Tecionophysm, 119 (1985) 435-449

Elsevier Science Publishers B.V., Amsterdam - Printed in The Netherlands

435

THE SUMBA ENIGMA: IS SUMBA A DIAPIRIC FORE-ARC NAPPE IN

PROCESS OF FORMATION?

M.G. AUDLEY-CHARLES

Department of Geological Sciences, University CoRege London, Gower Street, London, WClE 6BT {Great

Brrtarn)

(Received July 18, 1984; accepted January 22, 1985)

ABSTRACT

Audley-Charles, M.G., 1985. The Sumba enigma: Is Sumba a diapiric fore-arc nappe in process of

formation? In: N.L. Carter and S. Uyeda (Editors), Collision Tectonics: Deformation of Collisional

Lithosphere. Tectonophysia, 119: 435-449

The anomalous updomed rno~holo~c~ expression of Sumba island, its enigmatic lack of strong

Neogene deformation and the northward morphological indentation of southern Sumbawa and Flares

require explanation.

The stratigraphy of Sumba may be correlated with the Cretaceous to Miocene part of the Timor

allochthon. The sedimentary and eruptive rock succession in Sumba shows remarkable similarities with

the allochthonous Palelo, Wiluba and Cablac deposits of Timor. In both islands the Cretaceous parts of

these sequences are regarded as characteristic of fore-arc deposits built on thin continental crust.

The Timor nappe is interpreted as a 5 km thick tectonic flake of the Banda fore-arc thrust onto the

Australian continental margin in the mid-Pliocene collision. The postulated Sumba nappe has not yet

been thrust onto the Australian margin which, in the Sumba region, has not yet converged as close to the

arc as in the Timor area. The postulated Sumba nappe is interpreted as a diapiric elongated dome of the

Sunda fore-arc that is being squeezed by the converging margin of Australia against the volcanic islands

of Sumbawa and Flares.

The absence of indications on the seismic reflection profiles for the presence of the thrust fault of the

Sumba nappe may perhaps be explained by the thrusts being nearly horizontal within flat-lying strata.

The Savu thrust is correlated with the probably older (pre-Late Pliocene) Wetar Suture as a major

southward dipping lithospheric rupture. East of 124”E, this suture does not seem to have moved much

since the mid-Pliocene collision that emplaced the nappes on Timor. However, microearthquake data

suggest some activity is continuing.

INTRODUCTION

Sumba Island is 200 km long and 50 km wide, occupying a position between the eastern end of the Java Trench and the Sunda volcanic-arc islands of Sumbawa and Flares (Fig. 1). Sumba is a geological enigma because, as Chamaiaun et al. (1981) and Milsom et al. (1983) pointed out, it does not exhibit the features of the Neogene-Quaternary subduction zone of the Sunda Arc system (west of Sumba),

HO-1951/85/$03.30 0 1985 Elsevier Science Publishers B.V.

- - Inferred former pasitton of trench A Volc?lI~o@S now overridden by Australian plate

:I: ‘.‘.‘,~.~.‘_ Earthquakes ::.:.:::.

,~_‘.‘.‘. . Area underlain by Austrafmn ,‘,‘~“’ continent

a LA:: 7.0 ~~~~~~~,“~‘~‘~~~~ O-fOkit~

* MS 7.0 -‘A. -A

I Postulated boundary of x forearc nappes

Fjo -1llOkiir

Fig. 1. Map of earthquake epicenrcrs for shallow (h i iix) km) events taken from ISC bulletins for the

years 1964-1975. Limits of Australian continental crust based on Johnston and Howin (1%1), C‘hamalaun

et al. (1976} and geological mapping. Saw, FIoses and Wetar thrusts plotted from Silver et ai. (1983).

Position of Sumba nappe based on ~dth~rnetr~ (Hamilton. 1979). Location of tear faults based on

gwlogigiral mapping of Audfey-Char& (1968). Rosidi et af. ($979) and Earle (1981 f and iater unpu~l~shed

revision with A.J. Barber, 13.5. Carter and T. Charlton. Position of Timor nappe based on geological

mapping cited above.

nor does Sumba exhibit the features of the Pliocene collision zone of the Banda Arc

system (east of Sumba).

Sumba appears from Bouguer gravity anomaly of -t 160 to -t 200 mGal and

seismic refraction data to be underlain by continental-type crust estimated by

Chamalaun et al. (1981) to be at least 24 km thick. Cardwell and Isacks (1978)

interpreted the earthquake data to show that below Sumba the Benioff Zone is

continuous from the Sunda Arc to the Banda Arc. Chamalaun et at. (1981)

437

lndian ::-‘sCOTT ‘::yi:‘, j,

Ocean ,;+?TEA”:ij ._ (...

:_.. .:: ~..,~,..‘.~. :::‘. .,~...~.....

Fig. 2. Geological interpretation of Sumba and Timor nappes based on data in Fig. 1 and on the model of

Price and Audley-Charles (1983).

reinterpreted the earthquake data. They argued that west of 121”E (Figs. 1 and 2) the Benioff Zone appears to be typical of an oceanic subduction zone but that east of 121”E shallow earthquakes are almost absent and the Benioff Zone is less steep. Chamalaun et al. (1981) concluded that there is a sharp seismic discontinuity at about 121”E and that, in terms of earthquake pattern, Sumba is much more closely related to the Sunda Arc and is very different seismically from the Banda Arc. McCaffrey et al. (1984) reporting results of their microearthquake survey also drew attention to the significantly lower earthquake activity east of 124S’E. Whitford et al. (1977) pointed out that there are important differences in composition of the volcanic part of the Sunda and Banda Arcs and that these changes occur in the region opposite Timor between 123”E-127”30’E.

Palaeomagnetic data for Sumba were interpreted by Otofuji et al. (1980) to indicate Sumba had been derived from the Australian continental margin. Chamalaun

et al. (1981)+ reviewing the palaeomagnetic data for Sumba. concluded that the data

were insufficient to permit a confident interpretation of the palaeogeographiunf

origin of pre-Tertiary Sumba.

Since the review of Sumha’s geology by ~~~~anlaJaun et al. { 1981). two papers have

presented important new data: the sedimentological discoveries reported by Van der

Borch et al. (1983) have direct bearing on the Sumba’s pre-Tertiary origin, while the

marine geophysical work of Silver et aJ. (1983) has considerabte relevance for the

mode of late Tertiary evolution. These papers have led to the proposal put forward

here to correlate the pre-Eocene stratigraphy of Sumba with that of the PaIeJo

Group of Timor (Earie. 1983). This correlation has considerable structural, tectonic

and paJaeogeographi,cal implication because the Paleto Group of Timor is nJloch-

thonous (Barber et al., 1977: Rosidi et al.. 1979: EarJe, 1983).

STRATIGRAPHYOFSWMBA

The recent study by Von der Borch et al. (1983) has substantially improved our

knowledge of the pre-Tertiary rocks of Sumba. The dating of the oldest exposed

rocks as Cretaeeous (JocaJJy mid-Cr~ta~euus~ according to Chama~aun et al., f9Xf)

and discovery that they contain volcanogenic mudstones, sandstones, gravels and

diamictites intruded by andesite and dacite dykes is not only important in itself but

also because it invites comparison of the Sumba stratigraphical succession (Fig. 3)

with the Palelo Group in Timor (Earle, 1983).

Von der Borch et al. (I 983) concluded from their study that the pre-Tertiary rocks

of Sumba were deposited on thin continental crust and that Sumba represents part

of an uplifted fore-arc basin. Earle (1983) concluded from his study of the Mutis

n~etarno~hj~ complex and its overlying Palelo Group and associated Pateogene

succession that these altochthonous Timor deposits represent part of a fore-arc basin

floored by thin continental crust overthrust onto the Australian continental margin

by its Pliocene collision with the Java Trench.

Comparison of the pre-Tertiary to Early Miocene successions of Sumba (Van

Bemmefen, 1949; Von der Borch et al., 1983) with the Pafelo Group (Haiie et al.,

1979; Earle, 19831, Eocene Wiluba facies (Audiey-Charles and Carter, 1972) and

Cablac Limestone (Rosidi et al., 1979) of the allochthonous nappes of west Timor

(Fig. 3) reveals some strong simiiarities and some notable differences in these

successions in Sumba and Timor that are now separated by 400 km along strike.

The base of the sedimentary section in Sumba is not exposed. In West Timor the

base of the PaleJo Group is faulted against the Mutis Complex but an angular

unconformity is suspected (Haile et al., 1979). The Metan Formation of Timor is

regarded by HaiIe et al. (1979) and by Earle (1983) as probably Late Jurassic 0x1 the

basis of it occurring below the Early Cretaceous cherts. Rosidi et al. (1973) suggest a

different series of stratigraphical elements for the Palelo Group. I have fallowed

Ha& et al. and EarJe because their work was far more detaifed and because it

corresponds more closely to my own unpublished observations.

- 400hl . WEST

TIMOR

NON! FORMAlWN

,roan

Fig. 3. Stratigraphical successions (not to scale) in Sumba based on Van Bemmeien (1949). Chamalaun et al. (1981) and Von der Borch et al. (1983). West Timor stratigraphical section based on Tappenbeck (1940), Audley-Charles and Carter (1972), Rosidi et al. (19791, Earle (1983) and unpublished field work in 1983 by the author.

The lower part of the Noni Formation of Timor seems to be similar to the Lasipu Formation of Sumba (Fig. 3). Differences appear at the top of the Noni Formation where about 5 m of pelagic limestones and cherts occur. These have not been seen in Sumba. However, the red and green chert pebbles and cabbies in the Lasipu conglomerates are reminiscent of the Noni cherts of Timor.

There are far more dykes cutting the Lasipu Formation than in the Noni of Timor and no plutonic equivalent of the granodiorite intrusions cutting the Lasipu is known in Timor. However, the pre-Paleogene nature of these intrusions is common.

The presence of ignimbritic rocks of Early Paleogene age is common to both Sumba and Timor followed in both successions by neritic limestones with large Foraminifera of Middle and Upper Eocene age. Both successions display neritic limestones of Early Miocene age, but volcanism seems to have been much more important in Sumba during the Miocene than in the all~hthon of Timor.

The angular unconformity at the base of the Paieogene is a notable feature of both the Sumba and Timor allochthon successions. Similarly the unconformity at the base of the Miocene shallow water limestones is present in both.

440

Overall the sequence of the two successions (now 400 km apart along strike)

seems remarkably similar from late Mesozoic until the end of the Early Miocene.

The change in the Middle Miocene may correlate with the very early stage of the

collision process between the continental margin of Australia that began to under-

thrust the Banda fore-arc in MiddleeLate Miocene times in the Timor region. The

rocks of the Timor allochthon seem to be more deformed as might be expected from

their occurrence as part of a nappe that has travelled 50 km southward over the

para-autochthon.

STRUCTURE OF SUMBA AND MUTIS-PALELO GROUP OF TIMOR

According to Von der Borch et al, (1983) the structure of Sumba is very simple

and Van Bemmelen (1949) reports only erosional unconformities within the Ceno-

A SUMBA

S N

B TIMOR

NW Ml.MOLlO SE

QUE OUP

Fig. 4. A. Section across Sumba to show relative posltion of principal stratigraphical elements based on

Chamalaun et al. (1981) and Von der Borch et al. (1983). The basement below the Cretaceous is not

exposed in Sumba. Much minor faulting omitted. B. Section across Mt. Mollo in west Timor to show

relationship of Palelo Group to Mutis metamorphic basement and the position of the Wiluba facies and

Cablac Limestone based on Tappenbeck (1940). Rosidi et al. (1979). Earle (1983) and unpublished

geological map (1 : 50,000) of west Timor by Audley-Charles. Much minor normal faulting omitted.

441

zoic above the basal Tertiary angular unconformity. Figure 4 summarises the geological structure of Sumba which compares closely with the structural relation- ships of the Palelo-Wiluba-Cablac succession of the Timor allochthon (Tappen- beck, 1940; Carter et al., 1976; Barber et al., 1977; Rosidi et al., 1979; Earle, 1983).

Some of these relationships are revealed in the most detailed published map (scale 1: 25,000) of the Mollo Massif of Timor (Tappenbeck, 1940). The Rosidi et al. (1979) map of western Timor (scale 1 : 250,000) reveals a geometry compatible with that shown in,Fig. 4. However, the cross-section L-M drawn by Rosidi et al. (1979) through this Mollo Massif interprets the Mutis Complex as bounded by steeply northward dipping reverse faults. This reflects the interpretation Rosidi et al. have adopted for all their cross sections of western Timor, but the geometry implied by the outcrop-topography relationships throughout their map is incompatible in many places (including Mollo) with those steeply dipping reverse faults (c.f. also Tappen- beck’s more detailed map and section along the same line). In many parts of western Timor the Rosidi et al. (1979) mapping indicates their allochthon to be flat lying thrust sheets overlain by a terrain linking unconformable contact at the base of the Bobonaro Scaly Clay (Bobonaro Complex). That is also the structure mapped throughout eastern Timor (Audley-Charles, 1968).

EARTHQUAKES

Silver et al. (1983) published a map showing the earthquakes of shallow focus epicenters (O-100 km) in the Sumba-Timor region. These same data have been

@I JAVA

TRENCH FLORES El

- 1’1

A

ASTHENOSPHEf?E

,-200

-300

km 0 100 Xx) ?%400 km

Fig. 5. Interpretative N-S cross section located west Sumba (see Fig. 1). Earthquakes plotted from the seismicity window A of Chamalaun et al. (1981).

442

plotted (Fig. 1) with the thrusts mapped off-shore by Silver et al. (1983) together

with the principal submarine morphological features of the region. The fore-arc

nappes in Timor are based on geological mapping (Audley-Charles, 196X; Carter et

al.. 1976: Barber et al.. 1977: Kosidi et af.. 1979) interpreted in terms of the model

proposed by Price and Audlev-Charles (198.3). This model follows the views of

10

3c

5C

7c

I

------ 8” 9”

------ ----- ---_r__

IO rr” I?” s

FLORES SUMBA c S

l

*oo :‘= . l e :

0

200 400 I I

600 k

Fig. 6. interpretative N-S cross section through east Sumba based on the tectonic model of Price and

Audley-Charles (1983). Earthquakes plotted from ISC bulletins for years 1964-1980.

443

Carter et al. (1976), Earle (1983), Brown and Earle (1983) in regarding the alloch- thonous elements of Timor as derived from the basement of the Banda volcanic

fore-arc. This interpretation now appears to be strengthened by the correlation (Figs. 2, 3,4) of Sumba and the Timor allochthon.

McCaffrey et al. (1984,1985) in their microearthquake survey of the region found the epicenters distribution pattern similar to that reported by Silver et al. (1983). One feature of particular interest noted by McCaffrey et al. (1985, fig. 9) is their hypocenter plot onto a plane striking N25”W through Niki Niki in southwest Timor. This reveals a distinctly northward dipping Benioff zone below the north coast of Timor passing below the volcanic arc. However, below Timor their hypocenters have a flat or slightly southward dip as far as the south coast of Timor. There is no indication of the Benioff zone, which appears to be about 70 km below the north coast of Timor, extending upward to the surface in the Timor Trough as postulated by Hamilton (1979), Von der Borch (1979), SiIver et al. (1983) and other authors. The results of this microearthquake survey (McCaffrey et al., 1985) appear to support the crustal and lithospheric structure proposed in Figs. 5,6 and 7.

INI Q.I.U. Q.1.U. TECTONIC FLAKE e - &gREARC q

Former position

SOUTH BANM SEA ALOR WETAR STRAIT TIMOR /

of Java Trench I

-0

C’

-100

ASJHfNOSPHERE -200

WETAR THRUST ,-300

QIU REG/oNOF ~ATERNARY fSO-

0

km 0 100 200 I ,i 300 A;

Fig. 7. Interpretative N-S cross section through west Timor. Earthquakes plotted from seismicity window

B of Chamalaun et al. (1981). The section is based on the tectonic model of Price and Audley-Charles

(1983). The postulated presence of mantle and asthenosphere in the Wetar Suture seems to be required by

the very high positive gravity anomaly below the Wetar Strait (Milsom and Audley-Charles, 1985). The

discovery of ecologite on the north coast of Timor near Atapupu (close to this line of section) by S.

Tjokrosapoetro in the intertidal zone (pers. commun., 1983) is noted. Afthough this particular profile

shows no hypocenters beneath Timor, McCaffrey et al. (1985) have found microearthquake activity below

Timor.

444

McCaffrey et al. (1984) also noted the tendency for many hypocentrrs below

Timor to be less than 150 km deep indicating the strain is affecting the crust of this

region.

STRIKE-SLIP FAULTS

One feature to emerge from this study (Figs. 1 and 2) is the correlation between

shallow focus epicenters and tear faults across the strike in the Timor region. These

faults have been mapped onshore (Audley-Charles. 1968; Rosidi et al.. 1979: Earle,

19X1) and studies now being undertaken (‘I‘. Charlton. pers. commun.) of the

Australian facies para-autochthon exposed in western Timor are revealing the

importance of these post-overthrusting (post mid-Pliocene) strike-slip faults. The

~ollcentration of earthquake activity in the western part of this Wetar Strait appears

to correspond with a rapid shift in the position of the northern limit of the

Australian offshore continental crust reflected in the shape of the very steep offshore

margin of westernmost Timor. This is also the region where the allochthonous

nappes of Timor cut out westward against the Savu Sea.

We can suspect that these changes (Fi g. 2) may he related to tear faults cutting _ across the strike of western Timor in the region between Semau and Panlar islands.

Furthermore, on the basis of their microearthquake study McCaffrey et al. (1985)

suggested a major tear fault is present below 124”E and they drew attention to the

contrast in seismic characteristics on either side of 124OE.

The Savu Thrust has been mapped as an active feature (Silver et al.. i983). It is

regarded here as marking ,the northern limit of the Australian continentat margin

rocks. The interpretation (Figs. 1 and 2) that it continues eastward to the western

margin of Timor lacks the support of any direct published observation, but Michel

Latreille (pers. commun., 1984) has unpublished high quality seismic reflection

survey data that led him to interpret its presence as shown in Figs. 1 and 2. The high

rate of sedimentation at the vvestern margin of Timor may perhaps obscure its

recognition from lower quality seismic reflection data. The Savu Thrust is considered

to terminate against the strike slip fault below 124OE proposed by McC’affrey et al.

(in press) on the basis of their micro~arthquake survey. Their report of much less

active seismicity east of 124Ot; correlates with the low level of Quaternary movement

on the Wetar Suture regarded by Price and Audley-Charles f 1983) as a nli~i-Plio~eIle

structure.

EPICENTERS AND BATHYMETRY BETWEEN SUMBA AND THE VOLCANIC‘ ARC’

The 1000 m bathymetric contour off southern Sumbawa and Flares shows a

northward indentation by 30 km into the southern border of these volcanic islands.

This indentation can be clearly seen on Hamilton’s (1979) tectonic map of In-

donesia. The epicenters of the quakes at depths of 60.-100 km make a similarly

445

northward convex pattern north of Sumba (Fig. 1). The close relationship between

these details of submarine morphology and earthquakes of the Sumba region require

consideration in the light of the stratigraphical and structural similarities now

revealed between Sumba and the Timor allochthon.

DISCUSSION

The palaeogeography of Sumba

Chamalaun et al. (1981) in their recent review of the palaeogeography of Sumba

identified three possible origins: (a) the margin of N.W. Australia from which

Sumba was rifted in Jurassic times in association with the formation of the Wharton

Basin (Audley-Charles, 1975; Otofuji et al., 1980); (b) the margin of Sundaland in

the region between east Java, southeast Kalimantan and southwest Sulawesi from

which it was rifted presumably during the Cenozoic (Hamilton, 1979); (c) an

intra-Tethys position where it formed a micro-continent of uncertain affinity.

The available palaeomagnetic evidence is, according to Chamalaun et al. (1981)

not sufficiently strong to discriminate between these various models but they

consider it slightly favoured derivation from the northwest Australian margin.

The new stratigraphical and sedimentological data of Von der Borch et al. (1983)

particularly the indication that an active volcanic arc was close to and north of

Sumba during the Cretaceous, together with the palaeocurrents indicating Creta-

ceous volcanoclastic sediments moving southwest into Sumba argues strongly against

the northwest Australian margin origin and supports Hamilton’s proposal that in

Cretaceous times Sumba formed a part of the margin of southeast Sundaland. Such

a view corresponds with the proposed origin of the thrust sheets of the Timor

allochthon (Carter et al., 1976; Brown and Earle, 1983; Earle, 1983). The stra-

tigraphical and sedimentological correlation of Sumba with the Timor allochthon

(Fig. 3) reinforces the interpretation of their common derivation from southeast

Sundaland.

Deep structure and tectonic evolution of Sumba

One interpretative step taken by this paper (Figs. 2, 7, 8) has been to postulate

that the 2 km high morphological elongated dome of Sumba (400 km-150 km in

area), representing an uplifted part of the forearc basin (Von der Borch et al., 1983)

is a diapiric nappe in process of formation within the fore-arc. The postulated

Sumba nappe is correlated with the nappe of the Timor region of the Outer Banda

Arc on the basis of stratigraphy. Price and McClay (1981) favoured the role of

diapirism in generating nappes. The presence of the exposed pre-Paleogene pervasive

dykes in Sumba (Von der Borch et al., 1983) indicates a significant heat input and

implies a large-scale displacement of water which might be related to the end-creta-

ceous uplift of Sumba. The later repetition of such events related to continuing

subduction below this fore-arc region which. together with the squeezing effect of the

converging Australian continent. might be responsible for the post-Miocene uplift of

Sumba as a large diapir.

Another interpretative step has been to correlate the Savu Thrust with the Wetar

Suture. This suture is regarded by Audley-Charles (1983) and by Price and Audley-

Charles (1983) as representing the surface trace of the southward dipping thrust that

has ruptured the entire Australian lithosphere so that this Wetar suture carried the

Australian continent over the fore-arc accretionary wedge in Mid-Pliocene times

(Fig. 8). This inte~retation implies that the N-S strike-slip faults of the

Pantar-Semau region (McCaffrey et al., 1985) deveioped as tears in the evolving

fithospheric thrust. East of these tears the southward dipping Wetar Suture litho-

spheric thrust developed deeper below the fore-arc (Fig. 7) so that its surface trace is

much further north and closer to the volcanic arc. This led in Timor to the

emplacement of the tectonic flake of the fore-arc on the ruptured Australian margin.

West of this PantarSemau fault zone the southward dipping Savu thrust litho-

spheric rupture developed in a shallower position below the fore-arc so that it

reached the surface in a more southerly part of the accretionary wedge. The Savu

thrust may also have developed later than the Wetar Suture. These events have been

dated onshore Timor as mid-Pli[~cene (N20) by Carter et al. (1976) on the basis of

the age of the emplacement of the overthrust fore-arc nappe above the Australian

margin deposits.

An argument against the postulated Sumba Nappe is the apparent absence of

indications of thrusting in the marine seismic reflection surveys (Silver et al.. 1983).

This may be a less serious objection than it appears if the postulated thrusts are

horizontal within flat lying strata at shallow levels and steepen rapidly at depth

being concave downward. This would imply that the Sumba elongated dome is a

diapiric-type nappe whose flat-lying thrust near the perimeter of the nappe might be

exceedingly difficult to detect on the seismic reflection profiles. The dome-like shape

of the postulated Sumha nappe would seem to accord with the proposed shape of the

thrusts at depth and to accord with the compressive stresses that seem likely to be

exerted by the converging Australian continental margin colliding with the Java

Trench (Figs. 1, 2 and 6). Sumha appears to be under squeezing stresses between the

volcanic arc edifice of SumbawaaFlores and the converging Australian margin

interpreted here as having overridden the Java Trench and, west of Timor. to be

riding over the outer part of the fore-arc accretionary wedge.

SAVU THRUST AND WETAR SUTURE AS A MAJOR TECTONIC RUPTURE

Observations that support the correlation of the Wetar Suture with a major

lithospheric discontinuity are the gravity data of Chamalaun et al. (1976) and

Milsom and Richardson (1976). These papers showed that a fundamental change

from continental to much denser rock occurs at the north coast of Timor. Chamalaun et al. (1976) modelled their data to show continental crust terminating abruptly at

the north coast implying the presence of a fundamentaf lithospheric contact. The geomagnetic depth sounding of Chamalaun and White (1975) led them to suggest a subduction zone is present close to the north coast of Timor, although as Milsom et al. (1983) pointed out there is little support for this from other evidence. However, these observations suggest an important discontinuity is present immediately off- shore northern Timor. McCaffrey et al. (1985) have argued that their microearth- quake survey data indicate that the downgoing Australian plate has ruptured in the position and with the same sense of movement (south side up) as was postulated for the Wetar Suture by Price and Audley-Charles (1983), the only difference being that McCaffrey et al. interpret the rupture to dip more steeply.

The indications that this major structural discontinuity of the Wetar Suture dips south are the microearthquake data of McCaffrey et al. (1985), its correlation with the southward dipping Savu thrust of Silver et al. (1983) and the absence of a fore-arc accretionary wedge on-shore Timor where Australian continental margin deposits underlie the whole island cropping out from south to north coast (Audley- Charles, 1968; Rosidi et al., 1979). It is because these Permian to Early Pliocene Australian continental margin sediments accumulated at a passive margin so far

from any volcanic arc that they show so little resemblance to a fore-arc accretionary prism, although locally they are imbricated. The presence of Australian continental margin deposits on the north coast of Timor only 20 km from the volcanic-arc island of Atauro can be explained by the Wetar Suture being a southward dipping thrust that has carried the Austrahan margin northward over the Banda fore-arc accretion- ary wedge (Fig. 7).

THE SUMBA NAPPE

The Sumba nappe, unhke the Timor Nappe, has not been thrust over the Australian continental margin. On the contrary, the Sumba nappe appears to be slightly overridden in the southeast by the Australian margin (Figs. 1, 5 and 6). The enigmatic lack of strong Neogene deformation, the anomalous updomed mdrpho- logical expression and the northward morphological indentation of southern Sumbawa and Fiores require explanation. One approach is to regard Sumba as being in the process of being detached from its basement to form a diapiric nappe as the Australian margin converging with the volcanic islands of Flores and Sumbawa compresses the intervening Sunda fore-arc.

SIGNIFICANCE OF THE FLORES THRUST AND WETAR THRUST

These back-arc thrusts appear to be relatively young features (Silver et al,, 1983). They may be interpreted qualitatively as the result of strain in the Asian plate

44x

consequent upon continuing convergence of the Australian-Indian Ocean plate (Fig.

7).

ACKNOWLEDGEMENTS

I am grateful to Michel Latreille, John Milsom, Stan Murrell. Neville Price, Eli

Silver and Soebardjio Tjokrosapoetro for discussion, to the referees for suggesting

improvements, and to Janet Baker and Colin Stuart for help with the art work.

REFERENCES

Audley-Charles. M.G., 196X. The geology of Portuguese Timor. Geol. Sot. London. Mem., 4: I ~76.

Audley-Charles. M.G., 1975. The Sumba fracture: a maJor discontinuity between eastern and western

Indonesia. Tectonophysics 26, 213321X.

Audley-Charles. M.G.. 1983. Comments on “Analogous tectonic evolution of the Ordovician foredeeps.

southern and central Appalachians”. Geology. 1 1: 490-493.

Audley-Charles. M.G. and Carter. D.J., 1972. Palaeogeographical significance of some aspects 01

Palaeogene and early Neogene stratigraphy and tectonics of the Timor Sea region. Palaeogeogr.

Palaeoclimatol. Palaeoecol., 11: 247.-264.

Barber, A.J.. Audley-Charles, M.G. and Carter. D.J.. 1977. Thrust tectomcs in Timor. J. Geol. Sot. Aust..

24: 51-62.

Brown. M. and Earle, M.M., 1983. C‘ordierite-bearing schists and gneisaes from Trmor. eastern Indonesia:

I’ T condittons of metamorphism and tectonic implications. J. Metamorph. Gcol.. I : 1 X3 203.

Cardwell. R.K. and Isacks. B.L.. 197X. Geometry of the subducted lithosphere beneath the Banda Sea in

eastern Indonesia from seismicity and fault plane solutions. J. Geophya. Res.. X3: 2X25 2X3X.

Carter. D.J.. Audley-Charles. M.G. and Barber, A.J.. 1976. Stratigraphical analysis of island arc--con-

tinental margin collision in eastern Indonesia. J. Geol. Sot. London., 132: 179-19X.

C‘hamalaun. F.H. and White. A.. 1975. Electromagnetic induction at Dili, Portuguese Timor. J. Geophys..

41: 53775413.

C‘hamalaun. F.H., Lockwood. K. and White, A., 1976. The Bouguer gravity field and crustal structure of

eastern Timor. Tectonophysics. 30: 241.-259.

Chamalaun. F.H.. Grady, A.E.. Van der Borch, C.C. and Hartono. H.M.S.. 19X1. I‘he tectontc srgnifi-

cance of Sumha. Bull. Geol. Res. Develop. Centre, Bandung, 5: l-20.

Earle. M.M.. 1981. A study of Boi and Molo. two metamorphic massifs in Timor. eastern Indonesia.

Unpubl. Ph.D. thesis, Univ. of London.

Earle. M.M.. 19X3. Contrnental margin origin for Cretaceouz radiolarian cherts rn western Tumor. Nature.

305: l29- 130.

Grady. A.E. and Berry, R.F.. 1977. Some PalaeozoicMesozoic stratigraphic-structural relattonships tn

east Timor and their significance in the tectonics of Timor. J. Geol. Sot. Aust., 24: 203 -214.

Haile. N.S.. Barber. A.J. and Carter, D.J.. 1979. Mesozoic cherts on crystalline schists in Sulawesi and

Timor. J. Geol. Sot. London, 136: 65570.

Hamilton. W.. 1979. Tectonics of the Indonesian region. U.S. Geol. Surv., Prof. Pap.. 107X.

Hubhert, M.K. and Rubey. W.W.. 1959. Role of fluid pressure in mechanics of overthrust fauitmg. Pt. I.

Mechanrcs of fluid filled porous solids and its application to overthrust faulting. Bull. Geol. SOC. Am..

70: 1155166.

Johnston, C.R. and Bowin. C.0.. 19X1. Crustal reactions resulting from the mid-Pliocene to Recent

continent-island arc collision in the Timor region. Bur. Miner. Resour., J. Aust. Geol. Geophys.. 6:

223.--243.

449

McCaffrey, R. and 16 others, 1984. Preliminary results of the 1982 Microearthquake Survey in Nusa

Tenggara Timor, Indonesia. Bull. Geol. Res. Dev. Centre, Bandung, 10: 1-9.

McCaffrey, R., Molnar, P. and Roecker, S., 1985. Micr~arthquake seismicity and fault plane solutions

related to arc-continental collision in the eastern Sunda Arc, Indonesia. J. Geophys. Res., 451 l-4528.

Milsom, J. and Audley-Charles, M.G.. 1985. Post-collision isostatic readjustment in the southern Banda

Arc. In: J.G. Ramsay, M.P. Coward and A.C; Ries (Editors), Collision Tectonics. Geol Sot. London,

Spec. Publ., 11: 341-360.

Milsom, J. and Richardson, A., 1976. Implications of the occurrence of large gravity gradients in northern

Timor. Geol. Mijnbouw. 55: 175-178.

Milsom, J., Audley-Charles, M.G., Barber, A.J. and Carter, D.J., 1983. Geological-geophysical paradoxes

of the eastern Indonesian collision zone. In: T.W.C. Hilde and S. Uyeda (Editors). Geodynamics of

the Western Pacific-Indonesian Region. Geodyn. Ser., Am. Geophys. Union, 11: 401-411.

Otofuji. Y., Sasajima, S., Nishimuta, S., Hadiwisastra, S., Yokoyama. ‘I. and Hehuwat, F., 1980.

Palaeomagnetic evidence for the paleoposition of Sumba Island, Indonesia. In: S. Nishimura (Editor),

Physics and Geology of the Indonesian Island Arcs. Kyoto University Press, Kyoto, pp. 59-66.

Price, N.J. and Audley-Charles, M.G., 1983. Plate rupture by hydraulic fracture resulting in overthrusting.

Nature. 306: 572-575.

Price. N.J. and McClay, K.R., 1981. Introduction. In: K. McCIay and N.J. Price (Editors), Thrust and

Nappe Tectonics. Geol. Sot. London, Spec. PubI., 9: 1-5.

Rosidi, H.M.D., Suwitodirdjo, K. and Tjokrosapoetro. S.. 1979. Geologic map of the Kupang-Atambua

quadrangles, Timor. Geological Research and Development Centre, Bandung.

Silver, E.A., Reed, D., McCaffrey, R. and Joyodirwiryo, Y., 1983. Back arc thrusting in the eastern Sunda

arc, Indonesia: a consequence of arc-continent collision. J. Geophys. Res., 88: 7429-7448.

Tappenbeck, D., 1940. Geologie des Mollogebirges und einiger benachbarter Gebiete (Nierlandisch

Timor). Geol. Exp. Lesser Sunda Islands, 1: 1-105--Univ. Amsterdam.

Van Bemmelen, R.W., 1949. The Geology of Indonesia. Government Printing Office, The Hague, 732 pp.

Van der Borch, C.C., Grady, A.E., Hardjoprawiro, S., Prasetyo, H. and Hadiwisastra, S., 1983. Mesozoic

and Iate Tertiary submarine fan sequences and their tectonic significance. Sumba, Indonesia. Sedi-

ment. Geol., 37: 113-132.

Whitford, D.J., Compston, W., Nicholls, I.A. and Abbott, M.J., 1977. Geochemistry of late Cenozoic

Iavas from eastern Indonesia: role of subducted sediments in petrogenesis. Geology, 5: 571-575.