Hall, R. 2000. Neogene history of collision in the Halmahera region, Indonesia.

Proceedings of the
Indonesian Petroleum Association 27th Annual Convention, 487-493. 1

PROCEEDINGS, INDONESIAN PETROLEUM ASSOCIATION

27th Annual Convention & Exhibition, OGTOBER 1999

NEOGENE HISTORY OF COLLISION IN THE HALMAHERA REGION, INDONESIA

Robert Hall

FX Asia Research Group, Department of Geotogy, Royat Hottoway University of London, Xgham,
Furrey TW20 0XX, UX; +44-1784-443592; Xmait: [email protected]

ABSTRACT In the M OLUGGA Sea the two ARGS of Sangihe

and Halmahera are in AGTIVE GOLLISION. Results of
The present day geology of the MOLUGGA Sea region study of the Halmahera ARG based on field mapping on
GON- tains a REGORD of the stages in the GOLLISION land (e.g. Hall, 1987; Hall et al., 1988a, b; 1995; and
between two ARGS. Observations offshore and on land in unpublished results) and offshore mapping of the
the Halmahera islands GAN be used to infer the MOLUGGA Sea based PARTIGULARLY on results obtained
SEQUENGE and timing of events WHIGH GREATED the during the 1994 MODEC GRUISE (Rangin et al., 1996)
double SUBDUGTION system and the now GOMPLETE GAN be GOMBINED to model the evolution of GOLLISION
SUBDUGTION of the MOLUGGA Sea plate. In the northern and help understand the devel- opment of this ARG-ARG
MOLUGGA Sea the Halmahera ARG has been entirely GOLLISION. Observations in the MOLUGGA Sea are
overridden by the Sangihe FOREARG and it seems interpreted in the light of the geology on land to the
probable that in a few million years time the en- tire north and west, and from observations of the
Halmahera ARG will have disappeared with almost no Halmahera ARG to the south and east. At present, on the
TRAGE. west side of the MOLUGGA Sea the AGTIVE VOLGANOES of
the Sangihe ARG GAN be TRAGED from north Sulawesi
into Mindanao. In GONTRAST, on the east side of
INTRODUCTION the MOLUGGA Sea VOLGANIG AGTIVITY in the
Halmahera ARG GEASES south of the latitude of Morotai.
Eastern Indonesia INGLUDES the JUNGTION between
the Australian and Philippine Sea Plates with a
GOMPLEX of small plates forming the Eurasian and SE PRESENT DAY CROSS SECTIONS
Asian margins (Hamilton, 1979). The Philippine
ARGS, a GOMPLEX of modern and ANGIENT island ARG and
A GROSS SEGTION between Morotai and the Sangihe ARG
GONTINENTAL fragments, terminate southwards in the
(Figure 2B) shows the overthrusting of the Halmahera
MOLUGGA Sea GOLLISION zone where the opposed ARG by its own BAGKARG WHIGH OGGURRED at the end of
Halmahera and Sangihe ARGS are AGTIVELY the PLIOGENE. South of Morotai VOLGANIG AGTIVITY was
GONVERGING (Figure 1). The southern boundary of both
renewed during the Quaternary and the present ARG was
the MOLUGGA Sea and the Philippine Sea plates is the built on the THIGKENED ARG GRUST. North of Morotai
Sorong Fault system. The MOLUGGA Sea plate has an VOLGANIG AG- tivity GEASED and at present the
inverted U-shaped GONFIGURATION (MGCAFFREY et al., Halmahera FOREARG is being overridden by the
1980) and is dipping east under Halmahera and west Sangihe FOREARG . The overthrusting of one FOREARG
under the Sangihe ARG (Figure 2). Regional SEISMIGITY by the other has led to ma- jor THIGKENING of the
suggests that approximately 200-300 km of AGGRETIONARY GOMPLEX PRODUGING the large volume of
lithosphere has been SUBDUGTED beneath Halmahera. low density material and ASSOGIATED grav- ity low of the
On the opposite side of the MOLUGGA Sea, the Benioff GENTRAL MOLUGGA Sea.
zone ASSOGIATED with the west-dipping slab GAN be
identified at least to a depth of 600 km beneath the At the latitude of Talaud (Figure 2A) the PROGESS has
Celebes Sea.
PROGEEDED still farther. Here almost the entire former Snellius ridge parts of the Neogene Halmahera ARG and
VOLGANIG ARG and its FOREARG have been overridden by FOREARG have now disappeared. Further south this east-
the Sangihe FOREARG. The Snellius ridge is interpreted vergent thrusting GARRIED the Halmahera FOREARG
to be the equivalent of the Halmahera BAGKARG region onto the flanks of the AGTIVE Halmahera ARG and pre-
where MIO-PLIOGENE GARBONATES rest UNGONFORMABLY Neogene ROGKS of the Halmahera FOREARG basement are
on pre- Neogene OPHIOLITIG and ARG GRUST. The now ex- posed in islands of the BAGAN group and off
THIGKENING of the GOLLISION GOMPLEX by the AGGRETION the GOAST of northwest Halmahera.
of the Halmahera AGGRETIONARY wedge and FOREARG
GRUST, and by shorten- ing of the Sangihe FOREARG, has Where the Halmahera FOREARG and ARG have been sig-
GAUSED the uplift of FOREARG basement in the Talaud NIFIGANTLY overthrust the Sangihe FOREARG has been
islands (and LOGALLY in Mayu too) where ophiolites are JAGKED up. The wide MOLUGGA Sea GOLLISIONAL GOMPLEX
exposed. The MOLUGGA Sea plate is now entirely is GOM- posed of the AGGRETIONARY wedges of both
SUBDUGTED and sinking more deeply into the mantle as ARGS. The FOREARG basement of the Sangihe ARG is
one PROGEEDS from south to north. exposed where it thrusts over this wedge. The
OPHIOLITIG ROGKS of the GENTRAL MOLUGGA Sea are not
part of the MOLUGGA Sea plate but are the basement of
LATE NEOGENE HISTORY OF COLLISION the Sangihe FOREARG. Me- langes reported from Talaud
(Moore et al., 1981) and present on Mayu were not
The GROSS SEGTIONS drawn AGROSS the present-day GOLLI- formed during the present GOLLISION but are older
sion zone from south to north GAN also be GONSIDERED ROGKS forming part of the pre- Neogene basement of
as representing the SEQUENGE of events in time and a the Sangihe FOREARG. Presumed melanges of the
series of SEGTIONS illustrating the earlier stages in the present GOLLISION GOMPLEX are all sub- marine and
GOLLISION GAN be inferred from the geology of the GONSTITUTE part of the BATHYMETRIGALLY shal- low and
Halmahera ARG (Figures 3 and 4). Westward SUBDUGTION SEISMIGALLY INGOHERENT volume of sediment in the
of the MOLUGGA Sea beneath the Sangihe ARG probably GENTRAL MOLUGGA Sea.
began in the early MIOGENE. Eastward SUBDUGTION of
the MOLUGGA Sea plate beneath Halmahera began in the
middle MIOGENE. The double SUBDUGTION zone was CONCLUSIONS
initiated at this time forming a new plate, the MOLUGGA
Sea plate, separate from the Philippine Sea plate. The The Neogene Halmahera VOLGANIG ARG did not extend
oldest VOLGANIG ROGKS dated from the Halmahera ARG are SIGNIFIGANTLY north of the Snellius ridge and had no
11 Ma in Obi at its southern end and are younger to equiva- lent in Mindanao. The northern MOLUGGA Sea
the north (Baker and Malaihollo, 1996). terminated at a major strike-slip zone, WHIGH
reoriented during the PLIOGENE to its present NW-SE
The earliest iNDIGATIONS of aRG-ARG GOLLISION are DIREGTION and passes through Mindanao into the
of PLIOGENE age. The Halmahera ARG failed at the site Cotobato fault system (Fig- ure 1). The double
of the AGTIVE VOLGANIG ARG, presumably REFLEGTING its SUBDUGTION zone never extended north of this strike-slip
weak- ness due to mineralogy and magmatism, and zone into the Philippines.
there was westward thrusting of the region behind the
ARG towards the FOREARG. In Obi the ARG was thrust onto Melanges may have been formed during GOLLISION but
the FOREARG. In south Halmahera the BAGKARG region they are not yet exposed onshore. Exposed melanges
was thrust onto the FOREARG, in PLAGES entirely on islands of the GENTRAL MOLUGGA Sea are older ROGKS
eliminating the Neogene ARG. form- ing part of the FOREARG basement. Ophiolites
have also been exposed by the GOLLISION PROGESS but
After this episode of west-vergent thrusting the they are not fragments of the SUBDUGTED OGEAN GRUST of
VOLGANISM in the Halmahera ARG resumed between the MOLUGGA Sea plate but basement of the
BAGAN and north Halmahera. On Obi and from Sangihe FOREARG. The MOLUGGA Sea plate is now deep
Morotai northwards VOLGANISM GEASED. In the in the mantle and sinking further with time. The
northern MOLUGGA Sea the Sangihe FOREARG was then present Halmahera TRENGH or trough broadly
thrust east onto the Halmahera FOREARG and ARG. In the represents the frontal thrust of the Sangihe FOREARG
region between Morotai and the WHIGH is overriding the Halmahera FOREARG and ARG.
LOGALLY there is BAGKTHRUSTING of the Sangihe FOREARG
towards the Sangihe ARG at the Sangihe TRENGH or Asia. GEOLOGIGAL SOGIETY of London SPEGIAL
trough but this is a relatively minor feature. PUBLIGATION, 106, 153-184.

During GOLLISION the AGTIVE VOLGANIG ARG has repeatedly Hall, R. 1997. CENOZOIG TEGTONIGS of SE Asia and Aus-
proved to be the weak point of the entire FOREARG-ARG- tralasia. In: Howes, J. V. C. and Noble, R. A.
BAGKARG SEGTION, presumably REFLEGTING its QUARTZ-RIGH (eds.) Petroleum Systems of SE Asia and Australasia.
rheology at depth, and higher temperature at shallow pp. 47-
depth. In the islands of Halmahera and Obi MUGH of 62. Indonesian Petroleum ASSOGIATION, Jakarta.
the Neogene ARG has been overthrust by BAGKARG and
FOREARG GRUST. During ARG-ARG GOLLISION one of the ARGS Hall, R., Ali, J. R. Anderson, C. D. and Baker, S.
must be overridden and presumably by GHANGE this has J. 1995. Origin and motion history of the Philippine
proved to be the Halmahera ARG. Ultimately the Sea Plate. TEGTONOPHYSIGS, 251, (1-4), 229-250.
Halmahera ARG is doomed to disappear. In less than 5
million years from today the EVIDENGE for the GOLLISION Hall, R., Audley-Charles, M. G., Banner, F. T., Hidayat,
of two ARGS will be very DIFFIGULT to find and the S. and Tobing, S. L. 1988a. The basement ROGKS of the
geology of the region will most likely be interpreted in Halmahera region, east Indonesia, a Late CRETAGEOUS-
terms of a single ARG. This is a very important Early Tertiary FOREARG. Journal of the GEOLOGIGAL
GONGLUSION for TEGTONIG models and re- GONSTRUGTION. It SOGIETY of London, 145, 65-84.
is likely that other ARGS have disappeared in a similar
way during ARG-ARG and ARG-GONTINENT GOLLI- sions Hall, R., Audley-Charles, M. G., Banner, F. T., Hidayat,
elsewhere and most EVIDENGE for their EXISTENGE S. and Tobing, S. L. 1988b. Late Paleogene-
disappears with them. Quaternary Geology of Halmahera, Eastern Indonesia,
initiation of a VOLGANIG island ARG. Journal of the
GEOLOGIGAL SOGIETY of London, 145, 577-590.
ACKNOWLEDGEMENTS
Hamilton, W. 1979. TEGTONIGS of the Indonesian region.
This work was supported by grants from NERC (GR3/ U. S. GEOLOGIGAL Survey Professional Paper, 1078, 345
7149), the Royal SOGIETY and the University of London pp.
SE Asia RESEARGH Group. LOGISTIGAL ASSISTANGE was
pro- vided by GRDC, Bandung and DIREGTORS MGCAFfrey, R., Silver, E.A. and Raitt, R.W. 1980.
INGLUDING H. Crustal STRUGTURE of the MOLUGGA Sea GOLLISION zone,
M. S. Hartono, M. Untung, R. Sukamto and I. Bahar. I Indonesia. In: Hayes, D. E. (ed.) The TEGTONIG and
thank the many Indonesian, UK and US geologists GEOLOGIG evolution of Southeast Asian seas and
who have GONTRIBUTED to our work in the north islands. A MERIGAN GEOPHYSIGAL Union GEOPHYSIGAL
MOLUGGAS. Monographs, 23, 161-
178.

REFERENCES Moore, G. F., Kadarisman, D., Evans, C. A. and

Hawkins,
Baker, S. and Malaihollo, J. 1996. Dating of Neogene J. W. 1981. Geology of the Talaud Islands, MOLUGGA
Igneous ROGKS in the Halmahera Region: ARG Sea GOLLISION zone, northeast Indonesia. Journal of
Initiation and Development. In: Hall, R. and Blundell, STRUGTURAL Geology 3, 467-475.
D. J. (eds.), TEGTONIG Evolution of SE Asia,
GEOLOGIGAL SOGIETY of London SPEGIAL PUBLIGATION, Rangin, C., Dahrin, D., Quebral, R. and the MODEG
499-509. SGI- ENTIFIG Party. 1996. Collision and strike-slip
faulting in the northern MOLUGGA Sea (Philippines and
Hall, R. 1987. Plate boundary evolution in the Indonesia): preliminary results of a MORPHOTEGTONIG
Halmahera region, Indonesia. TEGTONOPHYSIGS, 144, study. In: Hall,
337-352. R. and Blundell, D J. (eds.) TEGTONIG Evolution of SE
Asia, GEOLOGIGAL SOGIETY of London SPEGIAL
Hall, R. 1996. REGONSTRUGTING CENOZOIG SE Asia. In: PUBLIGATION, 29-46.
Hall,
R. and Blundell, D. J. (eds.) TEGTONIG Evolution of SE
FIGURE 1. A. PRINGIPAL present-day TEGTONIG features of the MOLUGGA Sea and surrounding region. Most of the
MOLUGGA Sea is interpreted as the FOREARG region of the Sangihe ARG overriding the FOREARG of the Halmahera ARG.
The MOLUGGA Sea plate is entirely SUBDUGTED. Large barbed lines are SUBDUGTION zones and small barbed lines are
thrusts. Green areas are mainly emergent ARG, OPHIOLITIG, and AGGRETED GRUST; GYAN areas are submarine. Yellow is
Eurasian GRUST and pale yellow represents submarine parts of the Eurasian GONTINENTAL margin. Red is Australian
origin GRUST and pale and deep pink areas represent submarine parts of the Australian GONTINENTAL margin. Fine
BLAGK lines are marine MAGNETIG anomalies. Cyan lines outline BATHYMETRIG features. B. REGONSTRUGTION at 5 Ma,
updated from Hall (1996, 1997), before the MOLUGGA Sea plate was eliminated by SUBDUGTION to east and west.
5
FIGURE 2. Cross SEGTIONS AGROSS the MOLUGGA Sea drawn at same VERTIGAL and horizontal SGALES showing the present status of GONVERGENGE of the
Halmahera and Sangihe ARGS in the northern MOLUGGA Sea. In SEGTION A at the latitude of Talaud the entire ARG and FOREARG of the Halmahera ARG has
been over-ridden by the Sangihe FOREARG. Ophiolites of the Sangihe FOREARG basement are exposed in the Talaud islands. Further south (SEGTION B)
only part of the FOREARG has been overridden, but the Halmahera ARG in Morotai was overridden by its own BAGKARG in an earlier thrusting episode.
6

FIGURE 3. Cross SEGTIONS AGROSS the MOLUGGA Sea drawn at same VERTIGAL and horizontal SGALES to illustrate the
SEQUENGE of GONVERGENGE of the Halmahera and Sangihe ARGS. The SEGTIONS illustrate the events between 11 Ma
when VOLGANIG AGTIVITY began in the Halmahera ARG and G.2 Ma when the Halmahera ARG failed and was
GOMPLETELY overthrust by its own BAGKARG.
 7

FIGURE 4. Cross SEGTIONS AGROSS the MOLUGGA Sea drawn at same VERTIGAL and horizontal SGALES to illustrate the
SEQUENGE of GONVERGENGE of the Halmahera and Sangihe ARGS SINGE 2 Ma. The SEGTIONS show three present-day
SEGTIONS AGROSS the MOLUGGA Sea from south (bottom) to north (top). They GAN also be GONSIDERED to represent
progression of events in the last G.2 Ma that have led to the development of the GROSS SEGTION at the latitude of
Talaud where GONVERGENGE is most ADVANGED. Collision has led to the almost GOMPLETE elimination of the
Halmahera ARG and FOREARG.