Assessing effective provenance methods for fluvial

sediment in the South China Sea

PETER D. CLIFT

Department of Geology and Geophysics, Louisiana State University,

Baton Rouge, LA 70803, USA (e-mail: pclift@lsu.edu)

Abstract: Sediment is delivered by the rivers of SE Asia to the South China Sea where it providesan archive of continental environmental conditions since the Eocene. Interpreting this archive iscomplicated because sediment may be derived from a number of unique sources and the riversthemselves have experienced headwater capture that also affects their composition. A numberof methods exist to constrain provenance, but not all work well in this area. Anthropogenicimpacts, most notably agriculture, mean that the modern rivers contain more weathered materialsthan they did up until about 3000 years ago. The rivers have also changed their bulk chemistry andclay mineralogy in response to climate change, so that these proxies, as well as Sr isotopes, aregenerally unreliable provenance indicators. Nd isotopes resolve influx from Luzon, but manyother sources in SE Asia have similar values and clear resolution of end members can be difficult.Instead, thermochronology methods are best suited, especially apatite fission track, which showsmore diversity in the sources than either U–Pb zircon or Ar/Ar muscovite dating. Nonetheless,even fission track is best used as part of a multiproxy approach if a robust quantitative budgetis desired.

The South China Sea is one of the largest marginalbasins in the Western Pacific and has been the repos-itory of large volumes of clastic sediment deliveredto the continental margins by the large rivers thatdrain the East Asian continent. In theory, these sedi-mentary records could be used to decipher the his-tory of tectonism, surface uplift and environmentalevolution in this region. Such data are essential ifwe are to understand the relationships betweensolid Earth evolution and the development of cli-mate in the aftermath of the India–Asia collision,most notably the intensification of the East Asianmonsoon whose history of activity is still not wellunderstood (Sun & Wang 2005; Clift et al. 2014).However, if we are to read and interpret these sedi-mentary records then it is necessary to first under-stand where the sediment in each sub-basin wasderived from because changes in sediment sourcemay also result in changes in composition and min-eralogy, which could be mistaken for changes inenvironmental conditions when in reality theymerely reflects derivation from source bedrocks withdifferent bulk compositions. Unless the sedimentsource can be properly assigned then other datasets cannot be used to their full potential in under-standing how processes such as the intensifica-tion of the East Asian monsoon have impacted thecontinental environment over the last 65 millionyears. In this paper I review a number of the morecommonly applied methods for constraining the

source of sediment into this basin and explore theeffectiveness of each in allowing us to understandhow sediment is dispersed after its delivery tothe ocean.

Sediment provenance is particularly important inthe South China Sea because such methods havebeen used to track sediment transport within thebasin and subsequently to infer the influence of bot-tom currents, which are in turn controlled by theopening and closure of gateways (Lei et al. 2007),especially those into the Western Pacific, such asthe Bashi Straits between Taiwan and Luzon(Fig. 1). Furthermore, the three large rivers thatdrain into the basin, the Mekong, the Red (SongHong) and the Pearl Rivers have all been proposedto have experienced significant amounts of headwa-ter drainage capture, as a result of the eastwards tilt-ing of the Asian continent during the uplift of theTibetan Plateau (Brookfield 1998; Clark et al.2004). The timing of this reorganization is contro-versial (Clift et al. 2006a; Robinson et al. 2013;Zheng et al. 2013), but is also important for testingmodels of surface uplift in Tibet and surroundingregions. If we are not able to pinpoint the sourceof the sediment within the delta and submarine fansystems in the marginal seas then it is impossibleto fingerprint the influence of one river comparedto another and thus to isolate the potential impactof drainage capture. Developing robust provenancetools is the first stage in addressing this process.

From: Clift, P. D., Harff, J., Wu, J. & Yan, Q. (eds) River-Dominated Shelf Sediments of East Asian Seas.Geological Society, London, Special Publications, 429, http://doi.org/10.1144/SP429.3# 2015 The Author(s). Published by The Geological Society of London. All rights reserved.For permissions: http://www.geolsoc.org.uk/permissions. Publishing disclaimer: www.geolsoc.org.uk/pub_ethics

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

Most of the methods that I discuss are not par-ticularly novel, but represent standard methodolo-gies that are applied to many sedimentary basinsaround the world. The South China Sea repre-sents a special challenge because of the diversityof possible sources, but also provides an exampleof how, even in such a relatively complicated set-ting, effective sediment provenance reconstruc-tions can be achieved.

Geological setting

The tectonic origins of the South China Sea havebeen strongly debated, but it is reasonably clearthat the region began to experience significant con-tinental extension during the Eocene (Ru & Pigott1986; Franke 2013), culminating in the onset of sea-floor spreading around 28 Ma (Briais et al. 1993;Barckhausen et al. 2014). Seafloor spreading ceased

Fig. 1. Shaded bathymetric and topographical map of the South China Sea and surrounding continental areas showingthe major drainage systems that are feeding sediment into the basin together with the Molengraaff River, which wasactive during sea level below stands at least in the recent geological past (Voris 2000). The map is from GeoMapApp.Bathymetric contours are in 1000 m increments. VCH, Vietnamese Central Highlands; KM, Kontum Massif; KP,Khorat Plateau; SC, Song Chay Massif.

P. D. CLIFT

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

by 17 Ma, after which time the region has largelybeen affected by thermal subsidence, although dis-rupted by localized neotectonic activity, such as thevolcanism and uplift in Hainan island (Shi et al.2011). The most important tectonic event to haveinfluenced sediment supply to the basin after theend of extension has been the collision betweenthe oceanic island arc of Luzon with the passivemargin of China, as manifest in the island of Taiwan(Suppe 1984; Huang et al. 2006). Taiwan is one ofthe world’s great sediment sources and exceedsthe Ganges–Brahmaputra delta in supplying sedi-ment to the ocean despite its small size (Milliman& Syvitski 1992).

As well as receiving sediment from eastern andsouthern Taiwan, the South China Sea is suppliedby sediment from smaller rivers on the PhilippineIslands to the east, from Borneo to the south, but,more importantly, from three major rivers alongthe western and northern sides of the basin, andnamely the Mekong, Red and Pearl rivers (Fig. 1).These, respectively, carry loads of 160, 170 and69 Mt a21 (Milliman & Syvitski 1992; Le et al.2007). These numbers can only be considered as arough guide, as the budgets are typically only forthe suspended load, are not always taken near theriver mouth and may be affected by anthropogenicdisruption of the basin. Despite the large size ofsome of these drainage systems, the two largest riv-ers in SW Taiwan – the Tsengwen and the Kaopingrivers – have measured pre-modern sediment loadsof 31 and 49 Mt a21, respectively, and the long-termrecent discharge into the Taiwan Strait exceeds380 Mt a21 (Kao & Milliman 2008). It is note-worthy that Taiwan is also struck by many largetyphoons originating in the central Pacific. Thesetyphoons result in significant erosion and sedimentdischarge, with Typhoon Herb in 1996 alone beingresponsible for the discharge of 142 Mt into theSouth China Sea in a single week (Milliman &Kao 2005). Although Taiwan cannot have been asignificant sediment source before the uplift of thepresent ranges after around 6 Ma (Huang et al.2006), it is certainly worth considering the sedi-ment-producing potential of each possible sourcewhen making volumetric assessments of the contri-bution from different end members to the deeperpart of the South China Sea.

Basis of provenance

The ability to distinguish and estimate the amount ofsediment derived from a given source is mostlybased on the concept that the bedrock sources pro-viding the sediment are distinguished from oneanother in different parts of the potential sourcearea on the basis of their chemistry, geochronology

or tectonic evolution and that these differences aretransferred from the bedrock to the sediment in therivers. Southeast Asia is remarkably suitable forsuch provenance work because of the assemblageof a number of contrasting tectonic blocks or ter-ranes in this region (Fig. 2). These were largelybrought into juxtaposition during the Triassic Indo-sinian Orogeny (Carter et al. 2001; Lepvrier et al.2004; Carter & Clift 2008), with later additions,especially in the Tibetan headwaters of the largerivers during the final collision between India andEurasia. Because of their contrasting geologicalhistories, tectonic blocks shown in Figure 2 pro-duce sediment of different composition or geochro-nological age, which can then be detected in thesediments deposited in the South China Sea. Thegeological evolution of each of these blocks is rela-tively complicated, but it is possible to say thatessentially southern China, Cathaysia, represents atectonic block that collided with the Yangtze Cra-ton at approximately 800 Ma and that subsequentlythis was the host to a Mesozoic volcanic arc com-plex (Hutchison 1994; Fletcher et al. 2004). In con-trast, the central part of China is dominated by theancient crust of the Yangtze Craton, which itselfcollided with the North China Block during the Tri-assic (Hu et al. 2006). The Songpan Garze Terrane,which no longer provides sediment directly into theSouth China Sea, represents an accretionary com-plex formed during this collision between northand southern China (Zhou & Graham 1996; Huanget al. 2003; Enkelmann et al. 2007).

On the western side of the basin, the IndochinaPeninsula is dominated by a separate tectonicblock, but one that was also involved in the Indosi-nian Orogeny (Carter et al. 2001). Indochina hasundergone more recent deformation as a result ofthe emplacement of flood basalt sequences in theCentral Highlands of Vietnam during the Late Mio-cene (Carter et al. 2000). This is one of a number ofrather enigmatic volcanic provinces around thebasin, which provide isotopically unique sedimentfrom newly emplaced primitive volcanic sequences.Likewise, on the eastern side of the basin, the islandarc of Luzon has provided some sediment into thebasin in the recent geological past, although it isworth noting that its location to the east of themain deep-water basin is a relatively recent devel-opment following the northwards drift of the arcand the collision of the arc with the southern marginof mainland Eurasia (Fig. 3). During the latest Mio-cene, plate tectonic reconstructions show that Luzonwas positioned somewhat to the south of the basinand has only been able to supply sediment to thebasin in the last few million years (Hall 2002).

The rivers draining the island of Borneo areprobably the least well defined of any potentialsource around the sea, but they too have had a

PROVENANCE IN THE SOUTH CHINA SEA

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

limited impact on sediment flux into the ocean,partly because they were separated from the pre-sent basin by a palaeo-South China Sea until around16 Ma when the southern margin of the basin, dom-inated by the Dangerous Grounds, began to collidewith the northern margin of Borneo (Hutchison2005; Clift et al. 2008a). Even since that time, directsediment supply from Borneo into the deeper partsof the South China Sea has been restricted bythe tectonic topography of the Dangerous Grounds(Hutchison & Vijayan 2010), and particularlyby the long and deep North Borneo Trough thatseparates Borneo from the main part of the seaand which acts as a very effective sediment trap(Hutchison 2010), so that only plumes of suspendedrelatively fine-grained sediment can be transporteddeep into the basin.

Consideration of such plate tectonic reconstruc-tions is very important when making provenanceassessments because it is clearly impossible toderive sediment from a tectonic block that was notin the vicinity of the South China Sea at the timeof sedimentation. Conversely, convincing prove-nance data can help us to better define the tectonic

development of SE Asia by showing which blockswere present at any particular time.

Sediment mixing on the continental shelf

Provenance analysis of sediment in the deep basindoes not necessarily reflect the contribution of asingle river system to the overall basin budgetbecause of filtering of the signal through the conti-nental shelf. Changes in sea level and, therefore, inthe position of the river mouth relative to the conti-nental shelf edge affects how important that riverwill be in supplying sediment into the deep basin.For example, the Mekong River appears to havebeen important in supplying sediment into the deepSW part of the basin during sea-level lowstands, buthas been relatively cut-off following post-glacialsea-level rise (Szczucinski et al. 2013). The sameis true of the Red and Pearl rivers, with their widecontinental shelves. In contrast, areas where thecontinental shelf is very narrow, such as offshorecentral Vietnam, may be important sources of sedi-ment to the deep basin even during periods of

Fig. 2. Simplified tectonic terrane map of East and SE Asia showing the major blocks discussed in this paper and thecourses of the rivers. After Metcalfe (1996).

P. D. CLIFT

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

relatively high sea level (Schimanski & Stattegger2005; Lahajnar et al. 2007; Szczucinski et al. 2009).

Reworking on the continental shelf is impor-tant in influencing the net contribution to the deepbasin. Sediment originating from the MekongRiver catchment to the deep basin is now delivered,not from modern sediment discharge at the rivermouth, but from erosion of the lowstand Mekongdelta at the shelf edge (Dung et al. 2013). Newlydelivered sediment is, instead, sequestered close tothe coast. Geochemical data from South Asia hasreinforced the idea that the continental shelf is alocation in which sediment from different sourcesmay be mixed via longshore transport and thatreworking of older deposits, due to storms or bottomcurrents, may influence the composition of sedimentdelivered to the deep sea, and, in particular, this doesnot necessarily reflect the large river mouth in closeproximity (Limmer et al. 2012).

As a result, fingerprinting of sediments in mod-ern rivers can help to resolve the influence of

onshore basins in supplying sediment layers to thecontinental shelf and to the deep ocean. Sedimentdeposited in the offshore, even close to the rivermouth, cannot be considered a reliable fingerprintof provenance.

Clay minerals

Clay minerals can provide an important source ofsediment provenance data, especially in distal fine-grained sediment sequences where single grainmethods may not be applicable, most sand beingsequestered on the shelf in many systems. Themethod is based on the contrasting mineralogy ofsediments in the different river systems that cur-rently supply the basin and the documented changesin clay mineralogy of the modern seafloor acrossthe basin (Chen 1978). The method suffers frombeing only ‘semi-quantitative’, in that relative pro-portions of different key minerals are determined

Fig. 3. Palaeogeographical map of the South China Sea at 20 Ma showing that the Luzon arc was only beginning tobecome a potential sediment source region at this time. Prior to this, sediment flux from such primitive arc sources wouldhave been impossible. Modified from P. Clift et al. (2008a).

PROVENANCE IN THE SOUTH CHINA SEA

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

by a variety of methods such as X-ray diffractiondata, with different methods producing slightly dif-ferent estimates that may not all be completely inaccordance with one another (Holtzapffel 1985;Moore & Reynolds 1989; Hillier 2003). The accu-racy of the methods is not entirely clear and is bestapplied to resolving relatively larger differencesin assemblages (.5%). Smaller differences in claymineral assemblage cannot be considered robustand are not considered as effective provenance tools.

A number of studies have now highlighted thefact that the Pearl, Red and Mekong rivers, as wellas smaller rivers draining Borneo and the MalayPeninsula, are characterized by different clay min-eral assemblages that roughly correlate with theintensity of chemical weathering in the fluvialbasin (Liu et al. 2007b, 2012). Although some ofthe differences in clay mineral content are relatedto source rock compositions, much of the contrastis the result of a variable intensity of chemicalweathering, which in turn is linked to the topogra-phy. The intensity of the summer monsoon, whichprovides much of the moisture and related warmthto the area, also modulates the rate of chemicalweathering and, thus, the clay mineralogy (Westet al. 2005), accounting for the abundance of smec-tite and kaolinite in the more tropical regions of thesouthern South China Sea, especially the SundaShelf (Aoki 1976; Chen 1978).

Composition is important in the case of the riversdraining Luzon because these rivers are rich insmectite, which is a product of the chemical break-down of volcanic rocks (Liu et al. 2009) that areabundant in this volcanic arc, as well as in the rangesof central Vietnam (Jagodzinski 2005). In contrast,although the rocks of Taiwan originated as sedi-ments on the passive margin of China, these haveexperienced significant metamorphism and arenow dominated by illite and chlorite, which largelyrepresent the products of physical erosion of thehigh mountains in Taiwan rather than the productsof chemical weathering of pre-existing bedrock.Meanwhile, sediment delivered by the Pearl Riveris dominated by kaolinite with some illite and chlo-rite influx (Liu et al. 2007a). Further south, theMekong submarine delta, which is the primarydepocentre for the modern river, is dominated byillite with lesser amounts of smectite, kaolinite andchlorite (Szczucinski et al. 2013). Further illite issupplied from the rivers of northern Borneo (Liuet al. 2007c).

Differences in clay mineralogy of Holocene andrecent sediment in the deep basin, as well as thosefound in older deposits, have been used to separatethe different contributions from potential sources(Boulay et al. 2005; Liu et al. 2010b). Figure 4shows a triangular diagram and the type of mixingarrays that have been proposed in the past to

separate the influence of these different source ter-rains. In this particular example, we see that sedi-ment from Ocean Drilling Program (ODP) Site1144 lies close to the field of Taiwan, allowing theprovenance of this deposit to be constrained to thisisland (Hu et al. 2012). Similar approaches havebeen applied to modern sediment in the northernSouth China Sea (Z. Liu et al. 2003; J. Liu et al.2014) and used to infer sediment transport direc-tions, driven by bottom currents.

Unfortunately, this method is prone to problemsbecause it is based on the assumption that river claymineral assemblages have been the same in the pastas they are at the present. There seems little doubtthat environmental changes have caused changesin clay assemblages over long periods of geologicaltime (Clift et al. 2002; Clift et al. 2014; Wan et al.2007), largely as a result of changes in the monsoon,so that the end members cannot be considered sta-ble over long periods of time as the climate in SEAsia has evolved. Furthermore, the clay mineralevidence for the origin of the Holocene sedimentat ODP Site 1144 appears to show that the sedimentis not entirely derived from Taiwan, despite the factthat other proxies indicate that it would seem to bemost likely (Hu et al. 2012). Why then do thesedata not plot directly in the modern Taiwan field?The difference was attributed by Hu et al. (2012)to weathering of the sediment on the exposed conti-nental shelf during the Last Glacial Maximum andthen reworking during the Holocene before finalsedimentation. Sediment buffering between sourceand sink is often associated with chemical weather-ing (Lupker et al. 2012), the net result of which is atransformation of the clay mineral assemblage rela-tive to the modern river composition. The method isalso open to error if the sequences are affected byburial diagenesis, as may be the case in deep bore-holes where burial heating can be significant.

An additional complexity was recognized bySteinke et al. (2008), who showed that the clay min-eralogy at any one site on the Sunda Shelf waslargely controlled by sea level, which acted to dis-rupt the large drainage systems that existing duringthe Last Glacial Maximum. In particular, sedimentrich in kaolinite derived from the south, as well asfrom the exposed Sunda Shelf itself, is stronglyreduced during the Holocene as flux from theMekong increased relative to the southern sources,as their river mouths retreated from proximity tothe shelf edge. Moreover, these workers recognizedthat, as sea level rose, little of this material wasreaching the deep basin, but, rather, was sequesteredin terrestrial flood plains and submarine delta clino-forms. Furthermore, studies of the clays around theMekong delta shows that these differ from east towest in the modern system, partly reflecting thepreferential settling of some clays close to the river

P. D. CLIFT

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

mouth (Xue et al. 2014). Sediments from the east,on the South China Sea side of the delta, look likethe Mekong River itself, while those to the westhave much less illite and chlorite, and imply reducedfine particle inputs from the Mekong River towardsthe Gulf of Thailand.

Most seriously of all is the recognition that thecomposition of the modern rivers is not in equilib-rium and may bear little resemblance to the naturalstate of the river before a few thousand years ago. Ithas increasingly been demonstrated that human set-tlement has had a major impact on landscape evolu-tion and, in particular, has encouraged the rapiderosion of soils as agriculture became prevalentover the last several thousand years (Syvitski et al.2005; Syvitski & Kettner 2011). Data from thedelta of the Pearl River suggests that human settle-ment in southern China had become highly disrup-tive to the river by around 2500 years ago (Huet al. 2013) and similar patterns might be anticipatedfor Indochina. More recently, the development ofmore sophisticated industrial cultures has resultedin the damming of major river systems so that therate of sediment delivery to the ocean has beenreduced in recent times.

A number of studies have now shown that thealteration state of sediment reaching the deltasince the onset of agriculture has increased (Bayonet al. 2012; Hu et al. 2013), consistent with the

increased influence from anthropogenic contami-nants at that time (Zong et al. 2010). For example,the Pearl River shows a large dominance of kaolinitein the modern system, but cores taken in the rivermouth show that the situation only arose at around2500 ka and that before that time smectite wasmuch more important, as shown by the differencebetween modern Pearl River sediments and thosefrom the Holocene delta (Fig. 4). Thus, to assumethat much of the smectite in the South China Seawas derived from Luzon or from central Vietnamprior to 2500 ka would be incorrect, and beforethat time similar size variations can be linked to cli-mate change during the early Holocene when themonsoon was stronger (Wang et al. 2001; Dykoskiet al. 2005). Unless the impact of modern humandisruption to river systems and past climate changecan be accounted for, therefore, it seems that claymineralogy by itself is a rather unreliable prove-nance proxy unless simply applied to the presentday. Although in deep-water settings, where coarsersediment is not found, it may be one of the fewmethods that can even be attempted.

Bulk geochemistry

An alternative method to using weathering regimeas a tool for sediment provenance is to look at the

Fig. 4. Clay mineralogy of the Holocene sediments analysed from ODP Site 1144 (Hu et al. 2012) compared to modernriver and offshore sediment compositions in the possible source regions (Liu et al. 2010a). Holocene Pearl Riverdata are from Hu et al. (2013). The arrow shows how the Pearl River composition changes after 2.5 ka.

PROVENANCE IN THE SOUTH CHINA SEA

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

bulk geochemistry. This approach is potentially use-ful in sediment of many different grain sizes, includ-ing the muds that are not easily analysed using thethermochronology methods discussed below. Caremust, however, be exercised in comparing geo-chemical data between sediments of approximatelysimilar size as hydrodynamic sorting of mineralsand preferential alteration are methods by which dif-ferent grain sizes can end up with quite differentchemistries, despite coming from the same originalsource.

The premise of this particular method is that ele-ments that are particularly soluble in aqueous solu-tion are depleted in rocks and sediment, which hasexperienced more chemical alteration, and aremobilized and removed compared to immobile ele-ments such as aluminum or silicon (Nesbitt &Young 1982; Galy & France-Lanord 1999). Thismeans that rivers in warmer, wetter environmentstend to have more altered sediment in them thanthose in drier, colder places. Nonetheless, it has tobe recognized that topography is also importantbecause fast-flowing, high-energy rivers fromsteep mountains, such as Taiwan, tend to transportsediment quickly, leading to less chemical weather-ing. Liu et al. (2007b) surveyed the rivers of the

South China Sea and showed that the Pearl Rivertended to include sediment that was more weatheredthan that in the Mekong, which in turn is moreweathered than that in the Red River. This studywas able to do this using the geochemical proxy‘chemical index of alteration’ (CIA: Fig. 5), whichwas developed for use in soils (Nesbitt & Young1982). Despite showing some overlap, Liu et al.(2007b) did highlight differences between the riverswhen a limited grain size was considered. The CIAproxy was not developed for marine sediments andshould probably not be applied in this situation,but has, nevertheless, a long history of being usedto look at the alteration state of deep-water marinesediments.

Application of proxies such as the CIA requirethat only limited grain-size fractions should becompared with one another because of the effectsof the hydrodynamic sorting of different mineralspecies that has a dominant control on bulk chemis-try. Furthermore, marine sediments are often con-taminated by biogenic calcium, which needs to becorrected for if the proxy is to have any meaningwhatsoever (Singh et al. 2005). Likewise, sodiummay be increase in marine sediments as a result ofseawater in pores, which has to be flushed from

Fig. 5. Chemical index of alteration (CIA) plotted against silica content for sand and silt samples from the modernRed River (black dots) and older borehole samples from the Hanoi Basin labelled with depositional age (open circles)(Clift et al. 2008b). Fields for modern trunk Red, Mekong and Pearl rivers are from Liu et al. (2007b), and include onlyfine-grained sediments.

P. D. CLIFT

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

the sample if the analysis is to have any significantmeaning. However, with appropriate sample prepa-ration, these issues can be overcome.

Figure 5 shows how the data presented by Liuet al. (2007b), which was generated after decar-bonation, compares with a series of more sandysediments from the Red River alone (Clift et al.2008b). What is apparent is that more sandy mate-rial tends to have lower CIA values, but that the scat-ter of data is very significant. The distribution ofvalues within a single river is so great that it is prob-ably unrealistic to hope to find a characteristic valuefor any given river system. This also suggests thatusing CIA or other chemical weathering proxies asa way to track provenance is probably useless unlessthis is restricted to the present day and/or to a verylimited grain-size range. Geochemical studies ofthe Yellow Sea region do, however, show thatwith appropriate filtering for grain size and with cor-rection for biogenic carbonates that some degreeof provenance can be successfully achieved usingbulk sediment major-element chemistry (Kim et al.1999; Lim et al. 2013).

Other types of bulk sediment geochemical datahave been used in the past to track provenance,most notably rare earth elements (REEs). Theseare more uniformly immobile to aqueous chemicalweathering and might be expected to better preservethe average composition of the source bedrocks thatare generating the sediment. However, much of theupper continental crust tends to have a gentle enrich-ment in light REEs (LREEs) and to be often similarbetween different basins (Rudnick & Gao 2003).Total REE content is not useful because the REEsare strongly concentrated in a number of heavy min-erals so that the concentrations are largely a reflec-tion of the absolute abundance of these minerals inthe sample selected (Garcon et al. 2014). This inturn is related to hydrodynamic sorting during sedi-mentation, so that samples taken from differentparts of a single bedform might show a quite differ-ent REE content. Studies of the REEs in the RedRiver showed a wide scatter of REE concentrationsand ratios, but no systematic difference in REEcharacter between different tributaries, suggestingthat large tracts of the upper continental crust havequite similar degrees of enrichment and that, there-fore, large rivers with diverse catchment geologieswill often have a similar chemical character (Cliftet al. 2008b).

In contrast, some studies claim that key REEratios such as La/Yb or Gd/Yb, as well as the rela-tive europium anomaly, can be distinctive of dif-ferent source regions, although this will be lesstrue in larger river basins where more averaging canoccur (Vital & Stattegger 2000; Jung et al. 2012). Aswith other forms of chemistry, REE chemistrymay be affected by variations in grain size and

by fractionation of heavy minerals, and so are bestused on a only a limited size fraction

In general, REEs do not seem like promis-ing provenance proxies, with the exception ofpotentially finding sediment derived from theLuzon Arc, which, being a more primitive, mantle-derived piece of crust, would be associated withmore LREE-depleted compositions (i.e. low La/Yb values).

Bulk isotope character

Combined Nd and Sr isotopes have an establishedtrack record in terms of resolving provenance inmany basins worldwide and specifically in theSouth China Sea (Clift et al. 2002; Li et al. 2003).Nd, in particular, is generally recognized as beingimmobile to chemical weathering, and is not frac-tionated by erosion, weathering and transport(Goldstein & Jacobsen 1988). Sr, however, is moremobile, and is fractionated so that more weatheredsediments tend to have higher 87Sr/86Sr valuesthat represent both source composition and weath-ering intensity (Derry & France-Lanord 1996). Ndis a powerful provenance proxy, although, again,grain size may have an influence because Nd islargely dominated by monazite content (Garconet al. 2013). The method has most successfully beenapplied to fine-grained sediments, where it gives anestimate of the relative age of the continental crustfrom which the sediment is derived (Allegre &Ben Othman 1980), although it could be applied tocoarser materials provided that these data were com-pared with other coarse sediment measurements.

Figure 6 shows the range of measured Sr and Ndisotope ratios for a series of modern rivers draininginto the basin, together with selected analyses frommarine cores, largely in the northern part of the sea.It is clear that the modern Pearl River and some partsof the modern Red River are characterized by veryhigh 87Sr/86Sr values and might be resolvable inthis respect. However, this ignores the fact thatthese rivers are anthropogenically disrupted andare presently carrying much more altered sedimentthan has been typical for the Holocene.

Sediment in rivers draining Luzon is truly uniquein Sr and Nd isotopes, in having both very low87Sr/86Sr values and very high 1Nd values. BecauseNd is resistant to change during weathering, theinfluence of Luzon in providing sediment shouldbe easily resolved using this approach. However,many possible sources cluster around 87Sr/86Sr val-ues of 0.72 and 1Nd values of 210, including themodern Mekong and Red rivers, Taiwan, and theHolocene of the Pearl River. Not surprisingly,many of the cored sediments analysed for these iso-topes also plot in this region, suggesting that this

PROVENANCE IN THE SOUTH CHINA SEA

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

approach may not be the best at resolving sources inSE Asia. It is noteworthy that ODP Site 1144 showsoverlap with Taiwan and has the lowest 87Sr/86Srvalues, which do not seem unique to that sourcecompared to sources in southern China or Indo-china. The relative lack of variation in 1Nd valuesreflects the fact that many of the blocks in SE Asiahave similar crustal residence ages, but it shouldbe noted that when rivers have connected to sourcesin the much older Yangtze Craton in the past thenmuch lower 1Nd values were recognized, most nota-bly in the Red River delta (Song Hong-YinggehaiBasin) (Clift et al. 2006a).

Apatite fission track

Comparison of fission-track ages between sedi-ments and bedrock sources has been a powerfulmethod for sediment provenance for some time

(Hurford & Carter 1991; Carter 2007), and is onewith some of the greatest potential in the SouthChina Sea. The apatite fission-track method recordscooling of rocks through approximately 60–1258Cover timescales of 1–10 Ma (Green et al. 1989)and, provided that the sediment has not been buriedand reheated again since deposition, allows singlegrains to be tied to sources with unique exhuma-tion histories. This favours sedimentary systems inSE Asia because different parts of the marginshave experienced different deformation and erosionhistories. Fission-track methods are best applied tosediment that is fine sand or coarse in grain size,reflecting the need to measure track densities andthe 16 mm length of new tracks. However, coarsesilts can be used in extreme circumstances.

Luzon is less easily recognized in this approachbecause basalts that dominate the arc are not richin apatite but should be Oligocene or youngerif present, given the known range of magmatism

Fig. 6. Sr and Nd isotopic plot showing the variability in Holocene and modern sediments from ODP Site 1144 (Huet al. 2012) and in Lower Miocene–Recent sediments at ODP Site 1148 compared to modern potential sources aroundthe South China Sea. The diagram shows the general similarity of the sediment with modern Taiwanese rivers andbedrock samples (Chen & Lee 1990; Lan et al. 2002), and the differences with sediments in the modern Pearl River (Liuet al. 2007b) and with potential volcanic sources in the Philippine island of Luzon (Zhou et al. 2002). The Neogenesediments at ODP Site 1148 show an overlap with both Taiwanese rivers and the Holocene sediments of the Pearl RiverEstuary (Hu et al. 2013). Data from the Red and Mekong rivers are from Liu et al. (2007b) and Clift et al. (2008b).

P. D. CLIFT

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

(Defant et al. 1989). Figure 7 shows a range ofkernel density estimate (KDE) plots designed todemonstrate the most likely cooling ages for eachsource (Vermeesch 2012). Taiwan is unique in thissystem, with very young ages that post-date the startof collision at c. 5 Ma, making grains derived fromthis island easy to resolve within mixtures. Fission-track cooling ages from Cathaysia in southernChina, and from the Khorat Plateau, the Viet-namese Central Highlands and the Kontum Massifin Indochina have a similar range of central ages,especially around 30–60 Ma. This age reflects cool-ing driven by the break-up of the South China Seaand the subsequent erosional degradation of therifted margins. However, because this process iscommon, sediment derived from these differentsources cannot be resolved with this technique, withthe possible exception that the Central Highlandshas slightly younger ages than other parts of thewestern margin because of rejuvenation of thatregion during emplacement of volcanic sequencesat approximately 8 Ma (Carter et al. 2000).

An intermediate grouping of fission-track agesare those grains with cooling ages of c. 20 Ma.These are typical of bedrocks in the Red RiverFault Zone, but are found throughout the SE flankof the Tibetan Plateau, especially in the gorges ofSW China (Clark et al. 2005) and are not spe-cific to the fault zone. This cooling is related tothe deformation and rock uplift associated with thesoutheastward propagation of the Tibetan Plateauduring the Miocene (Schoenbohm et al. 2006).This explains why grains of this age dominate themodern Red River (Fig. 7) which derives its sedi-ment from this region (Clift et al. 2006b). In con-trast, the Mekong River has a subtly different agespectrum, with a younger peak at approximately14 Ma and a long tail of older ages. The youngeraged grains in the Mekong are believed to bederived from younger sources in its upper reachesin Tibet and the older ages from the lowlands ofIndochina, including the Khorat Plateau (Cliftet al. 2006b). For whatever reason, the fission-track ages of these two rivers contrast with each

Age (Ma)Age (Ma)0 20 40 60 80 100

0 20 40 60 80 100 0 20 40 60 80 100

Pro

babi

lity

Pro

babi

lity

Pro

babi

lity

0 20 40 60 80 100

Red River Fault Zone

Khorat Plateau

Kontum Massif/Central Highlands

Songpan GarzeGorges

0 20 40 60 80 100

Mekong RiverN = 19

Red River

0 20 40 60 80 100

N = 67

N = 15N = 30

N = 18 N = 34

0 20 40 60 80 100

Pro

babi

lity

TaiwanN = 64

0 20 40 60 80 100

CathaysiaN = 17

Fig. 7. Probability density diagrams for apatite fission-track ages from sands in the modern Mekong and Red riverscompared with those from possible source terrains in eastern Asia (Clift et al. 2006b). Data from the Ailao Shan/RedRiver Fault Zone is from Bergman et al. (1997) and Maluski et al. (2001). Data from the Khorat Plateau is fromRacey et al. (1997) and Upton (1999). Data from the Songpan Garze Terrane is from Reid et al. (2005). Data from theKontum Massif is from Carter et al. (2000). Data from Taiwan is from Fuller et al. (2006) and Willett et al. (2003).Data from Cathaysia is from Yan et al. (2009).

PROVENANCE IN THE SOUTH CHINA SEA

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

other and the presumed spectrum for the Pearl,which must be dominated by the older ages fromCathaysia.

Apatite fission-track thermochronology thusholds great promise in being able to distinguish sedi-ment from the three major rivers, as well as Taiwan,even if there is more ambiguity concerning grainsdated at 30–60 Ma, which may be eroded fromeither the basin’s western or northern margins.

Muscovite Ar/Ar dating

There is much less existing 40Ar/39Ar data frompotential bedrock sources around the South ChinaSea compared to fission-track data, yet this methodis also useful at pinpointing sources and can be com-plementary to the lower temperature method. The40Ar/39Ar method has been applied in severalAsian provenance studies (Najman et al. 1997;White et al. 2002; Szulc et al. 2006; Hoang et al.2010) and allows the age when each single micagrain cooled through an isotherm of around 3008C

to be determined (Hodges 2003). Triassic coolingages are common in SE Asia and are related to theIndosinian Orogeny (Huang et al. 2003; Lepvrieret al. 2004). However, there are important discrep-ancies to this general picture, and resolvable dif-ferences between sources around the northern andwestern edges of the basin that make this a goodprovenance tool. Single-grain mica dating requiressubstantial crystals and even multiple grain methodsmean that the method is really only applicable tosands, generally limiting its application to the prox-imal parts of a given sedimentary system.

Analysis of the modern Red and Mekong riversshows that the Mekong has a generally younger setof Indosinian micas than seen in the Red River;150–210 Ma compared to 210–250 Ma (Clift et al.2006b) (Fig. 8). This difference must reflect con-trasting cooling ages of Indosinian metamorphicrocks in the two drainage basins and when trans-ferred to the marine realm would allow sedimentfrom each river to be resolved within offshore depo-centres. Both rivers contain muscovite micas dat-ing to c. 20–30 Ma, similar to the rocks of the

0 50 100 150 200 250 300

Age (Ma)0 50 100 150 200 250 300

0 50 100 150 200 250 300 0 50 100 150 200 250 300

0 50 100 150 200 250 300 0 50 100 150 200 250 300

Age (Ma)

Pro

babi

lity

Pro

babi

lity

Pro

babi

lity N = 45

Qiangtang Block

Songpan GarzeN = 25

Red RiverN = 47

Mekong RiverN = 48

Ailao Shan/Red River Fault ZoneN = 82

Central VietnamN = 31

Fig. 8. Probability density diagrams for Ar/Ar ages of detrital muscovites within the Mekong and Red rivers comparedto known ranges from possible source regions. Central Vietnam data are from Lepvrier et al. (1997) and Nagy et al.(2000). Ailao Shan/Red River Fault Zone data are from Leloup et al. (1993, 2001), Harrison et al. (1996), P. L. Wanget al. (1998), Jolivet et al. (1999) and Maluski et al. (2001), and data from Songpan Garze are from Reid et al. (2005).Qiangtang Block data are from Kapp et al. (2000).

P. D. CLIFT

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

Red River Fault Zone and equivalent structuresacross SE Tibet and Indochina, such as the MaePing and Three Pagoda faults (P. L. Wang et al.1998; Morley 2002). As a result, these ages areonly partially source diagnostic, and are only relatedto sources from SE Tibet. The timing of motion onthese faults is commonly Oligo-Miocene, making itimpossible to resolve between the different sources.Only the Xianshuihe–Xiaojiang Fault of SW Chinahas continued rapid motion until recent times (E.Wang et al. 1998). Nonetheless, the presence of20–30 Ma micas does at least rule out erosionfrom southern China. 40Ar/39Ar data from southernChina–Cathaysia is sparse, but that which doesexist tends to be of Indosinian age, 207–215 Ma(Wang et al. 2005), but critically slightly youn-ger than dated basement rocks in central Vietnam,which tend to cluster at about 240 Ma (Lepvrieret al. 1997). This latter tectonic block is, however,marked by a minor population of micas dating at110–130 Ma, which is not found further north.

Luzon is not a significant source of muscovitebecause of its dominant basaltic oceanic arc charac-ter and any mica derived from the Central Rangesof Taiwan must have very young ages of ,5 Maif they have been reset by the orogeny on thatisland. If muscovite grains from Taiwan have notbeen reset – for example, in the lower structural lev-els of the thrust wedge – then they would carry thesame array of ages as seen in SE China and wouldnot be resolvable using this method. Although the40Ar/39Ar method is generally slower and moreexpensive than that of the fission track method, itcan form an important tool for quantifying erosionbudgets in the South China Sea because of thearray of unique sources that characterize some ofthe major rivers, especially along the western mar-gins of the basin.

Zircon U–Pb dating

U–Pb dating of zircon has become one of themost popular forms of provenance analysis, follow-ing the improvement in dating technology that nowallows large numbers of grains to be quickly andcheaply dated using laser ablation inductively cou-pled plasma mass spectrometry (ICP-MS) insteadof the more traditional mass spectrometry. Thismethod can only be applied to grains that are largeenough to be dated, which usually means .50 mm(coarse silt), as smaller grains are too small to be tar-geted by a laser ablation ICP-MS that dominatesthe approach. This means that the method is lessuseful in the most distal, deep-water deposits inthe basin centre.

U–Pb ages in zircon are considered to reflectthe time of zircon crystallization at around 7508C

(Hodges 2003), so that they may be interpretedto record the last growth phase in a rock’s his-tory. However, zircons are known to be resistantto physical abrasion during erosion and transport,as well as to chemical weathering, making themsusceptible to multiple phases of reworking (DeCel-les et al. 2004; Campbell et al. 2005). Althoughit is impossible to say precisely where a singlegrain with a single age is derived from, it is possi-ble to suggest the most likely source rock unit formany grains, making U–Pb dating a powerfulprovenance tool in this area. Because the differentpossible source terrains have unique age spectra, itis usually possible to at least exclude possibledominant sources based on the U–Pb age of thesediment.

As with the Ar/Ar ages, Triassic Indosinian agesare very common in many of the source regions andtend not to be diagnostic by themselves. Figure 9hshows that the Qiangtang Block of Tibet and itsequivalent terrane, Sibumasu, in SE Asia (Fig. 2)are particularly dominated by this age range andhave relatively few older grains, resulting in apotentially unique signature. Likewise, Red RiverFault Zone rocks yield Cenozoic ages that are syn-chronous with the motion on that fault (Fig. 9g),and which are effectively unknown outside thiszone and presumably equivalent fault zone rocksin the other major strike-slip zones of SE Tibet–SW China. In any case, observations from theMekong and Red rivers indicate that the faultzone rocks are not major contributors to the netflux to the ocean (Fig. 9a, b). A high proportion ofgrains dated between 700 and 1000 Ma is typicalof bedrock from the Yangtze Block. These grainsare also common in Cathaysia-derived sediment,but these have a slightly older peak at about900 Ma rather than about 800 Ma, and also con-tain significant populations clustered around 1800and 2400 Ma, which are present but are not so abun-dant in the Yangtze Block. The Songpan Garzeterrane shows intermediate character, having rela-tively few 700–1000 Ma grains, but commonapproximately 1800 and 2400 Ma grains, as wellas a strong 400–500 Ma population (Fig. 9d),which is unknown in the Yangtze Block, rare inCathaysia but is also known in Indochina in theKhorat Plateau, the Kontum Massif and the CentralHighland areas that fringe the western edge of theSouth China Sea.

Consideration of the spectra of zircon ages inmodern Red and Mekong river sediments showsthat while they share many of the same populationsthere are differences between these systems thatreflect the contrasting ages in the bedrock sources.The Red River has a much stronger peak at around800 Ma compared to the Mekong River, whichmay be interpreted to indicate more erosion from

PROVENANCE IN THE SOUTH CHINA SEA

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

Mekong River Red River

Cathaysia

Songpan Garze

Khorat Plateau, Kontum MassifCentral Highlands

0 500 1000 1500 2000 2500 3000

0 500 1000 1500 2000 2500 3000

0 500 1000 1500 2000 2500 30000 500 1000 1500 2000 2500 3000

Age (Ma)

0 500 1000 1500 2000 2500 3000

Qiantang Block-Sibumasu

0 500 1000 1500 2000 2500 3000

Pro

babi

lity

Pro

babi

lity

Pro

babi

lity

Age (Ma)

Caledonian

Jinningian Luliangian

Luliangian

Caledonian

Indosinian

Indosinian

Caledonian

Indosinian

Indosinian Indosinian

Caledonian

Luliangian

Luliangian

Basement

Basement

N = 106N = 118

Yangtze Craton

0 500 1000 1500 2000 2500 3000

JinningianN = 574 N = 905

Ailao Shan/Red River Fault Zone

0 500 1000 1500 2000 2500 3000

Pro

babi

lity

N = 20 N = 55

N = 530

N = 418

(a) (b)

(c) (d)

(e) (f)

(g) (h)

Fig. 9. Probability density diagrams showing the detrital U–Pb ages of zircon grains known from the major tectonicblocks in SE Asia. Data for Ailao Shan and Red River Fault Zone rocks are from Scharer et al. (1990, 1994), Zhang &Scharer (1999), Nagy et al. (2000) and Carter et al. (2001). Data for Cathaysia are from Li et al. (1989, 2001, 2005).Data for Indochina sources (including the Khorat Plateau, the Kontum Massif and the Central Highlands of Vietnam)are from Carter & Moss (1999), Carter et al. (2001), Nagy et al. (2001) and Carter & Bristow (2003). Data for theSongpan Garze Terrane are from Hu et al. (2005), Bruguier et al. (1997) and Weislogel et al. (2006, 2010). Data for thewestern Yangtze Craton are from Sun et al. (2009), Xu et al. (2008) and Yang et al. (2005). Data for the QiangtangBlock are from Roger et al. (2000, 2003).

Table 1. Summary of the different provenance methods reviewed in this paper and simplified results for thedifferent major source terrains that are supplying sediment to the modern South China Sea

Method Taiwan South China Indochina Luzon

Bulk sediment chemistry Variable Variable Variable VolcanicClay minerals Chlorite and illite Kaolinite, smectite

and illiteKaolinite, smectite

and illiteSmectite

Nd isotopes 29 to 211 28 to 211 29 to 213, except theCentral Highlands

+4 to +8

Apatite fission track ,3 Ma 30–60 Ma 17–23 Ma (S. ChayMassif), 27–50 Ma(Kontum)

,45 Ma

Zircon fission track ,5 Ma 90–120 Ma 90–550 Ma N/AAr–Ar muscovite ,6 Ma 200–220 Ma 160–250 Ma N/AU–Pb zircons ,500 and

c. 1800 Ma800–1500 Ma 240–450 and c. 1800 Ma ,45 Ma

P. D. CLIFT

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

the Yangtze Craton, consistent with the knownextent of the modern drainage. Furthermore, Cliftet al. (2006b) demonstrated that the Indosinianpeak in the Mekong River is slightly younger thanin the Red River, a difference that could be exploitedto resolve and quantify relative input from thetwo drainages.

Although rivers from Luzon have not yet beenanalysed for zircon U–Pb ages, those sources arenot considered to be important to the zircon budgetof the basin because of the relative lack of zircons inbasaltic and basaltic andesite lavas in that arc. Whenthey are present then they must be younger than theOligocene when that arc became active and thuscould only be potentially confused with sedimentfrom the Red River Fault Zones. Metamorphicrocks in Taiwan were not heated sufficiently toreset the U–Pb system in zircon during that arc–continent collision and are anticipated to yield typ-ical Cathaysian ages based on the sources to thatpart of the passive margin (Lan et al. 2014). Fortu-nately, Taiwan has a unique signature in terms oflower temperature thermochronology systems sothat its contribution is not dependent on U–Pb zir-con dating. It appears that U–Pb dating can be a use-ful component to a provenance study in the SouthChina Sea, but often yields mixed or ambiguous sig-nals because many of the major rivers share similaraged bedrocks.

Discussion and conclusions

In this review of the major provenance methodsnow applied in the South China Sea, it is clear thata single method is generally insufficient to quantifythe relative sediment flux contributions from multi-ple terrains into a given sediment. Table 1 shows asummary of the character of the sources on threesides of the basin. There is relatively little infor-mation about the character of sources in Borneo orpeninsular Malaysia, both of which may have beenmore important in the past when sea level waslower and when the Molengraaff River may be fedmaterial to the Sunda Shelf and into deep-waterbasin, at least at its SW end (Voris 2000; Steinkeet al. 2008; Hanebuth et al. 2011). When sea levelwas higher, sediment flux into the South ChinaSea from these sources was generally mini-mal. More work needs to be carried out to look atadditional methods that should be effective in thisbasin, such as heavy mineral studies, but at thetime of writing insufficient is known about the typ-ical mineralogy of the modern drainages to assesswhether this would actually work on ancientsequences or not. Only the continental shelf of Viet-nam has been investigated for heavy minerals(Jagodzinski 2005).

What is clear is that proxies that are stronglyaffected by chemical weathering make unreliableprovenance proxies for sediments predating the lastfew hundred years. The modern rivers are known tobe significantly disrupted in terms of bulk sedimentchemistry and clay mineralogy, so that these com-positions cannot be used to fingerprint ancient sedi-ments. The same is also true of Sr isotopes. Evenif the pre-modern composition can be fixed by look-ing at Holocene sediments under the most recentcover, it must be realized that the sediment compo-sition is influenced by the changing climate, espe-cially the intensity of the East Asian Monsoon.

Nd isotopes are only of limited use, except toresolve input from the young and primitive rocksof Luzon and in the case where rivers are drainingsediment out of central southern China where theancient crust of the Yangtze Craton results in verynegative 1Nd values. Nd compositions in Indochinaand southern China are very similar and largelyoverlap, resulting in serious ambiguity in terms ofresolving the effects of different sources. However,coherent changes may be detected and could beuseful in discerning some changes depending onthe size and heterogeneity of the basement rocksinvolved, since not all parts of Cathaysia or Indo-china are identical. Erosion of the Miocene volcanicrocks from the Vietnamese Central Highlandswould be anticipated to yield more positive eNd val-ues than is typical, for example, for the rest of theVietnamese coast.

The most useful provenance proxies are thoserelated to detrital thermochronology, but even hereit is generally most effective to use more than onemethod to achieve a robust result. Apatite fissiontrack is probably the best single method becauseof the contrasting timing of moderate amounts ofexhumation around the basin, which allow mostsources to be distinguished, although there is someoverlap between parts of Indochina and southernChina, not only in apatite but also in muscoviteAr/Ar. Fortunately, these sources can be separatedon the basis of U–Pb zircon dating. Provenance inthe South China Sea is complex, but it is not animpossible task and one that is essential to solve ifwe are to reconstruct the palaeogeography, drainageevolution and chemical weathering regimes in SEAsia during the Cenozoic. Because this is a classicarea for collision tectonics, associated drainagereorganization and the development of the mon-soon, the sedimentary archives in the South ChinaSea are invaluable to efforts to address these issuesand require provenance control if the records are tobe understood.

PDC thanks the Charles T. McCord Chair in PetroleumGeology at Louisiana State University for support toundertake this research.

PROVENANCE IN THE SOUTH CHINA SEA

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

References

Allegre, C. J. & Ben Othman, D. 1980. Nd–Sr isotopicrelationship in granitoid rocks and continental crustdevelopment: a chemical approach to orogenesis.Nature, 286, 335–342.

Aoki, S. 1976. Clay mineral distribution in sediments ofthe Gulf of Thailand and the South China Sea. Jour-nal of the Oceanographic Society of Japan, 32,169–174.

Barckhausen, U., Engels, M., Franke, D., Ladage, S.& Pubellier, M. 2014. Evolution of the South ChinaSea: revised ages for breakup and seafloor spread-ing. In: Pubellier, M., Franke, D., McIntosh, K.,Menier, D. & Li, C.-F. (eds) Evolution, Structure,and Sedimentary Record of the South China Sea andAdjacent Basins. Marine and Petroleum Geology, 58,599–611, http://doi.org/10.1016/j.marpetgeo.2014.02.022

Bayon, G., Dennielou, B., Etoubleau, J., Ponzevera,E., Toucanne, S. & Bermell, S. 2012. Intensify-ing weathering and land use in iron age Central Africa.Science, 335, 1219–1222, http://doi.org/10.1126/science.1215400

Bergman, S. C., Leloup, P. H., Tapponnier, P.,Scharer, U. & O’Sullivan, P. B. 1997. Apatite fis-sion Track Thermal History of the Ailao Shan-Red River Shear Zone, China. Paper presented at theEuropean Union of Geoscientists, Strasbourg, 23–27March 1997.

Boulay, S., Colin, C., Trentesaux, A., Frank, N. &Liu, Z. 2005. Sediment sources and East Asian mon-soon intensity over the last 450 ky: mineralogical andgeochemical investigations on South China Sea sedi-ments. Palaeogeography Palaeoclimatology Palaeoe-cology, 228, 260–277.

Briais, A., Patriat, P. & Tapponnier, P. 1993. Updatedinterpretation of magnetic anomalies and seafloorspreading stages in the South China Sea: implicationsfor the Tertiary tectonics of Southeast Asia. Journalof Geophysical Research, 98, 6299–6328, http://doi.org/10.1029/92JB02280

Brookfield, M. E. 1998. The evolution of the great riversystems of southern Asia during the CenozoicIndia-Asia collision; rivers draining southwards. Geo-morphology, 22, 285–312.

Bruguier, O., Lancelot, J. R. & Malavieille, J. 1997.U–Pb dating on single detrital zircon grains from theTriassic Songpan-Ganze Flysch (central China); prov-enance and tectonic correlations. Earth and PlanetaryScience Letters, 152, 217–231.

Campbell, I. H., Reiners, P. W., Allen, C. M., Nico-

lescu, S. & Upadhyay, R. 2005. He–Pb double datingof detrital zircons from the Ganges and Indus rivers;implication for quantifying sediment recycling andprovenance studies. Earth and Planetary Science Let-ters, 237, 402–432.

Carter, A. 2007. Heavy minerals and detrital fission-track thermochronology. Developments in Sedi-mentology, 58, 851–868, http://doi.org/10.1016/S0070-4571(07)58033-7

Carter, A. & Bristow, C. S. 2003. Linking hinter-land evolution and continental basin sedimentationby using detrital zircon thermochronology; a study of

the Khorat Plateau basin, eastern Thailand. BasinResearch, 15, 271–285.

Carter, A. & Clift, P. D. 2008. Was the Indosinian orog-eny a Triassic mountain building or thermotectonicreactivation event? Comptes Rendues de l’AcademieScientifique, Geoscience, 340, 83–93.

Carter, A. & Moss, S. J. 1999. Combined detrital-zirconfission-track and U–Pb dating: a new approach tounderstanding hinterland evolution. Geology, 27,235–238.

Carter, A., Roques, D. & Bristow, C. S. 2000. Denuda-tion history of onshore central Vietnam: constraints onthe Cenozoic evolution of the western margin of theSouth China Sea. Tectonophysics, 322, 265–277.

Carter, A., Roques, D., Bristow, C. & Kinny, P. D.2001. Understanding Mesozoic accretion in SoutheastAsia: significance of Triassic thermotectonism (Indosi-nian orogeny) in Vietnam. Geology, 29, 211–214.

Chen, C. H. & Lee, T. 1990. A Nd-Sr isotopic study onriver sediments of Taiwan. Proceedings of the Geolog-ical Society of China, 33, 339–350.

Chen, P. Y. 1978. Minerals in bottom sediments ofthe South China Sea. Geological Society of AmericaBulletin, 89, 211–222.

Clark, M. K., Schoenbohm, L. M. et al. 2004. Surfaceuplift, tectonics, and erosion of eastern Tibet fromlarge-scale drainage patterns. Tectonics, 23, TC1006,http://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 southeastern Tibet. Geol-ogy, 33, 525–528, http://doi.org/10.1130/G21265.1

Clift, P., Lee, J. I., Clark, M. K. & Blusztajn, J. 2002.Erosional response of south China to arc rifting andmonsoonal strengthening; a record from the SouthChina Sea. Marine Geology, 184, 207–226.

Clift, P. D., Blusztajn, J. & Nguyen, D. A. 2006a.Large-scale drainage capture and surface uplift ineastern Tibet-SW China before 24 Ma inferred fromsediments of the Hanoi Basin, Vietnam. Geophys-ical Research Letters, 33, http://doi.org/10.1029/2006GL027772

Clift, P. D., Carter, A., Campbell, I. H., Pringle, M.,Hodges, K. V., Lap, N. V. & Allen, C. M. 2006b.Thermochronology of mineral grains in the SongHong and Mekong Rivers, Vietnam. Geophysics, Geo-chemistry, Geosystems, 7, Q10005, http://doi.org/10.1029/2006GC001336

Clift, P., Lee, G. H., Nguyen, A. D., Barckhausen, U.,Hoang, V. L. & Sun, Z. 2008a. Seismic evidence for aDangerous Grounds mini-plate: no extrusion origin forthe South China Sea. Tectonics, 27, TC3008, http://doi.org/10.1029/2007TC002216

Clift, P. D., Hoang, V. L., Hinton, R., Ellam, R., Han-

nigan, R., Tan, M. T. & Nguyen, D. A. 2008b. Evolv-ing East Asian river systems reconstructed by traceelement and Pb and Nd isotope variations in modernand ancient Red River-Song Hong sediments. Geo-chemistry, Geophysics, Geosystems, 9, Q04039,http://doi.org/10.1029/2007GC001867

Clift, P. D., Wan, S. & Blusztajn, J. 2014. Recon-structing chemical weathering, physical erosionand monsoon intensity since 25 Ma in the northernSouth China Sea: a review of competing proxies.

P. D. CLIFT

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

Earth-Science Reviews, 130, 86–102, http://doi.org/10.1016/j.earscirev.2014.01.002

DeCelles, P. G., Gehrels, G. E., Najman, Y., Martin,A. J., Carter, A. & Garzanti, E. 2004. Detrital geo-chronology and geochemistry of Cretaceous-EarlyMiocene strata of Nepal: implications for timing anddiachroneity of initial Himalayan orogenesis. Earthand Planetary Science Letters, 227, 313–330.

Defant, M. J., Jacques, D., Maury, R. C., De Boer, J. &Joron, J. L. 1989. Geochemistry and tectonic settingof the Luzon arc, Philippines. Geological Society ofAmerica Bulletin, 101, 663–672, http://doi.org/10.1130/0016-7606(1989)101

Derry, L. A. & France-Lanord, C. 1996. NeogeneHimalayan weathering history and river 87Sr/86Sr;impact on the marine Sr record. Earth and PlanetaryScience Letters, 142, 59–74.

Dung, B. V., Stattegger, K., Unverricht, D., Phach,P. V. & Thanh, N. T. 2013. Late Pleistocene-Holocene seismic stratigraphy of the Southeast Viet-nam Shelf. Global and Planetary Change, 110B,156–169.

Dykoski, C. A., Edwards, R. L. et al. 2005. A high-resolution, absolute-dated Holocene and deglacialAsian monsoon record from Dongge Cave, China.Earth and Planetary Science Letters, 233, 71–86.

Enkelmann, E., Weislogel, A., Ratschbacher, L.,Eide, E., Renno, A. & Wooden, J. 2007. How wasthe Triassic Songpan-Ganzi basin filled? A provenancestudy. Tectonics, 26, TC4007.

Fletcher, C. J. N., Chan, L. S., Sewell, R. J., Camp-

bell, S. D. G., Davis, D. W. & Zhu, J. 2004. Basementheterogeneity in the Cathaysia crustal block, southeastChina. In: Malpas, J., Fletcher, C. J. N., Ali, J. R. &Aitchison, J. C. (eds) Aspects of the Tectonic Evolu-tion of China. Geological Society, London, SpecialPublications, 226, 145–155, http://doi.org/10.1144/GSL.SP.2004.226.01.08

Franke, D. 2013. Rifting, lithosphere breakup and volca-nism: comparison of magma-poor and volcanic riftedmargins. Marine and Petroleum Geology, 43, 63–87,http://doi.org/10.1016/j.marpetgeo.2012.11.003

Fuller, C. W., Willett, S. D. & Brandon, M. T. 2006.Formation of forearc basins and their influence on sub-duction zone earthquakes. Geology, 34, 65–68, http://doi.org/10.1130/G21828.1

Galy, A. & France-Lanord, C. 1999. Weathering pro-cesses in the Ganges-Brahmaputra basin and the river-ine alkalinity budget. Chemical Geology, 159, 31–60.

Garcon, M., Chauvel, C., France-Lanord, C.,Huyghe, P. & Lave, J. 2013. Continental sedimentaryprocesses decouple Nd and Hf isotopes. Geochimica etCosmochimica Acta, 121, 177–195.

Garcon, M., Chauvel, C., France-Lanord, C.,Limonta, M. & Garzanti, E. 2014. Which mineralscontrol the Nd–Hf–Sr–Pb isotopic compositions ofriver sediments? Chemical Geology, 364, 42–55,http://doi.org/10.1016/j.chemgeo.2013.11.018

Goldstein, S. J. & Jacobsen, S. B. 1988. Nd and Sr iso-topic systematics of river water suspended material;implications for crustal evolution. Earth and PlanetaryScience Letters, 87, 249–265.

Green, P. F., Duddy, I. R., Laslett, G. M., Hegarty,K. A., Gleadow, A. J. W. & Lovering, J. F. 1989.

Thermal annealing of fission tracks in apatite; 4,quantitative modelling techniques and extension togeological timescales. Chemical Geology; IsotopeGeoscience Section, 79, 155–182.

Hall, R. 2002. Cenozoic geological and plate tectonicevolution of SE Asia and the SW Pacific: computer-based reconstructions and animations. Journal ofAsian Earth Sciences, 20, 353–434.

Hanebuth, T. J. J., Voris, H. K., Yokoyama, Y., Saito,Y. & Okuno, J. I. 2011. Formation and fate of sedi-mentary depocentres on Southeast Asia’s SundaShelf over the past sea-level cycle and biogeographicimplications. Earth-Science Reviews, 104, 92–110,http://doi.org/10.1016/j.earscirev.2010.09.006

Harrison, T. M., Leloup, P. H., Ryerson, F. J., Tappon-

nier, P., Lacassin, R. & Wenji, C. 1996. Diachronousinitiation of transtension along the Ailao Shan-RedRiver shear zone, Yunnan and Vietnam. In: Yin, A.& Harrison, T. M. (eds) The Tectonic Evolutionof Asia. Cambridge University Press, Cambridge,110–137.

Hillier, S. 2003. Quantitative analysis of clay and otherminerals in sandstones by X-ray powder diffraction(XRPD). In: Worden, R. H. & Morad, S. (eds) Claymineral Cements in Sandstones. International Associa-tion of Sedimentologists, Special Publications, 34,213–251.

Hoang, L. V., Clift, P. D., Mark, D., Zheng, H. & Tan,M. T. 2010. Ar–Ar Muscovite dating as a constraint onsediment provenance and erosion processes in the Redand Yangtze River systems, SE Asia. Earth and Plan-etary Science Letters, 295, 379–389, http://doi.org/10.1016/j.epsl.2010.04.012

Hodges, K. 2003. Geochronology and thermochronologyin orogenic systems. In: Rudnick, R. (ed.) TheCrust. Elsevier, Amsterdam, 263–292.

Holtzapffel, T. 1985. Les mineraux argileux, prepara-tion, analyse diffractometriques et determination.Societie Geologique du Nord, Lille.

Hu, D., Boning, P. et al. 2012. Deep sea records ofthe continental weathering and erosion response toEast Asian monsoon intensification since 14 ka inthe South China Sea. Chemical Geology, 326–327,1–18, http://doi.org/10.1016/j.chemgeo.2012.07.024

Hu, D., Clift, P. D. et al. 2013. Holocene evolution inweathering and erosion patterns in the Pearl Riverdelta. Geochemistry, Geophysics, Geosystems, 14,2349–2368, http://doi.org/10.1002/ggge.20166

Hu, J., Meng, Q., Shi, Y. & Qu, H. 2005. SHRIMP U–Pbdating of zircons from granitoid bodies in the Songpan-Ganzi terrane and its implications. Acta PetrologicaSinica, 21, 867–880.

Hu, S., Raza, A. et al. 2006. Late Mesozoic and Ceno-zoic thermotectonic evolution along a transect fromthe north China craton through the Qinling orogeninto the Yangtze craton, central China. Tectonics, 25,TC6009, http://doi.org/10.1029/2006TC001985

Huang, C. Y., Yuan, P. B. & Tsao, S. H. 2006. Temporaland spatial records of active arc-continent collision inTaiwan: a synthesis. Geological Society of AmericaBulletin, 118, 274–288.

Huang, M., Maas, R., Buick, I. S. & Williams, I. S.2003. Crustal response to continental collisionsbetween the Tibet, Indian, South China and North

PROVENANCE IN THE SOUTH CHINA SEA

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

China blocks; geochronological constraints from theSongpan-Garze orogenic belt, western China. Journalof Metamorphic Geology, 21, 223–240.

Hurford, A. J. & Carter, A. 1991. The role of fissiontrack dating in discrimination of provenance. In:Morton, A. C., Todd, S. P. & Haughton, P. D. W.(eds) Developments in Sedimentary ProvenanceStudies. Geological Society, London, Special Publi-cations, 57, 67–78, http://doi.org/10.1144/GSL.SP.1991.057.01.07

Hutchison, C. S. 1994. Gondwana and Cathaysian blocks,Palaeotethys sutures and Cenozoic tectonics in South-east Asia. Geologische Rundschau, 83, 388–405,http://doi.org/10.1007/BF00210553

Hutchison, C. S. 2005. Geology of North-West Borneo:Sarawak, Brunei and Sabah. Elsevier, Amsterdam.

Hutchison, C. S. 2010. The North-West Borneo Trough.Marine Geology, 271, 32–43, http://doi.org/10.1016/j.margeo.2010.01.007

Hutchison, C. S. & Vijayan, V. R. 2010. What are theSpratly Islands? Journal of Asian Earth Sciences, 39,371–385, http://doi.org/10.1016/j.jseaes.2010.04.013

Jagodzinski, R. 2005. Petrography and Geochemistry ofSurface Sediments from Sunda and Vietnamese Shelves(South China Sea). Adam Mickiewicz UniversityPress, Poznan.

Jolivet, L., Maluski, H., Beyssac, O., Goffe, B.,Lepvrier, C., Phan, T. T. & Nguyen, V. V. 1999.Oligocene-Miocene Bu Khang extentional gneissdome in Vietnam: geodynamic implications. Geology,27,67.

Jung, H.-S., Lim, D., Choi, J.-Y., Yoo, H.-S., Rho, K.-C.& Lee, H.-B. 2012. Rare earth element compositionsof core sediments from the shelf of the South Sea,Korea: their controls and origins. Continental ShelfResearch, 48, 75–86, http://doi.org/10.1016/j.csr.2012.08.008

Kao, S. J. & Milliman, J. D. 2008. Water and sedimentdischarge from small mountainous rivers, Taiwan:the roles of lithology, episodic events, and humanactivities. Journal of Geology, 116, 431–448.

Kapp, P., Yin, A. et al. 2000. Blueschist-bearing meta-morphic core complexes in the Qiangtang Block revealdeep crustal structure of northern Tibet. Geology, 28,19–22.

Kim, G., Yang, H. S. & Church, T. M. 1999. Geochem-istry of alkaline earth elements (Mg, Ca, Sr, Ba) inthe surface sediments of the Yellow Sea. ChemicalGeology, 153, 1–10.

Lahajnar, N., Wiesner, M. G. & Gaye, B. 2007. Fluxesof amino acids and hexosamines to the deep SouthChina Sea. Deep Sea Research I, 54, 2120–2144.

Lan, C. Y., Lee, C.-S., Shen, J. J.-S., Lu, C. Y., Mertz-

man, S. A. & Wu, T.-W. 2002. Nd -Sr isotopic compo-sition and geochemistry of sediments from Taiwan andtheir implications. Western Pacific Earth Sciences, 2,205–222.

Lan, Q., Yan, Y. et al. 2014. Tectonics, topography, andriver system transition in East Tibet: insights from thesedimentary record in Taiwan. Geochemistry, Geo-physics, Geosystems, 15, 3658–3674, http://doi.org/10.1002/2014GC005310

Le, T. P. Q., Garnier, J., Gilles, B., Sylvain, T. &Minh, C. V. 2007. The changing flow regime and

sediment load of the Red River, Viet Nam. Journalof Hydrology, 334, 199–214, http://doi.org/10.1016/j.jhydrol.2006.10.020

Lei, S., Li, X. et al. 2007. Deep water bottom currentdeposition in the northern South China Sea. Sciencein China Series D: Earth Sciences, 50, 1862–2801,http://doi.org/10.1007/s11430-007-0015-y

Leloup, P. H., Harrison, T. M., Ryerson, F. J., Chen,W. J., Li, Q., Tapponnier, P. & Lacassin, R. 1993.Structural, petrological and thermal evolution of a ter-tiary ductile strike-slip shear zone, Diancang Shan,Yunnan. Journal of Geophysical Research: SolidEarth, 98, 6715–6743.

Leloup, P. H., Arnaud, N. et al. 2001. New constraintson the structure, thermochronology, and timing of theAilao Shan-Red River shear zone, SE Asia. Journalof Geophysical Research, 106, 6657–6671.

Lepvrier, C., Maluski, H., Vuong, N. V., Roques, D.,Axente, V. & Rangin, C. 1997. Indosinian NW-trending shear zones within the Truong Son belt(Vietnam) 40Ar/39Ar Triassic ages and Cretaceous toCenozoic overprints. Tectonophysics, 283, 105–127.

Lepvrier, C., Maluski, H., Vu, V. T., Leyreloup, A.,Phan, T. T. & Vuong, N. V. 2004. The Early TriassicIndosinian orogeny in Vietnam (Truong Son Belt andKontum Massif); implications for the geodynamicevolution of Indochina. Tectonophysics, 393, 87–118.

Li, W. X., Li, X. H. & Li, Z. X. 2005. Neoproterozoicbimodal magmatism in the Cathaysia Block of SouthChina and its tectonic significance. PrecambrianResearch, 136, 51–66.

Li, X., Tatsumoto, M., Premo, W. R. & Gu, X. 1989. Ageand origin of the Tanghu Granite, Southeastern China;results from U–Pb single zircon and Nd isotopes.Geology, 17, 395–399.

Li, X., Wei, G. et al. 2003. Geochemical and Nd isotopicvariations in sediments of the South China Sea; aresponse to Cenozoic tectonism in SE Asia. Earthand Planetary Science Letters, 211, 207–220.

Li, Z. X., Li, X. H., Zhou, H. & Kinny, P. D. 2001. Gren-villian continental collision in south China: newSHRIMP U–Pb zircon results and implications forthe configuration of Rodinia. Geology, 30, 163–166.

Lim, D., Choi, J. Y., Shin, H. H., Rho, K. C. & Jung, H. S.2013. Multielement geochemistry of offshore sedi-ments in the southeastern Yellow Sea and implica-tions for sediment origin and dispersal. QuaternaryInternational, 298, 196–206.

Limmer, D. R., Boening, P. et al. 2012. Geochemicalrecord of holocene to recent sedimentation on theWestern Indus continental shelf, Arabian Sea. Geo-chemistry, Geophysics, Geosystems, 13, Q01008,http://doi.org/10.1029/2011GC003845

Liu, J., Clift, P. D., Yan, W., Chen, Z., Chen, H.,Xiang, R. & Wang, D. 2014. Modern transport anddeposition of settling particles in the northern SouthChina Sea: sediment trap evidence adjacent to XishaTrough. Deep-Sea Research I, 93, 145–155, http://doi.org/10.1016/j.dsr.2014.08.005

Liu, Z., Trentesaux, A. & Clemens, S. C. 2003. Claymineral assemblages in the northern South ChinaSea: implications for East Asian monsoon evolutionover the past 2 million years. Marine Geology, 201,133–146.

P. D. CLIFT

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

Liu, Z., Colin, C., Huang, W., Chen, Z., Trentesaux,A. & Chen, J. F. 2007a. Clay minerals in surface sed-iments of the Pearl River drainage basin and their con-tribution to the South China Sea. Chinese ScienceBulletin, 52, 1101–1111.

Liu, Z., Colin, C., Huang, W., Le, K. P., Tong, S., Chen,Z. & Trentesaux, A. 2007b. Climatic and tectoniccontrols on weathering in south China and Indo-china Peninsula: clay mineralogical and geochemicalinvestigations from the Pearl, Red, and Mekong drai-nage basins. Geochemistry, Geophysics, Geosystems,8, Q05005, http://doi.org/10.1029/2006GC 001490

Liu, Z., Zhao, Y., Li, J. & Colin, C. 2007c. Late Quater-nary clay minerals off Middle Vietnam in the westernSouth China Sea: implications for source analysisand East Asian monsoon evolution. Science in China,Series D, Earth Sciences, 50, 1674–1684.

Liu, Z., Zhao, Y., Colin, C., Siringan, F. P. & Wu, Q.2009. Chemical weathering in Luzon, Philippinesfrom clay mineralogy and major-element geochemis-try of river sediments. Applied Geochemistry, 24,2195–2205.

Liu, Z., Colin, C. et al. 2010a. Clay mineral distributionin surface sediments of the northeastern South ChinaSea and surrounding fluvial drainage basins: sourceand transport. Marine Geology, 277, 48–60, http://doi.org/10.1016/j.margeo.2010.08.010

Liu, Z., Li, X., Colin, C. & Ge, H. 2010b. A high-resolution clay mineralogical record in the northernSouth China Sea since the Last Glacial Maximum,and its time series provenance analysis. Chinese Sci-ence Bulletin, 55, 4058–4068.

Liu, Z., Wang, H., Hantoro, W. S., Sathiamurthy, E.,Colin, C., Zhao, Y. & Li, J. 2012. Climatic and tec-tonic controls on chemical weathering in tropicalSoutheast Asia (Malay Peninsula, Borneo, and Suma-tra). Chemical Geology, 291, 1–12.

Lupker, M., France-Lanord, C. et al. 2012. Pre-dominant floodplain over mountain weathering ofHimalayan sediments (Ganga basin). Geochimica etCosmochimica Acta, 84, 410–432.

Maluski, H., Lepvrier, C. et al. 2001. Ar–Ar andfission-track ages in the Song Chay Massif: Early Tri-assic and Cenozoic tectonics in northern Vietnam.Journal of Asian Earth Sciences, 19, 233–248.

Metcalfe, I. 1996. Pre-Cretaceous evolution of SE Asianterranes. In: Hall, R. & Blundell, D. J. (eds) Tec-tonic Evolution of SE Asia. Geological Society, Lon-don, Special Publications, 106, 97–122, http://doi.org/10.1144/GSL.SP.1996.106.01.09

Milliman, J. D. & Kao, S.-J. 2005. Hyperpycnal dis-charge of fluvial sediment to the ocean: impact ofsuper-typhoon Herb (1996) on Taiwanese Rivers.Journal of Geology, 113, 503–516.

Milliman, J. D. & Syvitski, J. P. M. 1992. Geomorphic/tectonic control of sediment discharge to the ocean; theimportance of small mountainous rivers. Journal ofGeology, 100, 525–544.

Moore, D. & Reynolds, R. 1989. X-ray Diffraction andthe Identification and Analysis of Clay Minerals.Oxford University Press, Oxford.

Morley, C. K. 2002. A tectonic model for the Tertiaryevolution of strike-slip faults and rift basins in SEAsia. Tectonophysics, 347, 189–215.

Nagy, E. A., Scharer, U. & Nguyen, T. M. 2000. Oligo-Miocene granitic magmatism in central Vietnam andimplications for continental deformation in Indochina.Terra Nova, 12, 67–76.

Nagy, E. A., Maluski, H., Lepvrier, C., Scharer, U.,Phan, T. T., Leyreloup, A. & Vu, V. T. 2001. Geo-dynamic significance of the Kontum Massif in centralVietnam; composite 40Ar/39Ar and U–Pb ages fromPaleozoic to Triassic. Journal of Geology, 109,755–770.

Najman, Y. M. R., Pringle, M. S., Johnson, M. R. W.,Robertson, A. H. F. & Wijbrans, J. R. 1997. Laser40Ar/39Ar dating of single detrital muscovite grainsfrom early foreland-basin sedimentary deposits inIndia; implications for early Himalayan evolution.Geology, 25, 535–538.

Nesbitt, H. W. & Young, G. M. 1982. Early Proterozoicclimates and plate motions inferred from major ele-ment chemistry of lutites. Nature, 299, 715–717.

Racey, A., Duddy, L. R. & Love, M. A. 1997. Apa-tite fission track analysis of Mesozoic red beds fromnortheastern Thailand and western Laos. In: Dheera-

dilok, P., Hinthong, C.et al. (eds) Proceedingsof the International Conference on Stratigraphy andTectonic Evolution of Southeast Asia and the SouthPacific, Volume 1. Department of Mineral Resources,Bangkok, 200–209.

Reid, A. J., Wilson, C. J. L., Phillips, D. & Liu, S. 2005.Mesozoic cooling across the Yidun Arc, central-eastern Tibetan Plateau: a reconnaissance 40Ar/39Arstudy. Tectonophysics, 398, 45–66.

Robinson, R. A. J., Brezina, C. A. et al. 2013. Large riv-ers and orogens: the evolution of the YarlungTsangpo–Irrawaddy system and the eastern Himala-yan syntaxis. Gondwana Research, 26, 112–121,http://doi.org/10.1016/j.gr.2013.07.002

Roger, F., Leloup, P. H., Jolivet, M., Lacassin, R.,Trinh, P. T., Brunel, M. & Seward, D. 2000.Long and complex thermal history of the SongChay metamorphic dome (Northern Vietnam) bymulti-system geochronology. Tectonophysics, 321,449–466.

Roger, F., Arnaud, N. et al. 2003. Geochronologicaland geochemical constraints on Mesozoic suturing ineast central Tibet. Tectonics, 1037, 22. http://doi.org/10.1029/2002TC001466

Ru, K. & Pigott, J. D. 1986. Episodic rifting and subsi-dence in the South China Sea. American Associationof Petroleum Geologists Bulletin, 70, 1136–1155.

Rudnick, R. L. & Gao, S. 2003. The composition of thecontinental crust. In: Rudnick, R. L. (ed.) The Crust,Volume 3. Elsevier–Pergamon, Oxford, 1–64.

Scharer, U., Tapponnier, P., Lacassin, R., Leloup,P. H., Dalai, Z. & Shaocheng, J. 1990. Intraplate tec-tonics in Asia; a precise age for large-scale Miocenemovement along the Ailao Shan-Red River shearzone, China. Earth and Planetary Science Letters,97, 65–77.

Scharer, U., Zhang, L. S. & Tapponnier, P. 1994. Dura-tion of strike-slip movements in large shear zones: theRed River belt, China. Earth and Planetary ScienceLetters, 126, 379–397.

Schimanski, A. & Stattegger, K. 2005. Deglacial andHolocene evolution of the Vietnam Shelf; stratigraphy,

PROVENANCE IN THE SOUTH CHINA SEA

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

sediments and sea-level change. Marine Geology, 214,365–387.

Schoenbohm, L. M., Burchfiel, B. C. & Chen, L. 2006.Propagation of surface uplift, lower crustal flow, andCenozoic tectonics of the southeast margin of theTibetan Plateau. Geology, 34, 813–816, http://doi.org/10.1130/G22679.1

Shi, X., Kohn, B. et al. 2011. Cenozoic denudation his-tory of southern Hainan Island, South China Sea: con-straints from low temperature thermochronology.Tectonophysics, 504, 100–115.

Singh, S. K., Sarin, M. M. & France-Lanord, C. 2005.Chemical erosion in the eastern Himalaya; major ioncomposition of the Brahmaputra and d13C of dissolvedinorganic carbon. Geochimica et Cosmochimica Acta,69, 3573–3588.

Steinke, S., Hanebuth, T. J., Vogt, C. & Stattegger, K.2008. Sea level induced variations in clay mineralcomposition in the southwestern South China Seaover the past 17000 years. Marine Geology, 250,199–210.

Sun, W. H., Zhou, M. F., Gao, J. F., Yang, Y. H., Zhao,X. F. & Zhao, J. H. 2009. Detrital zircon U–Pb geo-chronological and Lu–Hf isotopic constraints on thePrecambrian magmatic and crustal evolution of thewestern Yangtze Block, SW China. PrecambrianResearch, 172, 99–126.

Sun, X. & Wang, P. 2005. How old is the Asian monsoonsystem? Palaeobotanical records from China. Palaeo-geography, Palaeoclimatology, Palaeoecology, 222,181–222.

Suppe, J. 1984. Kinematics of arc–continent colli-sion, flipping of subduction, and backarc spreading nearTaiwan. In: Tsan, S. F. (ed.) A Special Volume Dedi-cated to Chun-Sun Ho on the Occasion of His Retire-ment. Geological Society of China, Memoirs, 6,21–33.

Syvitski, J. P. M. & Kettner, A. J. 2011. Sediment fluxand the Anthropocene. Philosophical Transactions ofthe Royal Society of London, Series A: Mathematicaland Physical Sciences, 369, 957–975, http://doi.org/10.1098/rsta.2010.0329

Syvitski, J. P. M., Vorosmarty, C. J., Kettner, A. J. &Green, P. 2005. Impact of humans on the flux of ter-restrial sediment to the global coastal ocean. Science,308, 376–380.

Szczucinski, W., Stattegger, K. & Scholten, J. 2009.Modern sediments and sediment accumulation rates onthe narrow shelf off central Vietnam, South China Sea.Geo-Marine Letters, 29, 47–59.

Szczucinski, W., Jagodzinski, R. et al. 2013. Mod-ern sedimentation and sediment dispersal patternon the continental shelf off the Mekong River delta,South China Sea. Global and Planetary Change,110, 195–213, http://doi.org/10.1016/j.gloplacha.2013.08.019

Szulc, A. G., Najman, Y. et al. 2006. Tectonic evolutionof the Himalaya constrained by detrital 40Ar/39Ar,Sm/Nd and petrographic data from the Siwalik fore-land basin succession, SW Nepal. Basin Research,18, 375–391.

Upton, D. R. 1999. A Regional Fission Track Study ofThailand: Implications for Thermal History and Denu-dation. University of London, London.

Vermeesch, P. 2012. On the visualisation of detrital agedistributions. Chemical Geology, 312–313, 190–194,http://doi.org/10.1016/j.chemgeo.2012.04.021

Vital, H. & Stattegger, K. 2000. Major and trace ele-ments of stream sediments from the lowermost Ama-zon River. Chemical Geology, 168, 151–168.

Voris, H. K. 2000. Maps of Pleistocene sea levels in South-east Asia: shores, river systems and time durations.Journal of Biogeography, 27, 1153–1167.

Wan, S., Li, A., Clift, P. D. & Stuut, J.-B. W. 2007.Development of the East Asian monsoon: mineralogi-cal and sedimentologic records in the northern SouthChina Sea since 20 Ma. Palaeogeography, Palaeocli-matology, Palaeoecology, 254, 561–582.

Wang, E., Burchfiel, B., Royden, L., Chen, L.,Chen, J., Li, W. & Chen, Z. 1998. Late CenozoicXianshuihe-Xiaojiang, Red River, and Dali FaultSystems of Southwestern Sichuan and CentralYunnan, China. Geological Society of America,Boulder, CO.

Wang, P. L., Lo, C. H., Lee, T. Y., Chung, S. L. &Yem, N. T. 1998. Thermochronological evidencefor the movement of the Ailao Shan-Red Rivershear zone: a perspective from Vietnam. Geology,26, 887–890.

Wang, Y., Zhang, Y., Fan, W. & Peng, T. 2005. Struc-tural signatures and 40Ar/39Ar geochronology of theIndosinian Xuefengshan tectonic belt, South ChinaBlock. Journal of Structural Geology, 27, 985–998,http://doi.org/10.1016/j.jsg.2005.04.004

Wang, Y. J., Cheng, H., Edwards, R. L., An, Z. S., Wu,J. Y., Shen, C.-C. & Dorale, J. A. 2001. A high-resolution absolute-dated late Pleistocene Monsoonrecord from Hulu Cave, China. Science, 294,2345–2348.

Weislogel, A. L., Graham, S. A., Chang, E. Z.,Wooden, J. L., Gehrels, G. E. & Yang, H. 2006.Detrital zircon provenance of the Late TriassicSongpan–Ganzi complex: sedimentary record of colli-sion of the North and South China blocks. Geology, 34,97–100.

Weislogel, A. L., Graham, S. A., Chang, E. Z.,Wooden, J. L. & Gehrels, G. 2010. Detrital zir-con provenance from three turbidite depocenters ofthe Middle-Upper Triassic Songpan-Ganzi complex,central China: record of collisional tectonics, erosionalexhumation, and sediment production. GeologicalSociety of America Bulletin, 122, 2041–2062.

West, A. J., Galy, A. & Bickle, M. J. 2005. Tectonic andclimatic controls on silicate weathering. Earth andPlanetary Science Letters, 235, 211–228, http://doi.org/10.1016/j.epsl.2005.03.020

White, N. M., Pringle, M., Garzanti, E., Bickle, M.,Najman, Y., Chapman, H. & Friend, P. 2002.Constraints on the exhumation and erosion of thehigh Himalayan slab, NW India, from foreland basindeposits. Earth and Planetary Science Letters, 195,29–44.

Willett, S. D., Fisher, D., Fuller, C., Chao, Y.-E. &Yu, L.-C. 2003. Erosion rates and orogenic-wedgekinematics in Taiwan inferred from fission-track ther-mochronometry. Geology, 31, 945–948.

Xu, Y. G., Luo, Z. Y. & Huang, X. L. 2008. ZirconU–Pb and Hf isotope constraints on crustal melting

P. D. CLIFT

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from

associated with the Emeishan mantle plume. Geochi-mica et Cosmochimica Acta, 72, 3084–3104.

Xue, Z., Liu, J. P. et al. 2014. Sedimentary processes onthe Mekong subaqueous delta: clay mineral and geo-chemical analysis. Journal of Asian Earth Sciences,79A, 520–528, http://doi.org/10.1016/j.jseaes.2012.07.012

Yan, Y., Carter, A., Xia, B., Ge, L., Brichau, S. &Xiaoqiong, H. 2009. A fission-track and (U–Th)/Hethermochronometric study of the northern margin ofthe South China Sea: an example of a complex passivemargin. Tectonophysics, 474, 584–594.

Yang, J. H., Chung, S. L., Wilde, S. A., Wu, F. Y., Chu,M. F., Lo, C. H. & Fan, H. R. 2005. Petrogenesis ofpost-orogenic syenites in the Sulu orogenic belt, eastChina; geochronological, geochemical and Nd-Sr iso-topic evidence. Chemical Geology, 214, 99–125.

Zhang, L. S. & Scharer, U. 1999. Age and origin of mag-matism along the Cenozoic Red River shear belt,China. Contributions to Mineralogy and Petrology,134, 67–85.

Zheng, H., Clift, P. D., Tada, R., Jia, J. T., He, M. Y.& Wang, P. 2013. A Pre-Miocene birth to theYangtze River. Proceedings of the National Acad-emy of Sciences of the United States of America,110, 7556–7561, http://doi.org/10.1073/pnas.1216241110

Zhou, D. & Graham, S. A. 1996. Songpan–Ganzi Trias-sic flysch complex of the West Qinling Shan as a rem-nant ocean basin. In: Yin, A. & Harrison, M. (eds)The Tectonic Evolution of Asia. Cambridge UniversityPress, Cambridge, 281–299.

Zhou, W., Lu, X., Wu, Z., Deng, L., Jull, A. J. T.,Donahue, D. & Beck, W. 2002. Peat record reflectingHolocene climatic change in the Zoige Plateau andAMS radiocarbon dating. Chinese Science Bulletin,47, 66–70.

Zong, Y., Yu, F., Huang, G., Lloyd, J. M. & Yim,W. W.-S. 2010. Sedimentary evidence of Late Holo-cene human activity in the Pearl River delta, China.Earth Surface Processes and Landforms, 35, 1095–1102, http://doi.org/10.1002/esp.1970

PROVENANCE IN THE SOUTH CHINA SEA

by guest on September 4, 2015http://sp.lyellcollection.org/Downloaded from