Journal of Asian Earth Sciences 76 (2013) 266–282

Contents lists available at SciVerse ScienceDirect

Journal of Asian Earth Sciences

journal homepage: www.elsevier .com/locate / jseaes

Provenance and geochronology of Cenozoic sandstones of northernBorneo

1367-9120/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.jseaes.2013.02.033

⇑ Corresponding author. Tel.: +60 123313872.E-mail addresses: marco.vanhattum@shell.com, m.vanhattum@shell.com

(M.W.A. van Hattum).

M.W.A. van Hattum a,⇑, R. Hall a, A.L. Pickard b, G.J. Nichols a

a SE Asia Research Group, Department of Earth Sciences, Royal Holloway University of London, Egham, Surrey TW20 0EX, United Kingdomb University of Western Australia, Crawley, WA 6009, Australia

a r t i c l e i n f o

Article history:Available online 15 March 2013

Keywords:BorneoPalawanProvenanceHeavy mineralsZircon geochronology

a b s t r a c t

The Crocker Fan of Sabah was deposited during subduction of the Proto-South China Sea between theEocene and Early Miocene. Collision of South China microcontinental blocks with Borneo in the EarlyMiocene terminated deep water sedimentation and resulted in the major regional Top Crocker Unconfor-mity (TCU). Sedimentation of fluvio-deltaic and shallow marine character resumed in the late Early Mio-cene. The Crocker Fan sandstones were derived from nearby sources in Borneo and nearby SE Asia, ratherthan distant Asian and Himalayan sources. The Crocker Fan sandstones have a mature composition, buttheir textures and heavy mineralogy indicate they are first-cycle sandstones, mostly derived from nearbygranitic source rocks, with some input of metamorphic, sedimentary and ophiolitic material. The discrep-ancy between compositional maturity and textural immaturity is attributed to the effects of tropicalweathering. U–Pb ages of detrital zircons are predominantly Mesozoic. In the Eocene sandstones Creta-ceous zircons dominate and suggest derivation from granites of the Schwaner Mountains of southernBorneo. In Oligocene sandstones Permian–Triassic and Palaeoproterozoic zircons become more impor-tant, and are interpreted to be derived from Permian–Triassic granites and Proterozoic basement ofthe Malay Tin Belt. Miocene fluvio-deltaic and shallow marine sandstones above the TCU were mostlyrecycled from the deformed Crocker Fan in the rising central mountain range of Borneo. The provenanceof the Tajau Sandstone Member of the Lower Miocene Kudat Formation in north Sabah is strikingly dif-ferent from other Miocene and older sandstones. Sediment was derived mainly from granitic and high-grade metamorphic source rocks. No such rocks existed in Borneo during the Early Miocene, but potentialsources are present on Palawan, to the north of Borneo. They represent continental crust from SouthChina and subduction-related metamorphic rocks which formed an elevated region in the Early Miocenewhich briefly supplied sediment to north Sabah.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

The Eocene–Lower Miocene Crocker Fan of northern Borneo(Fig. 1) represents one of the largest Paleogene sedimentary depos-its of SE Asia. Despite preservation of several kilometres of silici-clastic turbidite sandstones and shales, their provenance islargely unknown. Sedimentation of Crocker Fan in the Eocene be-gan at about the same time as collision of India and Asia. Previousstudies of SE Asia have suggested material may have been derivedfrom distant sources which could have included the easternHimalayas (Hamilton, 1979; Hall, 1996; Hutchison, 1996; Métivieret al., 1999), whereas other studies have suggested that nearbysource areas, possibly Borneo itself, may have been important(Hutchison et al., 2000; William et al., 2003; Hall and Morley,

2004; van Hattum et al., 2006). The Early Miocene was a periodof collision of micro-continental fragments with the western edgeof Borneo, and cessation of deep marine sedimentation. Shortlyafter, and throughout the Neogene, vast amounts of siliciclasticsediments, estimated to be up to 12 km in thickness, were depos-ited in marginal basins on and around Borneo (Hamilton, 1979;Hall and Nichols, 2002). These sedimentary basins host importanthydrocarbon occurrences. The provenance of these sediments isalso poorly understood.

Although there has been work on the sedimentology of theCrocker Fan sediments, these have been concerned mainly withdepositional environment, and there have been few studies ofcomposition and textures. The provenance characteristics of thenorthern Borneo sandstones are discussed here based on analysisof detrital modes, heavy minerals, as well as determination ofdetrital zircon U–Pb ages. Neogene formations including theMeligan and Setap Shale Formations of SW Sabah and the KudatFormation (Fig. 2) of northern Sabah were also studied. Although

Fig. 1. Simplified regional geological map of Borneo and nearby SE Asia.

M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282 267

these formations were deposited at the same time (Early Miocene)in a similar depositional environment and climate, their sandstonecharacteristics are very different, and reflect important differencesin provenance.

The identification of source areas for the Crocker Fan and Neo-gene sandstones helps to constrain sediment availability and path-ways, and provides new insights in the tectonic setting of northernBorneo between the Eocene and Miocene. A unique feature of Bor-neo is that it experienced a humid tropical climate throughout theCenozoic. Tropical weathering during source erosion and sedimenttransport was found to have a more profound influence on thecharacteristics of sandstones than is often assumed. This studyhas identified some of the limitations of provenance studies onsandstones that have experienced tropical weathering.

2. Geological background

Northern Borneo (Fig. 1) lies in an area of complex Cenozoicconvergence at the Eurasian margin influenced by relative motionsof the Indian–Australian, Pacific and Philippine Sea plates(Hamilton, 1979; Hutchison, 1989; Hall, 1996, 2002, 2012). Thecontinental core of SE Asia, consisting of peninsular Malaysia,Thailand, SW Borneo and Sumatra, and shallow marine shelf areasbetween, is referred to as Sundaland (Fig. 1), and includesabundant granitic rocks. This composite region of continentalblocks is formed of fragments that separated from Gondwana atvarious times, drifted northward across the Tethys, and accretedto the Eurasian continent in successive stages during the LatePalaeozoic and Mesozoic (Metcalfe, 1996; Hall and Sevastjanova,

2012). Sundaland was an extensive continent during the Cenozoicand was largely emergent land during eustatic lowstands and gla-cial intervals. Despite the present flat surface and shallow seas cov-ering the Sunda Shelf it has been a tectonically active area, andincludes many deep sedimentary basins containing large volumesof clastic sediment (Hall and Morley, 2004).

The Indochina Block forms much of Indochina and extends fromthe continental shelf of Vietnam westwards across Cambodia tothe western uplifted margin of the Khorat Basin of Thailand(Hutchison, 1989). Granites and granodiorites occur throughoutthe Indochina block (Metcalfe, 1996), and span a wide age range.Palaeozoic granites in NE Indochina have ages which are predom-inantly Silurian, but other batholiths are younger, at 250–190 Ma.In the southern part of the Indochina block there are Jurassic andCretaceous plutons (Hutchison, 1989).

Lower Palaeozoic sediments are generally absent from Indo-china. The Indochina Block was emergent during the Devonianand Carboniferous, and locally continental sediments were depos-ited. Carboniferous–Permian limestones were deposited to thenorth and west of the Kontum Massif, and Permian shallow marineclastic sediments were deposited at the peripheries of the Indo-china landmass. Mesozoic sediments are widespread in Indochina,often within fault-bounded grabens, where they includesandstones, mudstones and volcaniclastics. Cenozoic sedimentsare typically continental and coal-bearing, and are developed inlocal depressions. Widely distributed Cenozoic basalts indicateepisodes of rifting (Hutchison, 1989).

The eastern Himalayas contain continental blocks that wereonce part of Gondwana, prior to Carboniferous–Permian separa-

Fig. 2. Onshore stratigraphy of NW Borneo and interpreted equivalent major unconformities offshore.

268 M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282

tion. These terranes are characterised by platform Palaeozoic suc-cessions, in places covering Precambrian basement, which is usu-ally not exposed (Hutchison, 1989). The eastern Himalayas, likesome areas in Indochina, have a high concentration of rift-, subduc-tion- and collision-related granite batholiths (Mitchell, 1979). Themajor magmatic arc of the Himalayas is the Gandise or Transhima-layan batholith system. It contains a variety of granites with agesfrom Cretaceous to Eocene (Maluski et al., 1983).

The India–Asia collision is widely considered to have initiated inthe Eocene although the exact timing remains controversial (e.g.Tapponnier et al., 1986; Besse and Courtillot, 1991; Rowley,1996; Aitchison et al., 2007; Najman et al., 2010; van Hinsbergenet al., 2011; Ali and Aitchison, 2012; White and Lister, 2012) andlarge volumes of sediment must have been transported from thecollision zone to the south or east. It has been suggested that thelarge SE Asian rivers draining from eastern Tibet to the Indochinacoast had different routes prior to the onset of the India–Eurasiacollision, and that before the late Miocene drainage in Asia was sig-nificantly different from the present day (Clark et al., 2004). Priorto the uplift of the Tibetan plateau (Clark et al., 2005) the Mekongand Salween may have been less important rivers than they are to-day, and much of the sediment from eastern Tibet was shed intothe Gulf of Tonkin by the palaeo-Red River.

Peninsular Malaysia includes rocks typical of a Palaeozoic con-tinental margin, which are intruded by abundant tin-bearingPermian–Triassic and minor Cretaceous granites (Cobbing et al.,1992). The granites are part of what is known as the SE Asian TinBelt (Garson et al., 1975; Beckinsale, 1979; Cobbing et al., 1992)which extends from Myanmar southwards to the Thai-Malay Pen-insula and further south into the Indonesian Tin Islands (Fig. 1).Liew and Page (1985) showed that the granites were derived from,and intrude, Proterozoic continental crust. Thermochronologicalstudies show exhumation of Tin Belt granites in the Cretaceous

and Cenozoic (Krähenbuhl, 1991; Kwan et al., 1992; Cottamet al., in press).

Borneo is suggested to be the product of Mesozoic andCenozoic accretion of ophiolitic, island arc and microcontinentalfragments onto Sundaland (Hamilton, 1979; Hutchison, 1989;Metcalfe, 1996; Hall et al., 2008; Hall, 2012; Hall andSevastjanova, 2012). Extensive granitoid plutons and associatedvolcanics form the Schwaner Mountains in southern Borneo.They intrude metamorphic rocks of the Pinoh Group. The igneousrocks yield radiometric ages ranging throughout the Cretaceous(Williams et al., 1988). In western Sarawak smaller Cretaceousgranitoids form relatively small, isolated intrusions in an arcuatebelt from the Indonesian border in the west to the NieuwenhuisMountains in the east (Tate, 2001). A thick succession of UpperCretaceous–Lower Miocene deep marine sediments of the RajangGroup and the Crocker Formation form much of the main moun-tain range of western, central and northern Borneo. There wassouthward subduction of the Proto-South China Sea beneaththe north Borneo margin from the Eocene to the Early Miocene.The Crocker Fan sediments were deposited at an active subduc-tion margin on the south side of the Proto-South China Sea.Subduction ceased in the Early Miocene when there was collisionof extended continental crust (Palawan, Reed Bank andDangerous Grounds blocks) and the NW Borneo margin (e.g.Hamilton, 1979; Holloway, 1982; Taylor and Hayes, 1983; Tanand Lamy, 1990; Hazebroek and Tan, 1993; Hall, 1996, 2002;Hutchison et al., 2000). This episode is referred to as the SabahOrogeny (Hutchison, 1996). Deep marine sedimentation stopped,but sedimentation resumed in the Early Miocene in a fluvio-deltaic to shallow marine setting. During the Miocene the centralmountains of Borneo became elevated and rapid removal ofmaterial by erosion in a humid tropical setting generated greatvolumes of sediment (Hall and Nichols, 2002).

M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282 269

Sediment provenance studies from Borneo are few and haveconcentrated on the Neogene strata (Tanean et al., 1996). Thereare two main lines of thought regarding the provenance of thePaleogene Crocker Fan: (1) sediments were derived from distantsource areas, probably exposed by the India–Eurasia collision,and were either transported over the Sunda Shelf by major riverssuch as the proto-Mekong (Hutchison, 1989, 1996; Hall, 1996;Métivier et al., 1999) and/or from mainland SE Asia longitudinallyalong the Borneo deep water margins into a subduction complex(Hamilton, 1979) or alternatively (2) the characteristics of the sed-iment, and traps and obstructions on the Sunda Shelf, are inter-preted to indicate that material was derived from nearby sources,possibly Borneo itself, exposed by local tectonics and quicklyeroded in a tropical setting (Hall, 2002; Morley, 2002; Williamet al., 2003). These two scenarios do not necessarily contradicteach other, and a combination of the two is possible.

3. Stratigraphy

The onshore stratigraphy of northern Borneo is shown in Fig. 2,based on published literature, new field studies (van Hattum,2005), and interpretation of the provenance and geochronologicalresults of this study. Cenozoic sedimentary deposits in northernBorneo can be divided into two parts: (1) Deep marine rocks ofthe Upper Cretaceous to Eocene Rajang Group (Sapulut Formation)

Fig. 3. Sample locations and contoured map with approximate ages inferred from micr

and the Eocene to Lower Miocene Crocker Fan (Trusmadi, Crockerand Temburong Formations). The thickness of the CrockerFormation in Sabah may locally exceed 10 km (Collenette, 1958;Hutchison, 1996) and the main palaeocurrent direction is NNE(Stauffer, 1967) suggesting a source area in the SSW. (2) Fluvio-deltaic to shallow marine sedimentation resumed in the EarlyMiocene and continued until at least the end of the Miocene. TheTop Crocker Unconformity (TCU) separates deep marine depositsfrom younger sedimentary rocks (van Hattum, 2005; Hall et al.,2008). Drainage patterns became similar to the present day fromthe late Early Miocene.

The age and stratigraphy of the Crocker Fan are still relativelypoorly known. Dating of the Crocker Fan based on biostratigraphyis difficult due to the paucity of age-determining fossil assem-blages (Liechti et al., 1960) and the internal stratigraphy is oftennot clear. All available biostratigraphic ages from western Sabah(Rutten, 1915, 1925; van der Vlerk, 1951; Stephens, 1956;Collenette, 1958, 1965; Liechti et al., 1960; Wilson, 1961; Wilsonand Wong, 1964; Jasin, 1991; Jasin et al., 1995) were compiledand were used to produce a contoured age map of the UpperCretaceous–Lower Miocene sedimentary rocks of western Sabah(Fig. 3). There is a distinct trend of westward younging withinthe Crocker Fan, away from land at the time.

The TCU has previously been correlated in different ways withunconformities offshore, typically in sequences deposited above

ofossils for the Upper Cretaceous to Miocene sedimentary rocks of western Sabah.

Fig. 4. Onshore traces of the Middle Miocene Deep Regional Unconformity (DRU) and the Lower Miocene Top Crocker Unconformity (TCU) in Sabah interpreted from SRTMimage and field mapping of the Meligan Formation by Wong (1962).

270 M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282

the Crocker Fan. On land it marks the termination of deep marinesedimentation and a change from deep water sediments and mel-

anges to fluviatile and shallow marine sediments. Offshore, thisunconformity was identified by earlier workers (e.g. Bol and van

Fig. 5. Simplified geological map of Kudat Peninsula, Sabah, after Lim and Heng (1985).

M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282 271

Hoorn, 1980; Levell, 1987; Hazebroek and Tan, 1993) as an uncon-formity below the Miocene hydrocarbon-producing strata of NWBorneo, but was left unnamed. Most studies of the offshore regionhave paid little attention to this important unconformity althoughyounger unconformities are widely discussed, for example, thewell-known Middle Miocene Deep Regional Unconformity (DRU)which has commonly been correlated (e.g. Levell, 1987) with Earlyor Middle Miocene collision, we believe incorrectly (Hall et al.,2008; Hall, in press). Younger regional unconformities, such asthe Intermediate and Shallow Regional Unconformity (IRU andSRU) are best known offshore (Hazebroek and Tan, 1993). Althoughdifficult to observe in the field, because of the difficulties of pene-trating into the rainforest interior of Sabah and Sarawak in deepriver valleys, both the TCU and DRU can be clearly separated onSRTM (Shuttle Radar Topography Mission) images (Fig. 4).

Neogene fluvio-deltaic to shallow marine sediments overlie theTCU. The Setap Shale Formation of SW Sabah (Fig. 2) is a monoto-nous marine succession of dark clay and shale with minor interca-lations of thin-bedded sandstone and siltstone (Wilson and Wong,1964). Towards the east the proportion of sand increases, and theSetap Shale Formation occasionally interfingers with the sandyMeligan Formation. Their age is Early Miocene. The lithology ofthe Setap Shale Formation resembles that of the older TemburongFormation of SW Sabah (Fig. 2), which is the upper part of theCrocker Fan. The Setap Shale and the Temburong Formations areoften difficult to tell apart in the field. However, the TemburongFormation is deformed together with the Crocker Formation andhas experienced low-grade metamorphism. The Setap Shale For-mation is non-metamorphic and generally less deformed. In theTemburong Formation Bouma sequences can be observed indicat-

ing deep marine turbiditic deposition, but the Setap Shale Forma-tion contains shallow-marine Skolithos burrows.

The Meligan Formation (Fig. 2) is a relatively uniform sandstonesuccession and resembles the Oligo-Miocene Nyalau Formation ofSarawak (Liechti et al., 1960). Lower Miocene pelagic foraminiferawere found in dark grey shales within the Meligan Formation(Wilson and Wong, 1962). The arenaceous microfauna, significantchanges of thickness and presence of large foresets, in combinationwith widespread carbonaceous matter, ripple marks and crossbed-ding, indicate a littoral environment and very shallow marineconditions. Towards the top of the formation, paralic conditionsare indicated by brackish-water faunas and an increase in lignites.The base of the Meligan Formation is generally conformable withthe partly older Setap Shale Formation, and locally the two forma-tions interfinger. The maximum thickness is reported to reach4500–5500 m, but these are composite figures and in any verticalcolumn the thickness does not exceed 3000 m (Liechti et al., 1960).

The Lower Miocene Kudat Formation (Fig. 2) also overlies the TCUand is a thick succession of predominantly sandy paralic to shallowmarine sediments covering almost the entire Kudat Peninsula andparts of the Bongaya Peninsula in northern Sabah (Fig. 5). The KudatFormation is the only formation in Sabah which can be subdividedinto members (Stephens, 1956). The lowest Tajau SandstoneMember consists of very proximal sandy debris flows that haveoccasionally been reworked as storm beds with hummocky crossstratification. The overlying Sikuati Member contains paralic sand-stones with layers of lignite. The formation was originally dated asEocene–Miocene (Stephens, 1956) but Clement and Keij (1958) latershowed that the Eocene and Oligocene ages were based either onbenthic microfauna that are not age-determining, or reworked

A

B

Fig. 6. (A) QFL plot of framework detrital modes showing compositions of the Lower Miocene sandstones of western Sabah and the average composition of the Crocker Fansandstones. The western Sabah sandstones generally have a quartzose mature composition. The least mature sandstones belong to the Tajau Sandstone Member of the KudatFormation. (B) QmFLt plot of framework detrital modes showing compositions of the Lower Miocene sandstones of western Sabah, and the average composition of theCrocker Fan sandstones. The highest compositional maturity is found in the sandstones of the Meligan Formation. The least mature sandstones belong to the Tajau SandstoneMember of the Kudat Formation.

272 M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282

larger foraminifera. They used numerous larger pelagic foraminiferato determine an Early Miocene age. The Kudat Formation is a time-equivalent of the Meligan and Setap Shale Formations of SW Sabahbut unfortunately the unrevised ages of Stephens (1956) are oftenquoted as depositional ages (e.g. Lim and Heng, 1985; Tongkul,1995; Petronas, 1999), leading to confusion over the origin and ageof the Kudat Formation.

4. Methods

Analysis of detrital constituents is routinely used to interpretthe provenance and palaeotectonic setting of sandstones (e.g. Dick-inson et al., 1983). Heavy mineral analysis is useful for making

provenance interpretations, as long as the provenance signal issuccessfully distinguished from the effects of hydraulic sortingand chemical destruction. Heavy mineral fractions in sandstonescan also be used to identify sediment pathways, sediment dispersalpatterns, sedimentary environments and for correlation of barrenstrata (Mange and Maurer, 1992).

Provenance analysis requires the freshest possible samples butexposure in Sabah can be poor because a humid tropical climateand lush vegetation cause outcrops to be eliminated or com-pletely overgrown within a matter of years. Fresh samples ofCenozoic sandstones were collected from river sections and coast-al outcrops, and new outcrops were discovered at sites of roadconstruction and urban development during fieldwork in 2001and 2002.

M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282 273

The Gazzi-Dickinson method of point counting was used forlight mineral modal analysis in order to reduce grain size influence(Tucker, 1988) and 300 grains were counted per sample. Grains lar-ger than 30 lm were counted as mineral constituents, and smallergrains were counted as matrix. Heavy mineral fractions were sep-arated by methods similar to the funnel separation method ofMange and Maurer (1992). The heavy mineral grains were countedin permanent grain mounts using the ribbon count method yield-ing number frequencies of minerals.

Zircons were separated from the heavy mineral fractions forvarietal studies and SHRIMP (sensitive high-resolution ion micro-probe) U–Pb dating was carried out at the SHRIMP II facility at Cur-tin University of Technology, Australia where we attempted toanalyse sufficient grains characterise the principal age populationsin sandstone samples (Dodson et al., 1988; Cawood et al., 1999).SHRIMP experimental techniques are discussed in detail by Smithet al. (1998).

5. Mineralogy

5.1. Light mineral detrital modes

The sandstones of the Crocker Fan have a mature quartzosecomposition (Fig. 6), and the most abundant clasts are plutonicquartz. Metamorphic and volcanic quartz are much less common,and volcanic quartz occurs only in the oldest samples of the Crock-er Fan. When feldspar is present, K-feldspar is the most abundanttype. It is most likely that the Crocker Fan sandstones have amostly acid plutonic source. Of the lithic fragments, radiolarianchert is never abundant but always present, forming up to 5% ofthe modal volume. This suggests that rocks of the ophiolitic base-ment were available for erosion. Other recognizable lithic frag-ments include schistose fragments, rare acid volcanic fragmentsand granite fragments in the coarser sandstones, and polycrystal-line quartzose grains that could have an igneous or metasedimen-tary origin. Carbonate clasts are rare. Other detrital constituentsinclude opaque grains, chlorite, muscovite and biotite.

During the Early Miocene quartzose sandstones of the LowerMiocene Setap Shale and Meligan Formations were deposited inSW Sabah above the Top Crocker Unconformity (TCU). These sand-stones are similar to the Upper Cretaceous–Lower Miocene Rajangand Crocker sandstones, but are more mature in composition andtexture (sorting and clast rounding). Their most important compo-nent is monocrystalline plutonic quartz and they appear to berecycled from the Crocker and Rajang sandstones.

The compositions of some sandstones of the Lower Miocene Ku-dat Formation of northern Sabah are quite different. Sandstones ofthe Tajau Sandstone Member are compositionally and texturallyimmature, with a mixed magmatic arc character (Fig. 6). They con-

Table 1Average heavy mineral contents of the main formations of western Sabah. The Tajau Sandsto its distinctive composition.

Mineral Average heavy mineral contents (%)

Trusmadi Fm Crocker Fm SEocene Eocene–L Miocene L

Zircon 27.2 36.9 4Tourmaline 46.7 38.5 3Rutile 6.7 6.0Cr-Spinel 2.2 1.1Garnet 2.8 7.0Apatite 5.2 3.9Pyroxene 0.7 0.8Amphibole 0.4 0.2Monazite 0.1 0.3Others 8.0 5.2

tain abundant K-feldspar and lithic clasts that suggest derivationfrom a nearby acid (meta)plutonic source. No potential sourcerocks are found in Sabah and the nearest potential source rocksare further north on Palawan. Limited palaeocurrent data fromthe Kudat Formation suggest that a source area to the north islikely. More mature sandstones of the Sikuati Member of the KudatFormation, like the Setap Shale and Meligan Formation sandstones,resemble recycled Crocker sandstones.

5.2. Textures

The textures of the Crocker sandstones are strikingly immature,resembling first-cycle sandstones. The shape of the grains is angu-lar to subangular, and there are few clasts with signs of sedimen-tary recycling. The sandstones tend to be poorly sorted, have amuddy matrix and very low porosity, making them of little interestas hydrocarbon reservoir sands. This is in contrast to their quartz-ose, compositionally mature character, which suggests a more ad-vanced degree of sedimentary recycling. The low-grademetamorphic quartzites of the Trusmadi Formation are denseand quartz-cemented.

The QFL and QmFLt plots (Dickinson et al., 1983; Dickinson andSuczek, 1979) suggest that most of the Crocker sandstones have arecycled orogenic source (Fig. 6) but an important limitation ofthese plots is that they mostly disregard climatic influence onsandstone composition; throughout the Neogene there has beena humid tropical climate on Borneo leading to high weatheringand erosion rates (Hall and Nichols, 2002). The interpretation ofstandard modal plots can be misleading for tropical sandstones be-cause light mineral compositions can be strongly altered by humidtropical weathering during erosion, transport and alluvial storageof sand (Suttner et al., 1981; Johnsson et al., 1988). This occursby the preferential destruction of feldspar and unstable lithic frag-ments relative to chemically stable quartz, therefore altering ordestroying the tectonic modal character. Tropical weathering canaccount for the discrepancy between textural and compositionaldata of the Crocker Fan sandstones. The textural immaturity willbe further highlighted by varietal zircon studies.

The Lower Miocene sandstones have a higher degree of grainrounding than the Crocker sandstones, consistent with sedimen-tary recycling. The Miocene sandstones are better sorted and havehigher porosities than those of the Crocker Fan.

5.3. Heavy minerals

The heavy minerals of all Crocker Fan sandstones are predomi-nantly those that are chemically stable. A summary of the contentsof detrital heavy minerals in the Crocker Fan sandstones is shownin Table 1. The heavy minerals are dominated by zircon and

tone Member is listed separately from the other members of the Kudat Formation due

etap-Meligam Fm Kudat Fm (Tajau) Kudat Fm (other)ower Miocene Lower Miocene Lower Miocene

6.9 24.4 42.83.2 7.8 25.97.0 2.3 10.44.4 0.0 2.03.2 34.0 4.61.0 9.8 7.40.6 7.6 1.60.0 0.1 0.30.2 1.1 0.43.5 12.9 4.6

Fig. 7. Chrome spinel-zircon (CZi) vs garnet-zircon (GZi) index value diagram ofwestern Sabah sandstones. CZi indicates the amount of ophiolitic material relativeto acid igneous material, GZi, indicates the amount of metamorphic materialrelative to acid igneous material. Throughout the Crocker Formation a small butsteady input of ophiolitic material, but a rather variable input of metamorphicmaterial.

274 M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282

tourmaline. Other common minerals are rutile, garnet, chromianspinel and apatite. There are also minor quantities of pyroxene,amphibole and monazite. Authigenic heavy minerals includebrookite and anatase. It is striking that detrital zircon and tourma-line usually show few signs of abrasion.

Apatite, a common mineral in acid igneous rocks, is not alwayspresent, and many apatite grains that do occur are often pitted orpartially dissolved, even in the freshest sandstone samples. Apatiteis unstable under conditions of acidic weathering (Morton, 1984),especially humid tropical conditions, and apatite dissolution mayhave occurred at any stage of source erosion, transport and depo-sition. Some samples of the Trusmadi Formation that contain noapatite do contain relatively fresh pyroxene, suggesting that theophiolitic material was less weathered than acid igneous material.

Chromian spinel is nearly always present, but never exceeds 4%of the total heavy mineral fraction. It shows very little abrasion,indicates that a local ophiolitic source must have been important.This was most likely the basement of northern Borneo, whichconsists mostly of Mesozoic ophiolites (Hutchison, 1978; Halland Wilson, 2000). Less stable ferromagnesian minerals are absentalthough clinopyroxenes and amphiboles are sometimespreserved.

The abundance of garnet is very variable but is usually subordi-nate to zircon and tourmaline. The garnets rarely show signs ofchemical destruction. The provenance of the garnet is not certain.Potential source rocks are the Pinoh Metamorphics of southernBorneo, intruded by the granites of the Schwaner Mountains(Pieters and Sanyoto, 1993), or the basement of the Tin Belt(Krähenbuhl, 1991).

The presence of detrital cassiterite, although rare, indicates thatthe Tin Belt probably supplied material to the Crocker Fan. Despitebeing chemically stable, cassiterite is a very brittle and densemineral. It is hydraulically separated from lower density sedimentat an early stage of transport (Hosking, 1971), and does not getre-entrained.

It is notable that volcanic material is absent in most of theCrocker Fan sandstones. The poorly dated Oligocene LabangFormation contains relatively large amounts of pyroxene and horn-blende, which suggest a greater input from intermediate and basicsource rocks than to other sediments of western Sabah. It containssome of the least mature sandstones of Sabah, with a possiblemagmatic arc provenance based on their modal compositions,and was probably deposited in a forearc position but closer tothe arc than the Crocker Fan.

Despite the compositional maturity of the heavy mineralassemblages of the Crocker Fan sandstones, their grain shapes indi-cate a low textural maturity, with a high proportion of unabraded,probably first cycle, zircons. Tourmalines have a similar character.A small proportion of strongly coloured zircons, mostly purplevarieties, are usually well rounded and probably have a polycyclichistory. The dominance of unabraded zircon and tourmaline sug-gests a proximal acid plutonic source. There are no Paleogene orolder granites in northern Borneo itself. The nearest potentialsource areas are the Cretaceous granites of the Schwaner Moun-tains (southern Borneo) and the mainly Permian–Triassic granitesof the Malay-Thai Tin Belt.

The heavy mineral assemblages of the Lower Miocene SetapShale and Meligan Formations are also compositionally very ma-ture and dominated by zircon and tourmaline (Table 1). Other min-erals include rutile, chrome spinel, garnet, apatite and pyroxene.The relatively high proportion of chromian spinel and the preser-vation of pyroxene in otherwise highly mature sandstones suggesta fresh ophiolitic source in addition to recycled sedimentary mate-rial. The similarity of the heavy mineral assemblages to those ofthe Crocker Fan and their slightly more rounded character suggestsrecycling.

In contrast, the Tajau Sandstone Member of the Lower MioceneKudat Formation of northern Sabah contains the compositionallyand texturally least mature sandstones of western Sabah. The hea-vy mineral assemblages are the most diverse of the Sabah sand-stones (Table 1). They are dominated by garnet, which can makeup nearly half of the entire heavy mineral fraction. Other mineralsinclude mostly unabraded zircon, tourmaline, unweathered apatiteand rutile, and appreciable amounts of chrome spinel, epidote,pyroxene, kyanite, and sillimanite, as well as small amounts ofstaurolite, corundum, sphene, monazite and olivine. Examples ofheavy minerals are shown in SEM images in Supplementary Docu-ment 1. The Tajau Sandstone Member is the only sedimentary uniton Sabah to include appreciable amounts of kyanite, suggesting ahigh-grade metamorphic source, and the assemblages includemany other metamorphic minerals. The high proportion of zircon,tourmaline and apatite shows that acid plutonic source rocks werealso important. The Tajau Sandstone Member heavy mineralassemblages could not be derived by recycling of Crocker sand-stones. Grains show very few signs of transport, and limited indi-cations of chemical destruction.

5.4. Heavy mineral indices

Indices of mineral pairs with similar hydraulic and diageneticbehaviour may reflect provenance, provided that they are stablewithin the diagenetic and weathering context of the study, andthat they have similar densities and grain sizes, which are the maincontrols on hydrodynamic behaviour. Morton and Hallsworth(1994) proposed a number of mineral indices that largely reflectprovenance characteristics. Of these indices, the apatite-tourmaline index (ATi) is likely to be affected by tropical weather-ing during erosion, transport and sedimentation, and maytherefore be unsuitable for provenance interpretations in thisstudy. The index values of GZi (garnet-zircon) and CZi (chromianspinel-zircon) are the most useful for provenance interpretationsof the western Sabah sandstones. The CZi index provides an indica-tion of the amount of ophiolitic material relative to acid igneousmaterial, and the GZi index gives an indication of the amount ofmetamorphic material relative to acid igneous material.

M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282 275

CZi values of the Crocker and Trusmadi Formations are typically2–7 (Fig. 7), indicating a minor but consistent input of ophioliticmaterial. The GZi values vary from 0 to 61. The large variation ingarnet contents in the Crocker Formation, suggests changing inputfrom metamorphic sources. CZi and GZi indices are not correlatedimplying ophiolitic and metamorphic source rocks in differentareas.

CZi values of the Setap Shale and Meligan Formations, whichwere interpreted above to be largely recycled from the Crocker For-mation, are generally higher than those of the Crocker sandstones,suggesting that they had an additional supply from a ophioliticsource. The Setap Shale and Meligan Formations show relativelylow GZi-values which can be accounted for simply by recyclingof Crocker sandstones (Fig. 7).

CZi values of the Kudat Formation sandstones are variable buttend to be low, especially in the Tajau Sandstone Member, suggest-ing a limited input of ophiolitic material. However, GZi values ofthe Kudat Formation sandstones show a large variation with thehighest values in the Tajau Sandstone Member (GZi = 73), confirm-ing an important metamorphic source. In contrast, there is littlegarnet in the mature quartzose and younger Sikuati Member ofthe Kudat Formation, which is interpreted to have been recycledfrom the Crocker Fan.

5.5. Zircon varietal studies

Zircon morphology is widely used in provenance studies todetermine the nature and genesis of the source rocks. Varietalstudies are helpful in low-diversity, ultramature heavy mineralassemblages, where conventional species-level heavy mineralstudies prove inconclusive because zircon is chemically ultrastableand is not affected by weathering or diagenesis. Zircons were stud-ied for shape, colour and, if present, internal structure. SEM imagesof typical morphologies with proportions of morphological types inthe Crocker, Meligan and Kudat Formations are shown in Supple-mentary Document 1.

The zircon varieties in different parts of the Crocker Fan aremostly similar. Sandstones contain a high proportion of euhedraland subhedral colourless zircons mixed with smaller amounts ofrounded colourless and sometimes purple zircons. They probablyall endured a comparable transport history. The high proportionof unabraded zircons makes it unlikely that they have travelledvery far by fluvial transport, and it is likely they have been derivedfrom relatively nearby source areas such as the Cretaceous granitesof the Schwaner Mountains and the Permian–Triassic granites ofthe Tin Belt rather than far away source areas such as the easternHimalayas or Indochina.

The zircon varieties in the Neogene formations of western Sa-bah are also similar, except for the Kudat Formation. Although itseems likely from light and heavy mineral studies that the MeliganFormation has been recycled from the Crocker Formation, the zir-cons from the Meligan Formation are not significantly morerounded than those of the Crocker Formation, probably indicatinga short transport path and/or rapid recycling.

The zircon varieties of the Tajau Sandstone Member of the Ku-dat Formation contain many more euhedral and subhedral zirconsthan other Lower Miocene sandstones, and very few subroundedand rounded zircons. This suggests a nearby acid plutonic or meta-morphic source, rather than sedimentary recycling.

6. Example sections

Provenance characteristics of the Oligocene part of the CrockerFormation near the west coast of Sabah were studied at outcropscale. Two outcrops had fresh continuous exposure of over 150 m

of turbidite sandstones and mudstones, one at Lok Kawi Heights(17 km south of Kota Kinabalu) and one at Sepangar Bay (12 kmNE of Kota Kinabalu). The sandstones all have a quartzose compo-sition. The two outcrops are briefly described below and illustratedin Supplementary Document 1 with composite diagrams that showlithostratigraphy, palaeocurrent measurements, abundance of themost prominent heavy minerals (zircon, tourmaline, rutile, chro-mian spinel, apatite and garnet), provenance-sensitive heavy min-eral indices, and zircon varieties.

Both sections are dominated by zircon and tourmaline, and con-tain relatively small amounts of rutile and chromian spinel. Thereare significant differences between the abundance of apatite andgarnet in the two sections. The Sepangar Bay section is overall con-siderably richer in apatite than the Lok Kawi Heights section. Bothsites were sampled below the modern-day weathering profilewithin months of excavation, so recent destruction of apatite is un-likely. The difference in apatite contents probably indicates differ-ent amounts of chemical weathering in the source areas of the twosections. In contrast, the difference in garnet contents is morelikely to indicate differences in source rocks. The garnet contentsin the Lok Kawi Heights section increase up section in sandstonesof similar grain size, and this probably records an increasing influxof metamorphic detritus.

The sandstones of the Sepangar Bay section appear to be de-rived mostly from acid plutonic rocks, a smaller proportion ofophiolitic rocks, and a small and variable proportion of metamor-phic rocks. The sandstones of the Lok Kawi Heights section havebeen derived primarily from acid igneous source rocks, but overtime (the time span the sections represent is unclear) an increasingamount of metamorphic source rocks became available. Smallamounts of ophiolitic material were available throughout.

The index values ATi (apatite-tourmaline), GZi (garnet-zircon),RuZi (rutile-zircon) and CZi (chrome spinel-zircon) show a largevariation within the outcrops, especially GZi in the Lok Kawi sec-tion, suggesting a variable but increasing input of metamorphicdetritus throughout the section. The ATi index may be influencedby weathering. Hurst and Morton (2001) point out that sedimenthomogenisation is more likely to occur on a shallow marine shelf,especially above the wave base, than in an alluvial basin, or in adeep marine basin. Deep-water sandstones with variable heavymineral indices of pairs of minerals with similar hydraulic behav-iour and chemical stability may have been derived directly fromalluvial basins, bypassing any contemporaneous marine shelf,while deep-water sandstones with homogeneous heavy mineral-ogy are inferred to be fed by sediment that originally accumu-lated on shallow marine shelves where it was mixed. Thevariable heavy mineral indices of the deep-marine OligoceneCrocker Formation at Sepangar Bay and Lok Kawi suggest thatthe sediments of the Crocker Formation have been derived byshelf bypass directly from an alluvial basin. Before collisions be-tween micro-continental blocks and western Borneo the shallowshelf area of northern Borneo may have been much more limitedthan at present.

The zircon varieties in the Sepangar Bay section show very littlevariation throughout the section. The zircon varieties suggest thatthe sediment has been derived from a mixed plutonic and sedi-mentary origin, indicated by the presence of both fresh euhedraland subhedral zircons as well as a relatively high proportion ofsubrounded and rounded zircons. The zircon varieties in the LokKawi Heights section are more varied than those of SepangarBay, and there are more euhedral and subhedral zircons. The high-est proportions of unabraded zircons coincide with high propor-tions of garnet. It is possible that the peaks of garnet andunabraded zircons represent first-cycle provenance pulses fromthe Pinoh Metamorphics and Schwaner Granites of southernBorneo.

Fig. 8. Probability plots of detrital zircon SHRIMP U–Pb ages from 6 Cenozoic western Sabah sandstone samples, with pie diagrams showing simplified source compositions ofthe Paleogene sandstones, based on heavy mineral characteristics and a principal component analysis of zircon age abundances. Provenance categories are Schwaner:Cretaceous granites of Schwaner Mountains; Tin Belt: Permian–Triassic granites of Malay-Thai Tin Belt; YB: Young basement; OB: Old Basement; Ophiolites: ophioliticbasement.

276 M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282

7. Geochronology

Detrital zircon U–Pb ages aid identification of the source areaand its age, and were determined for zircons from six Cenozoicsandstones. Samples were selected to represent a large area ofwestern Sabah, and to represent the Eocene to Miocene age of stra-ta. An important selection criterion was a reasonable constraint onthe sample depositional age. Detrital zircon ages were obtainedprimarily from the Crocker Fan (Crocker and Trusmadi Forma-tions), but ages were also obtained from one older Rajang Groupsample from the Sapulut Formation and strata overlying the TCUfrom the Kudat Formation (Fig. 8). In addition, two samples fromthe Sukadana Granite of the Schwaner Mountains, considered to

be a possible source area, were dated to compare their zircon ageswith detrital zircons.

Zircon U–Pb ages can be obtained from both 206Pb�/238U and207Pb�/206Pb ratios. The amount of 207Pb in Phanerozoic zirconscan be very small, and 206Pb�/238U ages are preferred to the207Pb/206Pb ages. In contrast, 207Pb�/206Pb ages are consideredto be more accurate for Precambrian zircons (Pickard et al.,2000). Cathodoluminescence images were used to select grains.Few display internal zoning or later overgrowth. There were nodifferences between core and rim ages, and no indication ofmetamorphic resetting. Zircons appear to be igneous, andtheir ages can be compared to potential igneous sourcerocks.

Fig. 9. Detrital zircon age groups based on all concordant SHRIMP ages from Paleogene western Sabah sandstones.

M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282 277

The Crocker Fan detrital zircon ages range from Eocene (50 Ma)to Archaean (2532 Ma). The different Paleogene samples displaysimilar age populations (Fig. 9). The most prominent is Cretaceous(Group A: ca. 77–130 Ma), the second is Permian–Triassic (GroupB: ca. 213–268 Ma) and there is a conspicuous Palaeoproterozoic(Group C: ca. 1750–1900 Ma). Smaller age groups include Jurassicand Silurian-Ordovician zircons. The Miocene Kudat sample wasdominated by Jurassic-Cretaceous zircons, with a significant num-ber of Palaeoproterozoic zircons. The individual samples are brieflydescribed below and a complete listing of age data is in Supple-mentary Document 2, Tables 1–6.

7.1. MVH02-264, Sapulut Formation, Early Eocene

This sample is from the Eocene part of the Sapulut Formation,the youngest part of the Rajang Group. 55 U–Pb zircon ages wereobtained and 45 zircons yielded concordant ages (Fig. 8). The mostimportant age group is Cretaceous. Other Phanerozoic populationsare Carboniferous, Devonian and Silurian. Jurassic zircons are ab-sent. There are few Precambrian zircons, but the most importantage group is Palaeoproterozoic. The Cretaceous zircons are colour-less and the least abraded. The Palaeozoic zircons display a higherdegree of rounding. The Precambrian zircons are often coloured,and tend to be the most rounded. The youngest age(56.1 ± 0.9 Ma) is of stratigraphical significance as it indicates themaximum depositional age of the rocks cannot be older than latestPaleocene.

7.2. MVH02-142, Trusmadi Formation, Middle Eocene

This sample is from the Trusmadi Formation, the oldest part ofthe Crocker Fan. 66 ages were obtained from this sample, of which48 were concordant (Fig. 8). The majority of the zircons yieldingdiscordant ages are Proterozoic. The most prominent age group isCretaceous. Other significant age groups are Triassic,

Carboniferous–Early Permian and Ordovician. The most importantgroup of concordant Precambrian zircons is Palaeoproterozoic. Inthis sample an Archaean zircon was encountered (2531.6 ±11.2 Ma), which is the oldest radiometric age reported fromBorneo. The oldest zircons tend to be coloured and/or rounded,and the younger Mesozoic zircons are mostly unabraded andcolourless. The youngest age (49.9 ± 1.9 Ma) is very important inan area where biostratigraphical evidence is nearly absent andindicates the maximum depositional age to be Early Eocene.

7.3. MVH02-271, Crocker Formation, Late Eocene

This sample is a texturally and compositionally immatureCrocker Formation sandstone. Of the 72 ages from this sample,59 are concordant (Fig. 8). The most important age group is EarlyCretaceous, which account for 44% of the ages. There is also a sig-nificant number of Late Cretaceous zircons. Other age groups areJurassic, Triassic and Palaeoproterozoic. Mesozoic zircons aremostly colourless and unabraded. Older Proterozoic zircons tendto be coloured and/or rounded.

7.4. MVH02-115, Crocker Formation, Oligocene

This sample was from one of the most northern outcrops of theCrocker Formation. Of the 70 ages 61 were concordant (Fig. 8). ThePermian–Triassic is represented by 20 ages. Other significant agegroups are Ordovician, Jurassic, Carboniferous and Devonian. Cre-taceous ages are nearly absent. There is a relatively large numberof Precambrian zircons, especially Palaeoproterozoic; 15 zirconsare older than 1600 Ma.

7.5. MVH02-116, Crocker Formation, Late Oligocene

This is the best-dated sample of the Crocker Formation based onLate Oligocene nannofossils (NP24-NP25) found nearby (E. Finch,

278 M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282

2002, pers. comm.) and represents the youngest part of the CrockerFan. Of the 66 zircon ages 52 are concordant (Fig. 8). There is a rel-atively wide spread of ages compared to other samples, and thereis also a high proportion of anhedral zircons. Cretaceous-Permianzircons form the main age group with a peak in the Late Permian.Other significant groups are Early Cretaceous, Carboniferous, Ordo-vician and Palaeoproterozoic.

7.6. MVH02-087, Kudat Formation, Tajau Sandstone Member, EarlyMiocene

The sample is a medium- to coarse-grained sandstone that con-tains 28% K-feldspar. The heavy mineral suite is relatively diverse,and is dominated by zircon, tourmaline, apatite and pyroxene, andcontains minor amounts of rutile, garnet, monazite, kyanite, silli-manite, staurolite, chloritoid, sphene, serpentine, corundum andolivine. Of the 57 U–Pb zircon ages, 50 are concordant (Fig. 8).The majority of the zircons are Jurassic-Cretaceous. Major peaksare Early Cretaceous (ca. 121 Ma) and Early Jurassic (ca. 181 Ma).Jurassic zircons are relatively rare in other Sabah sandstones. Aminor peak is Palaeoproterozoic (ca. 1867 Ma). The oldest zirconis dated at 2488.5 ± 16.3 Ma. The Mesozoic zircons are predomi-nantly unabraded and colourless. Coloured and/or rounded zirconsare mostly Precambrian, but occur throughout the age spectrum.

7.7. RT.C and RT.D, Sukadana Granite, Kalimantan, Cretaceous

Cretaceous granitoids and related volcanic rocks are widely dis-tributed in the southern part of the Schwaner Mountains of Kali-mantan. The main component of the Schwaner batholith in theKetapang area is the Sukadana Granite which includes a range ofrock types from monzonite to granite, with Cretaceous ages of127–79 Ma based on K–Ar dating of hornblende and biotite andU–Pb dating of zircons (Haile et al., 1977; de Keyser and Rustandi,1993). The results from the two samples analysed are summarisedbelow and described with a complete listing of age data in Supple-mentary Document 3.

Samples RT.C and RT.D were collected about 16 km apart in thesame medium- to coarse-grained monzogranite in the Ketapangarea. 30 zircons were dated from sample RT.C. They are nearly allLate Cretaceous, with two age peaks at 87.0 ± 0.8 Ma and83.8 ± 0.7 Ma, and a mean age of 84.7 ± 1.3 Ma. Most of the grainshave a simple internal structure. There is one Jurassic grain(151.9 ± 1.5 Ma) which the SEM cathodoluminescence image sug-gests is a core; no other inherited ages were found. 13 zircons weredated from sample RT.D. They were all Late Cretaceous and slightlyyounger than those in sample RT.C with two age peaks at84.0 ± 1.0 Ma and 79.7 ± 0.9 Ma, and a mean age of 81.7 ± 1.0 Ma,excluding one grain with a large error. SEM cathodoluminescenceimages of dated zircons from the sample show no older cores.

7.8. Age trends and provenance

There is a relationship between zircon ages, morphology andcolour. Cretaceous colourless zircons are the least abraded, sug-gesting limited transport. Permian–Triassic zircons are moreabraded than Cretaceous zircons, but colourless and unabradedzircons dominate, also suggesting a nearby source. The older, inparticular Palaeoproterozoic, zircons are generally coloured andwell-rounded, suggesting a long history of recycling.

In the Eocene sandstone samples unabraded Cretaceous zirconsdominate (Fig. 8). Their ages are similar to those of Cretaceousgranites of the Schwaner Mountains of southern Borneo (Fig. 1),which are the closest abundant granites. More distant Cretaceousgranites are probably currently submerged in the Sunda Shelf(Pupilli, 1973) and are known from the Malay peninsula (Cobbing

et al., 1992). Late Cretaceous zircons, which are more abundantthan Early Cretaceous zircons in the oldest sandstones, are similarto zircon ages from the Schwaner Mountains Sukadana Granite.Early Cretaceous zircons become more abundant than Late Creta-ceous zircons in the Upper Eocene–Oligocene sandstones, similarto the present-day distribution of Schwaner Mountains granites(Williams et al., 1988).

In the Oligocene sandstones Cretaceous zircons cease to domi-nate and Permian–Triassic ages become more common (Fig. 8).These zircons show slightly more rounding than Cretaceous zir-cons, probably reflecting a longer and/or higher energy transportpath. Ages of Permian–Triassic zircons resemble those of volumi-nous granites in the Malay-Thai Tin Belt, suggesting the mainsource of the Crocker Fan sandstones shifted from the SchwanerMountains to the Tin Belt during the Oligocene.

The proportion of usually well-rounded and coloured Palaeo-proterozoic and Neoarchaean detrital zircons strongly correlateswith the abundance of Permian–Triassic zircons, and they areinterpreted to represent metasedimentary continental basementof the Permian–Triassic Tin Belt granites of the Malay Peninsula.

Ages of detrital zircons in the Crocker Fan do not match those ofplutonic rocks in the eastern Himalayas or Indochina. Magmaticages in the eastern Himalayas older than the Crocker Fan are400–500 Ma, ca. 160 Ma, ca. 120 Ma and 40–70 Ma. The 400–500 Ma ages occur in zircon cores, and younger ages are thoughtto be related to Andean-style Gandise arc plutonism preceding In-dia-Asia collision (e.g. Maluski et al., 1983; Ding et al., 2001; Boothet al., 2004). If Indochina had been an important source more abun-dant Precambrian and Palaeozoic zircons would be expected but itis possible that some of the Palaeozoic zircons were ultimately de-rived from Indochina.

Ages of detrital zircon from the Tajau Sandstone Member aremostly Jurassic and Cretaceous. Jurassic zircons are very uncom-mon in other Sabah sandstones. This supports the idea that thesandstones were derived from the Palawan block, which was orig-inally part of South China, a region rich in Jurassic and Cretaceousgranites. Provenance studies of Cretaceous-Eocene Palawan sand-stones of similar composition to the Tajau Sandstone Member sug-gest that South China may have been the ultimate source (Suzukiet al., 2000). Although at present there are no exposures of Juras-sic-Cretaceous granites on Palawan, they may well be submergedoffshore of Palawan or in the area between Palawan and northernBorneo. Zircon cores from the Miocene Capoas granite in northernPalawan are Palaeoproterozoic, indicating melting of South Chinacontinental crust (Encarnación and Mukasa, 1997). The core agesare similar to the Palaeoproterozoic zircon ages in the Tajau Sand-stone Member, supporting derivation from Palawan. There are nozircons younger than Cretaceous in the Tajau Sandstone Member.High-grade subduction-related metamorphic rocks of Palawansuggested above to have provided kyanite and garnet to the KudatFormation sandstones are of Early Oligocene age, based on horn-blende and muscovite Ar–Ar dating, and there are no zircon agesreported from the metamorphic rocks (Encarnación et al., 1995).

8. Principal component analysis

Hypotheses about sources of the zircons were tested with prin-cipal component analysis. The zircon ages of the Sapulut, Trusmadiand Crocker Formations were subdivided into age categories and aprincipal component analysis was performed. The sample from theKudat Formation was excluded because of its potential Palawansource. A strong positive correlation between proportions ofPermian–Triassic and Palaeoproterozoic zircons supportsderivation of Palaeoproterozoic zircons from the Tin Belt. A strongnegative correlation between the proportions of Cretaceous and

M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282 279

Permian–Triassic zircons supports the suggestion of two indepen-dent sources which were Cretaceous granites of the SchwanerMountains and Permian–Triassic granites of the Tin Belt. The heavymineral assemblages and correlations were used to construct asimple model of five provenance groups. The relative proportionsas they appear in the individual samples are shown in Fig. 8.

(1) Cretaceous zircons show a weak positive correlation withCenozoic zircons, and they most likely represent Borneosources. Cenozoic zircons were probably derived fromnearby volcanic activity, and Cretaceous zircons from gran-ites in the Schwaner Mountains.

(2) Jurassic zircons show a strong positive correlation withPermian–Triassic zircons. The Permian–Triassic zircons rep-resent the Tin Belt granites. Jurassic igneous rocks are notknown from the Tin Belt, but it is possible that some acidvolcanic activity continued into the Early Jurassic.

(3) The proportions of Palaeozoic, Neoproterozoic and Mesopro-terozoic zircons show strong positive correlations with eachother. The source of this group is not clear. Although themetamorphic Pinoh Group of southern Borneo may havesupplied some of these zircons, there is no strong correlationwith Cretaceous zircons from the Schwaner Granites, whichintrude the Pinoh Group.

(4) The proportion of coloured and rounded Palaeoproterozoicand Archaean zircons strongly correlates with the propor-tion of Permian–Triassic zircons, and they probably repre-sent metasedimentary continental basement rocks of theTin Belt (Liew and Page, 1985).

(5) Ophiolitic material is always present but its abundance canbe estimated only from light and heavy mineral studies.

9. Discussion

Both the light and heavy mineral fractions of the Eocene–LowerMiocene deep marine Crocker Fan indicate the importance of first-cycle granitic source rocks. The light minerals are dominated byangular and poorly sorted igneous quartz and K-feldspar whenfeldspar is present. The heavy mineral fractions are dominated byunabraded zircon and tourmaline, suggesting limited sedimenttransport and a nearby source. The small but consistent presenceof chert fragments and unabraded chromian spinel indicates a con-tribution from the ophiolitic basement of northern Borneo. Thevariable amounts of garnet indicates supply from metamorphicrocks, which fluctuated with time. Volcanism, indicated by volca-nic quartz and zircons of ages similar to the depositional age ofthe sandstones, contributed small volumes of sediment in the Eo-cene. After the TCU sedimentation changed to shallow marineand fluvio-deltaic, and sediment recycling from the Crocker Fan be-came important.

There is a conspicuous discrepancy between the mature compo-sition and the immature texture of the Crocker Fan sandstones. Thetextures and zircon ages strongly suggest that the sandstones werederived from nearby sources in Borneo and the Tin Belt. The ma-ture composition can be explained by tropical weathering in a hu-mid climate that prevailed in Borneo and nearby SE Asiathroughout the Cenozoic (Morley, 1998). Thus, modal analysis ofsandstone composition (Dickinson et al., 1983) may be of limitedvalue for sandstones eroded and transported under humid tropicalconditions, because of the rapid preferential destruction of feldsparand unstable lithic fragments (Suttner et al., 1981). More reliableprovenance interpretations of tropical sandstones can be achievedby studies using combined petrographic (composition andtexture), heavy mineral and zircon age analysis. The intensity ofweathering in tropical climates can produce good qualityhydrocarbon reservoir sandstones in a short time. Although the

first-cycle sandstones of the Crocker Fan lack good reservoir prop-erties due to the relatively large amounts of clay-sized materialreducing porosity, the Miocene sandstones produced from recy-cling the Crocker Fan sandstones are highly quartzose sandstoneswith excellent reservoir properties.

It is suggested here that most material of the Crocker Fan wasderived from Borneo and nearby areas of SE Asia, rather thandistant sources in mainland Asia exposed as a results of theIndia–Eurasia collision (van Hattum et al., 2006). Fluvial transportfrom distal Indochina and eastern Himalayan sources would haveproduced greater rounding of material, including zircons, ratherthan the abundant unabraded grains that are observed. Further-more, detrital zircon ages of igneous origin do not match withigneous ages from these distant source areas. In contrast, the mostprominent detrital age groups do match well with Cretaceousgranites of the Schwaner Mountains of southern Borneo andPermian–Triassic granites and Precambrian basement of theMalay-Thai Tin Belt.

During deposition of the lower part of the Crocker Fan in the Eo-cene Cretaceous granites (Schwaner Mountains and adjacent Sun-da Shelf) contributed the majority of sediment. AFT data show thatgranites of the Schwaner Mountains were exhumed in the LateCretaceous (Sumartadipura, 1976), and could have produced largeamounts of sediment in the Eocene. Cretaceous AFT ages were alsoreported from Crocker Formation sandstones (Hutchison et al.,2000). Hall and Nichols (2002) suggested that northern Borneosediments were mostly derived from Borneo itself throughoutthe Neogene, and this study suggests that Borneo may have sup-plied large amounts of sediment since at least the Eocene. A depo-sitional and provenance model of the Crocker Fan at the end of theEocene based on the results of this study is shown in Fig. 10.

In the upper part of the Crocker Fan (Oligocene) the proportionof Permian–Triassic zircons increases, with ages corresponding tothose of the Tin Belt granites, as well as Palaeoproterozoic colouredand rounded zircons, probably from the Tin Belt basement. Zirconsof similar ages are relatively common in modern river sedimentsfrom the Malay peninsula (Sevastjanova et al., 2011, 2012). TheTin Belt is further away from northern Borneo than the SchwanerMountains, which may account for the slightly higher degree ofzircon rounding. AFT ages from the Tin Belt granites indicate amajor phase of exhumation in the Oligocene (ca. 24–33 Ma;Krähenbuhl, 1991; Kwan et al., 1992), and it is at this time thatthe Tin Belt granites probably became available for erosion andmaterial was transported to the Crocker Fan.

The heterogeneous character of indices of provenance-specificheavy mineral pairs (Morton and Hallsworth, 1999; Hurst andMorton, 2001) at outcrop scale suggests sediment supply into thedeep marine basin directly from an alluvial/fluvial system, by-passing any shallow marine shelf, which was probably much morelimited in width before Early Miocene closure of the Proto-SouthChina Sea. Large volumes of sediment derived from the SchwanerMountains of southern Borneo were being deposited in the Pro-to-South China Sea. The main drainage divide in southern Borneowas probably located considerably further south than its presentposition in the central Borneo mountains, which consist mainlyof Rajang Group and Crocker Fan deep marine sedimentary rocks.The history of emergence of this mountain range is poorly known,but followed Early Miocene collision of continental fragments withBorneo. Deep marine sedimentation of the Crocker Fan terminatedin the Early Miocene, during the final closure of the Proto-SouthChina Sea and collision of micro-continental fragments withBorneo.

Collision produced a major unconformity, the Top CrockerUnconformity, incising sediments of the Crocker Fan. Subsidenceand sedimentation resumed in the Early Miocene, with a drainagepattern similar to the present day. Fluvio-deltaic to shallow marine

Fig. 10. Depositional model for the Crocker Fan during the Late Eocene, showing the main source areas and transport paths.

280 M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282

siliciclastic sediments were deposited in western and northernSabah in a similar depositional and climatic setting, but with dif-ferent compositions reflecting different sources. The sandstonesof the Lower Miocene Meligan and Setap Shale of SW Sabah areprimarily a product of sedimentary recycling. The characteristicsof the quartzose sandstones are similar to those of the UpperCretaceous–Eocene Rajang Group and the Eocene–Lower MioceneCrocker Fan, but are compositionally and texturally more mature.The Crocker Fan started to supply large amounts of sediment tothe Lower Miocene basins. Local ophiolitic sources also continuedto supply sediment. The local generation of large amounts ofsediment is consistent with the ideas of Hall and Nichols (2002)that since the start of the Neogene Borneo itself supplied most ofthe material deposited in the basins on and around Borneo. Thesandstones of the Meligan and Setap Shale Formation are poten-tially excellent hydrocarbon reservoirs, and tropical weatheringprobably had a very important role in producing these highlyquartzose sandstones.

In contrast, sandstones of the Tajau Member of the KudatFormation are unlike any of the other Lower Miocene westernSabah sandstones. They are proximal debris flows and storm beds,and they are compositionally and texturally immature. The mainsources were granitic and high-grade metamorphic rocks from anearby area with smaller amounts of ophiolitic material. Therewere no fresh plutonic or metamorphic rocks exposed in northernBorneo during the Early Miocene. Potential metamorphic sourcerocks have been described by Encarnación et al. (1995) on Palawan,north of Kudat, where there are high-pressure subduction-relatedand high-temperature sub-ophiolitic metamorphic rocks. Theserocks contain garnet, kyanite, apatite, epidote, and abundantK-feldspar, which fits well with the detrital mineral assemblagesin the Tajau Sandstone Member. The metamorphic rocks onPalawan experienced peak metamorphic conditions in the earliest

Oligocene in a subduction setting, followed by rapid cooling andexhumation in the Early Miocene related to collision of the ReedBank and North Palawan continental fragments with northern Bor-neo and Palawan. In the Early Miocene Palawan started to supplysediment to northern Borneo in the Kudat area after the deep mar-ine basin between the areas was eliminated. The mature characterof the Sikuati Member of the Kudat Formation suggests that thePalawan source area was short-lived and recycling of the CrockerFan became more important from later in the Early Mioceneonwards.

10. Conclusions

The voluminous Eocene–Lower Miocene deep marine CrockerFan sediments were mostly derived from nearby acid plutonicsources on Borneo, the Malay Peninsula and the Sunda Shelf, by adrainage system different from today. Distant source areas inmainland Asia did not play a significant role. During the Eocenemostly Cretaceous material was deposited, probably from theSchwaner Mountains and adjacent areas of the Sunda Shelf, andduring the Oligocene an increasing amount of material was derivedfrom the Permian–Triassic Tin Belt granites and its Proterozoicmetasedimentary basement. Microcontinent collisions with Bor-neo in the Early Miocene terminated deep marine sedimentation,and changed the drainage pattern of Borneo and nearby SE Asia.

After closure of the Proto-South China Sea and cessation of deepmarine deposition of the Crocker Fan fluvio-deltaic to shallow mar-ine deposition occurred in basins on and around Borneo. Sand-stones of the Lower Miocene Setap Shale and Meligan Formationsof SW Sabah were mostly recycled from sediments of the RajangGroup and Crocker Fans on Borneo, now exposed in the mainmountain range of Borneo. A smaller amount of material was sup-plied by local ophiolitic sources.

M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282 281

The lowest Tajau Sandstone Member of the Lower Miocene Ku-dat Formation was mostly derived from fresh granitic and high-grade metamorphic rocks, from a nearby Palawan source. Palawanincludes rocks that rifted from South China and collided in theEarly Miocene; high-grade metamorphic rocks subducted in theOligocene became available for erosion at this time. This Palawansource was short-lived and sandstones of the overlying SikuatiMember of the Kudat Formation are more similar to the MeliganFormation and they were probably recycled from the CrockerFormation.

Tropical weathering can strongly influence the composition ofsandstones, and requires special attention when performing prov-enance studies. A wide array of different methods should be usedin order to achieve reliable provenance interpretations. Detritalgeochronology and heavy mineral analysis (including varietalstudies) yield valuable results, but detrital modes should be usedwith caution in tropical settings.

Acknowledgements

This project was sponsored by the SE Asia Research Group ofRoyal Holloway University of London, which is supported by a con-sortium of oil companies. The Economic Planning Unit of Malaysiaallowed us to collect samples in Sabah, and Allagu Balaguru helpedto conduct the fieldwork. Mark Barley’s help at the University ofWestern Australia is greatly appreciated. Discussions with JohnEncarnación helped a lot in reconstructing the provenance of theKudat Formation. We thank Theo van Leeuwen, D. Hendrawanand Rio Tinto Exploration Indonesia for help in obtaining granitesamples from Kalimantan.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.jseaes.2013.02.033.

References

Aitchison, J.C., Ali, J.R., Davis, A.M., 2007. When and where did India and Asiacollide? Journal of Geophysical Research 112, B05423. http://dx.doi.org/10.1029/2006JB004706.

Ali, J.R., Aitchison, J.C., 2012. Comment on ‘‘Restoration of Cenozoic deformation inAsia and the size of Greater India’’ by D.J.J. van Hinsbergen et al. Tectonics 31,TC4006.

Beckinsale, R.D., 1979. Granite magmatism in the tin belt of Southeast Asia. In:Atherton, M.P., Tarney, J. (Eds.), Origin of Granite Batholiths: GeochemicalEvidence. Shiva Publishing Limited, Orpington, UK, pp. 34–44.

Besse, J., Courtillot, V., 1991. Revised and synthetic polar wander maps of theAfrican, Eurasian, North American and Indian Plates, and true polar wandersince 200 Ma. Journal of Geophysical Research 96, 4029–4050.

Bol, A.J., van Hoorn, B., 1980. Structural style in western Sabah offshore. Bulletin ofthe Geological Society of Malaysia 12, 1–16.

Booth, A.L., Zeitler, P.K., Kidd, W.S.F., Wooden, J., Liu, Y., Idleman, B., Hren, M.,Chamberlain, C.P., 2004. U–Pb zircon constraints on the tectonic evolution ofsoutheastern Tibet, Namche Barwa Area. American Journal of Science 304, 889–929.

Cawood, P.A., Nemchin, A.A., Leverenz, A., Saeed, A., Ballance, P.F., 1999. U/Pb datingof detrital zircons: implications for provenance record of Gondwana marginterranes. Geological Society of America Bulletin 111, 1107–1119.

Clark, M.K., Schoenbohm, L.M., Royden, L.H., Whipple, K.X., Burchfiel, B.C., Zhang, X.,Tang, W., Wang, E., Chen, L., 2004. Surface uplift, tectonics, and erosion ofeastern Tibet from large-scale drainage patterns. Tectonics 23. http://dx.doi.org/10.1029/2002TC001402.

Clark, M.K., House, M.A., Royden, L.H., Whipple, K.X., Burchfiel, B.C., Zhang, X., Tang,W., 2005. Late Cenozoic uplift of the southeastern Tibet. Geology 33, 525–528.

Clement, J.F., Keij, J., 1958. Geology of the Kudat Peninsula, North Borneo(Compilation). Annual Report of the Geological Survey Department, GR783.

Cobbing, E.J., Pitfield, P.E.J., Darbyshire, D.P.F., Mallick, D.I.J. (Eds.), 1992. TheGranites of the South-East Asian Tin Belt. Overseas Memoir of the BritishGeological Survey, 10, 369 pp.

Collenette, P., 1958. The geology and mineral resources of the Jesselton-Kinabaluarea, North Borneo. Malaysia Geological Survey Borneo Region, Memoir 6, 194pp.

Collenette, P., 1965. The geology and mineral resources of the Pensiangan andUpper Kinabatangan area, Sabah. Malaysia Geological Survey Borneo Region,Memoir 12, 150 pp.

Cottam, M.A., Hall, R., Ghani, A.A., in press. Late Cretaceous and Paleogene tectonicsof the Malay peninsula constrained by thermochronology. Journal of AsianEarth Sciences.

de Keyser, F., Rustandi, E., 1993. Geology of the Ketapang Sheet Area, Kalimantan1:250,000. Geological Research and Development Centre, Bandung 42 pp.

Dickinson, W.R., Suczek, C.A., 1979. Plate tectonics and sandstone composition.American Association of Petroleum Geologists Bulletin 63, 2164–2182.

Dickinson, W.R., Beard, L.S., Brakenridge, G.R., Erjavic, J.L., Ferguson, R.C., Inman,K.F., Knepp, R.A., Lindeberg, F.A., Ryberg, P.T., 1983. Provenance of NorthAmerican Phanerozoic sandstones in relation to tectonic setting. GeologicalSociety of America Bulletin 94, 222–235.

Ding, L., Zhong, D., Yin, A., Kapp, P., Harrison, T.M., 2001. Cenozoic structural andmetamorphic evolution of the eastern Himalayan syntaxis (Namche Barwa).Earth and Planetary Science Letters 192, 423–438.

Dodson, M.H., Compston, W., Williams, I.S., Wilson, J.F., 1988. A search for ancientdetrital zircons in Zimbabwean sediments. Journal of the Geological Society,London 145, 977–983.

Encarnación, J., Mukasa, S.B., 1997. Age and geochemistry of an ‘anorogenic’ crustalmelt and implications for I-type granite petrogenesis. Lithos 42, 1–13.

Encarnación, J.P., Essene, E.J., Mukasa, S.B., Hall, C.H., 1995. High-pressure and -temperature subophiolitic kyanite-garnet amphibolites generated duringinitiation of Mid-Tertiary subduction, Palawan, Philippines. Journal ofPetrology 36, 1481–1503.

Garson, M.S., Young, F.B., Mitchell, A.H.G., Tait, B.A.R., 1975. The geology of the tinbelt in peninsular Thailand around Phuket, Phangnga and Takua Pa: Institute ofGeological Sciences, London, Overseas Memoir 1, 112 pp.

Haile, N.S., McElhinny, M.W., McDougall, I., 1977. Palaeomagnetic data andradiometric ages from the Cretaceous of West Kalimantan (Borneo), and theirsignificance in interpreting regional structure. Journal of the Geological Societyof London 133, 133–144.

Hall, R., 1996. Reconstructing Cenozoic SE Asia. In: Hall, R., Blundell, D.J. (Eds.),Tectonic Evolution of SE Asia, Geological Society of London Special Publication106, pp. 153–184.

Hall, R., 2002. Cenozoic geological and plate tectonic evolution of SE Asia and theSW Pacific: computer-based reconstructions, model and animations. Journal ofAsian Earth Sciences 20, 353–434.

Hall, R., 2012. Late Jurassic–Cenozoic reconstructions of the Indonesian region andthe Indian Ocean. Tectonophysics 570–571, 1–41.

Hall, R., in press. Contraction and extension in Northern Borneo driven bysubduction rollback. Journal of Asian Earth Sciences.

Hall, R., Morley, C.K., 2004. Sundaland basins. In: Clift, P., Wang, P., Kuhnt, W.,Hayes, D.E. (Eds.), Continent–Ocean Interactions within the East Asian MarginalSeas, Geophysical Monograph. American Geophysical Union 149, pp. 55–85.

Hall, R., Nichols, G.J., 2002. Cenozoic sedimentation and tectonics in Borneo:climatic influences on orogenesis. In: Jones, S.J., Frostick, L. (Eds.), Sediment Fluxto Basins: Causes, Controls and Consequences, Geological Society LondonSpecial Publication 191, pp. 5–22.

Hall, R., Sevastjanova, I., 2012. Australian crust in Indonesia. Australian Journal ofEarth Sciences 59, 827–844.

Hall, R., Wilson, M.E.J., 2000. Neogene sutures in eastern Indonesia. Journal of AsianEarth Sciences 18, 787–814.

Hall, R., van Hattum, M.W.A., Spakman, W., 2008. Impact of India–Asia collision onSE Asia: the record in Borneo. Tectonophysics 451, 366–389.

Hamilton, W., 1979. Tectonics of the Indonesian Region 1078, U.S. Geological SurveyProfessional Paper, 345 pp.

Hazebroek, H.P., Tan, D.N.K., 1993. Tertiary tectonic evolution of the NW SabahContinental Margin. Bulletin of the Geological Society of Malaysia 33, 195–210.

Holloway, N.H., 1982. North Palawan block – its relation to Asian mainland and rolein evolution of South China Sea. American Association of Petroleum GeologistsBulletin 66, 1355–1383.

Hosking, K.F.G., 1971. The offshore tin deposits of Southeast Asia. United NationsECAFE CCOP Technical Bulletin 5, 112–129.

Hurst, A., Morton, A.C., 2001. Generic relationships in the mineral–chemicalstratigraphy of turbidite sandstones. Journal of the Geological Society ofLondon 158, 401–404.

Hutchison, C.S., 1978. Ophiolite metamorphism in northeast Borneo. Lithos 11,195–208.

Hutchison, C.S., 1989. Geological Evolution of South-East Asia, 13, OxfordMonographs on Geology and Geophysics, Clarendon Press, 376 pp.

Hutchison, C.S., 1996. The ‘Rajang Accretionary Prism’ and ‘Lupar Line’ problem ofBorneo. In: Hall, R., Blundell, D.J. (Eds.), Tectonic Evolution of SE Asia, GeologicalSociety London Special Publication 106, pp. 247–261.

Hutchison, C.S., Bergman, S.C., Swauger, D.A., Graves, J.E., 2000. A Miocenecollisional belt in north Borneo: uplift mechanism and isostatic adjustmentquantified by thermochronology. Journal of the Geological Society of London157, 783–793.

Jasin, B., 1991. Some larger foraminifera and radiolaria from Telupid olistostrome,Sabah. Warta Geologi, Geological Society of Malaysia Newsletter 17, 225–230.

Jasin, B., Tahir, S., Harun, Z., 1995. Some Miocene planktonic foraminifera fromBidu-Bidu area, Sabah. Warta Geologi, Geological Society of MalaysiaNewsletter 21, 241–246.

Johnsson, M.J., Stallard, R.F., Meade, R.H., 1988. First-cycle quartz arenites in theOrinoco River basin, Venezuela and Colombia. Journal of Geology 96, 263–277.

282 M.W.A. van Hattum et al. / Journal of Asian Earth Sciences 76 (2013) 266–282

Krähenbuhl, R., 1991. Magmatism, tin mineralization and tectonics of the MainRange, Malaysian Peninsula: consequences for the plate tectonic model ofSoutheast Asia based on Rb–Sr, K–Ar and fission track data. Bulletin of theGeological Society of Malaysia 29, 1–100.

Kwan, T.S., Krahenbuhl, R., Jager, E., 1992. Rb–Sr, K–Ar and fission-track ages forgranites from Penang Island, West Malaysia – an interpretation model for Rb–Srwhole-rock and for actual and experimental mica data. Contributions toMineralogy and Petrology 111, 527–542.

Levell, B.K., 1987. The nature and significance of regional unconformities in thehydrocarbon-bearing Neogene sequences offshore west Sabah. GeologicalSociety of Malaysia Bulletin 21, 55–90.

Liechti, P., Roe, F.W., Haile, N.S., 1960. The Geology of Sarawak, Brunei and theWestern Part of North Borneo. Geological Survey Department, British Territoriesof Borneo, Bulletin 3, 360 pp.

Liew, T.C., Page, R.W., 1985. U–Pb zircon dating of granitoid plutons from the westcoast of Peninsular Malaysia. Journal of the Geological Society of London 142,515–526.

Lim, P.S., Heng, Y.E., 1985. Geological Map of Sabah, 1:500,000. Geological Survey ofMalaysia.

Maluski, H., Proust, F., Zuchang, X., 1983. Age du magmatisme calco-alcalin trans-himalayen du Tibet meridional: premiers resultats obtenus par la methode39Ar–40Ar. In: Mercier, J.L., Guangcen, L. (Eds.), Mission Franco-Chinoise au Tibet1980. CNRS, Paris France, pp. 335–339.

Mange, M.A., Maurer, H.F.W., 1992. Heavy Minerals in Colour. Chapman & Hall,London, 147 pp.

Metcalfe, I., 1996. Pre-Cretaceous evolution of SE Asian terranes. In: Hall, R.,Blundell, D.J. (Eds.), Tectonic Evolution of SE Asia, Geological Society LondonSpecial Publication 106, pp. 97–122.

Métivier, F., Gaudemer, Y., Tapponnier, P., Klein, M., 1999. Mass accumulationrates in Asia during the Cenozoic. Geophysical Journal International 137,280–318.

Mitchell, A.H.G., 1979. Rift-, subduction- and collision-related tin belts. Bulletin ofthe Geological Society of Malaysia 11, 81–102.

Morley, R.J., 1998. Palynological evidence for Tertiary plant dispersal in the SE Asianregion in relation to plate tectonics and climate. In: Hall, R., Holloway, J.D.(Eds.), Biogeography and Geological Evolution of SE Asia. Backhuys Publishers,pp. 211–234.

Morley, C.K., 2002. A tectonic model for the Tertiary evolution of strike-slip faultsand rift basins in SE Asia. Tectonophysics 347, 189–215.

Morton, A.C., 1984. Stability of detrital heavy minerals in Tertiary sandstones fromthe North Sea Basin. Clay Minerals 19, 287–308.

Morton, A.C., Hallsworth, C.R., 1994. Identifying provenance-specific features ofdetrital heavy mineral assemblages in sandstones. Sedimentary Geology 90,241–256.

Morton, A.C., Hallsworth, C.R., 1999. Processes controlling the composition of heavymineral assemblages in sandstones. Sedimentary Geology 124, 3–29.

Najman, Y., Appel, E., Boudagher-Fadel, M., Bown, P., Carter, A., Garzanti, E., Godin,L., Han, J., Liebke, U., Oliver, G., Parrish, R., Vezzoli, G., 2010. Timing of India–Asiacollision: geological, biostratigraphic, and palaeomagnetic constraints. Journalof Geophysical Research 115, B12416. http://dx.doi.org/10.1029/2010JB007673.

Petronas, 1999. The Petroleum Geology and Resources of Malaysia, Petronas, KualaLumpur, Malaysia, 665 pp.

Pickard, A.L., Adams, C.J., Barley, M.E., 2000. Australian provenance for UpperPermian to Cretaceous rocks forming accretionary complexes on the NewZealand sector of the Gondwanaland margin. Australian Journal of EarthSciences 47, 987–1007.

Pieters, P.E., Sanyoto, P., 1993. Geology of the Pontianak/Nangataman Sheet area,Kalimantan: 1:250,000. Geological Research and Development Centre, Bandung,Indonesia.

Pupilli, M., 1973. Geological evolution of South China Sea Area: tentativereconstruction from borderland geology and well data. In: ProceedingsIndonesian Petroleum Association 2nd Annual Convention, pp. 223–242.

Rowley, D.B., 1996. Age of initiation of collision between India and Asia: a review ofstratigraphic data. Earth and Planetary Science Letters 145, 1–13.

Rutten, L.M.R., 1915. Studien über Foraminiferen aus Ost-Asien. 9. Tertiäreforaminiferen von den Inseln Balambangan und Banguey, nördl. von Borneo.:Sammlungen Geologischen Reichsmuseums, Leiden, Series I, v 10.

Rutten, L.M.R., 1925. Over fossielhoudende tertiaire kalkstenen uit Britsch NoordBorneo. Verhandelingen Koninklijk Nederlands Geologisch en MijnbouwkundigGenootschap, Geologische Serie 8, 415–427.

Sevastjanova, I., Clements, B., Hall, R., Belousova, E.A., Griffin, W.L., Pearson, N.,2011. Granitic magmatism, basement ages, and provenance indicators in theMalay Peninsula: insights from detrital zircon U–Pb and Hf-isotope data.Gondwana Research 19, 1024–1039.

Sevastjanova, I., Hall, R., Alderton, D., 2012. A detrital heavy mineral viewpoint onsediment provenance and tropical weathering in SE Asia. Sedimentary Geology280, 179–194.

Smith, J.B., Barley, M.E., Groves, D.I., Krapez, B., McNaughton, N.J., Bickle, M.J.,Chapman, H.J., 1998. The Sholl Shear Zone, west Pilbara: evidence for a domainboundary structure from integrated tectonostratigraphic analyses, SHRIMP U–Pb dating and isotopic and geochemical data of granitoids. PrecambrianResearch 88, 143–171.

Stauffer, P.H., 1967. Studies in the Crocker Formation, Sabah. Geological Survey ofMalaysia, Borneo Region Bulletin 8, 1–13.

Stephens, E.A., 1956. The geology and mineral resources of the Kota Belud andKudat area, North Borneo. Malaysia Geological Survey Borneo Region, Memoir5, 137 pp.

Sumartadipura, A.S., 1976. Geologic Map of the Tewah Quadrangle, CentralKalimantan – 1:250,000. Geological Survey of Indonesia, Directorate ofMineral Resources, Geological Research and Development Centre, Bandung.

Suttner, L.J., Basu, A., Mack, G.H., 1981. Climate and the origin of quartz arenites.Journal of Sedimentary Petrology 51, 1235–1246.

Suzuki, S., Takemura, S., Yumul Jr., G.P., David Jr., S.D., Asiedu, D.K., 2000.Composition and provenance of the Upper Cretaceous to Eocene sandstonesin Central Palawan, Philippines: constraints on the tectonic development ofPalawan. The Island Arc 9, 611–626.

Tan, D.N.K., Lamy, J.M., 1990. Tectonic evolution of the NW Sabah continentalmargin since the Late Eocene. Bulletin of the Geological Society of Malaysia 27,241–260.

Tanean, H., Paterson, D.W., Endarto, M., 1996. Source provenance interpretation ofKutei Basin sandstones and the implications for the tectono-stratigraphicevolution of Kalimantan. In: Proceedings Indonesian Petroleum Association25th Annual Convention, pp. 333–345.

Tapponnier, P., Peltzer, G., Armijo, R., 1986. On the mechanism of collision betweenIndia and Asia. In: Coward, M.P., Ries, A.C. (Eds.), Collision Tectonics, GeologicalSociety of London Special Publication. Geological Society London SpecialPublication 19, pp. 115–157.

Tate, R.B., 2001. The Geology of Borneo Island. Geological Society of Malaysia.Taylor, B., Hayes, D.E., 1983. Origin and history of the South China Sea Basin. In:

Hayes, D.E. (Ed.), The Tectonic and Geologic Evolution of Southeast Asian Seasand Islands, Part 2. American Geophysical Union, Geophysical MonographsSeries 27, pp. 23–56.

Tongkul, F., 1995. The Paleogene basins of Sabah, East Malaysia. Bulletin of theGeological Society of Malaysia 37, 301–308.

Tucker, M., 1988. Techniques in Sedimentology. Blackwell Scientific Publications,Oxford, 394 pp.

van der Vlerk, I.M., 1951. Tabulation of determinations of larger foraminifera. In:Reinhard, M., Wenk, E. (Eds.), Geology of the colony of North Borneo BritishTerritories of Borneo, Geological Survey Department, Bulletin 1, pp. 137–146(Appendix II).

van Hattum, M.W.A., 2005. Provenance of Cenozoic Sedimentary Rocks of NorthernBorneo. PhD Thesis, University of London, 457 pp.

van Hattum, M.W.A., Hall, R., Pickard, A.L., Nichols, G.J., 2006. SE Asian sedimentsnot from Asia: provenance and geochronology of North Borneo sandstones.Geology 34, 589–592.

van Hinsbergen, D.J.J., Kapp, P., Dupont-Nivet, G., Lippert, P.C., DeCelles, P.G.,Torsvik, T.H., 2011. Restoration of Cenozoic deformation in Asia and the size ofGreater India. Tectonics 30. http://dx.doi.org/10.1029/2011TC002908.

White, L.T., Lister, G.S., 2012. The collision of India with Asia. Journal ofGeodynamics 56–57, 7–17.

William, A.G., Lambiase, J.J., Back, S., Jamiran, M.K., 2003. Sedimentology of the JalanSelaiman and Bukit Melinsung outcrops, western Sabah: is the West CrockerFormation an analogue for Neogene turbidites offshore? Bulletin of theGeological Society of Malaysia 47, 63–75.

Williams, P.R., Johnston, C.R., Almond, R.A., Simamora, W.H., 1988. Late Cretaceousto Early Tertiary structural elements of West Kalimantan. Tectonophysics 148,279–298.

Wilson, R.A.M., 1961. The geology and mineral resources of the Banggi Island andSugut River area, North Borneo. British Territories Borneo Region GeologicalSurvey Department, Memoir 15, 143 pp.

Wilson, R.A.M., Wong, N.P.Y., 1962. Labuan and Padas Valley Area, North Borneo(Memoir 17) – progress report. British Borneo Geological Survey Annual Report1962, 191–208.

Wilson, R.A.M., Wong, N.P.Y., 1964. The geology and mineral resources of theLabuan and Padas Valley areas, Sabah. Geological Survey of Malaysia, BorneoRegion, Memoir 17, 150 pp.

Wong, N.P.Y., 1962. Progress Report. British Geological Survey Borneo Region,Annual Report 1962, 200–207.