belt, rather than having beenrejuvenated in Tertiary time. It is concluded that the erosional rebound model explains manyof the recent geomorphological observations that attest to little landscape evolution since the earlyTertiary. The model adopted is of a mountain range that is initially in local isostatic equilibriumbut which responds regionally to the erosional unloading. The crust or lithosphere is modelledas a viscoelastic layer so that rebound is not instantaneous. The rate of erosion at any time isassumed to be proportional to elevation above sealevel with a time constant of the order 108 years.For a given present-day topography the elevations, erosion, rebound, stress and gravity can becomputed throughout time as a function of model parameters. The time by which the reboundphase became the dominant process is 200-250 Ma ago and the elevations at that time were about75% greater than present values. An erosion time constant of 200 Ma produces average Tertiaryerosion rates of a few metres/million years, rates that are consistent with geomorphologicalobservations in several areas of south-eastern Australia.

Key words: uplift, erosion, isostatic rebound, vertical movements, East Australian Highlands.

INTRODUCTION

The highlands of eastern Australia are unspecta-cular by world standards. Far from present-day plateboundaries, they appear to be more comparable toother intraplate ranges, such as the Appalachiansof eastern North America, than to the iriterplateAndean or Himalayan Cordilleras. It is exactly thisdifference that makes their study of interest, for theyindicate either a different mountain formationprocess or they represent a residue of an oldmountain system that marks a former zone ofcollisional tectonics.

Mountain chains evolve through two phases. Thefirst, the orogenic phase, is one of mountainconstruction and active tectonism. Uplift, driven byhorizontal compressive forces or by thermalprocesses, exceeds erosion, and absolute elevationsincrease with time. The second, the passive phase,is one of destruction and decay in which the drivingforce has expired to the point where erosional forcesdominate the shaping of the landforms. Materialis transported away from the highlands and a broadregional isostatic rebound sets in. Now there is nogrowth of the total topography, and absoluteelevations decrease through time until only vestigesof the former mountains remain. Vertical move-

ments occur during both phases; in response to thedynamic mountain building process during theactive phase, and in response to the erosion andisostatic rebound during the passive phase.Throughout the mountain's history, therefore, riverscontinue their search for new equilibrium baselevels. The difference is that during the first phase,absolute elevations increase with time but duringthe second phase, absolute elevations decrease withtime.

The essential question is whether the Australianhighlands are in a constructive or destructive phase.This is the question addressed in this paper for thesouth-eastern highlands, an area bounded by theSydney and Surat Basins to the NE and N, theDarling and Murray Basins to the W and theGippsland and Otway Basins to the S (Fig. 1);although the arguments developed in this papershould also be valid for other parts of the highlandsof eastern Australia.

That the southern highlands of Australia exist isperhaps the only statement in this paper with whichall readers will agree. That the highlands are madeup of crust that mostly predates Late Palaeozoictime is also well established. Apart from isolatedCainozoic sediments and volcanics and even smaller

253

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7

254 K. LAMBECK AND R. STEPHENSON

Quivnsltmd

EROMANGA

BASIN

Fig. 1 Schematic structural map of south-easternAustralia. The area of study is defined by the dashed lines.

amounts of Mesozoic volcanics, there are few rockswith ages younger than Carboniferous in the areaof the highlands defined above (Packham 1969).That the area has been subjected to a number ofdisruptive tectonic events in Palaeozoic time alsoappears to be beyond dispute because the area ischaracterized by intense deformation and meta-morphism, volcanism and plutonic intrusions.Whether this evolution is a result of a cratonizationof eastern Australia or whether it reflects a tectonicreworking of Precambrian crust is immaterial to thepresent argument. More relevant is that the pro-cesses leading to the formation of the fold beltexhibit all the elements required for the formationof substantial highlands; phases of compressivetectonics with crustal folding and underthrusting,volcanism, and major igneous intrusions (Solomon& Griffiths 1972; Packham 1969). Northwards, theNew England Fold Belt continued to evolve intoTriassic time. In between the two fold belts theSydney Basin formed and was uplifted during thefinal stabilization of the two fold belts in LateTriassic time.

Subsequent deformation of the highlandsappears to have been minor. Old faults can be tracedover long distances throughout the highlands, butit is not clear whether movements occurred on themin more recent time. The Berridale fault, forexample, is believed to have formed in Early-MiddleDevonian time, but subsequent movements on ithave been relatively minor (Lambert & White 1965).Igneous intrusions and volcanism have occurredthroughout Mesozoic and Cainozoic time but while

these occurrences are quite widespread, the volumeof the volcanics is small (Wellman & McDougall1974a; McDougall & Wellman 1976). Anothertectonic event that cannot be ignored is the openingof the Tasman Sea and the separation of NewZealand from Australia, a spreading process thatstarted at about 80 Ma ago and ended by about60 Ma ago (Weissel & Hayes 1977). A similarspreading event removed Australia from Antarcticastarting at about 100 Ma ago (Cande & Mutter1982) and is still continuing today.

Whether these postorogenic processes are alsoresponsible for the present highlands remains adebatable issue. If they are not, then the passiveevolutionary phase will have dominated thehighland evolution onwards from about 250-200 Ma ago. The magnitude of this earlytopography is impossible to assess in any quantita-tive way but, by comparison with modern-dayexamples of interplate collision zones, thetopography is likely to have been substantial. In ageneral way, a relation between elevation and degreeof deformation, or between elevation and intensityof volcanism, may be anticipated as may a relationbetween the elevation and the duration of theorogenic process: the more intense the orogeny andthe longer its duration, the more substantial willbe the final topography.

If the present highlands are largely an erosionalresidue of Late Palaeozoic or Early Mesozoicmountains, then the erosion process must be veryslow. This is in distinct opposition to the geomor-phological dogma which holds that "Little of theEarth's topography is older than Tertiary and mostof it is no older than Pleistocene" (Brown 1980;Thornbury 1962), and which undoubtedly shapedthe view that the eastern highlands were upliftedin Late Pliocene, Early Pleistocene time (Andrews1910; David 1932; Browne 1969). If erosion is asrapid as implied by the above statement, then thehighlands cannot be residues of Palaeozoicancestors. Instead they must have been rebuilt inmore recent times. Not all geomorphologicalopinion, however, supports such short erosionaltime constants.

TERTIARY REJUVENATION: GEOLOGICALARGUMENTS AND EVIDENCE

With the exception of some recent geomorpho-logical studies, the discussion of the easternAustralian highland geomorphology is dominatedby the notion of cycles of uplift and erosion in

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7

POST-PALAEOZOIC UPLIFT HISTORY OF SE AUSTRALIA 255

which periods of uplift are followed by erosion andpeneplanation in more or less rapid succession(King 1959; Browne 1969). In Browne's synopsis,three periods of uplift occurred in the past 40 Mawith the latest, the Kosciusko uplift, occurring inLate Pliocene, Early Pleistocene time (Andrews1910; David 1932). In this model extensive pene-plains have been postulated for Late Cretaceoustime and again for Miocene time. It was recognizedthat this reduction was not always complete and thatresidual elevated areas rose above the Early Tertiarypeneplain erosion surface, remnants of which arebelieved to be still present today and to reflect tracesof earlier cycles of denudation (Andrews 1914; King1959; Browne 1969). Nevertheless the Miocenepeneplain must have been extensive in this model,for the erosional surface has purportedly beenidentified throughout the present highland belt,near sealevel along the south coast of NSW and atmore than 2000 m elevation in the southern high-lands of NSW (King 1962). It is not always clearhow much of the present highland was believed tobe shaped during the Kosciusko uplift. E. D. Gill,G. Packham and M. J. Rickard (in Fairbridge 1975)quite clearly accept that the Kosciusko event upliftedthe entire divide, but D. A. Brown, in the samevolume, considers that this uplift may have beenmore localized. The concept of a youthful, highlandterrain remains accepted without question in moststandard treatises of Australian geology (Brown etal 1968; Fairbridge 1975) and in some regionalstudies, notably in the major review of the SydneyBasin by Mayne et al (1974). The evidence for recentuplift is wholly morphological, being based largelyon the assumption that various exposed silcrete andlaterite surfaces formed part of a single peneplainformed in Miocene or earlier time. The correlationof the widespread erosion surface is, however,questionable (Oilier 1978), as are their ages.

Not all geologists or geomorphologists haveaccepted the view of Tertiary rejuvenation. Craft(1933a) regarded the plateaux of the south-easternhighlands as being erosional features rather thanthe result of differential uplift of a common surface.For example, he considered that the Monaro regionformed in Late Palaeozoic or Early Mesozoic times,after the partial reduction of features created duringthe earlier Carboniferous Kanimblan Orogeny(Craft 1933b). Craft also considered that thesouthern part of the Sydney Basin had suffered noappreciable differential uplift since at least MiddleTertiary time. Opik (1958) argued that the presentlandscape of the Canberra area is a relic of the

Palaeozoic past, even in its minutest detail. Young(1974, 1977) noted the occurrence of well-developedduricrust surfaces at the feet of coastal escarpmentsas well as on the plateaux within the southern partsof the Sydney Basin, and suggested that these high-lands had reached their present elevations by at leastEarly Tertiary time (Young 1978, 1981, 1983). Thisobservation is endorsed by the recent work of Youngand Bishop (1980) and Bishop et al (1985) in theupper Lachlan Valley and by Young and McDougall(1985) in the Shoalhaven Valley, where basalts ofsimilar age are found over a substantial altituderange, indicating that many of the deep valleys werealready formed at the time of basalt eruption, andthat considerable topography already existed at thattime.

Over the past decade an important shift hasoccurred away from models of Late Tertiary rejuve-nation to models of continuous or episodic upliftinitiated in Late Mesozoic or Early Cainozoic timeand continuing today in response to unspecifiedmechanisms (Wellman & McDougall 1974a;McDougall & Wellman 1976; Wellman 1974,1979a).The reason for this shift was the K-Ar dating ofthe volcanics of the eastern highlands carried outby McDougall and colleagues. Early work (Cooperet al 1963; Dulhunty & McDougall 1966; Stipp &McDougall 1968) on New England volcanicspointed to ages ranging from Early Oligocene toUpper Miocene and similar results were obtainedfor the volcanics of southern and easternQueensland (Webb et al 1967; Webb & McDougall1968). Later results by Wellman and McDougall(1974a) and Young and Bishop (1980) along theDivide of NSW and by Wellman (1974) for theVictorian highlands, confirmed that volcanismoccurred throughout the Tertiary. What nowdominates the geological models is the associationof this volcanism with a major phase of regionaluplift; one that was initiated between the LateMesozoic and Late Oligocene and which continuestoday (Ferguson et al 1979; Smith 1982; Wellman1979a; Jones & Veevers 1982).

Wellman (1979a) attempted to quantify theTertiary uplift model by using estimates of the ratesof the downcutting of rivers through time. Caino-zoic basalts are sometimes found overlying rivergravels that lie well above the heights of the presentriver beds, and Wellman adopts this difference inelevations of the old and new beds as a measureof uplift that occurred since the eruption of thebasalts. Wellman argued that uplifts of up to1500 m have taken place since the oldest basalts

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


were emplaced some 60-70 Ma ago and that theseobservations are indicative of a dynamic, upliftingprocess; that mountain building is taking placetoday and that this process has been going on sinceLate Mesozoic time. But, is it also possible to arguethat these uplifts are merely a consequence of thedestruction of much older mountains by erosionalprocesses? That as mountains erode, there is a con-comitant rebound of their base such that areasoriginally at sealevel move up, even though themountains as a whole are reduced? This reboundneed not be instantaneous because the crust orlithosphere tends to respond to loading as a viscousor viscoelastic medium and rebound could thereforebe occurring now, even if erosion in the recent pastwas reduced or had ceased altogether.

A cardinal assumption made in Wellman'sanalysis is that river beds adjust themselves rapidlyso as to reach gradients that are in equilibrium withthe baselevel beyond the highlands. This assumptionhas not been widely accepted by geomorphologistswho argue that local control of base level might bemore important than regional control, and thatrivers will not incise instantaneously throughouttheir length in response to uplift. Moreover, basaltfills in upland valleys show that the Tertiaryevolution of the landscape has been much slowerthan postulated in Wellman's model (Young 1981;Bishop et al 1985; Taylor et al 1985; Young &McDougall 1985).

A further argument against significant Tertiaryuplift and landscape evolution lies in theobservation of Young and McDougall (1982) thatsubaerial basalts, which erupted near Ulladulla30 Ma ago, are still near sealevel today. Theyconclude that the present coastal plain existed atthat time and that the adjacent Tablelands hadreached their present elevation before mid-Oligocene time^ Similarly, an Early Mioceneestuarine deposit in the Sydney Basin is still nearsealevel today (Partridge et al 1979).

The fate of the eroded material is clearly animportant point in discussing the evolution of thehighlands, but this aspect has not received muchattention. Only Jones and Veevers (1982) haveattempted to relate the Cainozoic sediments in theMurray Basin to the history of the highlands. Butthe sedimentary history of the other flanking basinsall point to there having been a substantial sourceof material from an area now occupied by thehighlands. To the W of the Divide, in the MurrayBasin, Permian and Triassic sediments are foundonly in a few small basins underlying an otherwise

predominantly Cainozoic sediment sequence(Thornton 1976). More substantial deposits occurto the N, beneath the Eromanga and Surat Basinsand the most extensive sequences are found in theSydney Basin. It would appear that these LatePalaeozoic-Early Mesozoic basins formed in asynorogenic or immediate postorogenic environ-ment, at a time when the dominant process was stillone of constructive crustal evolution and mountainbuilding. This is particularly so for the SydneyBasin, where the early sediments originated largelyfrom the Lachlan Fold Belt and from the NewEngland Fold Belt to the N (Mayne et al 1974;Gostin & Herbert 1973).

Jurassic and Cretaceous sediments were depositedin the Surat Basin and a considerable part of themis believed to have originated from the presenthighlands (Exon 1976). Jurassic sediments in theSydney Basin are restricted to small outcrops alongthe NW margin, but larger deposits occur in theOxley Basin to the NW (Mayne et al 1974). At least2000 m of Upper Cretaceous sediments also occuron the Lord Howe rise, sediments that have beendeposited in a continental or shallow marineenvironment (Willcox et al 1980; Burns et al 1973)prior to the opening of the Tasman Sea. The Gipps-land and Otway Basins to the S appear to have beenfed largely from the N (Threlfall et al 1976), startingin the Late Jurassic for the Otway Basin and in theEarly Cretaceous for the Gippsland Basin (Colman1976; Ellenor 1976; Threlfall et al 1976). Jurassicand Cretaceous sediments are not extensive in theMurray Basin and this implies that an adequatedepositional environment had not yet developed,rather than an absence of a ready source of sedi-ments; i.e. that the underlying basement had noteroded sufficiently to become a major sediment trapuntil Late Mesozoic time. Veevers (1982), forexample, has suggested that the Late Cretaceoussediments on the Ceduna Plateau were deliveredfrom the eastern highlands via an ancient MurrayDarling drainage system. This continuous supplyof sediments from Permian to Cainozoic time,originating from a region occupied by the presenthighlands, is inconsistent with the concept of amajor peneplain occupying this same regionthrough Mesozoic and early Tertiary time.

The above arguments present a good case forpushing the start of uplift of the highlands furtherback, in time than has generally been done but theexact timing of this event remains uncertain. Theyoungest marine or shallow water sediments foldedinto the Lachlan Fold Belt are Upper Devonian and

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


their deformation occurred during the Carboni-ferous Kanimblan Orogeny (Packham 1969).Folding was, in general, not intense and intrusiverocks and volcanics are not extensive. This orogeny,like its Palaeozoic precursors, may have been arather drawn out sequence of events, involvingseveral orogenic phases. The time of final tectonismremains uncertain, but it would appear that theoldest age for the development of the presenthighlands is about 300 Ma ago. Some of theerosional surfaces of the southern highlands maytherefore be of Late Palaeozoic age, uplifted duringor soon after the Kanimblan orogeny as was, in fact,suggested by Craft (1933a; Andrews 1914).

Uplift of the Sydney Basin is believed to haveoccurred in Late Triassic time (Mayne et al 1974)and this may have been a response to the finalevolutionary steps of the Late Permian Hunter-Bowen orogeny of New England, particularly if thisbasin represents a foreland basin with two principalsediment sources, the Lachlan Fold Belt to the Sand the New England Fold Belt to the N. Alterna-tively, this uplift could have been part of a moregeneral uplift of the region, a proposal that hassome appeal in that it explains the continuity inlandscape across the western and south-westernmargins of the Sydney Basin. An appropriate timefor the formation of the ancestors of the presenthighlands is, therefore, between 300 and 200 Maago. Possibly the southern part of the highlands,at latitudes below the topographic low N ofCanberra, was formed by about 300 Ma ago, andthe central part, to the W of Sydney, did not takeon its final shape until later, under the influenceof the formation of the Sydney Basin and the NewEngland Fold Belt.

GEOPHYSICAL MODELS FOR UPLIFT

The mechanisms for the proposed Tertiary rejuvena-tion of the highlands of eastern Australia have rarelybeen discussed beyond being described by suchvague terms as epeirogenic or cymatogenic uplift.Alternatively, the uplift has been attributed tomantle heat sources or to seafloor spreadingwithout a closer examination of the mechanismsbeing made. A complete analysis of mechanicalmodels for uplift is beyond the scope of this paper,and one can only comment qualitatively on someof the proposed mechanisms. These include: (1) theheating of the crust from below by a heat sourcelocated in the mantle (a view espoused most recentlyby Smith 1982); (2) the consequence of the opening

of the Tasman Sea in Late Mesozoic-EarlyCainozoic time (Wellman 1979a; Oilier 1978); and(3) underplating (Wellman 1979b).

In the first class of models, the lithosphere movesover a heat source, a Tiotspot' or "plume' that isfixed, or only slowly moving, with respect to thelower mantle. Mechanical-thermal models for upliftresulting from the thermal expansion have beendiscussed recently by Sandwell (1982), Mareschal(1983) and Nakiboglu and Lambeck (1985a). Thebroad Hawaiian swell, of a width of about 1000-1500 km and an amplitude of about 1000-1500 m,provides a useful analogy for discussing suchmodels (Crough 1983). The thermal history of thisswell is sufficiently long for a steady thermal stateto be approached (Nakiboglu & Lambeck 1985a)and much greater amplitudes for the uplift cannotbe expected unless the velocity of the plate relativeto the heat source is reduced so that less of the heatconducted into the lithosphere is carried away bythe moving lithosphere. An equivalent amplitudein a subaerial environment would be about two-thirds of the above amount. These models consideruplift due to thermal expansion only. Verticalmovements may also result from a verticalmigration of mineralogical phase boundaries. Smith(1982), for example, has suggested that an effectiveuplift mechanism is the vertical migration of abasalt-eclogite phase change at the base of the crustin response to a small input of heat. The adoptedmodel is, however, grossly oversimplified in that itignores the experimental evidence for the inter-mediate pressure garnet-granulite phase assemblage(Ringwood & Green 1966; Green & Ringwood 1967;Ito & Kennedy 1971) which is stable over about5-10 kbar at upper mantle-lower crustaltemperatures (Ringwood 1975; Ferguson et al 1979).The suggested phase change will therefore take placeover a depth range of 15-30 km rather than beingsharp, and the temperature anomalies required toproduce the uplift must be considerably greater thanthose suggested by Smith, although not as great asthat required in the thermal expansion model.

Solutions of the thermal-mechanical models arecharacterized by long wavelength uplift profiles,similar to those observed on the ocean swellsorthogonal to the spreading direction; that is, ofa total width of 1000-1500 km. The phase changemodel, by requiring a significant thermal anomalyonly in the lower part of the crust and upper mantle,may be able to produce a somewhat shorter wave-length anomaly than the purely thermal expansionmodel and this needs to be examined further.

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


However, because a substantial part of the lowercrust must be heated, there is lateral diffusion ofheat, and, because the response of the cold uppercrust to the thermal stress not only decreases theamplitude but also increases the wavelength of theuplift (Nakiboglu & Lambeck 1985b), it is unlikelythat the resulting wavelength of the uplift will besignificantly less than the above amount.

Both mechanisms are reversible. Uplift by thermalexpansion is followed by subsidence, with the sametime constant, once the heat source has decayed orthe crust has moved away from it. Evidence for this,with a time constant of 50-100 Ma is seen at oceanicridges and in mid-ocean swells (Crough 1983). Thephase change mechanism is also reversible if theambient temperatures are in excess of about 400-500°C (Ito & Kennedy 1971). The time constant ofthis phase change is short compared with theconduction time constant and it is the latter thatwill again control the vertical movements.

It has been suggested that aspects of the thermal-mechanical uplift model may be relevant to easternAustralia (Smith 1982) because the Tertiary volca-nism is almost wholly confined to the highlandprovince, and because some of the volcanoes exhibitan increasing age progression from S to N that isconsistent with the known movements of the Indo-Australian Plate (Wellman & McDougall 1974b).However, the above cited evidence that the Tertiaryvolcanism was erupted on a dissected uplandsurface of elevation differences of at least 500-700 m, indicate that this mechanism has not beenvery important. The Tertiary volcanics, when takentogether, do not exhibit a systematic age progres-sion. Only the central volcanoes of Wellman andMcDougall (1974b), those for which the extrusionoccurred from a well-defined vent area rather thanfrom an area of diffuse dykes and pipe swarms, and'which also have associated silicic volcanism, exhibita systematic increase in age from S to N. It iscurious that if this volcanic trace reflects the hotspotthat led to the formation of the highlands, nocentral-type volcanoes are found in the mostelevated parts of the south-eastern highlands. It isalso curious that, if the central volcanoes resultfrom a single hotspot, the direction of this tracechanges by about 45°, or is offset, just where thedirection of the coastline also changes. Eitherseveral heat sources are required (Sutherland 1981)or the Tertiary volcanism is predominantlycontrolled by the state of stress in the continent(Pilger 1982), possibly with an overprinting by the

single hotspot trace suggested by Wellman andMcDougall (1974b).

The analogy with the Hawaiian ridge indicatesthat the quantity of heat required to explain theeastern highlands as a thermal expansion effect canonly be injected into the crust if the heat source wascomparable or even larger than that below theHawaiian ridge. Alternatively, the movement of theplate over the heat source was relatively slow. Thefirst of these requirements appears to beinconsistent with the small volume of the totalCainozoic surface volcanics erupted in the easternhighlands (Wellman & McDougall 1974a), a volumethat represents only about 20% of a typicalHawaiian shield volcano. The second requirementis inconsistent with the seafloor spreading modelsfor Cainozoic time which postulate relatively rapidmovement of Australia for the past 50 Ma (Weissel& Hayes 1972; Cande & Mutter 1982). Thewavelength of the present highlands also presentssome difficulty. No known set of model parametersfor the thermal-mechanical uplift model canprovide a profile that rises more than 2000 m in adistance of as little as 150 km from the coast to theaxis of the Snowy Mountains, as is required bygeological interpretations discussed above. However,the western profile of the highlands, of a gradualreduction in height over a distance of some 500 km,is more commensurate with this model. With eitherthe thermal uplift model or the migrating phase-boundary model there should be evidence forsubsidence for those parts of the highlands thathave moved away from the heat source. This hasnot been observed. One possibility could be thatthe geothermal gradients are low so that the phasechange is not wholly reversible, but this isinconsistent with geotherms deduced fromobservations qf heat flow and geothermometry forsouth-eastern Australia (Sass & Lachenbruch 1979;Wass & Hollis 1983).

Neither mechanism offers a satisfactory explana-tion for a Tertiary or even Late Mesozoic rejuvena-tion of the highlands. Possibly these mechanismsare relevant if they occurred prior to the breakawayof Australia from Antarctica, at a time when platemovements were slow, and at a time prior to theopening of the Tasman Sea so that this latter eventcould produce the present E-W asymmetry of thehighlands, and prior to the ages of the oldestsediments on the Lord Howe Rise. Furthermore, inthe absence of evidence for subsidence, thesemechanisms would have to be irreversible.

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


Topographies larger than those formed by heatingthe crust from below will result if actual riftingoccurs, for now heat is also injected into thelithosphere from the newly-formed ridge axis. Thetypical ocean ridge elevations of 4 km suggest thatrelief of up to 3 km could be expected on land.Maximum uplift would have occurred near thepresent eastern margin, tapering eastwards. Sucha process was proposed by Oilier (1978) who alsodrew the analogy between the eastern highlands andthe East African rift system (Oilier 1982). The lattersystem developed in the Early Tertiary, although thebroad uplift may have preceded the actual riftingby a considerable time interval (Baker et al 1972;King 1978). The African plate is believed to havebeen relatively stationary in Tertiary time and thiscould produce the significant thermal doming andthe very substantial surface volcanism that isobserved; one that dwarfs the Australian volcanism.The condition of relative immobility of the platedoes not hold for Tertiary Australia. The analogy,therefore, is not valid.

Rifting along the eastern margin, by splitting ofa substantial part of the eastern half of the uplifteddome and sinking this part below sealevel andbeyond close scrutiny, has been proposed to explainthe E-W asymmetry of the present highlands (Oilier1978). The probable date for the initiation of thisrifting is about 80 Ma ago (Weissel & Hayes 1977),but this episode would not have left a long-termmark on the new marginal topography because anyassociated uplift must have been eroded down toform a coastal plain by Oligocene and EarlyMiocene times in order to form the environmentfor the deposition of the near sealevel basalts andsediments reported by Partridge et al (1979) andYoung and McDougall (1982). The rifting processitself is unlikely to have contributed significantlyto the formation of the highlands themselves: afterall, if it had, why is there no comparable highlandtopography along either the southern margin of theAustralian continent or the once-adjacent Antarcticmargin?

THE EROSION-REBOUND MODEL

The above geological and geophysical argumentsleave a sufficient number of unanswered questionsfor mechanisms other than post-Palaeozoic rejuve-nation to be seriously considered. In particular, itis worthwhile to examine the hypothesis that thepresent topography is a residue of a much olderhighland. Have the highlands, formed in Palaeozoic

time, been subsequently shaped by the passiveprocesses of erosion and rebound, instead of byactive tectonism as is implied by the Mesozoic-Tertiary uplift models?

The starting point of the erosion-rebound modelis a topographic relief formed in the geological pastby an unspecified mechanism or mechanisms. Withtime, the crust underlying the mountain range willhave cooled and strengthened so that it respondsin a regional manner to any change in surface loads.As erosion of the elevated areas takes place and newsediments are deposited away from the highlands,so is the load on the crust redistributed and theoriginal zero height datum shifts with time: it movesupwards where erosion is dominant and downwardswhere large-scale sedimentation takes place. Themodel is, therefore, analogous to the rebound ofthe Fennoscandian and Canadian Shields inresponse to the unloading by melting of thePleistocene ice sheets, but whereas the glacialrebound model is largely controlled by mantlerheology, the erosional rebound, occurring on amuch longer time scale, is controlled by crustal orlithospheric rheology.

Mathematically, the lithosphere is modelled asa thin viscoelastic Maxwell plate overlying anincompressible fluid substratum. This platesupports an initial topography q0 that iscompensated locally by, for example, an Airy localisostatic model. The governing equation is:

DAAw + k(— + w) = 3 +T T

(1)

where w is the deflection of the plate, positivedownwards; k is the buoyancy force, equal to emg;where g is gravity, and Qm the density of the mantlesubstratum. A is the Laplacian operator 92( )/3x2

+ 32( )/dy2, where x, y are orthogonal axes inthe horizonal plane, w and q denote differentiationof w and q with respect to time. D and T specifythe rheology of the plate. D is an effective flexuralrigidity and r is an effective relaxation timeconstant. In terms of a plate thickness (T), Young'smodulus (E), rigidity, ji and Poisson's ratio {v):

D = ET3/12(1 -

and

T = 7J/V

- v) (2)

(3)

The function q(xy ; t) specifies the load acting on theplate at any time (t). A particularly simple model

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


has been adopted to describe the denudation inwhich the rate of erosion (e) at time (t) is propor-tional to the height (h) at the same location andat the same time, or

e ( x , y ; t) = - h ( x , y ; t )/re (4)

where re is the erosion time constant (cf.Scheidegger 1961, p. 101; Sleep 1971). As it is thechange in topography that controls the rebound,the right hand side of (1) is replaced by

(5)

where QC is the density of the crust.The equations (l)-(5) are based on a number of

assumptions of which the principal ones concernthe rheology, the erosion model, and the initialconditions. These are discussed below. In addition,there are the usual assumptions relating to the useof the thin plate theory to an inhomogeneous crustor lithosphere.

Rheology

The asthenosphere is assumed to be fluid, anassumption that is readily justified since therelaxation time of this layer is of the order 103-104

years, much less than the geological time scales of1O7-1O8 years under discussion here. Thelithosphere is assumed to consist of a Maxwell soliddescribed by the two parameters D and T (equations2 and 3). The initial response to a surface load istherefore elastic but, with time, the flexural stressesrelax until the local isostatic state is reached(Beaumont 1978; Lambeck & Nakiboglu 1981). Thismodel represents a gross simplification of the reallithosphere since the parameters E, /*, T, V are depthdependent, and a more realistic model comprisesa series of elastic layers overlying viscoelastic layers.The upper part of the crust may be considered tobehave as an elastic layer whose strength increaseswith depth down to about 20-25 km. Much of theintraplate seismicity, including that of easternAustralia (Lambeck et al 1984), is confined to theupper crust and stress relaxation occurs here bymacroscopic processes. The lower crust is generallydevoid of seismicity, and this is frequently inter-preted in terms of a region that deforms by ductileflow when subject to a stress difference (Chen &Molnar 1983). Such an interpretation for south-eastern Australia is consistent with the evidence for

a high geothermal gradient (Sass & Lachenbruch1979) and the pressure-temperature relationsdeduced from lower crustal and upper mantlexenoliths: estimates of temperatures (T) in the lowercrust range from 880°C at 35 km depth to 980°Cat 55 km depth (Wass & Hollis 1983) compared witha solidus Ts of about 1200°C. That is, T/Ts(expressed in °K) is about 0.80 and ductile flow islikely to be important throughout the lower crust.

The upper mantle is sometimes considered to bea region of high strength which deforms elasticallyrather than viscously (Chen & Molnar 1983). Themost relevant flow laws for upper mantle materialsare those of Chopra and Paterson (1981, 1984) forthe temperature range of 1200-1400°C for both wetand dry polycrystalline olivine. The mantletemperatures beneath south-eastern Australia areestimated to be approximately 1240°C at 70 kmdepth (Ferguson et al 1979) and even for dry olivinethe effective relaxation constant is at most 102

years for a stress-difference of 10 MPa. In this partof the mantle at least, there is no evidence for astrong layer responding elastically to stress-differences of the order of 10 MPa or greater.

Layered elastic-over-viscoelastic models have beenexamined by Lambeck and Nakiboglu (1981) andin surface loading problems there is little todistinguish observationally between solutions of thesimple homogeneous Maxwell layers and the morecomplex layered models. What is important is thatthe parameters D and T are considered as effectiveor equivalent parameters, representing depth-averaged values. The simple models have beenfound satisfactory in analysing a number of tectonicproblems (Walcott 1970; Sleep & Snell 1976;Beaumont 1978; Lambeck 1983; Stephenson 1984).In these studies the parameters D and r are usuallynot independent but their product D T is generallywell determined; approximately 1024-1025Nm Mafor continental lithosphere (Sleep & Snell 1976;Beaumont 1978; Lambeck 1983; Stephenson 1984)and D = 1023Nm and T = 25 Ma has beenadopted. (This latter constant is an order ofmagnitude greater than that adopted in the first twoof the above-cited papers.) For a relaxed shearmodulus n = 2 x 1010Pa and v = 0.25, theeffective thickness (T) of the plate is about 33 km.The effective viscosity is 1025Pa s.

Erosion model

The erosion model (4) represents a gross simplifica-tion of erosion processes and its use is largely

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


governed by its mathematical convenience (Sleep1971; Scheidegger 1961, p. 101). The model can onlybe claimed to describe an effective erosional process,one that is spatially and temporally averaged overthe area as a whole. Local variations associated withlithology or vegetation are ignored, as are climate-related variations through time. The dependence ofthe erosion rate on slope is also not considered inthat Te is assumed to be independent of wavelengthin part for want of more information and in partfor a need to keep to a minimum the total numberof parameters in the model so that the predictedresults can be realistically evaluated. Therefore,concern here is not with the detailed erosion oftopography, but with the average conditions overareas of 1000 km2 and greater, or with topographicfeatures described by wavelengths fromapproximately 60-70 km to up to several hundredkilometres. From a geomorphological viewpoint, abetter erosion model is one in which the rate oferosion is proportional to regional relief (that isheight differences between valleys and ridges)because river gradients are probably less controlledby their heights above sealevel than by their heightsabove local base levels. However, because theaverage conditions dealt with are over areas of about1000 km2, areas of large mean height are generallyalso areas of large relief.

Erosion models such as (4) define the physicalerosion process. Chemical or solution erosion ismuch less dependent on elevation and will not besignificant in regions of steep terrain. Instead, itbecomes important in areas of little relief, such aselevated plateaux or lowlands. Elevated areas of lowrelief may therefore still be subject to substantialerosion but by chemical rather than by physicalprocesses. Whether this chemical process remainssignificant, thereby encouraging uniform denuda-tion of plateaux, depends on climatic factors, forif the process leads to the formation of a lateriticcover subsequent chemical erosion rates may bemuch reduced.

The eastern margin of the highlands is charac-terized by a coastal escarpment, and this is wherethe greatest erosion has occurred, by scarp retreat,during Tertiary times. Here the erosion model isclearly not appropriate and the predictions oferosion and uplift along this margin can only beapproximate measures of areally averaged rates.

The adopted time constant re is of the order of108 years, very long when compared with otherstudies of detailed erosion processes. For example,the predicted erosion rate of terrain at 1000 m

elevation is about 10 m/Ma for re = 108 years, and

this compares with erosion rates of about 50 m/Mafrequently found in the detailed local studies. Thisdifference may be because the latter studies areusually carried out in restricted areas of high ratesof erosion caused by local geological and climaticconditions. It may also be a consequence ofdifferent meanings of erosion rates in these localstudies from the more regional estimates. Forexample, erosion rates based on estimates of cliffrecession are indicative mainly of local land slippingand gravitational instabilities. The actual rate oferosion viewed on a larger scale depends more onthe ability to transport the debris away from the footof the cliff. Furthermore, the present-day estimatesof rates of erosion may be strongly influenced bythe climatic changes of Pleistocene time duringwhich interval temporary storages of erodedmaterials formed on footslopes and are nowbeing worked further down the system. Erosionrates, averaged over times of 106 years and longerand over areas of 103km2 and greater, are thereforelikely to be significantly less than the estimatesbased on these local studies.

Gilluly (1964) has estimated the present erosionrates for the eastern United States to be approxi-mately 20 m/Ma years, including clastic andsolution loads in about equal parts. For an averageterrain elevation of 1000 m, re = 50 Ma. Hisestimates for Cainozoic and Mesozoic rates are onlymarginally less than this. England and Richardson(1980) examined the effect of erosion on surfaceheat flow and adopted erosion time constants of50-200 Ma for the initially rapid phase of erosion.They note that denudation rates for active orogenicbelts, such as the alps, may initially be very rapidbut they also note the persistence of mountainsformed during the Hercynian, Caledonian and olderorogenies. Stephenson (1984) adopted longer timeconstants of 275-375 Ma for the older terrains ofNorth America, in recognition that the timeconstant may itself be a function of the age of theeroding structure. It would be appropriate toconsider a time dependent erosion constant, one inwhich erosion is initially more rapid than proposedhere. The adopted value is largely scaled by theestimates of Tertiary erosion and, as such, estimatesof total erosion since Late Palaeozoic time are likelyto be minimal rates.

Time constants in the interval 150-250 Ma areconsidered for south-eastern Australia. A value of200 Ma means that areas at 1000 m elevation areeroding at a spatial and time averaged rate of

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


approximately 5 m/Ma. Thus, on average, approxi-mately 150 km3 of material/1000 km2 of area hasbeen eroded since Oligocene time. For terrain at anaverage elevation of 500 m, these figures are reducedby a factor of 2. The local isostatic rebound inresponse to this unloading will be about two-thirdsof these erosion estimates and the change inabsolute elevation is only about 50 m at 1000 melevation in the past 30 Ma.

Initial conditions

The initial conditions required to solve equation (1)with (5) are two-fold. First, the epoch t0 has to beestablished; the time at which the passive destructivephase becomes the dominant factor in shaping thehighlands. The second is the stress-state at that time.

If the mountains are a residue of the PalaeozoicLachlan Fold Belt, then to equals or is youngerthan the end of the Kanimblan Orogeny, or approxi-mately 300 Ma. To allow for a thermal equilibriumto be reached after this and earlier orogenies, andto avoid the earliest period of decay when erosionprocesses may have been significantly more effectivethan later, a more appropriate value for to isperhaps 250 Ma. The uplift of the Sydney Basinsuggests that to =. 200 Ma for the highlandsadjacent to this basin [for that region W of Sydneyand N of the topographic low that lies betweenCanberra and Goulburn (Fig. 2)]. Values of between

Fig. 2 Contour map of present-day average elevations(at 250 m intervals). The heights have been averaged into32 x 33 km areas before being contoured. Coastline isshown by dashed line.

150 and 250 Ma ago for the starting time are there-fore appropriate.

If the initial mountains are a consequence ofcompressional tectonics accompanied by extensiveigneous intrusions, then the resulting mountains canbe expected to be in a state of local or near localisostasy. If the mountains are a consequence of athermally driven uplift, then the state of isostasymay be a more regional one because the colderupper crust will support the thermal stresses overan area that is larger than the heat source. Forwavelengths of uplift that exceed the thicknesses ofthe crust by a factor of 4 or 5, however, there willbe little to distinguish between the local and regionalstress states (Nakiboglu & Lambeck 1985b). There-fore, the topography at time t0 will be assumed tohave been locally compensated. This choice is notan important one insofar as the rebound iscontrolled mainly be the unloading of the crust.Where it has some consequence is in the calculationof stress and gravity.

MODEL PREDICTIONS

With the load specified by equations (4) and (5),equation (1) can be solved for the deformation w(t)through time, subject to the requirement that theelevation at the present time (t = tp) equals theobserved values. Other values that can be computedonce equation (1) is solved include the temporalevolution of elevation, the amounts and rates oferosion, gravity g(t) and stress ffjj(t). Because of theirregular geometry of this topography, the solutionof equation (1) is sought in the wavelength domainin which heights and other quantities are trans-formed into the two-dimensional Fourier domain.The solution of the Fourier transformed form ofequation (1) is transformed back into the spatialdomain (Stephenson 1984).

Figure 2 illustrates the mean elevations, for areasof 1000 km2, at time t = tp. Elevations have beenset to zero in the offshore region, thereby avoidingthe complications arising from the rifting andspreading in the Tasman Sea. Standard procedureshave been used to compute the Fourier spectra(Stephenson & Lambeck 1985). Table 1 summarizesthe various parameters adopted in solving equations(l)-(5). Stephenson and Lambeck (1985) adoptedan origin time of t0 = 180 Ma BP and an erosiontime constant of re = 150 Ma. These parameterswere chosen to give uplift rates for Tertiary time thatwere consistent with Wellman's estimates based onrates of river downcutting but, as discussed above,

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


Table 1 Parameters used to define the erosional reboundmodel. The secondary parameters define the equivalentviscoelastic layer.

Primary parametersStarting time t0Erosion time constant reEffective flexural rigidity DEffective relaxation

constant TDensity of substratum

Density of upper crust

Spatial resolution

Secondary parametersRelaxed shear modulusEquivalent elastic thick-

ness of plateEquivalent viscosity of

plate

250 Ma BP200 Ma1023 Nm25 Ma

6 m = 3.3 x 103 kg

m"3

e c = 2.7 x 103 kg

m31.5 km latitude x

33.3 km longitude

H = 2 x 10'° PaH = 33 km

)j = 1.6 x 1025 Pa s

the assumptions inherent in these observationalestimates have been questioned. Instead, parametersare now used that are more consistent with therecent geomorphological and geological studies ofYoung, Bishop and Taylor.

The predicted mean elevations, averaged for theapproximately 1000 km2 areas, are illustrated inFig. 3 for time to (t = 0), corresponding to 250 MaBP. Because phase shifts are not permitted in thismodel, the past topography is similar in form tothe present terrain, only of greater height. Thetopographic low predicted for the region imme-diately to the W of the highest elevation is largelya consequence of the Fourier analysis method notbeing able to adequately model steep gradientswhen the sampling interval is relatively large. Thezero height contour at t0 lies further westward thanit does now, and the central part of the Sydney Basinas well as the onshore Gippsland basin would havebeen shallow-water environments at t0, even in theabsence of sealevel fluctuations. With time, thiscontour moves away from the topographic axis, aconsequence of the regionality of the isostaticresponse. This is seen more clearly in Fig. 4 for asection from Albury to Bega. The rebound is of amore regional character than is the erosion and thisresults in a "broadening' of the topography withtime. Lowlands originally flanking the originalhighlands have now been uplifted. In some areassuch as the Monaro Plain, rebound has equallederosion and there has been little change in theabsolute elevations through time.

Fig. 3 Contour map of average elevations at t0 (250 Maago). Contours are at 500 m intervals (except for the 250 mcontour to the W). Shaded areas represent regions thatare below present-day sealevel.

The evolution of landscape, erosion and reboundis initially quite rapid, but during Tertiary timeevolution throughout the region is much reduced.This is illustrated in Fig. 5 for several localities. Forboth the Kosciusko region and the Victorian high-land region, for example Mt Feathertop, the initialerosion rates of approximately 12-14 m Ma"1

(averages for 1000 km2) decay to approximately6-8 m Ma"1 at present. For the Monaro Plain, theerosion rates have remained nearly constant at4-5 m Ma"1 . Somewhat lower rates are predictedfor the Upper Lachlan Valley area studied by Bishopet al (1985) and in the Sassafras area, near thesouthern margin of the Sydney Basin, studied byYoung and McDougall (1985). Erosion rates maychange quite rapidly from one 1000 km2 area tothe next, as can be seen for the results at twodifferent locations along the Snowy River inVictoria. For most areas the erosion rates haveevolved little through Cainozoic time, and the arealdistribution of these average rates for the past50 Ma is illustrated in Fig. 6. These values representaverage rates over areas of about 1000 km2 and thedistribution of erosion rates within the blocks canbe quite variable. Likewise, the erosion timeconstant is an average value for the past 250 Maand departures from it could be significant. Theareal distribution of the total predicted erosion since250 Ma BP is illustrated in Fig. 7. The estimatedvolume of eroded sediments is 1.5 x 105 km3, ofwhich approximately 3.5 x 104 km3 were removed

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


Fig. 4 Evolution of elevation, erosion and rebound along the section from Albury to Bega (Fig. 2) at intervals of50 Ma from 250 to 100 Ma BP, and at intervals of 25 Ma from 100 Ma BP to the present. For each epoch elevationsare given on the left and erosion and rebound on the right. Rebound is generally positive and erosion is predominantlynegative.

3- ot

P'wdgifl. Rongg jNimitobal) _

150 rOO

Time (Ma BP)

!MI FeathertopMitta-Mitto R.

, Snowy K.-Deddkk R. Junction

Fig. 5 Model-predicted erosion and uplift rates through time at selected sites in NSW (left) and Victoria (right).

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


Fig. 6 Model-predicted average rates of erosion duringCainozoic time (units of m/Ma).

Fig. 7 Distribution of total amount of material erodedfrom highlands since 250 Ma ago (contour interval is500 m).

in Cainozoic time, 4.2 x 104 km3 during theCretaceous and 3.3 x 104 km3 during the Jurassic(Fig. 8). As mentioned previously the volumes forMesozoic sediments are likely to be lower limits ifthe erosion time constants were greater during theearly part of the erosion history than during Tertiarytime.

Predicted uplift rates are also given at selectedsites in Fig. 5 and in most localities these rates haveremained nearly constant through time, with

maximum rates being about 5 m Ma"1 . Whileerosion rates for the more elevated areas areoriginally relatively rapid, the crust is notresponding immediately to this unloading becauseof the viscosity introduced into the lithosphericrheology and the maximum rebound rates will occurat a later time. The areal distribution of mean upliftrates for the past 50 Ma is illustrated in Fig. 9.

DISCUSSION

It is important to recognize that the model providesonly a broad framework for discussing the geo-logical evidence. Erosion rates will vary through

Fig. 8 Cumulative amount of total erosion with timefrom 250 Ma ago to the present.

148°

Fig. 9 Model predicted average rate of regional isostaticrebound during Cainozoic time (in units of m/Ma).

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


time and with location in a more irregular way thanis predicted here because the latter values representaverage rates over an area of about 1000 km2. Thismay not be adequate to model reliably the evolutionalong the eastern margin where erosion is mainlyby scarp retreat. Examples are illustrated in Fig. 10.In Fig. 10a erosion is by river incision only and theelevated plateau surface remains unchanged. Thearea-average rate of erosion will be e = (6A/A)e;,where e\ is the rate of river incision and 5A/A is thefraction of area occupied by eroding river valleys.The regional uplift rate of the area will be aboutep

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


sediments from the northern NSW and Queenslandhighlands as well as from western Victoria, andbecause the volume of sediments in a number ofthe flanking basins is poorly known. For an orderof magnitude comparison, the Murray Basin, forexample, contains about 4.2 x 104 km3 ofterrigenous Cainozoic sediments (Bishop 1985) ofwhich a significant fraction would have originatedfrom the area of study. The predicted amount oferoded Cainozoic sediments is 3.5 x 104 km3

(Fig. 8) and there is qualitative agreement.

Gravity has been computed by Stephenson andLambeck (1985) and the predicted field is consistentwith observations. The stress state of the crustthrough time is more difficult to evaluate becausethe stresses are more a function of the choice ofrheology and initial stress state than is thedisplacement field. The principal characteristic ofthe predicted stress field is one of tension in thenear-surface crust and compression in the deepercrust with maximum stress-differences reaching1 kbar. Should failure occur, it will be by normalfaulting near the surface and by thrust faulting atdepth. Because of the introduction of the viscouselement in the rheology, the maximum stressdifferences due to the rebound did not occur whenthe topography was greatest and preliminarycalculations show that maximum stresses actuallyoccurred in Tertiary time, and that the present-daystress differences are still near their maximum values(Stephenson & Lambeck 1985). A preliminaryexamination of these stresses in relation to theseismicity of the region has been made by Lambecket al (1984) and McQueen (1985).

It is inevitable that any eroding mountain rangewill be subjected to isostatic rebound of a regionalcharacter. The essential question here is whether thisis the dominant process that has been shaping thesouth-eastern highlands of Australia since LatePalaeozoic or Early Mesozoic time. It has beenconcluded that it is, that the present highlands arevery much an erosional residue of Late Palaeozoicor Early Mesozoic mountain building processes,and that the landscape has evolved only slowly sincethen. There does not appear to be strong evidencefor, nor a need for, significant rejuvenation of thehighlands in Late Mesozoic or Tertiary time.

This conclusion has important repercussions fortraditional geomorphological models of erosionwhich hold that erosion time constants are of theorder of 106-107 years rather than 108 years asproposed here. The conclusion also has significancefor studies of the basins peripheral to the highlands,

for not only does the model provide a ready sourceof sediments throughout Late Palaeozoic andMesozoic time, the rebound also perturbs thehistory of the basin margins.

ACKNOWLEDGMENTS

We thank P. Bishop, J. Chappell, I. McDougall,C. D. Oilier, G. Taylor, P. Wellman and R. W. Youngfor helpful comments on this manuscript.

REFERENCES

ANDREWS E. C. 1910. Geographical unity of easternAustralia in late and post Tertiary time. J. Roy. Soc.N.S.W. 67, 251-350.

ANDREWS E. C. 1914. The Tertiary and post-Tertiaryhistory of New South Wales. Handbook for NewSouth Wales. Brit. Ass. Advanc. Sci. 518-526.

BAKER B. H., MOHR P. A. & WILLIAMS L. A. J. 1972.

Geology of the Eastern Rift System of Africa. Geol.Soc. Am. Spec. 136, 67.

BEAUMONT C. 1978. The evolution of sedimentary basinson a viscoelastic lithosphere: theory and examples.Geophys. J. R. Astr. Soc. 55, 471-497.

BISHOP P. 1985. Southeast Australian late Mesozoic andCenozoic denudation rates: a test for late Tertiaryincreases in continental denudation. Geology 13,479-482.

BISHOP P. & YOUNG R. W. 1980. Discussion: on the

Cainozoic uplift of the southeastern Australianhighland. Geol. Soc. Aust., J. 27, 117-119.

BISHOP P., YOUNG R. W. & MCDOUGALL I. 1985. Stream

profile change and long-term landscape evolution—Early Miocene and modern rivers of the EastAustralian highland crest, central New South Wales,Australia. J. Geol. 93, 455-474.

BROWN D. A., CAMPBELL K. S. W. & CROOK K. A. W.

1968. The Geological Evolution of Australia and NewZealand, Pergamon Press, Oxford.

BROWN E. H. 1980. Historical geomorphology. Zeit.Geomorph. Suppl. 36, 9-15.

BROWNE W. R. 1969. General notes. In Packam G. M.ed., The Geology of New South Wales, Geol. Soc.Aust., J. 16, 559-569.

BURNS R. E. & ANDREWS J. E. et al 1973. Initial Reportsof the Deep Sea Drilling Project, No. 21. USGovernment Printing Office, Washington.

CANDE S. C. & MUTTER J. C. 1982. A revised identifica-tion of the oldest sea-floor spreading anomaliesbetween Australia and Antarctica. Earth Planet. Sci.Lett. 58, 151-160.

CHEN W. P. & MOLNAR P. 1983. Focal depths of intra-continental and intraplate earthquakes and theirimplications for the thermal and mechanicalproperties of the lithosphere. J. Geophys. Res. 88,4183-4214.

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


CHOPRA P. N. & PATERSON M. S. 1981. The experimental

deformation of dunite. Tectonophysics 78, 453-473.CHOPRA P. N. & PATERSON M. S. 1984. The role of water

in the deformation of water. J. Geophys. Res. 89,7861-7876.

COLMAN J. R. 1976. Gippsland Basin, on-shore. Monogr.Aust. Inst. Mining Metallurgy 7, 34-41.

COOPER J. A., RICHARDS J. R. & WEBB A. W. 1963. Some

potassium-argon ages in New England, New SouthWales. Geol. Soc. Aust., J. 10, 313-316.

CRAFT F. A. 1933a. The coastal tablelands and streamsof New South Wales. Proc. Linn. Soc. N.S.W. 58,437-460.

CRAFT F. A. 1933b. The surface history of Monaro,N.S.W. Proc. Linn. Soc. N.S.W. 58, 229-244.

CROUGH S. T. 1983. Hotspot swells. Ann. Rev. EarthPlanet. Sc. 11, 165-193.

DAVID T. W. E. 1932. The Geology of the

Commonwealth of Australia. No. 2. Edward Arnold,London.

DULHUNTY J. A. & MCDOUGALL I. 1966. Potassium-

argon dating of basalts in the Coonabarabran-Gunnedah District, New South Wales. Aust. J. Sci.28, 393-394.

ELLENOR D. W. 1976. Otway Basin. Monogr. Aust. Inst.Mining Metallurgy 7, 82-91.

EXON N. F. 1976. The Geology of the Surat Basin inQueensland. Bur. Min. Res. Aust. Bulletin No. 166.

ENGLAND P. C. & RICHARDSON S. W. 1980. Erosion and

the age dependence of continental heat flow.Geophys. J. R. Astr. Soc. 62, 421-437.

FAIRBRIDGE R. W. ed. 1975. Encyclopedia of EarthSciences, 8. Dowden, Hutchinson and Ross,Stroudsberg, Pennsylvania.

FERGUSON J., ARCULUS R. J. & JOYCE J. 1979. Kimberlite

and kimberlitic intrusives of southeastern Australia:a review. Bur. Min. Res. J. Aust. Geol. Geophys. 4,227-241.

GILLULY J. 1964. Atlantic sediments, erosion rates, andthe evolution of the continental shelf: somespeculations. Geol. Soc. Am. Bull. 75, 483-492.

GOSTIN V. A. & HERBERT C. 1973. Stratigraphy of the

upper carboniferous and lower permian sequence,southern Sydney Basin. Geol. Soc. Aust., J. 20, 49-70.

GREEN D. H. & RINGWOOD A. E. 1967. An experimental

investigation of the gabbro to eclogite transformationand its petrological applications. Geochim.Cosmochim. Acta. 31, 767-833.

HACK J. T. 1960. Interpretation of erosional topographyin humid temperate regions. Am. J. Sci. 258A, 80-97.

HACK J. T. 1979. Rock control and tectonism: theirimportance in shaping the Appalachian Highlands.Geol. Surv. Prof. Pap. 1126-A-J, Dept. Int.,Washington, B1-B17.

ITO K. & KENNEDY G. C. 1971. An experimental study ofthe basalt-garnet granulite-eclogite transition. In TheStructure and Physical Properties of the Earth'sCrust. Monograph 14, pp. 303-314. Am. Geophys.Union Geophys.

JONES J. G. & VEEVERS J. J. 1982. A Cainozoic history

of Australia's Southeast Highlands. Geol. Soc. Aust.,J. 29, 1-12.

KING B. C. 1978. Structural and volcanic evolution of theGregory Rift Valley. In Bishop W. W. ed., GeologicalBackground to Fossil Man, Scottish Academic Press,Edinburgh.

KING L. C. 1959. Denudational and tectonic relief insoutheastern Australia. Trans. Geol. Soc. SouthAfrica 62, 113-138.

KING L. C. 1962. The Morphology of the Earth. Oliverand Boyd, Edinburgh, 726 pages.

LAMBECK K. 1983. Structure and evolutionn of theintracratonic basins of central Australia. Geophys.J. R. Astr. Soc. 74, 843-886.

LAMBECK K. & NAKIBOGLU S. M. 1981. Seamount loading

and stress in the ocean lithosphere 2. Viscoelastic andelastic-viscoelastic models. J. Geophys. Res. 86,6961-6984.

LAMBECK K., MCQUEEN H., STEPHENSON R. & DENHAM

D. 1984. The state of stress within the Australiancontinent. Annales Geophysicae 2, 723-741.

LAMBERT I. B. & WHITE A. J. R. 1965. The Berridale

wrench fault: a major structure in the SnowyMountains of New South Wales. Geol. Soc. Aust.,J. 12, 25-33.

MARESCHAL J.-C. 1983. Uplift and heat flow followingthe injection of magmas into the lithosphere.Geophys. J. 73, 109-127.

MAYNE S. J., NICHOLAS E., BIGG-WITHER A. L., RASIDI

J. S. & RAINE M. J. 1974. Geology of the SydneyBasin. Bulletin No. 149. Bur. Min. Res. Aust.

MCDOUGALL I. & WELLMAN P. 1976. Potassium-argon

ages for some Australian Mesozoic igneous rocks.Geol. Soc. Aust., J. 23, 1-9.

MCQUEEN H. W. S. 1985. Vertical movements and stressacross continental margins. Bull. R. Soc. NewZealand (in press).

NAKIBOGLU S. M. & LAMBECK K. 1985a. Thermalevolution and isostasy of ocean lithosphere over amantle heat source. J. Geophys. Res. 90, 2985-2994.

NAKIBOGLU S. M. & LAMBECK K. 1985b. Comments onthermal isostasy. J. Geodynamics 2, 51-65.

OLLIER C. D. 1978. Tectonics and geomorphology of theeastern highlands. In Davies J. L. & WilliamsM. A. J. eds, Landform Evolution in Australasia, pp.5-47. Australian National University Press, Canberra.

OLLIER C. D. 1982. The great escarpment of easternAustralia: tectonic and geomorphic significance.Geol. Soc. Aust., J. 29, 13-23.

OPIK A. A. 1958. The Geology of the Canberra CityDistrict. Bulletin No. 32. Bur. Min. Res., Australia,97 pages.

PACKHAM G. H. (ed) 1969. The geology of New SouthWales. Geol. Soc. Aust., J. 16, 654.

PARTRIDGE A. D., BENSON J. M. & BELL D. 1979.

Unwrapping Little Bay: an early Miocene estuarinedeposit at Little Bay, Sydney, Abstract. Dept. Geol.12th Newcastle Symposium, Univ. Newcastle, NSW.

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


PILGER R. H. 1982. The origin of hotspot traces: evidencefrom eastern Australia. J. Geophys. Res. 87,1825-1834.

RINGWOOD A. E. & GREEN D. H. 1966. An experimental

investigation of the gabbro-eclogite transformationand some geophysical implications. Tectonophysics3, 383-427.

RINGWOOD A. E. 1975. Composition and Petrology ofthe Earth's Mantle. McGraw-Hill, New York, 618pages.

SANDWELL D. T. 1982. Thermal isostasy: response of amoving lithosphere to a distributed heat source. J.Geophys. Res. 87, 1001-1014.

SASS J. H. & LACHENBRUCH A. H. 1979. Thermal regimeof the Australian continental crust. In McElhinnyM. W. ed., The Earth: Its Origin, Structure andEvolution, pp. 301-351. Academic Press, London.

SCHEIDEGGER A. E. 1961. Theoretical Geomorphology.Springer-Verlag, Berlin, 333 pages.

SLEEP N. H. 1971. Thermal effects of the formation ofAtlantic continental margins by continental break-up. Geophys. J. 24, 325-350.

SLEEP, N. H. & SNELL N. S. 1976. Thermal contractionand flexure of mid-continent and Atlantic marginalbasins. Geophys. J. 45, 125-154.

SMITH A. G. 1982. Late Cenozoic uplift of stablecontinents in a reference frame fixed to SouthAmerica. Nature, 296, 400-404.

SOLOMON M. & GRIFFITHS J. R. 1972. Tectonic evolutionof the Tasman orogenic zone, Eastern Australia.Nature, 237, 3-6.

STEPHENSON R. 1984. Flexural models of continentallithosphere based on the long term decay oftopography. Geophys. J. R. Astr. Soc. 77, 385-413.

STEPHENSON R. & LAMBECK K. 1985. Erosion-isostatic

rebound models for uplift: an application to south-eastern Australia. Geophys. J. R. Astr. Soc. 82, 31-55.

STIPP J. J. & MCDOUGALL I. 1968. Potassium-argon agesfor the Nandewar Volcano, near Narrabri, New SouthWales. Aust. J. Sci. 31, 84-85.

SUTHERLAND F. L. 1981. Migration in relation to possibletectonic and regional controls in Eastern Australianvolcanism. J. Volcano. Geotherm. Res. 9, 181-213.

TAYLOR G., TAYLOR G. R., BINK M., FOUDOULIS C.,

GORDON I., HEDSTROM J., MINELLO J. & WHIPPY F.

1985. Sub-basaltic topography of northern Monaroand its implications. Aust. J. Earth Science (in press).

THORNBURY W. D. 1962. Principles of Geomorphology,Wiley, New York, 618 pages.

THORNTON R. C. N. Murray Basin and associatedinfrabasins. In Leslie R. B., Evans H. J. and KnightC. L. eds, Economic Geology of Australia and PapuaNew Guinea, 3, Monogr. Aust. Inst. MiningMetallurgy, No. 7.

THRELFALL W. F., BROWN B. R. & GRIFFITH B. R. 1976.

Gippsland Basin, Off-shore. In Leslie R. B., EvansH. J. & Knight C. L. eds, Economic Geology ofAustralia and Papua New Guinea, 3, Monogr. Aust.Inst. Mining Metallurgy No. 7, pp. 41-67.

VEEVERS J. J. 1982. Australian-Antarctic depression fromthe mid-ocean ridge to adjacent continents. Nature295, 315-317.

WALCOTT R. I. 1970. Flexural rigidity, thickness, andviscosity of the lithosphere. J. Geophys. Res. 75,3941-3954.

WASS S. Y. & HOLLIS J. D. 1983. Crustal growth in

southeastern Australia—evidence from lower crustaleclogitic and granulite xenoliths. J. MetamorphicGeol. 1, 25-45.

WEBB A. W. & MCDOUGALL I. 1968. The geochronology

of the igneous rocks of eastern Queensland. Geol.Soc. Aust., J. 15, 313-346.

WEBB A. W., STEVENS N. C. & MCDOUGALL I. 1967.

Isotopic age determinations on Tertiary volcanic rocksand intrusives of southeastern Queensland. Proc. R.Soc. Qld. 79, 79-92.

WEISSEL J. K. & HAYES D. E. 1972. Magnetic anomalies

in the southeast Indian Ocean. In AntarcticOceanography II, The Australian-New ZealandSector Antarctic Res. Serial No. 19, pp. 165-196. Am.Geophys. Union, Washington.

WEISSEL J. K. & HAYES D. E. 1977. Evolution of the

Tasman Sea reappraised. Earth Planet. Sci. Lett. 36,77-84.

WELLMAN P. 1974. Potassium-argon ages on theCainozoic volcanic rocks of eastern Victoria,Australia. Geol. Soc. Aust., J. 21, 359-376.

WELLMAN P. 1979a. On the Cainozoic uplift of thesoutheastern Australian highland. Geol. Soc. Aust.,J. 26, 1-9.

WELLMAN P. 1979b. On the isostatic compensation ofAustralian topography. Bur. Min. Res. J. Aust. Geol.Geophys. 4, 373-382.

WELLMAN P. & MCDOUGALL I. 1974a. Potassium-argonages of the Cainozoic volcanic rocks of New SouthWales. Geol. Soc. Aust., J. 21, 247-272.

WELLMAN P. & MCDOUGALL I. 1974b. Cainozoic igneousactivity in eastern Australia. Tectonophysics 23,49-65.

WILLCOX J. B., SYMONDS P. A., HINZ K. & BENNETT D.

1980. Lord Howe Rise, Tasman Sea: preliminarygeophysical results and petroleum prospects. Bur.Min. Res. J. Aust. Geol. Geophys. 5, 225-236.

YOUNG R. W. 1974. A major error in southeastAustralian landform studies. Proc. Int. Geograph.Un. Reg. Conf., pp. 283-288. Geographical Society,Palmerston North, New Zealand.

YOUNG R. W. 1977. Landscape development in theShoalhaven River catchment of southeastern NewSouth Wales. Z. Geomorph. N. F. 21, 262-283.

YOUNG R. W. 1978. The study of landform evolution inthe Sydney region: a review. Aust. Geographer14, 72-93.

YOUNG R. W. 1981. Denudational history of the south-central uplands of New South Wales. Aust.Geographer 15, 77-88.

YOUNG R. W. 1983. The tempo of geomorphological

Dow

nloa

ded

By:

[Aus

tralia

n N

atio

nal U

nive

rsity

] At:

04:0

8 18

Dec

embe

r 200

7


change: evidence from southeastern Australia. J.Geology 91, 221-230.

YOUNG R. W. & BISHOP P. 1980. Potassium-argon ageson Cainozoic volcanic rocks in the Crookwell-Goulburn area, New South Wales. Search 11,340-341.

YOUNG R. W. & MCDOUGALL I. 1982. Basalts and

silcretes on the coast near Ulladulla, southern NewSouth Wales. Geol. Soc. Aust., J. 29, 425-430.

YOUNG R. W. & MCDOUGALL I. 1985. The age, extent

and geomorphological significance of the Sassafrasbasalt, southeastern New South Wales. Aust. J. EarthSciences 32, 323-331.

(Received 30 March 1985; accepted 12 September 1985)

australian journal of earth sciencespeople.rses.anu.edu.au/lambeck_k/pdf/108.pdfpost-palaeozoic...

Documents