Download pdf - pownall_hall_2014 Seram Banda model.pdf

7212019 pownall_hall_2014 Seram Banda modelpdf

httpslidepdfcomreaderfullpownallhall2014-seram-banda-modelpdf 117

IPA14-G-305

PROCEEDINGS INDONESIAN PETROLEUM ASSOCIATIONThirty-Eighth Annual Convention amp Exhibition May 2014

NEOGENE EXTENSION ON SERAM

A NEW TECTONIC MODEL FOR THE NORTHERN BANDA ARC

Jonathan M Pownall

Robert Hall

ABSTRACT

It has long been argued that Seramrsquos Neogenetectonic evolution was dominated by thrusting andcrustal shortening due to collision of the northernBanda Arc with the Australian passive continentalmargin This collision model considers voluminous

peridotites which crop out in western and centralSeram and on Ambon to be part of a largedismembered ophiolite that was allegedly obductedfrom the Banda Sea with associated granitoidsexplained as the melting beneath the ophioliteduring emplacement

However new field geochronological andthermobarometric data question this model with

potentially important consequences for the

interpretation of sedimentary basin development both onshore and in the Seram Trough Based primarily on structural mapping of western Seramwe interpret the peridotites to representsubcontinental lithospheric mantle (SCLM) beneathSeram which was rapidly exhumed beneath low-angle detachment faults during a period of extremecrustal extension

Mantle exhumation is linked to widespread crustalanatexis and the formation of an extensivemigmatite complexmdashthe Kobipoto Complexmdash

which records ultrahigh (UHT) temperatures of~950degC at ~8 kbar pressure (25ndash30 km depth)Granitic rocks on Seram and Ambon and unusualcordierite + garnet dacites (ambonites) also exposedon Ambon all preserve evidence for theirgeneration in the Kobipoto Complex attemperatures far in excess of that predicted by theophiolite model SHRIMP UndashPb zircon dating hasrevealed that melting and UHT metamorphism ofthe Kobipoto Complex was underway by 16 Mawith granitoids produced episodically across Seramand Ambon from 16 Ma (Kobipoto Mountains

central Seram) until 35 Ma (Ambon)

Royal Holloway University of London

We propose that Seram has experienced a history ofextreme extension by detachment faulting that is

best explained by eastward rollback of the Bandaslab into the Banda Embayment since 16 Ma

INTRODUCTION

The island of Seram (Figures 1 and 2) originated as part of the Sula Spur (Klompeacute 1954) a promontoryof Australian crust located at the leading edge of theAustralian Plate (Figure 3) that collided with part ofSulawesi after subduction of the eastern Ceno-Tethys beneath the North Sulawesi-Philippines-Halmahera volcanic arc (Audley-Charles et al1972 Hamilton 1979 Bowin et al 1980 Hall1996 2002 2011 2012 Charlton 2000 Spakmanand Hall 2010) The AustralianndashSE Asian collision

which began at c 25 Ma (Hall 2011) is stillunderway

However there are many contrasting models for the Neogene tectonic evolution of the region especiallyregarding the nature of subduction around theBanda Arc (see discussions in Hall and Wilson2000 Milsom et al 2001 Pownall et al 2013)and the affinities of and relationships betweendifferent tectonic units comprising Seram and thesurrounding islands (eg Audley-Charles et al1979 Tjokrosapoetro and Budhitrisna 1982

Linthout et al 1989 Kemp and Mogg 1992Linthout and Helmers 1994 Hill 2005 Pownall etal 2013)

Seram has commonly been interpreted as a fold-and-thrust belt resulting from arc-continent collision(eg Audley-Charles et al 1979 Bowin 1980Jongsma et al 1989 Charlton 2004 Hill 2005)Audley-Charles et al (1979) proposed atectonostratigraphic scheme for Seram that involvedallochthonous units of Asian affinity having beenthrust over Australian para-autochthonous

sequences Ultramafic rocks which crop out overwestern and central Seram and on Ambon(see Figure 2) were originally ascribed to an



allochthonous thrust sheet and were thereforeinterpreted to comprise part of an extensiveophiolite that had been thrust from the Banda SeaWork by Linthout et al (1989) Linthout andHelmers (1994) Sopaheluwakan (1994) andMonnier et al (2003) built further on this lsquoSeramOphiolitersquo concept arguing that its obduction overthe metamorphic lsquobasementrsquo of the TehoruFormation had induced sub-ophiolitemetamorphism and the generation of cordierite-

bearing granites Thus the concept of Seramrepresenting a collisional complex dominated bylong-distance thrusting was developed Indeedfold-and-thrust-belts have been described fornorthern and eastern Seram (Zillman and Paten1975 Tjokrosapoetro and Budhitrisna 1982 Kempand Mogg 1992 Hill 2005) and the Seram Trough

(Pairault et al 2003ab Teas et al 2004) but goodevidence for this collisional tectonic style inwestern Seram is lacking

Hall and Wilson (2000) questioned the ophioliteconcept for Seram they suggested that theultramafic rocks may have instead been derivedfrom the subcontinental mantle and that extensionmay have been more important in the tectonicevolution of western Seram than collision andthrusting Spakman and Hall (2010) later providedfurther evidence for an extension-dominated history

for the island in the context of plate reconstructionsutilising new tomographic models for the Bandaregion More recently Pownall et al (2013) mappedout low-angle detachment faults in western Seram(Figure 4) that were shown to have facilitated theexhumation of subcontinental lithospheric mantle

providing evidence for significant lithosphericextension having occurred across the islandFurthermore Pownall et al (2013) found noevidence for significant thrusting in western Seram

THE GEOLOGY OF SERAM

Field relations on Seram and Ambon have previously been described by several authors (Valk1945 Germeraad 1946 van der Sluis 1950Audley-Charles et al 1979 Tjokrosapoetro andBudhitrisna 1982) The summary here is based theobservations by these authors and new fieldobservations by Pownall et al (2013)

Western Seram mainly comprises lower-greenschistto upper-amphibolite facies phyllites schists andgneisses of the Tehoru Formation Garnet mica-schists are widespread which are often intercalatedwith amphibolites Scarce kyanite-grade schists(containing also staurolite and garnet) represent the

highest grade part of the complex however largeareas are of a low metamorphic grade and preserveoriginal sedimentary structures (Audley-Charles etal 1979 Tjokrosapoetro and Budhitrisna 1982Linthout et al 1989 Pownall et al 2013)

The Tehoru Formation is associated with complexesof peridotites and cordierite + garnet-bearinggranitic rocks (see next section) The termlsquocordierite granitersquo has long been used in theliterature to describe the granitic lithologies (egPriem et al 1978) however these rocks containabundant mm-scale (and more rarely cm-scale)schlieren of granulite-facies mineral assemblages(sillimanite + spinel + cordierite) that demonstratethe rocks are part of a migmatite complex and havenot migrated from their source region to form

separate granitoid plutons (Pownall et al 20132014)

These leucosome-rich migmatites (diatexites) andassociated peridotites comprise part of the KobipotoComplex which has been exhumed beneath low-angle lithospheric detachment faults mapped on theKaibobo Peninsula (Figure 4 see Pownall et al2013) in response to major lithospheric extension

Gneisses adjacent to the Kobipoto Complex peridotites that crop out on the Kaibobo Peninsula

previously interpreted by Linthout et al (1989) andLinthout and Helmers (1994) to comprise ametamorphic sole certainly have been overprinted

by sillimanite-grade metamorphism and intensiveshearing However we interpret these metamorphicrocks to form the hanging wall above the exhumedKobipoto Complex not the footwall As outlined byPownall et al (2013) the Taunusa Complex is anoverprint of Tehoru Formation rocks in response tohigh-temperature exhumation of the underlyingKobipoto Complex

In central Seram the dominantly left-lateral KawaShear Zone (KSZ) is oriented roughly parallel to thestrike of the detachment faults in western Seram(Figure 5) and there is evidence from fault-boundedlenses of serpentinised peridotites that areincorporated within the KSZ that these faults mayhave re-activated similar lithospheric-scaledetachments (Pownall et al 2013) The KobipotoMountains are a left-lateral pop-up structure thatlinks with the KSZ and the Kobipoto Complexrocks that crop out in the mountains have evidently

been exhumed by left-lateral transpression

The 3 km high Manusela Mountains bounded alongtheir southern margin by the major strike-slip Kawa



Fault (Figure 2) are formed from the UpperTriassic Manusela Mountain Limestones (Martini etal 2004) Between them and the KobipotoMountains to their north are outcrops of theTriassic-Jurassic Kanikeh Formation deep-waterclays siltstones and micaceous carbonaceoussandstones (Tjokrosapoetro and Budhitrisna 1982)which are separated from the Tehoru Formation byslates and marbles of the Saku Formation

In northern Seram the Mid-Upper Jurassic toMiocene Nief Beds are exposed comprising a

passive margin sequence of deep water carbonatesand turbidites (Kemp and Mogg 1992) The NiefBeds are highly imbricated (Kemp and Mogg 1992Hill 2005) and are interpreted to rest on the KolaShales suggested by Hill (2005) to be cut by a

major basal deacutecollement beneath which the BirdrsquosHead may have underthrust northern Seram

Much of northern and eastern Seram is covered byextensive meacutelanges of the Salas Block Clay(interpreted by Audley-Charles et al (1979)Tjokrosapoetro and Budhitrisna (1982) andOrsquoSullivan et al (1985) as an olistostrome) andPlio-Pleistocene sedimentary deposits of the WahaiBeds and Fufa Formation (Zillman and Paten1975) Uplifted Quaternary reefs flank the island

UHT METAMORPHISM IN RESPONSE TOMANTLE EXHUMATION

The Kobipoto Complex consists of peridotites and agranulite-facies migmatite complex comprisinghighly residual aluminous granulites(Pownall et al 2014) stromatic metatexites andmelt-rich granitic diatexites (described previouslyas lsquocordierite granitesrsquo) All rocks in the migmatitecomplex are characterised by aluminous sillimanite+ spinel (plusmn corundum) bearing assemblages whichrepresent the residual component the cordierite

granites are peppered with schlieren of sillimanite +spinel as well as schollen of metatexite Thecordierite granites in the Kobipoto Mountains areequivalent to those in western Seram (eg on theKaibobo Peninsula) and on Ambon which aresimilarly associated with peridotites

In the Kobipoto Mountains the peridotites aremainly spinel lherzolites (Pownall et al 2013)although plagioclase lherzolites have been reported

by Monnier et al (2003) from Ambon and from theKaibobo Peninsula western Seram

It has been suggested that the peridotites of easternSeram are ophiolites which were emplaced at low

temperature (Charlton 2004) but this is ruled out by the extremely high temperatures recorded bymetamorphic rocks at their contacts The peridotitesare also unlike the depleted harzburgites typicallyfound in ophiolite complexes and are dominantlyfertile lherzolites without products of mantle partialmelting including basaltic dyke complexes andlarge-volume gabbros A subcontinentallithospheric mantle origin (SCLM) is much morelikely

Widespread partial melting was accompanied byultrahigh-temperature (UHT ge 900degC)metamorphism of the residuum peak PndashTconditions experienced by spinel + quartz andsapphirine-bearing granulites are inferred to be~950degC at a pressure of ~8 kbar based on phase

equilibria modelling (Pownall et al 2014) Thisdemonstrates the existence of an extremely high(apparent) geothermal gradient of sim120degC kbarndash1(~35degC kmndash1) Diatexites of the KobipotoComplex are always found in direct associationwith voluminous spinel lherzolites (whether they bein the Kobipoto Mountains the Kaibobo Peninsulathe Wallace Mountains the Hoamoal Peninsula oron Ambon) these hot mantle rocks are interpretedto be the driver of UHT metamorphism and meltingappears to have been short-lived event Assumingthat the granulites were close to being in thermalequilibrium with their heat source in order to haveexperienced metamorphism at 950degC theneighbouring lherzolites must have been close totypical subcontinental lithospheric mantle (SCLM)temperatures of sim1300degC

Most authors who supported the idea of the lsquoSeramOphiolitersquo applied the concept exclusively towestern Seram and Ambon and never attempted toexplain isolated outcrops of peridotites exposed inthe Kobipoto Mountains or in eastern Seram around

Gunung Langget (eg Linthout et al 1989 1996Sopaheluwakan 1994 Linthout and Helmers 1994Monnier et al 2003) If further proof were neededthat the lherzolites are not part of an obductedophiolite then their juxtaposition against theKobipoto Complex without cooling below uppermantle temperatures is the evidence In theKobipoto Mountains if the lherzolites had beenobducted from the Banda Sea to have generated thelsquocordierite granitesrsquo by anatexis within ametamorphic sole (cf Linthout et al 1989Linthout and Helmers 1994) they would have hadto have been thrust for gt 50 km over the TehoruFormation and the cold carbonates of the ManuselaFormation without any significant cooling



Sensitive High Resolution Ion Microprobe(SHRIMP) U-Pb dating of zircons from KobipotoComplex granulites sampled from the KobipotoMountains has revealed a ~16 Ma age for UHTmetamorphism (Pownall et al 2014) Importantlythis age is significantly younger than the Permian(or Precambrian) age that has often been assumedfor the complex (see Figure 2 and references inPownall et al 2013) Furthermore the presence ofTriassic detrital zircon grains demonstrates a

protolith age that is also younger than these earlierestimates Cordierite diatexites present on theKaibobo Peninsula have been dated at ~55 Ma(40Ar39Ar biotite age Linthout et al 1996) andvoluminous cordierite diatexites and related granites

present on Ambon are younger still at ~33 Ma(RbndashSr age Priem et al 1978) Cordierite + garnet

dacites (or lsquoambonitesrsquo) which are in part sourcedfrom the Kobipoto Complex were erupted onAmbon shortly before ~24 Ma (KndashAr age for thegroundmass Honthaas et al 1999)

INTERPRETATION OF THE SERAM

TROUGH

Hamilton (1979) presented the first seismic profilesacross the Seram Trough He interpreted the zone asthe surface trace of southward subduction of theBirdrsquos Head describing Seram as an accretionary

prism of Australian-derived terranes chamfered offthe top of the subducting slab ndash a model adopted byseveral later authors (Bowin et al 1980Tjokrosapoetro and Budhitrisna 1982 Hartono andTjokrosapoetro 1984 OrsquoSullivan et al 1985) Itwas also later suggested that the Seram Trough is

primarily a zone of strike-slip faulting associatedwith the Tarera-Aiduna and Sorong fault zones(Linthout et al 1997 Milsom et al 1983)

However in recent years these interpretations have been challenged both by greatly improved seismic

profiles (Pairault et al 2003ab Teas et al 2009Darman and Reemst 2012) and high-resolutionmultibeam bathymetry (Teas et al 2009) whichhave led to the interpretation of the Seram Troughas a foredeep basin flanking the Seram fold-and-thrust belt Pairault et al (2003ab) described theSeram Trough as a very young zone of thrusting inthe Australian continental margin resulting fromsouthward underthrusting of the Birdrsquos Head

beneath Seram A major deacutecollement is clearlyimaged on seismic sections shot perpendicular tothe trough along which a series of steep reflectorsinterpreted to be reverse faults within the imbricatedTriassic to Miocene Nief Beds are shown toterminate A Permian to Pliocene sedimentary

succession overlying Australian lsquobasementrsquo isshown to dip southwards into the Seram Troughwhere it is seemingly overthrust by lsquoSeramrsquo bedsriding the deacutecollement A cross-section published

by Hill (2005) depicts this interpretation with thedetachment fault having propagated through theKola Shales as previously mentioned AlthoughPairault et al (2003b) strongly favoured a foredeepmodel for the Seram Trough they concluded asubduction model cannot be ruled out entirely dueto the poor quality of seismic data at depth

Teas et al (2009) presented high-resolution bathymetry of the Seram Sea which clearly showsan EndashW-trending offshore fold-and-thrust beltdeveloped immediately south of the majordeacutecollement In agreement with Pairault et al

(2003ab) the authors concluded that thedeformation is due to convergence of the BirdrsquosHead with Seram and that no large-scale strike-slipfaulting is present However in the eastern offshorefold-and-thrust belt Teas et al (2009) identifiedmany localised EndashW oriented sinistral shear zoneswhich they attributed to a possible diffusetermination of the left-lateral Tarera-Aiduna FaultZone which runs south of the Birdrsquos Head The fold

belt narrows at the western end (Teas et al 2009Darman and Reemst 2012) where deformation isdominated by detachment faulting and block

rotation rather than the thin-skinned deformation inthe east Teas et al (2009) suggested that theevolution of the fold belt was dictated by therelative movement between the Birdrsquos Head andAustralia calculated by Stevens et al (2002) to beoccurring along an azimuth of 252deg The SeramTrough has been infilled with unconformablehorizontally-bedded sediments during thePleistocene (Darman and Reemst 2012)

BANDA SUBDUCTION

The Wadati-Benioff Zone for the subducted slabaround the Banda Arc as inferred from seismicity(eg Cardwell and Isacks 1978 McCaffrey 19881989 Das 2004) and tomographic models (egWidiyantoro and van der Hilst 1997 Spakman andHall 2010) is shown to have acquired a highlyconcave spoon-shaped or prow-shaped morphologywithin the 180deg-curved arc Early models proposedthat the Banda Arc formed by the bending of anoriginally straight subduction zone through 180deg(Audley-Charles et al 1972 Katili 1975) but mostauthors now agree that slab rollback has played arole in the formation of the tightly-curved arc (egHall 1996 Charlton 2000 Hall and Wilson 2000Harris 2006 Spakman and Hall 2010) although



opinions are still divided over whether one or twoslabs are involved Those authors that have invokedtwo separate subduction zones (eg Cardwell andIsacks 1978 McCaffrey 1988 1989 Das 2004Špičak et al 2013) have often done so in order toexplain a lineament of high-seismicity across thesouthern margin of the arc that is interpreted to linkwith the Tarera-Aiduna Fault Zone to the south ofthe Birdrsquos Head (West Papua) it has been arguedthat this zone represents a suture between twoseparate slabs with opposing dip directionsHowever a single-slab model arguably is bettersupported in the context of plate reconstructions forthe region (eg Hall 2011 2012) and the high-seismicity zone can be explained by a tear in asingle albeit highly contorted slab

The interpretation of seismicity data (Das 2004)combined with the Spakman and Hall (2010)tomographic model were used to construct the 3Ddiagram of the Banda Arc (Figure 6) The slabappears to be detached from the northern margin ata point between Buru and Seram and further westit has fully collapsed into the mantle to form a flat-subducted surface at 660 km depth The slab is thusinterpreted to have torn from the northern marginsupported by the presence of an aseismic zoneduring eastward rollback of the arc into a pre-existing D-shaped oceanic embayment (Spakman

and Hall 2010)

Present-day seismicity

Recent (1976-present) shallow-level (lt 80 kmdepth) earthquakes recorded for Seram and thenorthern Banda Arc catalogued in the GlobalCentroid-Moment-Tensor (CMT) database(Ekstroumlm 2006) were plotted using the UNAVCO(2012) Integrated Data Viewer (IDV) ndash see Figure7 The northern coast of Seram and the southernextent of the Seram Sea is currently the most

seismically-active region with earthquakestypically recording high-angle thrust faults withcomponents of right-lateral transpression In centralSeram several focal mechanisms can be interpretedas displaying strike-slip motions on fault planesstriking ESEndashWNW roughly parallel to the KawaShear Zone and the detachment faults in westernSeram A single earthquake in the ManuselaMountains recorded left-lateral strike-slip motion

but two earthquakes further west indicate right-lateral strike-slip A number of focal mechanismsshowing ENEndashWSW-striking extensional faulting

perhaps correlate with the youngest set of normalfaults interpreted in Figure 5 (shown in blue) Aseries of earthquakes in the Banda Sea indicate

purely north- or south-directed thrusting Thecluster of earthquakes around the eastern coast ofBuru mark the site of the largest shallow-depthearthquake in the datasetmdashan Mw 67 left-lateralstrike-slip earthquake just 13 km below the surface

It can be seen in Figure 7 that present-daydeformation across northern and eastern Seram is inresponse to maximum principal stress directionsthat are oriented roughly perpendicular to the SeramTrough with the pressure axes (shown in white)rotating clockwise to a more EndashW-alignedorientation further to the east This shortening isresponsible for the formation of the fold-and-thrust

belt in the southern Seram Trough although this isarguably a very recent feature

DISCUSSION A NEW TECTONIC MODEL

FOR THE NORTHERN BANDA ARC

As outlined by Pownall et al (2013 2014) a majorextensional episode is interpreted to have thinnedthe lithosphere and to have exhumed thesubcontinental lithospheric mantle beneath the

protolith of the Kobipoto Complex at temperaturessufficient to have induced UHT (gt 900degC)metamorphism From the phase equilibriamodelling described by Pownall et al (2014) a

depth of 25ndash30 km (based on a pressure of sim8 kbar)can be considered as the depth of heating to UHTconditions which is consistent also with theequilibration of the lherzolites in the Cr-spinelstability field at a pressure of 7ndash18 kbar (Gasparik1984) As demonstrated on the Kaibobo Peninsulamantle exhumation was in part facilitated by low-angle detachment faults however at deeperstructural levels upon transition into the ductileregime it is also plausible that exhumation andextension by pure shear may have also operatedAlternatively a combination of brittle normal shear

and ductile pure shear (cf Lister and Davis 1989)may best explain crustal extension in this core-complex-type scenario However the very highsim35degC kmndash1 apparent geothermal gradient and theexhumation of hot SCLM demonstrates a heat flowwhich surpasses that of typical lsquohot crustrsquo corecomplex models (cf Whitney et al 2013)

The initial mechanism for Kobipoto Complexexhumation in the Kobipoto Mountain is unclear

because the rocks have subsequently been exhumed by transpression and incorporated into the KobipotoMountains pop-up structure Field relations inWestern Seram may therefore more clearlydemonstrate the style of extensional tectonics In



Figure 8 the extension is depicted primarily as slipalong low-angle detachment faults (as mapped onthe Kaibobo Peninsula see Pownall et al 2013)that flatten and disappear into the weak melt-richductile zone which in turn has been exhumed bythe next detachment fault in the sequence (shownmore clearly in the 2D schematic cross-section ofFigure 9) The net effect would have been similar toa boudinaging of the crust as it stretched withindividual blocks (or giant boudins) having rotated

backwards into a flatter (thinner in cross-section)orientation This mechanism permits both hotSCLM and lower-crustal granulite-faciesmigmatites to be exhumed beneath the detachmentfaults as the blocks rotate and also accounts forhow Tehoru Formation rocks in the hanging wall(that are rotating downwards onto the upward-

rotating footwall of the next block) have beenoverprinted by localised heating and shearing by theKaibobo Complex granulites and lherzolites

A further implication of this model is that the protoliths of the Kobipoto Complex migmatites andthe Taunusa Complex gneisses are very likely in

both instances the Tehoru Formation therefore allmetamorphic complexes on Seram may share acommon origin Furthermore following this modelTaunusa Complex gneisses located in the hangingwall are equivalent to upper-amphibolite-facies

rocks that presumably separate the KobipotoComplex migmatites from lsquonormalrsquo TaunusaFormation schists in the respective footwall

That only localised anatexis is observed in thehanging wall demonstrates that the lherzolites andmigmatites must have been cooled by severalhundred degrees as they were exhumed beneath thedetachment faults

Spakman and Hall (2010) proposed that slabrollback into the Banda Embayment which initiated

at sim15 Ma caused the Sula Spur to extend andfragment as the slab rolled back eastwards (Figure10) Slab rollback is certainly the most probablemechanism to have driven this extensional episodeon Seram and the proposed initiation time of thisrollback by Spakman and Hall (2010) and Hall(2011 2012) coincides fairly well with the age ofUHT metamorphism Perhaps due to therequirement of having to extend the crust inadvance of UHT conditions being produced(Schenker et al 2012) that the rollback initiationmay have occurred at a few Myr earlier than

previously suggested and also that Seram and Burumay have originated further west bordering the SEarm of Sulawesi before 16 Ma

The strike-slip Kawa Shear Zone (KSZ) hasevidently operated most recently in the sinistralsense as is apparent from viewing the digitalelevation models (DEMs) of the island (see Figure5) However as described by Pownall et al (2013)

both left- and right-lateral kinematic indicators are present when investigated in the field indicatingthat the history of the KSZ is very likely morecomplex The incorporation of thin lenses and

boudins of serpentinised spinel peridotites withinthe wide fault core of the main Kawa Faultdemonstrates it is a major lithospheric-scale featureFurthermore the main section of the KSZ that runsthe length of the Wai Kawa into Teluk Taluti (nearTehoru Village) has an identical strike (120degndash300deg)to the Kaibobo Detachment suggesting that

possibility that the KSZ could have formed through

reactivation of similarly-oriented detachment faultsin central Seram By this mechanism peridotitesthat were initially exhumed along the base of thedetachment could then be incorporated into theshear zone after it reactivated with a dominantlystrike-slip motion Repeated cycles of strike-slipfaulting could then have further exhumed the

peridotites by lsquoshufflingrsquo them upwards from depth(cf Lister and Forster 2009)

This proposal for the reactivation of low-angledetachment faults as lithospheric strike-slip features

is illustrated in Figure 8 Feasibly the KSZ mayhave operated with both dip-slip and strike-slipcomponents to its motion during its early historyLinthout et al (1991) suggested that the KSZ actedas a right-lateral antithetic shear that facilitated ananticlockwise rotation of the island eastwards intothe northern limb of the Banda Arc however thisinterpretation would seem unlikely given theevidence presented here for a dominantly left-lateralhistory

Furthermore this interpretation does not make

sense in the context of rollback-driven extensionAs previously stated the 16 Ma age for UHTmetamorphism (suggesting that extension musthave initiated gt 16Ma) would locate Seram againstthe SE Arm of Sulawesi at this time not along the

NE margin of the Banda Embayment as implied byLinthout et al (1991 1997) Eastward slab rollbackis consistent with it having driven EndashW-strikingleft-lateral shear as would have been required todrag Seram towards to its present location furthereast This migration may or may not have beenaccompanied by an anticlockwise rotation throughabout 90deg as proposed by Haile (1978) in light ofhis reconnaissance palaeomagnetic survey althoughif Seram had undergone significant rotation an



anticlockwise direction would be probably mostlikely as proposed by Lonergan and White (1997)to explain the development of the Betic-Rif Arc inthe western Mediterranean

As demonstrated by modern-day seismicity (see previous section and Figure 7) the KSZ still showssigns of both left- and right-lateral strike-slipfaulting however most of the seismicity isconcentrated around the northern Seram coast inresponse to EndashW-striking thrust faulting As shown

by high-resolution bathymetry for the Seram Sea(Teas et al 2009) this fold-and-thrust-belt is alsoaffected by left-lateral transpression Thegeometries and orientation of transpressional andtranstensional structures in central Seram (ie theKobipoto Mountains and lsquoCentral Valleyrsquo) are

similar to those imaged offshore Thus active left-lateral shear across this portion of the northernBanda Arc can be interpreted However the fold-and-thrust belt off northern Seram is possibly veryyoung and not representative of the dominantlyextensional tectonic history that has affected theisland over the last 16 Ma

CONCLUSIONS

Our observations do not support models that invokecollisions of arcs with the Australian margin

emplacement of hot ophiolites nor thrusting that began in the Late Miocene At the present dayseismicity and structures on Seram and in the SeramTrough do indicate north-directed thrusting but wesuggest the thrusting began only during thePliocene Seram preserves a record of extremecrustal extension which continued during the

Neogene until about 2 million years ago

Subduction rollback into the Banda embayment isindicated by mantle structure and is supported bythe presence of young UHT rocks exhumed on

Seram intrusive peridotite-granite contacts and theexhumation of subcontinental mantle lithosphere

East Indonesia has been widely interpreted formany years in terms of collision and slicing ofcontinental fragments from New Guinea Theseinterpretations need to be reconsidered Seram is aspectacular example of prolonged extension driven

by subduction which is increasingly recognised to be of major importance in eastern Indonesia

The importance of extension cannot be over-emphasised but at present it is incorporated in fewstructural and tectonic models in a significant wayIt was this extension that fragmented the former

Sula Spur created the deep oceanic basins of theBanda Sea and left extended remnants ofcontinental crust throughout the region fromSulawesi to the Banda forearc islands Extension ismanifested on land in many areas by unusual andvery young igneous and metamorphic rocks butoffshore by deep sediment-filled basins which areof major interest to the hydrocarbon industry Ourknowledge of the basin-forming mechanisms ratesof vertical motions and heatflow implications is stilllimited and more work is required to betterunderstand the implications of regional extension

Seram also illustrates one other important feature ofIndonesian geology the need for many more newfield-based investigations The unusual UHT rocksof Seram are known elsewhere in the world (with

one or maybe two known exceptions) only fromPrecambrianndashPalaeozoic terranes and consequentlymechanisms for their formation are the subject ofcontroversy and speculation Seram offers anexample of a field study in an understudied areayielding surprising and exciting results which willcontribute to new thinking about tectonic models oforogenic belts with implications for basins andhydrocarbons There are many other parts ofIndonesia where new field-based studies offersimilar opportunities

ACKNOWLEDGEMENTS

We thank our Indonesian collaborators YasintoPriastomo and Ramadhan Adhitama (InstitutTeknologi Bandung) and local guides for fieldassistance Ian Watkinson is also acknowledged formany useful discussions both in the field and in theoffice This research has been funded by the SEAsia Research Group Royal Holloway Universityof London

REFERENCES

Audley-Charles MG Carter DJ and MilsomJS 1972 Tectonic Development of EasternIndonesia in Relation to Gondwanaland Dispersal

Nature Physical Science 239 35ndash39

Audley-Charles MG Carter DJ Barber AJ Norvick MS and Tjokrosapoetro S 1979Reinterpretation of the geology of Seramimplications for the Banda Arcs and northernAustralia Journal of the Geological Society 136547ndash566

Bowin C Purdy GM Johnston C Shor GLawver L Hartono HMS and Jezek P 1980



Arc-continent collision in Banda Sea region AAPGBulletin 64 868ndash915

Cardwell RK and Isacks BL 1978 Geometryof the subducted lithosphere beneath the Banda Seain eastern Indonesia from seismicity and fault planesolutions Journal of Geophysical Research 832825ndash2838

Charlton TR 2000 Tertiary evolution of theEastern Indonesia Collision Complex Journal ofAsian Earth Sciences 18 603ndash631

Charlton TR 2004 The petroleum potential ofinversion anticlines in the Banda Arc AAPGBulletin 88 565ndash585

Darman H and Reemst P 2012 Seismicexpression of geological features in Seram SeaSeram Trough Misool-Onin Ridge andSedimentary Basin Berita Sedimentologi 23 28ndash 34

Das S 2004 Seismicity gaps and the shape of theseismic zone in the Banda Sea region fromrelocated hypocenters Journal of GeophysicalResearch 109 B12303

Ekstroumlm G 2006 Global CMT Web Page

httpwwwglobalcmtorg

Gasparik T 1984 Experimental study ofsubsolidus phase relations and mixing properties of

pyroxene in the system CaO-Al2O3-SiO2Geochimica et Cosmochimica Acta 48 2537ndash2545

Germeraad J H 1946 Geology of central Seranin Rutten L and Hotz W eds Geological

petrographical and palaeontological results ofexplorations carried out from September 1917 tillJune 1919 in the island of Ceram De Bussy

Amsterdam 135 pp

Haile NS 1978 Palaeomagnetic evidence for therotation of Seram Indonesia Journal of Physics ofthe Earth 26 S191ndash198

Hall R 1996 Reconstructing Cenozoic SE Asia inHall R and Blundell D eds Tectonic Evolutionof Southeast Asia Geological Society of LondonSpecial Publication 106 153ndash184

Hall R 2002 Cenozoic geological and platetectonic evolution of SE Asia and the SW Pacificcomputer-based reconstructions model and

animations Journal of Asian Earth Sciences 20353ndash431

Hall R 2011 Australia-SE Asia collision platetectonics and crustal flow in Hall R CottamMA and Wilson MEJ eds The SE AsianGateway History and Tectonics of the Australia-Asia Collision Geological Society of LondonSpecial Publication 355 75ndash109

Hall R 2012 Late Jurassic-Cenozoicreconstructions of the Indonesian region and theIndian Ocean Tectonophysics 570ndash571 1ndash41

Hall R and Wilson MEJ 2000 Neogene suturesin eastern Indonesia Journal of Asian EarthSciences 18 781ndash808

Hamilton W 1979 Tectonics of the IndonesianRegion USGS Professional Paper 1078 345 pp

Harris R 2006 Rise and fall of the Eastern GreatIndonesian arc recorded by the assembly dispersionand accretion of the Banda Terrane TimorGondwana Research 10 207ndash231

Hartono HMS and Tjokrosapoetro S 1984Preliminary account and reconstruction ofIndonesian terranes Proceedings of Indonesian

Petroleum Association 13th Annual Convention185ndash226

Hill KC 2005 Tectonics and Regional Structureof Seram and the Banda Arc Proceedings ofIndonesian Petroleum Association 30th AnnualConvention 559ndash578

Honthaas C Maury R Priadi B Bellon H andCotten J 1999 The PliondashQuaternary Ambon arcEastern Indonesia Tectonophysics 301 261ndash281

Jongsma D Woodside JM Huson W SuparkaS and Kadarisman D 1989 Geophysics andtentative late Cenozoic seismic stratigraphy of theBanda Arc-Australian continent collision zonealong three transects Netherlands Journal of SeaResearch 24 205-229

Katili JA 1989 Review of past and presentgeotectonic concepts of Eastern Indonesia

Netherlands Journal of Sea Research 24 103ndash129

Kemp G and Mogg W 1992 A reappraisal ofthe geology tectonics and prospectivity of SeramIsland Eastern Indonesia Proceedings of



Indonesian Petroleum Association 21st AnnualConvention 521ndash552

Klompeacute THF 1954 The structural importance ofthe Sula Spur (Indonesia) Indonesian Journal of

Natural Sciences 110 21ndash40

Linthout K Helmers H Sopaheluwakan J and Nila E S 1989 Metamorphic complexes in Buruand Seram northern Banda Arc NetherlandsJournal of Sea Research 24 345ndash356

Linthout K Helmers H and Andriessen P1991 Dextral strike-slip in Central Seram and 3-45Ma RbSr ages in pre-Triassic metamorphics relatedto early Pliocene counterclockwise rotation of theBuru-Seram microplate (E Indonesia) Journal of

Southeast Asian Earth Sciences 6 335ndash342

Linthout K and Helmers H 1994 Plioceneobducted rotated and migrated ultramafic rocks andobduction-induced anatectic granite SW Seram andAmbon Eastern Indonesia Journal of SE AsianEarth Sciences 9 95ndash109

Linthout K Helmers H Wijbrans JR and vanWees JDAM 1996 40Ar39Ar constraints onobduction of the Seram ultramafic complexconsequences for the evolution of the southern

Banda Sea in Hall R and Blundell D edsTectonic Evolution of Southeast Asia GeologicalSociety of London Special Publication 106 455ndash 464

Linthout K Helmers H and Sopaheluwakan J1997 Late Miocene obduction and microplatemigration around the southern Banda Sea and theclosure of the Indonesian Seaway Tectonophysics281 17ndash30

Lister GS and Davis GA 1989 The origin of

metamorphic core complexes and detachment faultsformed during Tertiary continental extension in thenorthern Colorado River region USA Journal ofStructural Geology 11 65ndash94

Lister G and Forster M 2009 Tectonic modeswitches and the nature of orogenesis Lithos 113274ndash291

Lonergan L and White N 1997 Origin of theBetic-Rif mountain belt Tectonics 16 504ndash522

Martini R Zaninetti L Lathuilliegravere B Cirilli SCorneacutee JJ and Villeneuve M 2004 UpperTriassic carbonate deposits of Seram (Indonesia)

palaeogeographic and geodynamic implicationsPalaeogeography PalaeoclimatologyPalaeoecology 206 75ndash102

McCaffrey R 1988 Active tectonics of the easternSunda and Banda arcs Journal of GeophysicalResearch 93 163ndash182

McCaffrey R 1989 Seismological constraints andspeculations on Banda Arc tectonics NetherlandsJournal of Sea Research 24 141ndash152

Milsom J Audley-Charles MG Barber AJ andCarter DJ 1983 Geological paradoxes of theeastern Indonesia collision zone in Hilde WCand Uyeda S eds Geodynamics of the WesternPacific-Indonesian Region AGU Geodynamic

Series 401ndash411

Milsom J 2001 Subduction in eastern Indonesiahow many slabs Tectonophysics 338 167ndash178

Monnier C Girardeau J Permana H RehaultJ-P Bellon H and Cotten J 2003 Dynamicsand age of formation of the Seram-Ambonophiolites (Central Indonesia) Bulletin de laSocieacuteteacute Geacuteologique de France 174 529ndash543

OrsquoSullivan T Pegum D and Tarigan J 1985

Seram oil search past discoveries and future oil potential Proceedings of Indonesian PetroleumAssociation 14th Annual Convention 3ndash20

Pairault AA Hall R and Elders CF 2003aTectonic evolution of the Seram Trough IndonesiaProceedings of Indonesian Petroleum Association29th Annual Convention 355ndash370

Pairault AA Hall R and Elders CF 2003bStructural styles and tectonic evolution of the SeramTrough Indonesia Marine and Petroleum Geology

20 1141ndash1160

Pownall JM Hall R and Watkinson IM 2013Extreme extension across Seram and Amboneastern Indonesia evidence for Banda slab rollbackSolid Earth 4 277ndash314

Pownall JM Hall R Armstrong RA andForster MA 2014 Earthrsquos youngest knownultrahigh-temperature granulites discovered onSeram eastern Indonesia Geology 42 doi101130G352301

Priem HNA Andriessen PAM Boelrijk NAIM Hebeda EH Hutchinson CS



Verdurmen EAT and Versschure RH 1978Isotopic evidence for a middle to late Pliocene ageof the cordierite granite on Ambon IndonesiaGeologie en Mijnbouw 57 441ndash443

Schenker FL Gerya T and Burg J-P 2012Bimodal behavior of extended continentallithosphere Modeling insight and application tothermal history of migmatitic core complexesTectonophysics 579 88ndash103

Sopaheluwakan J 1994 Tectonic evolution of theBanda Arc East Indonesia Southern Tethyan crustobduction metamorphism and fragmentation ofEastern Gondwanaland Prosiding TridasawarsaPuslitbang Geoteknologi LIPI 157ndash162

Spakman W and Hall R 2010 Surfacedeformation and slab-mantle interaction duringBanda arc subduction rollback Nature Geoscience3 562ndash566

Špičaacutek A Matě jkovaacute R and Vaněk J 2013Seismic response to recent tectonic processes in theBanda Arc region Journal of Asian Earth Sciences64 1ndash13

Stevens CW McCaffrey R Bock Y GenrichJF Pubellier M and Subarya C 2002 Evidence

for Block Rotations and Basal Shear in the WorldrsquosFastest Slipping Continental Shear Zone in NW

New Guinea in Stein S and Freymueller J edsPlate Boundary Zones AGU Geodynamic Series30 87ndash99

Teas PA Decker J Orange D and Baillie P New insight into the structure and tectonics of theSeram Trough from SEASTEEP high resolution

bathymetry Proceedings of Indonesian Petroleum

Association 33rd Annual Convention IPA09-G-091 1-18

Tjokrosapoetro S and Budhitrisna T 1982Geology and tectonics of the northern Banda ArcBulletin of the Indonesian Geological Research andDevelopment Centre 6 1ndash17

UNAVCO 2012 UNAVCO IDV Web Page(accessed2012) httpfacilityunavcoorgsoftwareidv

Valk W 1945 Contributions to the geology ofWest Seran in Rutten L and Hotz W edsGeological petrographical and palaeontologicalresults of explorations carried out from September1917 till June 1919 in the island of Ceram De

Bussy Amsterdam 104 pp

van der Sluis J P 1950 Geology of East Seran inRutten L and Hotz W eds Geological

petrographical and palaeontological results ofexplorations carried out from September 1917 tillJune 1919 in the island of Ceram De BussyAmsterdam 67 pp

Whitney DL Teyssier C Rey P and BuckWR 2013 Continental and oceanic corecomplexes Geological Society of America Bulletin

125 273ndash298

Widiyantoro S and van der Hilst R 1997 Mantlestructure beneath Indonesia inferred from high-resolution tomographic imaging GeophysicalJournal International 130 167ndash182

Zillman NJ and Paten RJ 1975 Exploration and petroleum prospects Bula Basin SeramProceedings of Indonesian Petroleum Association4th Annual Convention 129ndash148



Figure 1 - Map of eastern Indonesia Seram is marked by the star The location of Figure 6 is shown by the box



Figure 2 - Geological sketch map of western and central Seram and Ambon (after Pownall et al 2013)The stratigraphic columns shown below (which can be used as a key to the map) show a

comparison between previous stratigraphic proposals Equivalent (and approximatelyequivalent) units are shown in the same colours Note that the stratigraphic position of themetamorphic units relates to the protolith age and not necessarily the age of metamorphism



Figure 3 - Australia-Asia collision at 25ndash23 Ma Figure modified slightly from Hall (2011) withwesterncentral Seram and Buru positioned further west (although the exact configuration of theSula Spur is unknown) Purplemdashoceanic crust gt 120 Ma bluemdashoceanic crust lt 120 Magreenmdashvolcanic arcs yellowmdashSE Asia redmdashAustralia



Figure 4 - (a) Geological map of the Kaibobo Peninsula western Seram (after Pownall et al 2013) (b)Cross-section through the Kaibobo Peninsula along profile XndashXprime ndashXPrime (c) Annotated panoramic

photograph taken from the top of Gunung Ailapia showing the Kobipoto Complex located beneath the Kaibobo Detachment



Figure 5 - Structural sketch map of Seram (from Pownall et al 2013) showing detachment faults (green)

strike-slip faults (red) high-angle normal faults (blue) and faults within the KobipotoMountains strike-slip pop-up structure (magenta) inferred from field observations and DEMinterpretation The relative timings of these three faulting episodes is indicated Detachmentfaults strike normal to an inferred NNEndashSSW initial extension direction (1) possible subsequentreactivation of detachment faults in central Seram by WNWndashESE left-lateral shear (2) forms theKawa Shear Zone and Kobipoto Mountains pop-up late-stage normal faulting in response toslight NndashS extension (3) cross-cuts all other structures

Figure 6 - 3D model of the Banda Arc (after Pownall et al 2013) based on earthquake hypocentrelocations (Das 2004) and seismic tomography (Spakman and Hall 2010) The box is drawn

between 1225ndash132degE and 2ndash10degS to 660 km depth



Figure 7 - Seismicity of the northern Banda Arc Pressure (white) and tension (red) axes are shown forearthquakes shallower than 80 km depth CMT solutions were obtained from the Global CMTdatabase at globalcmtorg (Ekstroumlm 2006) and plotted using the UNAVCO Integrated Data

Viewer (UNAVCO 2012) The dashed box shows the limits of the area of interest (1260ndash 1317degE 18ndash55degS)

Figure 8 - Schematic 3D block model (from Pownall et al 2014) demonstrating the generation ofultrahigh-temperature (UHT) conditions (red) by extreme extension accommodated by

lithospheric detachment faults and their subsequent strike-slip reactivation See Figure 9 for a2D version



Figure 9 - Schematic 2D model (from Pownall et al 2013) proposed for Seram to explain mantle

exhumation by extreme extension driven by Banda Arc slab rollback The relationship betweenthe Kobipoto Complex the Taunusa Complex and the Tehoru Formation is shown Thedetachment faults may have subsequently been reactivated as left-lateral strike-slip faults asobserved in the Kawa Fault Zone See Figure 8 for a 3D version

Figure 10 - Tectonic reconstructions for the Banda region at (A) 15 Ma (B) 7 Ma (C) 2 Ma showing slabrollback into the Banda Embayment (adapted from Hall 2012 Pownall et al 2014) Yellowstar indicates the location of Seram Oceanic crust is shown in purple (older than 120 Ma) and

blue (younger than 120 Ma) submarine arcs and oceanic plateaus are shown in cyan volcanicisland arcs ophiolites and material accreted along plate margins are shown in green



allochthonous thrust sheet and were thereforeinterpreted to comprise part of an extensiveophiolite that had been thrust from the Banda SeaWork by Linthout et al (1989) Linthout andHelmers (1994) Sopaheluwakan (1994) andMonnier et al (2003) built further on this lsquoSeramOphiolitersquo concept arguing that its obduction overthe metamorphic lsquobasementrsquo of the TehoruFormation had induced sub-ophiolitemetamorphism and the generation of cordierite-

bearing granites Thus the concept of Seramrepresenting a collisional complex dominated bylong-distance thrusting was developed Indeedfold-and-thrust-belts have been described fornorthern and eastern Seram (Zillman and Paten1975 Tjokrosapoetro and Budhitrisna 1982 Kempand Mogg 1992 Hill 2005) and the Seram Trough

(Pairault et al 2003ab Teas et al 2004) but goodevidence for this collisional tectonic style inwestern Seram is lacking

Hall and Wilson (2000) questioned the ophioliteconcept for Seram they suggested that theultramafic rocks may have instead been derivedfrom the subcontinental mantle and that extensionmay have been more important in the tectonicevolution of western Seram than collision andthrusting Spakman and Hall (2010) later providedfurther evidence for an extension-dominated history

for the island in the context of plate reconstructionsutilising new tomographic models for the Bandaregion More recently Pownall et al (2013) mappedout low-angle detachment faults in western Seram(Figure 4) that were shown to have facilitated theexhumation of subcontinental lithospheric mantle

providing evidence for significant lithosphericextension having occurred across the islandFurthermore Pownall et al (2013) found noevidence for significant thrusting in western Seram

THE GEOLOGY OF SERAM

Field relations on Seram and Ambon have previously been described by several authors (Valk1945 Germeraad 1946 van der Sluis 1950Audley-Charles et al 1979 Tjokrosapoetro andBudhitrisna 1982) The summary here is based theobservations by these authors and new fieldobservations by Pownall et al (2013)

Western Seram mainly comprises lower-greenschistto upper-amphibolite facies phyllites schists andgneisses of the Tehoru Formation Garnet mica-schists are widespread which are often intercalatedwith amphibolites Scarce kyanite-grade schists(containing also staurolite and garnet) represent the

highest grade part of the complex however largeareas are of a low metamorphic grade and preserveoriginal sedimentary structures (Audley-Charles etal 1979 Tjokrosapoetro and Budhitrisna 1982Linthout et al 1989 Pownall et al 2013)

The Tehoru Formation is associated with complexesof peridotites and cordierite + garnet-bearinggranitic rocks (see next section) The termlsquocordierite granitersquo has long been used in theliterature to describe the granitic lithologies (egPriem et al 1978) however these rocks containabundant mm-scale (and more rarely cm-scale)schlieren of granulite-facies mineral assemblages(sillimanite + spinel + cordierite) that demonstratethe rocks are part of a migmatite complex and havenot migrated from their source region to form

separate granitoid plutons (Pownall et al 20132014)

These leucosome-rich migmatites (diatexites) andassociated peridotites comprise part of the KobipotoComplex which has been exhumed beneath low-angle lithospheric detachment faults mapped on theKaibobo Peninsula (Figure 4 see Pownall et al2013) in response to major lithospheric extension

Gneisses adjacent to the Kobipoto Complex peridotites that crop out on the Kaibobo Peninsula

previously interpreted by Linthout et al (1989) andLinthout and Helmers (1994) to comprise ametamorphic sole certainly have been overprinted

by sillimanite-grade metamorphism and intensiveshearing However we interpret these metamorphicrocks to form the hanging wall above the exhumedKobipoto Complex not the footwall As outlined byPownall et al (2013) the Taunusa Complex is anoverprint of Tehoru Formation rocks in response tohigh-temperature exhumation of the underlyingKobipoto Complex

In central Seram the dominantly left-lateral KawaShear Zone (KSZ) is oriented roughly parallel to thestrike of the detachment faults in western Seram(Figure 5) and there is evidence from fault-boundedlenses of serpentinised peridotites that areincorporated within the KSZ that these faults mayhave re-activated similar lithospheric-scaledetachments (Pownall et al 2013) The KobipotoMountains are a left-lateral pop-up structure thatlinks with the KSZ and the Kobipoto Complexrocks that crop out in the mountains have evidently

been exhumed by left-lateral transpression

The 3 km high Manusela Mountains bounded alongtheir southern margin by the major strike-slip Kawa


























TROUGH













BANDA SUBDUCTION






and Hall 2010)














































CONCLUSIONS












ACKNOWLEDGEMENTS


REFERENCES














httpwwwglobalcmtorg





Amsterdam 135 pp







































Series 401ndash411







20 1141ndash1160
























125 273ndash298




























































TROUGH













BANDA SUBDUCTION






and Hall 2010)














































CONCLUSIONS












ACKNOWLEDGEMENTS


REFERENCES














httpwwwglobalcmtorg





Amsterdam 135 pp







































Series 401ndash411







20 1141ndash1160
























125 273ndash298










































TROUGH













BANDA SUBDUCTION






and Hall 2010)














































CONCLUSIONS












ACKNOWLEDGEMENTS


REFERENCES














httpwwwglobalcmtorg





Amsterdam 135 pp







































Series 401ndash411







20 1141ndash1160
























125 273ndash298







































and Hall 2010)














































CONCLUSIONS












ACKNOWLEDGEMENTS


REFERENCES














httpwwwglobalcmtorg





Amsterdam 135 pp







































Series 401ndash411







20 1141ndash1160
























125 273ndash298
































































CONCLUSIONS












ACKNOWLEDGEMENTS


REFERENCES














httpwwwglobalcmtorg





Amsterdam 135 pp







































Series 401ndash411







20 1141ndash1160
























125 273ndash298









































CONCLUSIONS












ACKNOWLEDGEMENTS


REFERENCES














httpwwwglobalcmtorg





Amsterdam 135 pp







































Series 401ndash411







20 1141ndash1160
























125 273ndash298












































httpwwwglobalcmtorg





Amsterdam 135 pp







































Series 401ndash411







20 1141ndash1160
























125 273ndash298
























































Series 401ndash411







20 1141ndash1160
























125 273ndash298























































125 273ndash298

















































































































































































