Mantle structure and tectonic evolution of the region north and...

INTRODUCTION

The northern Australia plate boundary (Figure 1) is a com-plex and actively deforming region which includes some ofthe fastest relative plate motions on Earth (Bevis et al.1995; Tregoning et al. 1998). Present-day plate motionsoffer some insight into the recent history of the region buteven a cursory study of the region shows that presentmotions provide only a snapshot of the tectonic history.Plate motion models indicate that Australia has movedrapidly north since the Eocene and that this movement hasbeen accompanied by subduction and collision events. Toreconstruct the region in the past requires an interpretationof the evidence remaining and the nature of subductionand collision means that some of the evidence is destroyed.Thus, there are many different tectonic models and distin-guishing between them can be difficult.

Tomographic methods can provide new informationwhich can help to test and improve tectonic models.Implicitly tectonic reconstructions predict where lithos-phere has been subducted through geological time andconsequently how much lithosphere has been consumed in

plate-convergence zones. In general terms, subductedlithosphere produces a strong temperature anomaly in theupper mantle (McKenzie 1970) which causes slabs to beseismologically detectable as regions with relatively fastseismic wave speeds. With seismic tomography the pre-served record of former plate convergence can be made vis-ible, albeit within spatial resolution limits. Thus,tomography models provide an additional and independentsource of information on tectonic evolution which can beexploited to test tectonic reconstructions (de Jonge et al.1994; Bunge & Grand 2000; Hafkenscheid et al. 2001).Conversely, plate-tectonic models of a region can help indeveloping new interpretations of imaged mantle structure(Richards & Engebretsen 1992; van der Hilst 1995; van derVoo 1999a, b; Wortel & Spakman 2000) The comparisonbetween tectonic and tomographic models often impliesnew constraints on tectonic evolution, which, however, relyon a correct interpretation of tomographic results.

The interpretation of anomalously fast regions in termsof a subducted slab is straightforward if the slab is seismi-

Geol. Soc. Australia Spec. Publ. 22, and Geol. Soc. America Spec. Pap. 372 (2003), 361–381

Mantle structure and tectonic evolution of the region north and east of AustraliaR. HALL1* AND W. SPAKMAN2

1 SE Asia Research Group, Department of Geology, Royal Holloway University of London, Egham, Surrey TW200EX, UK.

2 Vening Meinesz Research School of Geodynamics, Faculty of Earth Sciences, Universiteit Utrecht,Budapestlaan 4, 3584CD Utrecht, Netherlands.

Tomographic images of the mantle beneath the region extending from the Molucca Sea eastwardto Tonga, and from the Australian craton north into the Pacific, reveal a number of distinctive highseismic-velocity anomalies. The anomalies can be interpreted as subducted slabs and the positionsof the slabs can be compared to predictions made by tectonic models for the region. Severalstrong anomalies are due to present-day subduction and the slab lengths and positions are consis-tent with Neogene subduction at the Tonga and the New Hebrides Trenches, where the anomaliessuggest rapid rollback of subduction hinges since about 10 Ma, and beneath the New Britain andHalmahera Arcs. There are several generally flat-lying deeper anomalies which are not related topresent subduction. Beneath the Bird’s Head and Arafura Sea is an anomaly which we interpret tobe the result of north-dipping subduction beneath the Philippines–Halmahera Arc between 45 and25 Ma. A very large anomaly, which extends from the Papuan peninsula to the New Hebrides andfrom the Solomon Islands to the east Australian margin, is interpreted as the result of south-dippingsubduction beneath the Melanesian Arc between 45 and 25 Ma. Our interpretation implies that aflat-lying slab can survive for many tens of millions of years at the bottom of the upper mantle. Thereis a huge anomaly in the lower mantle which extends from beneath the Gulf of Carpentaria toPapua. We suggest this is a slab subducted before 45 Ma, which may be correlated with aCretaceous slab beneath the Australian–Antarctic Discordance or an early Cenozoic slab sub-ducted north of Australia. The anomaly is located above the position where there must have beena change in polarity in subduction at the boundary between the north- and south-dipping sub-duction zones north of Australia between 45 and 25 Ma. All of these have been overridden byAustralia since 25 Ma. One subduction system predicted by the tectonic models, the Marumuni Arcof Papua New Guinea, is not seen on the tomographic images.

KEY WORDS: island arcs, Melanesia, New Guinea, plate tectonics, subduction, tomography.

* Corresponding author: [email protected]

cally active. However, interpretation of aseismic regionswith relatively fast seismic wave speeds as (remnants of)subducted slabs is generally problematical because thermaland compositional heterogeneity may be caused by othermantle processes/structure not associated with subduction.Tomography provides a snapshot of mantle structure anddynamics from which the space–time evolution of subduc-tion cannot be derived without invoking other information.For instance, dynamic modelling (Zhong & Gurnis 1995)predicts that, depending on a number of factors, slab mor-phology and angle of subduction may change during thesubduction process. Because of varying subduction rates,variable surface area of an oceanic basin subducted, andpossible lateral variation of mantle viscosity on a globalscale, the correspondence between the depth of subductionand time a slab takes to reach that depth may not be sim-ple. Furthermore, the plate-like character of subduction inthe upper mantle may be completely lost in the lower man-tle. Interpretation is additionally complicated by a varietyof possible imaging errors some of which are extremely dif-ficult to identify. All these factors may complicate interpre-tations of high wave-speed anomalies in the deeper mantle,in particular in regions where the correspondence betweenpresent-day surface tectonics and deeper mantle structureis lost as a result of tens to hundreds of millions of years ofplate-tectonic evolution and mantle convection. This is

where plate-tectonic reconstructions come into play andcomparing predictions of such reconstructions with imagedmantle structure is at present the most powerful tool avail-able for the interpretation of tomographic models and fortesting the reconstructions.

In this paper we discuss tomography and tectonic inter-pretations and compare predictions of subduction withimaged mantle structure of a region extending from theMolucca Sea eastward to Tonga, and from the Australiancraton north into the Pacific, in order to understand theevolution of the northern Australian plate margin duringthe Cenozoic. Primarily, we invoke the tectonic model forthe interpretation of the tomographic results which exhibitseveral enigmatic, spatially large, wave-speed anomalies atgreat depth in the mantle. We have deliberately avoidedextending the region of interest further west to the NorthWest Shelf and east Indonesia because it is too compli-cated and will require a separate discussion.

PRESENT-DAY TECTONIC SETTING

Currently there is a long and complex boundary separatingthe Australian Plate from oceanic plates of the Pacific(Figure 2) which is marked by a zone of shallow seismicitystretching from the Bird’s Head of New Guinea to the

362 R. Hall and W. Spakman

Figure 1 Digital elevation model showing principal surface features of the northern Australian plate margin based on Smith & Sandwell(1997) showing their bathymetry combined with topography from GTOPO30.

Tonga–Kermadec Arc (Figure 3). To the north of this bound-ary are the Philippine Sea, the Caroline and Pacific Plates.To the south is the Australian Plate but within its marginsand in the boundary zone with the Pacific are a number ofmicroplates. Johnson and Molnar (1972) identified many ofthese and the principal tectonic features of this boundaryand later workers have identified additional small plates,small spreading centres and backarc basins within thiscomplex zone.

At the eastern edge of the Australian Plate (Figure 2) thePacific Plate is subducting west beneath theTonga–Kermadec Arc, behind which are the spreading cen-tres of the Lau Basin. Further west is the New Hebrides Arcwhere subduction is in almost the opposite direction,towards the northeast. In between the New Hebrides andthe Tonga Arcs are the spreading centres of the North FijiBasin. In the sector between the Solomons and easternNew Guinea subduction is essentially north-directed, at ahigh rate beneath the New Britain Arc and at a lower ratebeneath the Solomons. To the south of this subductionzone, within the Australian Plate, is the actively spreadingWoodlark Basin (Figure 4). To the north of the New BritainArc is the Bismarck Sea where the spreading centre of theManus Basin links in a complex way to strike-slip bound-

aries to the east in the Solomons, where the Pacific andAustralian Plates are moving past one another, and to thewest into a diffuse zone within the northern New Guineamargin which continues west to the Bird’s Head. GPS mea-surements suggest that the Bird’s Head is now moving withthe Pacific Plate and therefore the Australia–Pacific bound-ary must pass through New Guinea south of the Bird’sHead to link into the complexities of the Banda Sea andsoutheast Asia.

At the present day the plate tectonics of the region isclearly complex and to some extent confused since in someareas there is no clear connection between plate bound-aries. For example, at the east end of the Manus spreadingcentre (Figure 4) the link through New Ireland and into theSolomons is a complex zone where GPS measurements(Tregoning et al. 1998) indicate very high rates ofPacific–Australia motion but where there are no easilyidentifiable major faults on which the motion has occurred.In the Bismarck Sea region and to the south there are sev-eral bathymetric troughs which are interpreted as activeand inactive trenches. The New Britain Trench has clearseismicity down to several hundred kilometres butalthough the Manus Trench, the Trobriand Trough and theMoresby–Pocklington Trough may have some deformation

Australian Plate tomography and tectonics 363

Figure 2 Tectonic map of the same area as Figure 1 with principal geographical locations referred to in the text. Barbed lines are activesubduction zones and double lines are active spreading centres. Other lines are major faults. Small black filled triangles are volcanoesfrom the Smithsonian database (http://www.nmnh.si.edu/gvp/), and bathymetry is from the GEBCO digital atlas (IOC et al. 1997).Bathymetric contours are at 200, 2000, 4000 and 6000 m.


Figure 3 Seismicity of the region selected from the data set of Engdahl et al. (1998) used in the tomographic modelling. The maps showthe same areas as the tomographic images of Figure 6. Top: hypocentres between 0 and 100 km. Bottom: hypocentres between 500 and700 km. Grey lines, major plate boundaries; black lines, great circle segments of the tomographic slices presented in Figure 9.

associated with them, and may have been trenches in thepast, they appear not to be active plate boundaries at thepresent day. There is a similar confused situation in west-ern New Guinea and the Bird’s Head region (Figure 2)where to the north of the Bird’s Head there is an activespreading centre in the Ayu Trough between the PhilippineSea Plate and the Caroline Plate. At the south end of thisspreading centre there are no clear plate boundaries. TheNew Guinea Trench is a relatively shallow and sediment-filled feature which is much shallower than a normaloceanic trench and to the south of it is a poorly defined,broadly south-dipping, zone of diffuse seismicity whichextends only to depths of 100–200 km. The region south ofthe Ayu Trough appears to be a wide zone of intraplatedeformation without a simple clear plate boundary. Thecomplexity of present tectonics in part reflects very rapidmotion along the Pacific–Australia plate boundary, its vari-able geometry from east to west, accommodated for exam-ple by the creation of new plates or by short-livedsubduction of small plates, and intra-plate deformation.

CENOZOIC TECTONIC MODELS

In order to interpret the tomographic images of the region itis helpful to have tectonic models to which the images canbe compared. Below we summarise the features of differentmodels of the Cenozoic tectonic development that have

been proposed and attempt to identify aspects of thesemodels which could distinguish them using tomographicresults.

Cenozoic tectonics

Plate-tectonic models for the region fall into three groups.First, there are some regional plate models which show themajor plates of southeast Asia, the western Pacific andAustralia although few cover the entire region of interest inthis paper from New Guinea to the Tonga–Kermadec Arc(Crook & Belbin 1978; Hamilton 1979; Kroenke 1984;Wells 1989; Jolivet et al. 1989; Rangin et al. 1990; Smith1990; Daly et al. 1991; Yan & Kroenke 1993; Lee & Lawver1995; Hall 1996, 1997, 1998, 2002). Second, there aremany local plate models which deal with small sections ofthe Australian margin (Jaques & Robinson 1977; Taylor1979; Johnson & Jaques 1980; Cooper & Taylor 1987;Pigram & Davies 1987; Hill & Hegarty 1987; Hill et al.1993; Struckmeyer et al. 1993; Abbott 1995; Musgrave &Firth 1999; Hill & Raza 1999; Charlton 2000; Taylor et al.2000). Finally, there are many plate-tectonic cartoonswhich typically show cross-sections or maps of subductionzones at different stages, but which do not give a completehistory (Johnson 1976; Falvey 1978; Falvey & Pritchard1982; Ridgeway 1987; Richards et al. 1990; Benes et al.1994; Petterson et al. 1997; Monnier et al. 1999; Weiler &Coe 2000). Here we summarise the key features of different


Figure 4 Geographical features of the Bismarck Sea and Woodlark Basin and surrounding regions. Barbed lines are active subductionzones and double lines are active spreading centres. Other lines are major faults. Small black filled triangles are volcanoes from theSmithsonian database (http://www.nmnh.si.edu/gvp/), and bathymetry is from the GEBCO digital atlas (IOC et al. 1997). Bathymetriccontours are at 200, 2000, 4000 and 6000 m.

groups of models and divide the Australian margin into twomajor segments. The first includes New Guinea andextends to the east of the Papuan peninsula towards theSolomons, and the second extends from the Solomons east-wards to the Tonga–Kermadec Arc. We call these the NewGuinea segment and the Melanesian segment.

New Guinea segment

In the New Guinea segment models can essentially bedivided into two. The majority of models advocate north-ward-subduction of oceanic crust north of Australia begin-ning in the early Cenozoic leading to collision of theAustralian margin with a south-facing arc in the lateCenozoic. The second category of models advocates thatthe northern Australian margin was an active margin andthere was southward-subduction beneath this marginbefore the active north Australia margin collided with anarc. The evolution of the New Guinea margin is discussedin more detail by Hall (2002) and Hill & Hall (2003).

Many models show a single arc–continent collision.Variations include the opening of marginal basins withinthe Australian margin and their subsequent subduction,and multiple subduction zones. In some models these sub-duction zones have a consistent polarity, for example,many models suggest more than one subduction zonenorth of Australia dipping north, but in other models thereare subduction zones which have opposing polarities. Morecomplex models propose or imply multiple arc–continentcollisions (Pigram & Davies 1987; Daly et al. 1991; Lee &Lawver 1995) with accretion of arc fragments to the north-ern New Guinea margin at several periods during theCenozoic. Suggested dates of these events also vary.

A common theme in many of the models, particularlythose which involve multiple subduction zones and multi-ple collisions, is the diachronous collision of a volcanic arcwith the northern New Guinea margin. This is generallyagreed to have started earliest in the west and progressedeastwards to the present collision between the NewBritain–Finisterre Arc and New Guinea. The collision iscommonly interpreted to have followed subduction to bothnorth and south leaving beneath Papua a doubly vergentsubducted oceanic slab (Cooper & Taylor 1987; Pegler et al.1995) similar to the inverted U-shape of the Molucca Seaslab (Hamilton 1979; Cardwell et al. 1980; McCaffrey1982). The Solomon Sea is usually interpreted as the rem-nant of this ocean (Richards et al. 1990; Daly et al. 1991;Lee & Lawver 1995).

Melanesian segment

There are fewer models for the Melanesian segment (Crook& Belbin 1978; Kroenke 1984; Wells 1989; Smith 1990; Yan& Kroenke 1993; Hall 1998, 2002) and many are in cartooncross-section form or deal with only parts of this segment(Falvey & Pritchard 1982; Ridgeway 1987; Petterson et al.1997; Musgrave & Firth 1999; Taylor et al. 2000). There isbroad agreement that from the early Cenozoic there wassouth- or southwest-directed subduction beneath theSolomons and this continued until the Solomon Arc col-lided with the Ontong Java Plateau. Pacific–Australia plateconvergence by subduction in the Solomons region then

ceased and subduction transferred to other locations. Themain difference between models is in the proposed site ofinitiation of the arc, and the timing of arc collision with theOntong Java Plateau. Some models suggest that theMelanesian Arc was rifted away from the Australian conti-nental margin by southward-subduction in the earlyCenozoic, whereas others suggest the arc was initiated bysouthward-subduction in an intra-oceanic setting north ofthe Australian margin. Initiation of subduction at theAustralian margin would have led to formation of a largemarginal basin behind the Melanesian Arc and implies verysignificant rollback of the subduction hinge during thePalaeogene. The alternative intra-oceanic arc modelrequires that an ocean existed before subduction was initi-ated. In both models almost all the ocean crust postulatedwas subducted during the Neogene. The intra-oceanic arcmodels imply this crust was Mesozoic – Early Eocenewhereas the rollback models imply it wasEocene–Oligocene.

Most models accept that arc–plateau collision tookplace in the Miocene but may have occurred over anextended period. The early stage is thought to have short-ened part of the leading edge of the Ontong Java Plateaumargin and is often referred to as a soft collision. Whencontraction of the Ontong Java Plateau could no longercontinue there was thrusting of the Solomons Arc over theplateau during a hard collision phase. It is generallyaccepted that the Solomons were transferred to the PacificPlate or partially coupled to the Pacific Plate in the Early toMiddle Miocene and therefore the boundary ceased to be asubduction zone and instead became a broadly strike-slipfault zone with the possibility of later local subduction tothe southwest or to the northeast on either side of theSolomons. The most recent stages in the development ofthe Melanesian segment in the last 10–12 million yearsinvolved the initiation of northeast-dipping subductionbeneath the New Hebrides Arc, the relatively rapid rotationof the New Hebrides Arc to form the North Fiji Basin, andsignificant rollback of the subduction hinge as the PacificPlate was subducted west beneath the Tonga–KermadecArc.

New Guinea and Melanesian boundary

Because nearly all the reconstructions of the region focuson either the New Guinea or the Melanesian regions therehas been little attention paid to the change of polarity insubduction which is implied at about the boundarybetween the New Guinea and the Melanesian segments. Ina few models (Lee & Lawver 1995; Jolivet et al. 1989) thechange is shown schematically but the changing length ofthe section between the north- and south-dipping subduc-tion zones is not shown, nor is it clear exactly how thisregion is to be interpreted. Falvey and Pritchard (1982) sug-gested that before 45 Ma there was southeast-dipping sub-duction beneath the New Britain – New Ireland Arc whichwas orthogonal to the Australian margin. The southeast-dipping subduction bends round into the conventionalsouthwest-dipping subduction beneath the Solomons. Theypostulated rapid rotation of the New Britain – New Irelandsection of the arc at about 30 Ma followed by developmentof northeast-dipping subduction to give the present config-



Figure 5 Tectonic reconstructions at 40 and 10 Ma (Hall 2002). At 40 Ma Australia was moving north and south-dipping subductionbeneath the Melanesian Arc had just begun, whereas there was north-dipping subduction beneath the Philippines–Halmahera Arc eastof Papua. At 10 Ma eastward subduction of the Molucca Sea had begun beneath Halmahera. The New Guinea terranes moved in awide left-lateral strike-slip zone. Subduction of the Solomon Sea was underway beneath eastern New Guinea margin to form theMaramuni Arc. To the east of New Guinea subduction began on the east side of the Solomon Sea to initiate the New Hebrides Arc.


Figure 6 Twelve depth images for the southwest Pacific/Australia region. The sections are taken from the global tomographic model ofBijwaard & Spakman (2000). Section depths are indicated in the lower left corner of each panel. Colour contouring depicts the anom-alous P-wave speed relative to the average speed at depth. The reference model is ak135 (Kennett et al. 1995) Negative (positive) anom-alies are coded in red (blue) colours and represent relatively slow (fast) regions with respect to the average ak135 propagation speeds.The limits of the contouring scale vary between –X% and +X% where X=3 (a), 2.5 (b), 1.5 (c–i), and 0.75 (j–l).


uration. During part of this interval the former southeast-dipping subduction zone acted as a transform. Their modelis schematic but is the only model in which there is a sub-duction zone orthogonal to the Australian margin.

The tectonic model

Hall (1997, 1998, 2002) has described a plate-tectonicmodel which covers the whole of the region of interest inthis paper, between the Molucca Sea and theTonga–Kermadec Arc. It synthesises a range of data, fromspreading histories obtained from small ocean basins tomore qualitative information such as that obtained fromland geology. It therefore has features in common withmany of models outlined above but differs from all of themin postulating a mainly strike-slip plate-boundary zone innorthern New Guinea during the Neogene. Computer ani-mations of this model are in Hall (2002). Because we canexamine the model at 1 million year steps for the wholeregion of interest we have used it to compare its predictionswith features observed in tomographic images. The essen-tial features of this model are as follows.

From the Mesozoic the north Australian margin was apassive continental margin produced by Jurassic rifting.There was diachronous arc–continent collision of intra-Pacific arcs with this passive margin emplacing ophioliticrocks in between New Guinea and New Caledonia regionsin the Eocene. At about 45 Ma, after the collision, there wasa plate reorganisation; new north-dipping subductionbegan about 2000 km north of Australia beneath thePhilippines–Halmahera Arc. At about 45 Ma new south-ward subduction began at the northeast and eastAustralian margin forming the Melanesian Arc betweenNew Britain and Tonga (Figure 5).

Subduction beneath the Philippines–Halmahera Arccontinued until about 25 Ma. During this period thePhilippines–Halmahera Arc remained in approximately thesame position. In the Melanesian segment rollback of thesubduction hinge between 45 and 25 Ma led to the forma-tion of a wide backarc basin from the Solomon Sea to theSouth Fiji Basin. The subduction direction changed tosouthwest-dipping as the Melanesian Arc rotated north.From about 40 Ma the Caroline Sea opened as a backarcbasin by rollback of the Pacific subduction hinge at the east-ern margin of the Philippine Sea Plate. The South CarolineArc was east of this backarc basin and was the site of for-mation of arc terranes now found in northern New Guinea.

At about 25 Ma arc–continent collision occurred whenthe Philippines–Halmahera Arc collided with the Australianmargin and this terminated the north-dipping subduction.The zone of collision stretched from Sulawesi to easternNew Guinea. At about the same time the Ontong JavaPlateau collided with the Melanesian Arc. This led to ter-mination of the southwest-dipping subduction system.

After these collisions the arcs between the Philippinesand Melanesia became broadly a single system whichrotated clockwise at the leading edge of the Pacific Plate,accommodated by intra-plate subduction at the easternedge of the Philippine Sea Plate. The South Caroline Arcterranes, which are now found in north New Guinea,moved along the north Australian margin in a complexstrike-slip zone (Figure 5). There was therefore no signifi-

cant subduction at the north Australian margin in IrianJaya. Further east, Pacific–Australia convergence wasaccommodated by development of new subduction zones:there was southwest-dipping subduction beneath eastPapua forming the Maramuni Arc and northeast-dippingsubduction beneath the New Hebrides. The two opposedsubduction systems were linked by a transform crossing theSolomon Sea.

When Maramuni subduction ceased the New HebridesTrench propagated west to start new north-directed subduc-tion beneath the Solomons and New Britain Arcs. Slab-pullforces at the New Britain Trench then led to the formation ofthe Woodlark spreading centre at the former Solomon Seatransform. Very young Woodlark Basin lithosphere is nowbeing subducted at the South Solomon trench. Throughoutthis period west-dipping subduction of the Pacific Plate con-tinued at the Tonga–Kermadec Trench with rapid rollback ofthe subduction hinge in the last 10 million years.

SEISMIC TOMOGRAPHY MODELS

The mantle structure of the entire region is only imaged inglobal mantle models. Recent developments in data analy-sis and tomographic techniques have led to a new genera-tion of global models that are capable of resolving smallerdetails of mantle structure (van der Hilst et al. 1997; Grandet al. 1997; Bijwaard et al. 1998; Ritsema et al. 1999;Megnin & Romanowicz 2000; Bijwaard & Spakman 2000).In particular, the P-wave-speed model of Bijwaard et al.(1998), based on the dataset of Engdahl et al. (1998),shows details of slab morphology comparable to those pre-viously seen only in studies of smaller regions of Australiaand the southwest Pacific (Abers & Roecker 1991; van derHilst 1995).

Imaging small details on a global mantle scale has beenpossible by using a special cell-parameterisation techniquewhich allows the variation of cell dimensions as a functionof the local degree of data density (Spakman & Bijwaard2001). The smallest cells used have dimensions of 0.6° lat-erally and 35 km in depth. In this paper we use an improvedtomographic model (Bijwaard & Spakman 2000) which wasobtained by incorporating in addition 3-D ray tracing in the tomographic analysis in an attempt to account for (de-)focussing effects due to 3-D mantle structure.

The tomographic results for the region are displayed asselected layer sections down to a depth of 1040 km inFigure 6 and sensitivity tests for spatial resolution in Figure7. Figure 8 shows the principal positive anomalies on thetomographic images of Figure 6 and some vertical depthsections are shown in Figure 9.

Model confidence

Sensitivity tests (Spakman & Nolet 1988; Bijwaard et al.1998, Bijwaard & Spakman 2000) are extremely importantfor the interpretation of the tomographic results as theyindicate the scale or features which can be resolved andhence interpreted. Because of the importance of these tests,which have been conducted for a variety of models, we pro-vide below an explanation of them which we hope will sat-isfy both the non-technical and expert reader.


Seismic travel-time tomography is an inversion tech-nique that produces a 3-D mantle model of seismic-wavespeed from surface observations of the travel times of seis-mic waves. For a variety of reasons (e.g. data errors) themodel can only be considered as an estimate of true Earthstructure. One of the major difficulties in tomography is thedetermination of model errors. Techniques exist to estimatemodel errors for relatively small-size inverse problems(Menke 1989), but these methods are not yet applicable toa huge inverse problem of the type discussed here. For largeinverse problems one has to resort to so-called sensitivityanalysis with synthetic wave-speed models (Spakman &Nolet 1988). First, a synthetic model is created, for exam-ple, one which contains block-shaped anomalies of alter-nating sign, like a 3-dimensional checkerboard pattern.Next, the same seismic rays used in the actual data inver-sion are used to compute synthetic travel times from thesynthetic model. The last step is to invert the synthetic datain the same way as the real data inversion. This leads to anestimate (the tomographic view) of the synthetic model. Itcan be shown that the relation between the synthetic modeland its tomographic image is given by the same spatial res-olution operator as for the true Earth and its tomographicimage derived from real data. Thus, comparison of the syn-thetic model and its tomographic image can lead to infer-ences of spatial resolution errors (image blurring orsmearing) that are equally applicable to the actual tomo-graphic problem.

In Figure 7 we display examples of sensitivity analysisfor our region. The synthetic models are constructed fromisolated blocks (some with an irregular shape) with differ-ent spatial dimensions and which have designated syn-thetic wave-speed anomalies of alternating sign. In Figure 7the outlines of these blocks are plotted with a black contourline. The colour coding depicts the tomographic image ofthe synthetic model. If the synthetic model was perfectlyrecovered each of the black outline volumes (areas on thelayer plots shown) would contain either entirely red orentirely blue and all the intervening area would be shadedin the central neutral colour on the colour scale (palegreen). If the colour is blurred outside the contoured areas,and the colour ‘leaks out’ of the blocks this indicates thatthe synthetic model is less well recovered. In regions wherethe tomographic image correlates well with the syntheticmodel the conclusion is often made that the spatial resolu-tion is high and thus the tomographic image determinedfrom the real data is also of high quality. There is a risk ofthis reasoning being wrong (Lévêque et al. 1993) but thisrisk can be made small by conducting sensitivity tests witha range of different block sizes.

Figure 7 shows that spatial resolution is highly variablein the mantle volume of interest. In the shallow mantledetails on the scale of 1.2° are detectable in the region ofhigh seismicity close to the Australia–Pacific plate bound-ary (200 km) but in other regions this scale of resolutionis absent. However, at a similar depth (230 km) resolutionon a scale of 2.4° is possible over much more of theregion. Similarly, near the lower part of the upper mantle(530 km) blocks on the scale of 2.4° are well resolved overmost of the central region close to the Australia–Pacificplate boundary but, for example, beneath the Coral Searesolution is not so good although blocks on the scale of

4.2° are well resolved (595 km). At greater depths (980and 1040 km) resolution is on the scale of 3–4°. At alldepths it is not possible to resolve anything within theregion beneath the Pacific in the northwest part of themap. Generally, the information contained in Figure 7(and at other levels than those shown in Figure 7) isextremely helpful in selecting those parts of the actualtomographic model and the spatial scales for which inter-pretation can be attempted.

FEATURES VISIBLE IN TOMOGRAPHIC IMAGES

For the purposes of description and discussion we numberthe principal positive anomalies on the tomographic images(Figure 8). Several are directly associated with existing sub-duction (A1–A4). A number of more enigmatic wave-speedanomalies (A5–A8) are found in the deeper parts of theupper mantle and lower mantle. Here we first we identifythe anomalies and in the next section we discuss theirinterpretation.

A1. Positive P-wave speed anomalies associated withsubduction are readily identified in the Tonga–Kermadecregion where the slab is visible to depths of 1500 km. Theslab has a flat-lying portion below Tonga in the north whichis seismically active (Figures 6g–i, 9h). Below the westernpart of the flat-lying slab subduction can be followed intothe lower mantle where it has a dominant north–southstrike. The upper and lower mantle anomalies are wellresolved (Figure 7) and confirm the earlier findings of vander Hilst (1995).

A2. To the northwest, subduction at the New HebridesTrench yields a fragmented image (Figure 6a–f), but theslab is well delineated by intermediate and deeper seismic-ity (Figure 9g, h). To the northeast, below the North FijiBasin, a flat-lying positive anomaly in the upper mantlecorresponds to a region of deep seismicity (Figures 3b, 6g–i,9g). Although weak positive anomalies are imaged below700 km poor spatial resolution does not warrant an inter-pretation of slab penetration into the lower mantle belowthe North Fiji Basin.

A3. Further to the northwest, below the SolomonIslands segment of the plate boundary, tomographic evi-dence for an upper mantle slab is absent above 500 km inthe central and southern part (Figures 6a–f, 9f). Positiveanomalies appear only below 500 km with seismic activityin the southeast (Figure 3, bottom). Clear evidence for anupper mantle slab is found in the northwest part of theSolomon segment where slab geometry follows the cusp inthe plate boundary toward the New Britain Arc (Figures6a–i, 9e). The subducting Solomon Sea slab is seismicallyactive down to 600 km and north dipping. Laterally, theSolomon Sea slab suddenly terminates beneath centralNew Guinea and no upper mantle slab is imaged belowwest New Guinea (Figures 6a–g, 9c, d).

A4. Still further to the west, subduction-related anom-alies are found in the upper mantle beneath Sulawesi andHalmahera (Figure 6a–e). Subduction of the Molucca Seabelow both Halmahera to the east and Sulawesi and theSangihe Arc to the west shows a slab with a clear invertedU-shape (Figure 9a) as previously recognised byWidiyantoro and van der Hilst (1997). Below the Caroline



Figure 7 Sensitivity ‘spike’-test results for assessing image quality. The six panels show inversion results of the retrieval of synthetic‘spike’ models of anomalous seismic wave speed (Bijwaard & Spakman 2000) for some selected depths and for different spike (or block)size. See text for full explanation. The square-like black contour lines indicate the outline of single anomaly blocks with a –5% or +5%amplitude. Synthetic block-anomalies are in all directions surrounded by 0%-synthetic anomalies. This is different from the so-calledchecker-board test (Lévêque et al. 1993) and allows a much better detection of image smearing. The colour contouring shows theretrieval of the synthetic spike pattern. The contour limits are –2.5% and +2.5% which is a compromise between showing large-ampli-tude spike response and small-amplitude resolution artefacts. In the well-recovered areas many of the block amplitudes exceed the con-touring limits but 100% recovery of spike amplitudes is rarely obtained. Image smearing is an expression of lack of spatial resolutionwhich is probably small in the well-recovered regions and is large in the poorly and not-recovered regions. The first two panels (200,230 km) show test results for spike sizes of 1.2 and 2.4° at the top of the upper mantle. The two panels in the centre (595, 530 km)show the spike response for 2.4 and 4.2° spikes near the bottom of the upper mantle. The last two panels (1040, 980 km) give the resultsfor 3.0 and 4.2° spikes in the lower mantle. Notice that poor recovery of small blocks at a given depth does not imply poor recovery oflarger blocks. Anomalous mantle structure with a spatial dimension of about 2.4° near 600 km depth cannot be detected below the east-ern margin of Australia but structures with larger dimensions (4.2°) probably can be detected.

Plate this anomaly broadens with depth and appears tomerge with a large anomaly which is continuous into thelower mantle (Figure 9a). The Sangihe slab dips to thenorthwest (Figure 9a) and its anomaly also apparently linkswith a deeper anomaly, which is part of a huge positiveanomaly that underlies a large part of southeast Asia(Widiyantoro & van der Hilst 1997; Bijwaard et al. 1998;Rangin et al. 1999).

A5. A prominent positive anomaly is imaged below theCaroline Ridge (Figure 6g–k). It broadens with depth and isprimarily present in the upper mantle (Figure 9a, b) exceptfor its northern limit which corresponds to the deepMariana subduction.

A6. Another extensive positive anomaly is found betweendepths of 500 and 700 km with an elongate shape extendingfrom north of the Bird’s Head, beneath the Arafura Sea, tothe Gulf of Carpentaria (Figures 6g–j, 9c, d). Towards thenorth, this anomaly appears to merge with that associatedwith the young subduction of the Halmahera slab (A4).

A7. An important positive anomaly is imaged below 500km and predominantly in the upper mantle which extendsfrom the Papuan peninsula to the New Hebrides and fromthe Solomon Islands to the east Australian margin (Figure6g–i). Because of the large lateral extent of this flat anomaly(Figure 9f, h) it can be identified in many other global man-tle models including those with a nominally lower spatial res-olution (Ritsema et al. 1999; Megnin & Romanowicz 2000).

A8. There is a huge positive anomaly which stands outvery clearly between 710 and about 1100 km and extendsfrom beneath the Gulf of Carpentaria to Papua with a broadlynorth-northeast–south-southwest trend (Figures 6j–l, 9e).

From the sensitivity test (Figure 7) we infer that theanomalies A5–A8 are spatially sufficiently resolved to war-rant interpretation.

INTERPRETATION OF TOMOGRAPHIC FEATURESIN TERMS OF TECTONICS

We briefly comment on anomalies associated with theAustralian Plate, then deal with anomalies clearly related topresent-day subduction, and finally discuss other mantleanomalies which we interpret in terms of older subduction.

Lithosphere under Australia

The tomographic model shows a positive anomaly beneathmuch of northern Australia (Figures 6a, b, 9c, d). There ishowever a significant variation in lithosphere velocitiesfrom fast to slow from west to east New Guinea and fromwest to east across Queensland. This trend is better imagedin surface-wave tomography models and interpreted asslower wave speed below the Palaeozoic fold belts of east-ern Australia and increasing wave speeds toward the westbelow the Proterozoic and Archaean shields (Simons et al.1999; Debayle & Kennett 2000)

Young and current subduction systems

A1. At the Tonga Trench the Pacific Plate subducts westat a high rate, decreasing southwards towards NewZealand (Figure 9g, h). The age (>120 Ma) of the subducted


Figure 8 The anomalies discussed in the text. A1–A4 (green) are dipping anomalies in the upper mantle, A5–A7 (blue) are broadly flatanomalies near to the base of the upper mantle, and A8 (red) is a strong anomaly in the lower mantle. The anomaly labelled A1 andshown in red is that associated with the Tonga slab in the lower mantle and the outline is that identified on the 1040 km depth layerof Figure 6l. Grey lines, major plate boundaries; black lines, great circle segments of the tomographic slices presented in Figure 9.


Figure 9 Eight vertical sections through the tomographic model of Bijwaard & Spakman (2000) to a depth of 1500 km. All cross-sec-tions are taken along a great circle segment (straight line in each map) of 30°. Section coordinates (starting point, azimuth) are givenin the map. Colours in the map denote the Earth’s topography (ETOPO5). Each map has a small inset showing the location of the sec-tion. As in Figures 6 and 7 colours in the sections denote the anomalous P-wave velocity structure. It should be noted that anomalyamplitudes vary with depth from many percent in the top of the mantle to typical peak values around 0.5% at 1000 km or deeper(Bijwaard & Spakman 2000). All sections are contoured between –1% and +1% and relatively strong colours in the uppermost mantleor weak colours in the lower mantle should not be interpreted as an indication of significance or insignificance of imaged structure.White dots in the sections represent hypocentres (taken from Engdahl et al. 1998) of earthquakes that occurred within 50 km of thesection plane. Positions of sections are shown in Figures 6 and 8.

lithosphere and high rate of subduction (>10 cm/y) result ina deep negatively buoyant slab of lithosphere. Van der Hilst(1995) has shown that there are distinctive high-velocityanomalies beneath the Tonga–Kermadec Arc which makethe subducted slab visible down to more than 1200 km.Beneath the Tonga Arc the slab dips uniformly at about 50°and becomes almost horizontal in the transition zone anduppermost lower mantle (Figure 9h). It then sinks into thelower mantle several hundred kilometres further west. Vander Hilst (1995) suggested that the shape of the subductedslab is due to rapid rollback of the subduction hinge at the

northern end of the Tonga–Kermadec Arc where at least2100 km of slab is imaged, in contrast to much slower roll-back at the southern end of the system where <1000 km ofslab is visible. Assuming a constant rate of subductionthese dimensions suggest that the anomalies are due tosubduction since 25 Ma and that most rollback occurredsince about 10 Ma (Figure 10).

A2. Beneath the North Fiji Basin there are high velocityanomalies visible in the upper mantle (Figure 9g). The slabdips steeply in a northeasterly direction and becomesalmost horizontal in the transition zone. The tomographic



Figure 10 Tectonic reconstructions from 45 to 5 Ma (Hall 2002), omitting 25 Ma for layout reasons. The reconstructions use theIndian–Atlantic hotspot frame of Müller et al. (1993). On each map the present positions of the deep mantle anomalies (A6–A8) areshown in yellow. There is a good correspondence of the anomaly A6 with the predicted position of a slab produced by north-dippingsubduction beneath the east Philippines–Halmahera Arc between 45 and 25 Ma. There is also a good correspondence between anom-aly A7 and the predicted position of a slab produced by south-dipping subduction beneath the Melanesian Arc between 45 and 25 Ma.The younger anomalies are shown in pale green (A1–A5). Note in particular the deep part of the Tonga slab (anomaly A1 on the 1040km layer) compared to the position of the anomaly in the upper mantle. These suggest an almost stationary trench from about 25 to 10Ma and fast rollback since 10 Ma at the Tonga Trench.

results do not exclude slab penetration into the lower man-tle because of poor resolution. Beneath the northern part ofthe North Fiji Basin are deep earthquakes at 550–650 kmwhich have been attributed to events in the flat-lying slab(Okal & Kirby 1998). The seismicity and tomography sug-gest a similar interpretation to that of the Tonga subductionwith rapid rollback causing development of a kink and flatsegment in the subducted slab. The tectonic model (Hall2002) predicts about 1000 km of subduction since 15 Maat the New Hebrides Trench (Figure 10). Assuming a con-stant subduction rate the orientation and length are consis-tent with a slab subducted at the New Hebrides Trenchsince about 12 Ma, and clockwise rollback of the subduc-tion hinge during this period.

A3. Beneath the Solomons there is no seismic or tomo-graphic evidence of a significant length of subducted slab(Figure 9f). Seismicity to depths of 200–300 km suggestssubduction locally to north and south on the south andnorth sides of the Solomons respectively. There are a fewscattered events at depths greater than 500 km. At the west-ern end of the Solomons the subducted slab of the NewBritain Arc becomes visible and is associated with seismic-ity to depths of 450 km (Figure 9h). Just east of the longi-tude of New Ireland there is a marked arcuate kink in thehigh-velocity anomaly at depths below 200 km whichbecomes smoother at greater depths (Figure 6a–f). Thismay be due to deformation of the slab in the mantle. In thetectonic model (Hall 2002) the New Britain subduction isinterpreted to have propagated northwest from the NewHebrides Trench after subduction ceased at the MaramuniTrench. An alternative possibility suggested by the tomog-raphy is that subduction initiated beneath the New BritainArc and propagated east with the eastern Solomon Searemaining partly coupled to the Pacific. This would accountfor the absence of seismicity and a high-velocity anomalybeneath the Solomons. However, the absence of seismicityand anomaly beneath the Solomons may be due to the veryyoung age of the Woodlark Basin crust being subducted;south of New Georgia the Woodlark spreading centre isbeing subducted and therefore the subducted slab may behotter than the mantle into which it is sinking. North of theWoodlark Rise, where a high-velocity anomaly is very clear,the subducted slab descending beneath Bougainville is sig-nificantly older, and is probably Oligocene (Figure 10).

A4. Beneath the Molucca Sea region high-velocityanomalies are clearly visible to the west and east (Figure9a) and these were interpreted by Rangin et al. (1999) asthe Molucca Sea slab subducted during the Neogene.Seismicity shows the west-dipping Sangihe slab is identifi-able to at least 650 km and the east-dipping Halmaheraslab to about 200 km (Cardwell et al. 1980; McCaffrey1982). Tomography suggests the Halmahera slab goesdeeper to at least 400 km (Rangin et al. 1999). Taking intoaccount the dip on the slabs the total length of subductedlithosphere is therefore about 1500 km, very close to thelength predicted by the tectonic model (Hall 2002) sincethis subduction began on the Sangihe side of the MoluccaSea at 25 Ma. As both the Sangihe and Halmahera slabsare still connected to the surface, and the southern bound-ary of the Molucca Sea, which is the Philippine Sea –Australia plate boundary, is also moving northward, thecontinuity of these slabs implies that the entire inverted U-

shaped Molucca Sea Plate is moving north in mantle withAustralia. On some sections the Molucca Sea Plate anom-alies appear to merge with deeper anomalies and Rangin etal. (1999) postulated that both the Sangihe and Halmaheraslabs can be traced deeper into the mantle. They connectedthe Sangihe slab to a broad high-velocity zone beneath theCelebes Sea at depths between 700 and 1000 km, andjoined the Halmahera slab to another broad high-velocityzone beneath the Bird’s Head at depths between 400 and600 km. However, if this is the case there has been about3500 km of slab subducted (Rangin et al. 1999) whichwould be much greater than the amount of subduction pre-dicted by the tectonic model (Hall 2002) or other models(Jolivet et al. 1989; Rangin et al. 1990) and would placeHalmahera much further east than suggested on any recon-structions of any age. In contrast, we therefore interpretthese broad and flat anomalies as the remnants of a slabsubducted beneath the Philippines–Halmahera Arcbetween 45 and 25 Ma (Figure 10).

Older subduction systems

A5. Centred on the Caroline Ridge and extending southtowards the Bismarck Sea is a broad high-velocity anomalyat depths between 560 and 710 km (Figure 9b, d). This isthe southern end of a Pacific high-velocity anomaly trend-ing approximately north–south which continues to thenorth beneath the eastern Philippine Sea Plate. We inter-pret this anomaly to be the result of Pacific subductionbeneath the eastern margin of Philippine Sea Plate since 25Ma. This interpretation implies rollback of the Pacific sub-duction hinge and overriding of the subducted slab by thePhilippine Sea Plate as shown in the tectonic model (Hall2002) (Figure 10).

A6. At depths between 500 and 710 km is an extensive,elongate high-velocity anomaly extending from north of theBird’s Head, beneath the Arafura Sea, to the Gulf ofCarpentaria (Figure 9a–d). Towards the north this anomalyappears to merge with the young Halmahera slab anomalyeast of the Molucca Sea. However, the Halmahera slab hasa clear consistent dip to the east and the anomaly below500 km appears to be separated from it and flat lying. Thedeeper anomaly also extends much further south. It alsobecomes deeper to the north, to the north of Halmahera.We interpret this anomaly to be the result of northwardsubduction between 45 and 25 Ma beneath thePhilippines–Halmahera Arc north of Australia. Accordingto the tectonic model (Hall 2002) this would have been thesubduction zone at which approximately 2000 km ofoceanic lithosphere was subducted between 45 and 25 Ma(Figure 10). The length of slab estimated from the tomo-graphic cross-sections is of a similar order.

A7. At depths from about 500 to 700 km there is anextensive high-velocity anomaly extending from thePapuan peninsula to the New Hebrides and from theSolomons to the east Australian margin (Figure 9f, h).Below the Coral Sea and the eastern margin of Australiathis anomaly is not well defined at scales of a few hundredkilometres. The segment under Queensland may not beconnected to the extensive anomaly that is imaged frombeneath the Papuan peninsula to the New Hebrides. Weinterpret the latter anomaly to be the result of south- and


southwest-directed Pacific subduction at the MelanesianArc between 45 and 25 Ma. Ma. As noted above, the for-mation of a large marginal basin behind the MelanesianArc implies very significant rollback of the subductionhinge during the Palaeogene. By analogy with the present-day Tonga system, rapid rollback would produce an exten-sive area of flat-lying slab which would be expected to sinkto the base of the mantle after subduction ceases. A largehigh-velocity anomaly is consistent with both the Yan andKroenke (1993) model and the models which predict signif-icant rollback although the amount is different in the twomodels. Figure 10 shows there is excellent agreementbetween the prediction of the Hall (2002) tectonic modeland the tomography. If this interpretation is correct, a flat-lying slab can survive for many tens of millions of years atthe bottom of the upper mantle. Slowly increasing buoy-ancy due to temperature increase may prevent slumpinginto the lower mantle.

A8. One of the clearest anomalies in the region is anextensive volume of low velocities with a north-north-east–south-southwest trend visible between 920 and 1040km which extends from beneath the Gulf of Carpentaria toPapua (Figure 9d, e). Young subduction in the Tonga Archas caused a slab to enter the lower mantle but elsewherein the region there is little evidence for an active subductionzone which has penetrated so deep. The very high rate ofsubduction and the great age of the subducted Pacificlithosphere beneath Tonga may be the cause of such deeppenetration (van der Hilst 1995). Even the slabs postulatedabove to have been subducted between 45 and 25 Ma(anomalies 5 to 7) do not extend significantly into the lowermantle. This anomaly is therefore unusual. We have anumber of hypotheses for this anomaly. Gurnis et al. (2000)have suggested that an approximately north–south seismicanomaly beneath the Australian–Antarctic Discordancesouth of Australia may be the result of a slab in the mantlesubducted east of Australia during the Cretaceous and sub-sequently overridden by Australia as it has moved east. Ouranomaly 8 is at a similar depth and along strike from theGurnis et al. (2000) Australian–Antarctic Discordanceanomaly and may have the same explanation if this sub-duction zone had continued north with a similar orienta-tion. The anomaly is located above the position where theremust have been a change in polarity in subduction at theboundary between the New Guinea and the Melanesiansegments, as noted above. Based on palaeomagnetic argu-ments Falvey and Pritchard (1982) suggested that in thisregion before 45 Ma there was southeast-dipping subduc-tion which was orthogonal to the Australian margin. Thus asecond possibility is that the anomaly represents a slabsubducted before 45 Ma. This would be consistent with itsdepth, and the orientation of subduction. A final possibilityis that the anomaly represents a slab subducted between45 and 25 Ma which has sunk to a grater depth than otherslabs subducted during the same period. The Tonga regionshows that this is quite possible if there is high rate of sub-duction and/or the age of the slab is great. The tectonicmodel (Hall 2002) suggests this region was a zone of veryrapid subduction in this period in which there was fast roll-back of at least two subduction hinges (Figure 10), onerelated to subduction beneath the Melanesian Arc and asecond related to subduction beneath the Caroline Arc. We

prefer the first or second hypotheses since to produce thisanomaly with slab penetration a more or less stationarysubduction zone would probably be required.

Subduction systems which are not seen

Besides identifying mantle anomalies with old subductionsystems the tomography also offers the opportunity to dis-tinguish between models which postulate different subduc-tion histories. As reviewed above, most authors postulatenorth-dipping subduction north of Australia during thePalaeogene. We see no evidence for the south-dipping sub-duction suggested in some interpretations. If this hadoccurred there should be distinctive high-velocity anom-alies beneath the craton of western Australia (Figure 9c, d).Their absence and the presence of high-velocity anomaliesfurther north (anomaly A6) cause us to prefer north-dippingsubduction with subducted slabs now overridden byAustralia.

We see no evidence of continuation of subduction at theNew Britain Arc very far to the west (Figure 9c, d). Thisimplies the ‘doubly vergent system’ did not extend signifi-cantly west of Papua New Guinea and there has been noNeogene subduction beneath western New Guinea. This isconsistent with the absence of subduction-related volcanicactivity in Irian Jaya (Hall 2002) and the unusual chemicalcharacter of the volcanic rocks that are known (Housh &McMahon 2000).

There is one subduction zone which is widely postu-lated, and included in the tectonic model (Hall 2002)(Figure 10), but for which we see little evidence in thetomography, and that is the Maramuni Arc (Dow 1977;Rogerson & Williamson 1985; Cullen 1996; McDowell et al.1996; Hill & Raza 1999). Southwest-dipping subduction iscommonly suggested to have occurred during the Miocenebeneath the Papuan peninsula. In the tectonic model (Hall2002) the trench would have advanced as the AustralianPlate moved north from about 20 Ma to 7 Ma (Figure 10)and would have overridden the subducted slab. It wouldtherefore be expected that the subducted slab wouldbecome flat-lying in the upper mantle and the slab shouldbe visible at relatively shallow depths beneath the CoralSea and Queensland. About 1000 km of subduction is pre-dicted by the tectonic model (Hall 2002) but no clearanomaly is visible. It is possible that part of anomaly A7under Queensland is due to this subduction.

There are several possible explanations. First, the lithos-phere being subducted was relatively young. According tothe tectonic model (Hall 2002) the age of the lithospherearriving at the trench would have been about 20 millionyears greater than the age of its subduction. Visibility ontomographic images would have been reduced because theseismic-wave speed contrast with the ambient upper man-tle would be smaller. There is no lithosphere of this age cur-rently being subducted in the region with which to make acomparison but south of the Solomons, lithosphere of theWoodlark Basin, which is less than 10 Ma in age, is beingsubducted and no high-velocity anomaly is seen. A secondpossibility is that the resolution of tomographic model inthe region beneath the Coral Sea and Queensland may beinadequate. This is possible but one would expect to see1000 km of slab because resolution is of the order of 4°. A


third possibility is that the tectonic interpretation is wrong:for example volcanism was of shorter duration than postu-lated based on the few K–Ar dates (Page 1976; Rogerson &Williamson 1985); the volcanic activity generally inter-preted as subduction-related was due to another cause(some is reported to be shoshonitic); or that more subduc-tion occurred on the north side of the Solomon Sea and lesson the south side at the Maramuni Arc [for example Abbott(1995) questions the existence of the Maramuni Arc].

CONCLUSIONS

The rapidly improving tomographic images offer significantopportunities to distinguish between, test and improve tec-tonic interpretations of the region. In order to test models itis important to be able to examine a model at closelyspaced intervals of time, and it is an advantage to be ableto see how trenches move. Most tectonic models are not yeteasy to compare to tomographic results because they areincomplete, or because there are large gaps in time betweenreconstructions.

We have compared the tomographic images to the pre-dictions of one model, that of Hall (2002) and in generalfound good agreement. The positions of most interpretedsubduction zones are consistent with high-velocity mantleanomalies as are the predicted slab lengths. For the regionof northern Australia and immediately north in the Pacificwe conclude that there is evidence for north-directed sub-duction north of Australia and an absence of evidence forsouth-directed subduction beneath New Guinea between45 and 25 Ma. As Australia has moved north it has over-ridden former north-dipping subduction zones and theeffects of such a passage may be evident in the stratigraphy.For example, Müller et al. (2000) identified tectonic subsi-dence in Oceanic Drilling Program sites off northeastAustralia to be the result of dynamic topography induced asthe Australian margin passed over a subducted slab burialground. We agree with the suggestion but prefer to interpretmantle anomalies in this region as the result of Palaeogenesouth-directed subduction rather than north-directed sub-duction. Our interpretation of a huge area of flat slabunderlying the Coral Sea toward the New Hebrides impliesthat a flat-lying slab can survive for many tens of millionsof years at the bottom of the upper mantle.

We see no evidence for significant subduction beneathmost of New Guinea since 25 Ma. This is consistent withthe strike-slip model of Hall (2002) and it is consistent withother geological evidence, for example, the absence ofNeogene subduction-related volcanic activity in Irian Jaya.However, it is necessary to be cautious. One widelyaccepted subduction system, the Maramuni Arc of PapuaNew Guinea, is not visible on the tomographic images. Thiscould be because the lithosphere being subducted was rel-atively young, the resolution of tomographic model is inad-equate, or the tectonic interpretation is wrong. In manycritical areas, the tomographic model still has insufficientresolution. Slabs may not be imaged even if they are theredue to a lack of sampling by ray paths. Tomography maynot be able to resolve sufficient detail to distinguishbetween different models that are similar but differ indetail, the images may not reveal short-lived subduction

systems, and it may well be impossible to image younglithosphere being subducted. All of these points need to beconsidered in a region in which the tectonic setting is capa-ble of changing rapidly as indicated by the youth of severallarge marginal basins, and suggested by the complexity ofthe region. It is also essential to have a dynamic view. Slabscan move in mantle, as illustrated best by the Molucca SeaPlate which must have moved north with the AustralianPlate, and some deform in the mantle. Nonetheless, tomog-raphy can clearly make an important contribution by test-ing and improving tectonic models, and in some casesquestioning assumptions in a region where there is still arelatively limited knowledge of the surface geology.

ACKNOWLEDGEMENTS

Part of this work (W. Spakman) was conducted under theprogrammes of the Vening Meinesz Research School forGeodynamics (Utrecht University) and of the NetherlandsResearch Centre of Integrated Solid Earth Sciences.Financial support to R. Hall has been provided by NERC,the Royal Society, the London University Central ResearchFund, and the Royal Holloway SE Asia Research Groupsupported by a number of industrial companies.

REFERENCES

ABBOTT L. D. 1995. Neogene tectonic reconstruction of theAdelbert–Finisterre–New Britain collision, northern Papua NewGuinea. Journal of Southeast Asian Earth Sciences 11, 33–51.

ABERS G. A. & ROECKER S. W. 1991. Deep structure of an arc-continentcollision; earthquake relocation and inversion for upper mantle Pand S wave velocities beneath Papua New Guinea. Journal ofGeophysical Research 96, 6379–6401.

BENES V., SCOTT S. D. & BINNS R. A. 1994. Tectonics of rift propaga-tion into a continental margin: western Woodlark Basin, PapuaNew Guinea. Journal of Geophysical Research 99, 4439–4455.

BEVIS M., TAYLOR F. W., SCHUTZ B. E., RECY J., ISACKS B. L., HELU S.,SINGH R., KENDRICK E., STOWELL J., TAYLOR B. & CALMANT S. 1995.Geodetic observations of very rapid convergence and back-arcextension at the Tonga Arc. Nature 374, 249–251.

BIJWAARD H. & SPAKMAN W. 2000. Nonlinear global P-wave tomogra-phy by iterated linearized inversion. Geophysical JournalInternational 141, 71–82.

BIJWAARD H., SPAKMAN W. & ENGDAHL E. R. 1998. Closing the gapbetween regional and global travel time tomography. Journal ofGeophysical Research 103, 30055–30078.

BUNGE H-P. & GRAND S. P. 2000. Mesozoic plate-motion history belowthe northeast Pacific Ocean from seismic images of the subductedFarallon slab. Nature 405, 337–340.

CARDWELL R. K., ISACKS B. L. & KARIG D. E. 1980. The spatial distrib-ution of earthquakes, focal mechanism solutions and subductedlithosphere in the Philippines and northeast Indonesian islands.In: Hayes D. E. ed. The Tectonic and Geologic Evolution ofSoutheast Asian Seas and Islands, pp. 1–35. AmericanGeophysical Union Geophysical Monograph 23.

CHARLTON T. R. 2000. Tertiary evolution of the Eastern IndonesiaCollision Complex. Journal of Asian Earth Sciences 18, 603–631.

COOPER P. & TAYLOR B. 1987. Seismotectonics of New Guinea; a modelfor arc reversal following arc continent collision. Tectonics 6,53–67.

CROOK K. A. W. & BELBIN L. 1978. The Southwest Pacific area duringthe last 90 million years. Journal of the Geological Society ofAustralia 25, 23–40.

CULLEN A. G. 1996. Ramu Basin, Papua New Guinea: a record of LateMiocene terrane collision. American Association of PetroleumGeologists Bulletin 80, 663–685.

DALY M. C., COOPER M. A., WILSON I., SMITH D. G. & HOOPER B. G. D.


1991. Cenozoic plate tectonics and basin evolution in Indonesia.Marine and Petroleum Geology 8, 2–21.

DEBAYLE E. & KENNETT B. L. 2000. The Australian continental uppermantle: structure and deformation inferred from surface waves.Journal of Geophysical Research 105, 25423–25450.

DE JONGE M. R., WORTEL M. J. R. & SPAKMAN W. 1994. Regional scaletectonic evolution and the seismic velocity structure of the lithos-phere and upper mantle: the Mediterranean region. Journal ofGeophysical Research 99B, 12091-12108.

DOW D. B. 1977. A geological synthesis of Papua New Guinea.Bureau of Mineral Resources Bulletin 201.

ENGDAHL E. R., VAN DER HILST R. & BULAND R. P. 1998. Global tele-seismic earthquake relocation with improved travel times andprocedures for depth determination. Bulletin of the SeismologicalSociety of America 88, 722–743.

FALVEY D. A. 1978. Analysis of paleomagnetic data from the NewHebrides. Bulletin of the Australian Society of ExplorationGeophysics 9, 117–123.

FALVEY D. A. & PRITCHARD T. 1982. Preliminary paleomagnetic resultsfrom northern Papua New Guinea: evidence for large microplaterotations. In: Watson S. T. ed. Transactions of the 3rd Circum-Pacific Energy and Mineral Resources Conference, Honolulu, pp.593–600. American Association of Petroleum Geologists, Tulsa.

GRAND S. P., VAN DER HILST R. & WIDIYANTORO S. 1997. Global seismictomography: A snapshot of convection in the Earth. GSA Today 7,1–7.

GURNIS M., MORESI L. & MÜLLER R. D. 2000. Models of mantle convec-tion incorporating plate tectonics: the Australian region since theCretaceous. In: Richards M. A., Gordon R. G. & van der Hilst R. D.eds. The History and Dynamics of Global Plate Motions, pp. 211–238.American Geophysical Union Geophysical Monograph 121.

HAFKENSCHEID E., BUITER S. J., WORTEL M. J., SPAKMAN W. & BIJWAARD

H. 2001. Modelling the seismic velocity structure beneathIndonesia: a comparison with tomography. Tectonophysics 333,35–46.

HALL R. 1996. Reconstructing Cenozoic SE Asia. In: Hall R. &Blundell D. J. eds. Tectonic Evolution of SE Asia, pp. 153–184.Geological Society of London Special Publication 106.

HALL R. 1997. Cenozoic tectonics of SE Asia and Australasia. In:Howes J. V. C. & Noble R. A. eds. Proceedings of the InternationalConference on Petroleum Systems of SE Asia and Australia, pp.47–62. Indonesian Petroleum Association Jakarta.

HALL R. 1998. The plate tectonics of Cenozoic SE Asia and the distri-bution of land and sea. In: Hall R. & Holloway J. D. eds.Biogeography and Geological Evolution of SE Asia, pp. 99–131.Backhuys Publishers, Leiden.

HALL R. 2002. Cenozoic geological and plate tectonic evolution of SEAsia and the SW Pacific: computer-based reconstructions, modeland animations. Journal of Asian Earth Sciences 20, 353–434.

HAMILTON W. 1979. Tectonics of the Indonesian region. USGSProfessional Paper 1078.

HILL K. C., GREY A., FOSTER D. & BARRETT R. 1993. An alternativemodel for the Oligo-Miocene evolution of northern PNG and theSepik–Ramu Basin. In: Carman G. J. & Carman Z. eds. PetroleumExploration in Papua New Guinea, Proceedings of the 2nd PapuaNew Guinea Petroleum Convention, Port Moresby, pp. 241–259.Papua New Guinea Chamber of Mines, Port Moresby.

HILL K. C. & HALL R. 2003. Mesozoic–Cenozoic evolution ofAustralia’s New Guinea margin in a west Pacific context.Geological Society of Australia Special Publication 22 andGeological Society of America Special Paper 372, 265–289.

HILL K. C. & HEGARTY K. A. 1987. New tectonic framework for PNGand the Caroline plate: implications for the cessation of spreadingin back-arc basins. PACRIM ‘87 Congress, pp. 179–182.Australasian Institute of Mining and Metallurgy, Melbourne.

HILL K. C. & RAZA A. 1999. Arc–continent collision in Papua Guinea:constraints from fission track thermochronology. Tectonics 18,950–966.

HOUSH T. & MCMAHON T. P. 2000. Ancient isotopic characteristics ofNeogene potassic magmatism in western New Guinea (Irian Jaya,Indonesia). Lithos 50, 217–239.

IOC, IHO & BODC 1997. GEBCO-97: The 1997 Edition of the GebcoDigital Atlas. British Oceanographic Data Centre, Birkenhead.

JAQUES A. L. & ROBINSON G. P. 1977. The continent/island-arc collisionin northern Papua New Guinea. BMR Journal of AustralianGeology & Geophysics 2, 289–303.

JOHNSON R. W. 1976. Late Cainozoic volcanism and plate tectonics atthe southern margin of the Bismarck Sea, Papua New Guinea. In:Johnson R. W. ed. Volcanism in Australasia, pp. 101–116. Elsevier,Amsterdam.

JOHNSON R. W. & JAQUES A. L. 1980. Continent–arc collision and rever-sal of arc polarity: new interpretations from a critical area.Tectonophysics 63, 111–124.

JOHNSON T. & MOLNAR P. 1972. Focal mechanisms and plate tectonicsof the southwest Pacific. Journal of Geophysical Research 77,5000–5032.

JOLIVET L., HUCHON P. & RANGIN C. 1989. Tectonic setting of WesternPacific marginal basins. Tectonophysics 160, 23–47.

KENNETT B. L. N., ENGDAHL E. R. & BULAND R. 1995. Constraints onseismic velocities in the Earth from traveltimes. GeophysicalJournal International 122, 108–124.

KROENKE L. W. 1984. Cenozoic tectonic development of the SouthwestPacific. CCOP/SOPAC Technical Bulletin 6.

LEE T. Y. & LAWVER L. A. 1995. Cenozoic plate reconstruction of south-east Asia. Tectonophysics 251, 85–139.

LÉVÊQUE J-J., RIVERA L. & WITTLINGER G. 1993. On the use of thechecker-board test to assess the resolution of tomographic inver-sions. Geophysical Journal International 115, 313–318.

MCCAFFREY R. 1982. Lithospheric deformation within the Molucca seaarc–arc collision: evidence from shallow and intermediate earth-quake activity. Journal of Geophysical Research 87, 3663–3678.

MCDOWELL F. W., MCMAHON T. P., WARREN P. Q. & CLOOS M. 1996.Pliocene Cu–Au bearing igneous intrusions of the Gunung Bijih(Ertsberg) district, Irian Jaya, Indonesia. Journal of Geology 104,327–340.

MCKENZIE D. P. 1970. Temperature and potential temperaturebeneath island arcs. Tectonophysics 10, 357–366.

MEGNIN C. & ROMANOWICZ B. 2000. The three-dimensional shearvelocity structure of the mantle from inversion of body, surfaceand higher-mode waveforms. Geophysical Journal International143, 709–728.

MENKE W. 1989. Geophysical Data Analysis: Discrete Inverse Theory.Academic Press, San Diego.

MONNIER C., GIRARDEAU J., PUBELLIER M., POLVE M., PERMANA H. &BELLON H. 1999. Petrology and geochemistry of the Cyclops ophi-olites (Irian Jaya, East Indonesia): consequences for the Cenozoicevolution of the north Australian margin. Mineralogy andPetrology 65, 1–28.

MÜLLER R. D., LIM V. S. L. & ISERN A. R. 2000. Late Tertiary tectonicsubsidence on the northeast Australian passive margin: responseto dynamic stratigraphy. Marine Geology 162, 337–352.

MÜLLER R. D., ROYER J-Y. & LAWVER L. A. 1993. Revised plate motionsrelative to hotspots from combined Atlantic and Indian Oceanhotspot tracks. Geology 21, 275–278.

MUSGRAVE R. J. & FIRTH J. V. 1999. Magnitude and timing of NewHebrides Arc rotation: paleomagnetic evidence from Nendo,Solomon Islands. Journal of Geophysical Research 104,2841–2853.

OKAL E. A. & KIRBY S. H. 1998. Deep earthquakes beneath the FijiBasin, SW Pacific: Earth’s most intense deep seismicity in stag-nant slabs. Physics of the Earth and Planetary Interiors 109,25–63.

PAGE R. W. 1976. Geochronology of igneous and metamorphic rocksin the New Guinea Highlands. Bureau of Mineral ResourcesBulletin 162.

PEGLER G., DAS S. & WOODHOUSE J. H. 1995. A seismological study ofthe eastern New Guinea and the western Solomon Sea regionsand its tectonic implications. Geophysical Journal International122, 961–981.

PETTERSON M. G., NEAL C. R., MAHONEY J. J., KROENKE L. W., SAUNDERS

A. D., BABBS T. L., DUNCAN R. A., TOLIA D. & MCGRAIL B. 1997.Structure and deformation of north and central Malaita, SolomonIslands: tectonic implications for the Ontong Java PlateauSolomon arc collision, and for the fate of oceanic plateaus.Tectonophysics 283, 1–33.

PIGRAM C. J. & DAVIES H. L. 1987. Terranes and the accretion historyof the New Guinea orogen. BMR Journal of Australian Geology &Geophysics 10, 193–212.

RANGIN C., JOLIVET L. & PUBELLIER M. 1990. A simple model for the tec-tonic evolution of southeast Asia and Indonesia region for thepast 43 m.y. Bulletin de la Société géologique de France 8,889–905.


RANGIN C., SPAKMAN W., PUBELLIER M. & BIJWAARD H. 1999.Tomographic and geological constraints on subduction along theeastern Sundaland continental margin (South-East Asia). Bulletinde la Société géologique de France 170, 775–788.

RICHARDS J. P., CHAPPELL B. W. & MCCULLOCH M. T. 1990. Intraplatetype magmatism in a continent island arc collision zone: Porgeraintrusive complex, Papua New Guinea. Geology 18, 958–961.

RICHARDS M. A. & ENGEBRETSEN D. C. 1992. Large-scale mantle con-vection and the history of subduction. Nature 355, 437–440.

RIDGEWAY J. 1987. Neogene displacements in the Solomons islandarc. Tectonophysics 133, 81–93.

RITSEMA J., VAN HEIJST H-J. & WOODHOUSE J. H. 1999. Complex shearwave velocity structure imaged beneath Africa and Iceland.Science 286, 1925–1928.

ROGERSON R. & WILLIAMSON A. 1986. Age, petrology and mineraliza-tion associated with two Neogene intrusive types in the easternhighlands of Papua New Guinea. Buletin Persatuan GeologiMalaysia 20, 487–502.

SIMONS F. J., ZIELHUIS A. & VAN DER HILST R. D. 1999. The deep struc-ture of the Australian continent from surface wave tomography.Lithos 48, 17–43.

SMITH R. I. 1990. Tertiary plate tectonic setting and evolution ofPapua New Guinea. In: Carman G. J. & Carman Z. eds. PetroleumExploration in Papua New Guinea, Proceedings of the 1st PNGPetroleum Convention, Port Moresby, pp. 229–244. Papua NewGuinea Chamber of Mines, Port Moresby.

SMITH W. H. F. & SANDWELL D. T. 1997. Global seafloor topographyfrom satellite altimetry and ship depth soundings. Science 277,1956–1962.

SPAKMAN W. & BIJWAARD H. 2001. Optimization of cell parameteriza-tions for tomographic inverse problems. Pure and AppliedGeophysics 158, 1401–1423.

SPAKMAN W. & NOLET G. 1988. Imaging algorithms, accuracy and res-olution in delay time tomography. In: Vlaar N. J., Nolet N., WortelM. J. R. & Cloetingh S. A. P. L. eds. Mathematical Geophysics: ASurvey of Recent Developments in Seismology and Geodynamics,pp. 158–188. Reidel, Dordrecht.

STRUCKMEYER H. I. M., YEUNG M. & PIGRAM C. J. 1993. Mesozoic toCainozoic plate tectonic and palaeogeographic evolution of theNew Guinea region. In: Carman G. J. & Carman Z. eds. PetroleumExploration in Papua New Guinea, Proceedings of the Second PNGPetroleum Convention, Port Moresby, pp. 261–290. Papua NewGuinea Chamber of Mines, Port Moresby.

TAYLOR B. 1979. Bismarck Sea: evolution of a back-arc basin. Geology

7, 171–174.TAYLOR G. K., GASCOYNE J. & COLLEY H. 2000. Rapid rotation of Fiji:

paleomagnetic evidence and tectonic implications. Journal ofGeophysical Research 105, 5771–5781.

TREGONING P., LAMBECK K., STOLZ A., MORGAN P., MCCLUSKY S. C., VAN

DER BEEK P., MCQUEEN H., JACKSON R. J., LITTLE R. P., LAING A. &MURPHY B. 1998. Estimation of current plate motions in PapuaNew Guinea from Global Positioning System observations.Journal of Geophysical Research 103, 12181–12203.

VAN DER HILST R. 1995. Complex morphology of subducted lithos-phere in the mantle beneath the Tonga Trench. Nature 374,154–157.

VAN DER HILST R. D., WIDIYANTORO S. & ENGDAHL E. R. 1997. Evidencefor deep mantle circulation from global tomography. Nature 386,578–584.

VAN DER VOO R., SPAKMAN W. & BIJWAARD H. 1999a. Mesozoic sub-ducted slabs under Siberia. Nature 397, 246–249.

VAN DER VOO R., SPAKMAN W. & BIJWAARD H. 1999b. Tethyan sub-ducted slabs under India. Earth and Planetary Science Letters 171,7–20.

WEILER P. D. & COE R. S. 2000. Rotations in the actively collidingFinisterre Arc Terrane: paleomagnetic constraints on Plio-Pleistocene evolution of the South Bismarck microplate, north-eastern Papua New Guinea. Tectonophysics 316, 297–325.

WELLS R. E. 1989. The oceanic basalt basement of the SolomonIslands arc and its relationship to the Ontong Java Plateau:insights from Cenozoic plate motion models. In: Vedder J. G. &Bruns T. R. eds. Geology and Offshore Resources of Pacific IslandsArcs—Solomon Islands and Bougainville, Papua New GuineaRegions, pp. 7–22. Circum-Pacific Council for Energy and MineralResources Earth Science Series 12.

WIDYANTORO S. & VAN DER HILST R. D. 1997. Mantle structure beneathIndonesia inferred from high-resolution tomographic imaging.Geophysical Journal International 130, 167–182.

WORTEL M. J. R. & SPAKMAN W. 2000. Subduction and slab detach-ment in the Mediterranean–Carpathian region. Science 290,1910–1917.

YAN C. Y. & KROENKE L. W. 1993. A plate tectonic reconstruction of theSW Pacific 0-100 Ma. Proceedings of the Ocean Drilling Program,Scientific Results 130, 697–709.

ZHONG S. & GURNIS M. 1995. Mantle convection with mobile, faultedplate margins. Science 267, 838–843.

Received 15 August 2001; accepted 12 June 2002


Date post:	17-Jul-2020
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

Mantle structure and tectonic evolution of the region north and...

Documents