Journal of Petrology Advance Access originally published online on February 22, 2007
Journal of Petrology 2007 48(4):729-752; doi:10.1093/petrology/egl080
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Published by Oxford University Press (2007).
Lamproitic Rocks from a Continental Collision Zone: Evidence for Recycling of Subducted Tethyan Oceanic Sediments in the Mantle Beneath Southern Tibet
1Department of Resources, Shijiazhuang University of Economics, Huaian East Road 136, Shijiazhuang, Hebei 050031, China
2Institute of Geology, Chinese Academy of Sciences, Beijing 100037, China
3Department of Earth Sciences, Laurentian University, Sudbury, Ont., Canada
4Institute of Mineral Resources, Chinese Academy of Geological Sciences, Beijing 100037, China
5Hebei Bureau of Land and Resources, Shijiazhuang, Hebei 050051, China
RECEIVED FEBRUARY 9, 2005; ACCEPTED DECEMBER 21, 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL SETTINGANALYTICAL METHODSPETROGRAPHY AND MINERAL...GEOCHEMISTRYDISCUSSIONGEODYNAMIC IMPLICATIONSREFERENCES |
|---|
Major and trace element, Sr–Nd–Pb isotope and mineral chemical data are presented for newly discovered ultrapotassic lavas in the Tangra Yumco–Xuruco graben in southern Tibet. The ultrapotassic lavas are characterized by high MgO, K2O and TiO2, low Al2O3 and Na2O contents, and also have high molar K2O/Al2O3, molar (K2O + Na2O)/Al2O3 and K2O/Na2O ratios. Their high abundances of incompatible trace elements such as large ion lithophile elements (LILE) and light rare earth elements (LREE) reach the extreme levels typical of lamproites. The lamproites show highly radiogenic 87Sr/86Sr (0· 7166–0· 7363) and unradiogenic 143Nd/144Nd (0· 511796–0· 511962), low 206Pb/204Pb (18· 459–18· 931), and elevated radiogenic 207Pb/204Pb (15· 6732–15· 841) and 208Pb/204Pb (39· 557–40· 058) ratios. On the basis of their geochemical and isotopic systematics, the lamproites in south Tibet have a distinct magma source that can be differentiated from the sources of potassic lavas in the east Lhasa and Qiangtang blocks. Their high Nb/Ta ratios (17· 10–19· 84), extremely high Th/U ratios (5· 70–13· 74) and distinctive isotope compositions are compatible with a veined mantle source consisting of partial melts of subducted Tethyan oceanic sediments and sub-continental lithospheric depleted mantle. Identification of the lamproites and the delineation of their mantle source provide new evidence relevant for models of the uplift and extension of the Tibetan plateau following the Indo-Asia collision. Metasomatism by partial melts from isotopically evolved, old sediment subducted on the young Tethyan slab is an alternative explanation for Precambrian Nd and Pb model ages. In this model, differences in isotopic composition along-strike are attributed to differences in the type of sediment being subducted, thus obviating the need for multiple metasomatic events over hundreds of million years. The distribution of lamproites, restricted within a north–south-trending graben, indicates that the initiation of east–west extension in south Tibet started at
25 Ma. KEY WORDS: lamproites; subducted oceanic sediment; Tibetan active continental collision belt
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL METHODSPETROGRAPHY AND MINERAL...GEOCHEMISTRYDISCUSSIONGEODYNAMIC IMPLICATIONSREFERENCES |
|---|
The processes responsible for the uplift and extension of the Tibetan plateau following the Indo-Asia collision have long been the subject of debate (e.g. Yin, 2000
). Proposed models for the geodynamic evolution of the Tibetan plateau include convective removal of sub-continental lithospheric mantle (SCLM; Houseman et al., 1981; Turner et al., 1993, 1996; Williams et al., 2001, 2004), slab break-off (Miller et al., 1999; DeCellers et al., 2002; Kohn & Parkinson, 2002; Mahéo et al., 2002), and intracontinental subduction (Arnaud et al., 1992; Meyer et al., 1998; Tapponnnier et al., 2001; Ding et al., 2003; Guo et al., 2006). Of critical importance for the evaluation of these models is their prediction of the spatial and temporal distribution of post-collisional magmatism on the Tibetan plateau, the duration and conditions of partial melting, and the nature and location of the magma source regions.
Widespread Neogene potassic and ultrapotassic volcanism in southern Tibet has been interpreted as the product of partial melting induced by convective thinning of the lithosphere beneath the Tibetan plateau (Turner et al., 1996; Williams et al., 2001, 2004; Chung et al., 2005). The magmas were considered to be derived from distinct SCLM sources (Turner et al., 1993, 1996; Miller et al., 1999; Williams et al., 2001, 2004). To account for their Nd–Sr–Pb isotopic signatures, it has been suggested that the Tibetan mantle lithosphere experienced a multi-stage history of metasomatism during Precambrian times, as inferred from Neoproterozoic Nd (0· 9–1· 3 Ga) and even older Pb (2· 2–3· 5 Ga) model ages (Miller et al., 1999). However, the probability that enriched Precambrian mantle lithosphere has remained chemically isolated and physically intact beneath Tibet for almost 1 Gyr appears unlikely when the complex Phanerozoic tectonic and magmatic evolution of Tibet is considered (Ding et al., 2003). As an alternative, Ding et al. (2003) proposed that the Tibetan mantle lithosphere could have been metasomatized much more recently (during Cenozoic times) by fluids or melts derived from ancient continental crust subducted beneath Tibet during the Indo-Asian collision. In their model, the Precambrian Nd and Pb model ages inferred for the Tibetan mantle lithosphere may reflect not the time when the lithosphere became isolated from asthenospheric convection, but rather inheritance of the isotopic signatures from fluids and melts derived from subducted ancient continental crust. One possible problem with this interpretation is that potassic volcanism in the Lhasa block appears to have started before the collision of the Indian continent (Coulon et al., 1986), hence before continental crust could have reached a depth to contribute to magmatism.
In our view, there is one central observation that helps to constrain the source of the ultrapotassic magmatism. Here we draw attention to the fact that true ultrapotassic rocks have chemical and isotopic characteristics that are distinct from those of the potassic volcanic rocks. The ultrapotassic lavas from the Lhasa block may therefore have a distinct, well-defined magma source. Miller et al. (1999) suggested that the post-collisional ultrapotassic lavas in SW Tibet have a clear affinity with lamproites from SE Spain and were derived from a separate region of the SCLM. We report major and trace element and Sr–Nd–Pb isotope data for newly discovered outcrops of ultrapotassic lavas from the Tangra Yumco–Xuruco graben in southern Tibet. The samples are chemically identical to the post-collisional ultrapotassic volcanic rocks previously identified from all over the Lhasa block (Miller et al., 1999; Williams et al., 2001, 2004; Ding et al., 2003; Nomade et al., 2004), and, therefore, are likely to be cogenetic. The objectives of this study are to define their petrological and geochemical characteristics, and to provide constraints on their mantle source. Our new data suggest that these lavas are geochemically distinct lamproites, and represent the first examples of lamproite magmas identified from the Tibetan continental collision zone.
|
GEOLOGICAL SETTING |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL METHODSPETROGRAPHY AND MINERAL...GEOCHEMISTRYDISCUSSIONGEODYNAMIC IMPLICATIONSREFERENCES |
|---|
The geodynamic setting of southern Tibet is particularly complex, as a result of the convergence between the Indian and Asian plates. From the mid-Miocene (25 Ma), there was a renewal of magmatic activity
25 Myr after the Indo-Asian collision (Mahéo et al., 2002). Since then, the Lhasa block has become the site of widespread, but small-volume, eruptions of ultrapotassic–potassic lavas, adakites and a lesser amount of more felsic magmas (Mahéo et al., 2002; Gao et al., 2003). In spite of the fact that post-collisional potassic volcanic rocks are widespread in the Lhasa terrane (e.g. Xungba, East Jarga, Shiquanhe, Namling, Maquiang, Daggyai Tso, Pabbai Zong, Zabuye, Wuyu, Dajia Co; Turner et al., 1996; Miller et al., 1999; Williams et al., 2001, 2004; Zhao et al., 2001; Ding et al., 2003; Nomade et al., 2004), there are only few outcrops of true ultrapotassic rocks, restricted to the western part of the Lhasa block (Fig. 1).
|
The recent tectonic style in southern Tibet is marked by north–south-trending normal faults and grabens accommodating east–west extension (Molnar & England, 1978
). The Tangra Yumco–Xuruco graben is one of a series of north–south-trending rifts cutting across the Lhasa block within the western part (Figs 1 and 2). The graben is about 300 km long and 40 km wide. North–south-trending active fault belts limit both its eastern and western margins. Two lakes (Tangra Yumco and Xuruco) lie within the rift. The flanks of the rift are dominated by widespread outcrops of voluminous Linzizong volcanic rocks and contemporaneous granitoid intrusions. Isolated occurrences of ultrapotassic–potassic lavas are associated with the rift flanks (Fig. 2). The total area with an ultrapotassic–potassic lava cover is more than 150 km2. Potassic lavas occur close to Chazi in the southern segment of the rift (Fig. 2); these have an 40Ar/39Ar age range from 13· 3 ± 0· 4 Ma to 8· 2 ± 0· 5 Ma (Ding et al., 2003). Two occurrences of ultrapotassic lavas are located at Chazi and Mibale, respectively (Fig. 2). The Chazi ultrapotassic lavas occur on the eastern flanks of the rift, and overlie Linzizong volcanic rocks. There are no exposed field relationships between the Chazi ultrapotassic lavas and the nearby potassic lavas. The Mibale ultrapotassic lavas are underlain by sandstones of the Cretaceous Jingzhushan Formation, and also occur on the eastern flanks of the rift (Fig. 2). These lavas have yielded a K–Ar age of 19· 04 ± 0· 97 Ma (Liao et al., 2002). A third occurrence of ultrapotassic lavas is a volcanic ‘neck’ at Wenbu on the western side of the rift, which has yielded 40Ar/39Ar sanidine ages from 21· 5 ± 0· 3 Ma to 17· 8 ± 0· 3 Ma (Ding et al., 2003). It should be noted that all these geochronological data are consistent with the age range (18–13 Ma) of north–south-trending ultrapotassic dykes at the southern end of the rift (Williams et al., 2001).
|
|
ANALYTICAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL METHODSPETROGRAPHY AND MINERAL...GEOCHEMISTRYDISCUSSIONGEODYNAMIC IMPLICATIONSREFERENCES |
|---|
Only fresh rock samples, visibly free of alteration, were selected for analysis. Major elements were determined by gravimetry (wet chemistry) and atomic absorption spectrometry (AAS). Sr and Nd isotope compositions of powered samples were obtained by thermal ionization mass spectrometry (TIMS) (MAT-262) at the Isotope Geology Laboratory, Chinese Academy of Geological Science (Beijing). Repeat measurements of NBS987 yielded 87Sr/86Sr = 0· 71025 ± 2 (2
) and the accuracy of the Rb/Sr ratio was better than 0· 1% (four repeats). Strontium isotope mass fractionation was corrected using 88Sr/86Sr = 8· 37521. The mean 143Nd/144Nd ratio in J&M metal yielded 0· 511125 ± 8 (2), and the accuracy of the Sm/Nd ratio was better than 0· 1% (six repeats). Neodymium isotope fractionation was corrected using 146Nd/144Nd = 0· 7219. Initial 
Nd values and 87Sr/86Sr ratios were calculated for an age (t) = 19 Ma. Pb-isotope ratios were determined by strictly pyrometer-controlled TIMS from Pb purified with miniaturized HBr–HCl chemistry from an aliquot of the trace element digestion. The reproducibility of the NBS SRM 981 standard was: 206Pb/204Pb = 16· 942 ± 0· 008; 207Pb/204Pb = 15· 496 ± 0· 011; 208Pb/204Pb = 36· 716 ± 0· 036; 207Pb/206Pb = 0· 914622 ± 0· 00018; 208Pb/206Pb = 2· 16714 ± 0· 0011.
For trace element and Pb isotope compositions, a crushed aliquot was powdered in an agate mill. Trace element concentrations were obtained by solution inductively coupled plasma-mass spectrometry (ICP-MS) at ACQUIRE, University of Queensland, following the analytical procedure described by Kamber et al. (2003). The compositions of phenocryst minerals in three samples were determined using a CAMECA SX51 electron microprobe at the Institute of Geology, Chinese Academy of Sciences. Analytical conditions were similar to those described by Ding et al. (2003).
|
PETROGRAPHY AND MINERAL CHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL METHODSPETROGRAPHY AND MINERAL...GEOCHEMISTRYDISCUSSIONGEODYNAMIC IMPLICATIONSREFERENCES |
|---|
All the ultrapotassic–potassic rocks from Mibale and Chazi are highly porphyritic with phenocrysts of 1–3 mm dimension. The petrography of the ultrapotassic lavas is summarized in Table 1. The lavas are characterized by phenocrysts of phlogopite, clinopyroxene, sanidine and rare olivine, with accessory apatite and ilmenite, similar to the typical mineralogical assemblage of sanidine–phlogopite lamproites from central Italy (Conticelli & Peccerillo, 1992
) and SE Spain (Venturelli et al., 1984). Phlogopite and clinopyroxene are the most abundant mafic phases. The groundmass consists mainly of microcrystalline leucite, diopside, sanidine, phlogopite and varying amounts of black glass. Fe–Ti oxides with variable grain size occur as both phenocrysts and in the groundmass. All the phlogopite phenocrysts show a flow-oriented arrangement. All the samples from both Chazi and Mibale contain two generations of phlogopite. Small phlogopite I grains exhibit a distinctive erosional, ‘polished’ margin that is interpreted to have been caused by abrasion during turbulent magmatic emplacement. In most of the samples, phlogopite I grains are corroded and occur as cores to phlogopite II phenocrysts. Phlogopite II occurs as both large euhedral platy phenocrysts and groundmass microlites. Some of the phlogopite II phenocrysts display distinct compositional zones of oriented overgrowth on phlogopite I. Most of the phlogopite phenocrysts have a thin dark rim of fine-grained magnetite.
|
The two generations of phlogopite exhibit significant differences in their composition (Table 2). The distinctive compositional feature of phlogopite I is its high Ti content (6· 07–8· 39%), which is characteristic of lamproites (Stephen & Taylor, 1992
). In contrast, phlogopite II has a lower TiO2 content with a wider variation range (1· 76–3· 82 wt %), similar to the chemistry of groundmass phlogopite. Micas in the studied ultrapotassic lavas have, on average, similar Mg-number values (73–91) to phlogopites from lamproites (70–90; Conticelli & Peccerillo, 1992). In general, phlogopite II has a higher Mg-number (mostly >85) than phlogopite I (mostly <80). Both types of phlogopite have low Al2O3 abundances, but phlogopite I grains have distinctively lower Al2O3 contents (10· 80–12· 88 wt %) than phlogopite II grains (12· 12–14· 16 wt %).
|
The strong optical zoning from phlogopite I cores to phlogopite II rims in some phenocrysts reflects compositional variation. Most phlogopite I grains contain abundant inclusions of Ti-oxides along their cleavages. The regular orientation of the inclusions suggests that they have exsolved during cooling. The compositional variation of the phlogopite phenocrysts is broadly similar to that of micas from the Cancarix and Middle Table Mountain lamproites (Mitchell & Edgar, 2002
). Experimental data (Mitchell & Edgar, 2002) indicate that these micas represent quench phases, and the micas increase their Ti contents with increasing pressure.
Most of the clinopyroxene phenocrysts are diopside–endiopside (representative compositions are given in Table 2). Individual clinopyroxene grains are typically homogeneous, but some show an increase in Mg-number from core to rim. Generally, clinopyroxene is characterized by low Ti and Al contents—a feature common to clinopyroxene in lamproites from other areas (e.g. SE Spain; Venturelli et al., 1984). Pigeonite with high FeO (up to 20· 9%) and K2O contents (2%) occurs rarely. Some of the clinopyroxene phenocrysts have fine exsolution lamellae of orthopyroxene.
Rare olivine occurs as small undeformed grains, and may be replaced by pseudomorphs of other silicates or carbonates. The olivine has high FeO (28· 93–36· 54 wt %) and low MgO contents (25· 69–32· 63 wt %) and Mg-number (0· 55–0· 67). Titaniferous magnetite is commonly the only Fe–Ti oxide phase present as phenocrysts and in the groundmass. One sample (Tl/17) has Cr-spinel grains in its groundmass. High-Cr spinels (Cr2O3 34· 79 wt %) also occur in the Xungba ultrapotassic lavas (Miller et al., 1999).
Sanidine is the sole feldspar that crystallized in these rocks. It is present as both phenocrysts and in the groundmass. Sanidine phenocrysts have an Or content range from 81· 9 to 58· 6%, and show an increase in Na2O and a decrease in Or content from core to rim (Table 3). Like sanidines from lamproites of SE Spain (Venturelli et al., 1984), the studied feldspars are frequently characterized by appreciable amounts of FeO and TiO2 (FeO 0· 25–1· 8 wt %, TiO2 0· 07–0· 49 wt %; Table 3). This is consistent with the occurrence of small Fe–Ti oxide inclusions in the feldspars.
|
|
GEOCHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL METHODSPETROGRAPHY AND MINERAL...GEOCHEMISTRYDISCUSSIONGEODYNAMIC IMPLICATIONSREFERENCES |
|---|
Major and compatible trace elements
Bulk-rock major and trace element compositions for ultrapotassic lavas from the Mibale and Chazi areas are given in Table 4. Conventionally, potassic rocks are defined as those in which K2O exceeds Na2O (wt % or molar). Foley et al. (1987
) defined ultrapotassic igneous rocks as having high contents of K2O >3 wt %, MgO >3 wt %, and K2O/Na2O >2 (wt %). Most of the samples meet these criteria for ultrapotasic rocks. Some samples (Tl/06 and Tl/59) from Mibale have lower MgO (<3%) and higher SiO2 contents (Table 4), and are more properly classified as potassic. The ultrapotassic lavas are characterized by low concentrations of Al2O3 (10· 87–13· 71 wt %), CaO (4· 04–7· 12 wt %) and Na2O (1· 33–2· 78 wt %), high K2O (except for sample Tl/10, 6· 34–9· 17%) and TiO2 (1· 06–1· 68%), and variable SiO2 contents (53· 33–60· 07 wt %). As the ultrapotassic lavas have high K2O and low Al2O3 contents, they have high molar K2O/Al2O3 ratios (except for sample Tl/10, 0· 60–0· 8) and molar (K2O + Na2O)/Al2O3 ratios (0· 78–1· 02) and variable molar K2O/Na2O (1· 33–4· 23). The lavas also have high Mg-numbers (0· 58–0· 73), and high Ni (124–338 ppm) and Cr (239–564 ppm) contents. Like the ultrapotassic rocks from other parts of the Lhasa block (Miller et al., 1999; Williams et al., 2001, 2004; Ding et al., 2003; Nomade et al., 2004), the studied ultrapotassic lavas display decreasing MgO and CaO contents and increasing Al2O3 and K2O contents with increasing SiO2 (Fig. 3). In general, the compatible trace elements Ni, Co, Cr, V and Sc correlate positively with MgO content or Mg-number (Fig. 4a).
|
|
|
Incompatible trace elements
One of the most striking features of the ultrapotassic lavas is their high incompatible trace element abundances, including large ion lithophile elements (LILE) and light rare earth elements (LREE). Their abundances reach the extreme levels typical of many lamproites. All ultrapotassic lava samples are considerably enriched in LREE (Ce: 303–490 times) relative to chondrite, whereas heavy REE (HREE) are less enriched (Yb: 4· 32–10· 08 times chondrite), resulting in CeN/YbN ratios that range from 35 to 101 (Table 4). All samples have similar REE patterns, which tend to flatten out in both the LREE and HREE and exhibit pronounced negative Eu anomalies (Fig. 5a and b). Similar REE patterns have been previously reported for lamproites from SE Spain (Nixon et al., 1984
; Contini et al., 1993) and central Italy (Rogers et al., 1985). Like ultrapotassic lavas from SW Tibet (Miller et al., 1999), the negative Eu anomalies in the Mibale and Chazi ultrapotassic lavas appear unrelated to plagioclase fractionation.
|
All the investigated ultrapotassic lava samples have high or very high abundances of Cs (4· 3–52· 7 ppm), Rb (391–939 ppm), Ba (2134–3931 ppm), Sr (660–1564 ppm), Th (130–224 ppm), Pb (85–166 ppm) and Zr (790–1101 ppm; Table 3). Thorium and the LILE, particularly Rb, are significantly enriched relative to high field strength elements (HFSE) and HREE. Like the ultrapotassic lavas from SW Tibet (Miller et al., 1999
), ultrapotassic primitive mantle-normalized abundance patterns (Fig. 5c and d) are characterized by negative Ba, Ta, Nb, Sr, P and Ti anomalies, which occur despite the high absolute abundances of these elements.
The trace element patterns are remarkably similar to those of the lamproites from SE Spain (Venturelli et al., 1984) and central Italy (Conticelli & Peccerillo, 1992). They are also broadly similar to those of post-collisional potassic lavas from northern Tibet (Turner et al., 1996) although negative Ba anomalies are more pronounced, and Rb and Th are far more enriched in the ultrapotassic lavas relative to the potassic lavas from northern Tibet. Low Ba/Th and pronounced negative Ba anomalies are observed in all the Mediterranean lamproite provinces (Peccerillo, 1999). The HFSE geochemistry of all the lamproite samples from this study area is characterized by less pronounced negative Nb–Ta and Ti anomalies, and more pronounced, positive, Zr and Hf anomalies (Fig. 5c and d) than those of ultrapotassic rocks from SW Tibet (Miller et al., 1999). All samples show pronounced positive Pb anomalies relative to Ce and Pr (Fig. 5c and d).
Pb–Sr–Nd isotope characteristics
Lead isotope ratios were obtained for all samples, and Nd and Sr isotope data were obtained for all samples from Mibale and five representative samples from Chazi (listed in Table 5 and plotted in Figs 6 and 7).
|
|
|
There are several important features of the radiogenic isotope systematics. First, in all three systems, the samples have very high 207Pb/206Pb, 208Pb/206Pb, radiogenic Sr and unradiogenic Nd. Second, there is a clear distinction in Sr and Pb isotope composition between the Mibale and Chazi ultramafic rocks. The former have consistently lower Sr isotope initial ratios (87Sr/86Sr = 0· 7166–0· 7258) than the lavas from Chazi (87Sr/86Sr = 0· 7308–0· 7363). The Mibale lavas have higher 207Pb/204Pb (15· 7709–15· 8407) and lower 206Pb/204Pb (18· 4590–18· 5770) than the Chazi samples (207Pb/204Pb = 15· 6732–15· 7340; 206Pb/204Pb = 18· 7567–18· 9310). There is no clear distinction in 208Pb/204Pb or in Nd isotope composition. Additionally, there are significant covariations in Pb/Pb and in Sr vs Pb space. These characteristics are not unique to our studied sample suites but extend to all Tibetan ultapotassic and even some potassic rocks, which invites the following comparison.
In conventional Pb isotope diagrams (Fig. 6a and b), all the post-collisional ultrapotassic–potassic rocks from the Lhasa and Qiangtang blocks form three fields. The first field is defined by ultrapotassic lavas from Xungba, Wenbu and Mibale. These have low 206Pb/204Pb with a rather small variation from 18· 420 to 18· 603 (Fig. 6a). Their 207Pb/204Pb (15· 698–15· 841) and 208Pb/204Pb (39· 271–40· 08) ratios are very radiogenic and plot well above the sediment evolution line of Kramers & Tolstikhin (1997). The second field comprises those rocks from the central Lhasa and Qiangtang blocks, which define a Pb isotope array with a shallow slope (Fig. 6a and b). The third field, which is intermediate in composition, is defined by potassic rocks from western Lhasa.
The first group of ultrapotassic lavas defines very steep arrays in both Pb isotope diagrams. Such a steep array in uranogenic Pb space cannot be produced by the closed-system decay of U, as this would result in highly radiogenic 206Pb/204Pb ratios as well as high 207Pb/204Pb values (Nelson, 1992). It is worth noting that at the unradiogenic end, this steep trend line extends to more depleted mid-ocean ridge basalt (MORB). In fact, it merges with the shallow array of ultrapotassic–potassic rocks from the central Lhasa and Qiangtang blocks at the more depleted isotopic compositions of Eocene (48 Ma) Dazi Na-rich calc-alkaline basalts (Fig. 6a and b, Gao et al., 2006). In these diagrams, the distinct steep trend may be a binary mixing line between a highly enriched source and a depleted mantle source. The principal observation to which we attach significance is that the ultrapotassic lavas from the Mibale, Xungba and Wenbu areas are isotopically unrelated to the potassic lavas from the central Lhasa and Qiangtang blocks.
The ultrapotassic rocks from the other outcrops (e.g. Pabbai Zong, Chazi and Shiquanhe) of the western Lhasa block have similar Pb isotope compositions to potassic rocks from the western Lhasa block (Fig. 6a and b). Compared with the potassic rocks of the eastern Lhasa and Qiangtang blocks, ultrapotassic–potassic rocks from the western Lhasa block have higher 207Pb/204Pb and 208Pb/204Pb ratios (Fig. 6a and b).
In Sr–Nd isotopic space, both types of the ultrapotassic samples plot in the enriched quadrant (Fig. 7) defining fields separate from the potassic rocks from central Lhasa and the Qiangtang block. The ultrapotassic lavas have highly radiogenic 87Sr/86Sr (0· 7106–0· 7239) and low Nd(t) ratios (–11· 9 to –16· 03; Miller et al., 1999; Williams et al., 2001; Ding et al., 2003; this study). Their range overlaps with the field defined by lamproites from SE Spain (Nelson, 1992). Although all ultrapotassic–potassic volcanic rocks from the western Lhasa block have low Nd(t) ratios, their Sr isotopic compositions define a large range (87Sr/86Sr 0· 70714–0· 73994). In terms of Sr isotope ratios, the ultrapotassic rocks from Chazi, Pabbai Zong and Shiquanhe overlap with potassic rocks from the western Lhasa block (Miller et al., 1999; Ding et al., 2003; Williams et al., 2004).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL METHODSPETROGRAPHY AND MINERAL...GEOCHEMISTRYDISCUSSIONGEODYNAMIC IMPLICATIONSREFERENCES |
|---|
Lamproitic affinity of ultrapotassic rocks in southern Tibet
Rocks belonging to the lamproite clan (Woolley et al., 1996
) exhibit a very wide range of modal mineralogy and geochemistry but may be divided into two broad types termed olivine lamproites and leucite lamproites. The latter group, also termed SiO2-rich lamproites, is characterized by the common presence of leucite, sanidine and phlogopite, and has intermediate SiO2 (46–60· 5 wt %) and relatively low MgO (11· 47–3· 24 wt %) contents (Altherr et al., 2004, and references therein). Table 6 lists the main geochemical characteristics of the ultrapotassic rocks from southern Tibet for comparison with lamproites from central Italy (Conticelli & Peccerillo, 1992); the compositions of the ultrapotassic lavas from the two areas are geochemically similar. Based on the classification for ultrapotassic rocks of Foley et al. (1987) and Foley (1992a), those from southern Tibet correspond to the chemical end-member group 1 and are therefore called leucite lamproites (Woolley et al., 1996). The SiO2 contents of the lamproite samples from the Lhasa block range from 51· 27 to 60· 62 wt % (Table 6) and this range generally overlaps with that of most Mediterranean occurrences, such as central Italy (56· 1–58· 1 wt %; Conticelli & Peccerillo, 1992) and Spain (46· 1–60· 5 wt %; Nixon et al., 1984). Samples from southern Tibet display variable Al2O3 contents (10· 87–14· 37 wt %) resulting in CaO–Al2O3 characteristics that are intermediate between those of lamproite and Roman type rocks, typical of active orogenic areas (e.g. SE Spain, central Italy, NW Alps, Corsica) with variable Al2O3 contents (7· 5–14· 8). Likewise, the K2O/Al2O3 ratios and TiO2 contents of lamproites from active orogenic areas tend to be significantly lower than those of lamproites from relatively stable continental settings. In this and many other respects, the southern Tibetan ultrapotassic rocks compare closely with those from the Mediterranean ‘lamproite’ provinces.
|
Separate magma source for lamproites and evidence for binary mixing
Previous studies have suggested that post-collisional ultrapotassic rocks are temporally and spatially associated with calc-alkaline to potassic rocks in southern Tibet (Miller et al., 1999
; Williams et al., 2001, 2004; Ding et al., 2003; Nomade et al., 2004; Chung et al., 2005). However, it has previously been noted that the ultrapotassic lavas from the western Lhasa block exhibit some important differences in mineralogy and geochemistry from the potassic rocks of the eastern Lhasa and Qiangtang blocks (Miller et al., 1999; Williams et al., 2004). In the southern Tibetan lamproites, MgO, CaO, TiO2 and P2O5 are negativly correlated with SiO2, whereas K2O and Al2O3 are positivly correlated with SiO2 (Fig. 3). These trends may reflect fractionation of olivine, phlogopite, clinopyroxene, spinel, apatite and Ti-bearing oxides during the petrogenesis of the lamproites. Despite this evidence for fractional crytsallization, a number of lines of evidence indicate that the potassic rocks from the eastern Lhasa block could not be generated by differentiation of parental lamproites.
First, the combined Sr and Nd isotope compositions of lamproites and potassic rocks from the eastern Lhasa and Qiangtang blocks clearly define two distinct isotopic populations (Fig. 7). The potassic lavas from the eastern Lhasa and Qiangtang terranes have relatively unradiogenic Sr and variably radiogenic Nd, resulting in a steep array, whereas the lamproites and associated potassic rocks from the western part of the Lhasa block cluster around the field of typical lamproites from other active orogenic areas (Fig. 7) with very radiogenic Sr and unradiogenic Nd.
Second, the isotopic difference also extends to the Pb isotope systematics (Fig. 6), where three different arrays can be recognized among post-collisional ultrapotassic and potassic rocks from the Lhasa terrane. The Pb isotope array for the lamproites is distinctive from that of the potassic rocks, indicating that these rocks could not be derived from each other via assimilation plus fractional crystallization (AFC) processes. Most of the mafic samples have much higher Rb content (550–880 ppm) than that of continental crust (bulk continental crust 32 ppm, upper continental crust 112 ppm; Taylor & McLennan, 1995). Thus, crustal contamination would only dilute the LILE and LREE contents of the ultrapotassic lavas. The Mibale lamproite with the most radiogenic Pb isotope signatures (sample Tl/10; 206Pb/204Pb = 18· 518, 207Pb/204Pb = 15· 841, 208Pb/204Pb = 40· 058; Table 5) is also one of the most mafic (MgO 11· 8 wt %; Table 4). These isotopic arrays therefore cannot be explained by crustal contamination.
Third, some of key trace element pairs of the lamproites indicate that the lamproites have a separate magma source from the potassic rocks of the eastern Lhasa block. For example, all of the potassic rocks from the eastern Lhasa block have lower Dy/Yb and Ce/Pb and ratios than the lamproites (Fig. 8a and b). High Dy/Yb ratios of the primitive lamproites probably reflect residual garnet in the source of melting, because crystal fractionation involving garnet in the lower crust could only increase the Dy/Yb ratios of more evolved differentiation products. The difference in Ce/Pb may also be related to the sources of these magmas, indicating that the potassic rocks may have originated from a mantle region more strongly enriched with fluid. South Tibetan lamproites also exhibit very high Nb/Ta ratios (17· 1–19· 84), much higher than most potassic rocks (Nb/Ta = 12· 66–17· 2; Miller et al., 1999; Zhao et al., 2001; Williams et al., 2004). Because Nb and Ta are not significantly decoupled during mantle melting, the distinctive high Nb/Ta ratios of the lamproites imply a special source from which this characteristic was inherited. Furthermore, the Th/La ratios, which are also robust, are high in the lamproites (0· 65–1· 41), ruling out significant contamination by crustal material.
|
Fourth, the eastern Lhasa potassic rocks and some of western Lhasa potassic rocks have lower incompatible trace element contents than the lamproites (Figs 4 and 5e). In the primitive mantle-normalized abundance diagram (Fig. 5e), these potassic rocks lack the pronounced negative Ba, Sr and P anomalies that are typical characteristics of the lamproites. Although these trace element differences could, in part, reflect different extents of fractionation, it is curious that such fractionation should have affected only some incompatible element concentrations and not others. The potassic rocks from the eastern Lhasa block have also lower incompatible trace element concentrations than any of the lamproitic samples with high MgO (Fig. 4b–d). Therefore, derivation from the lamproites through fractional crystallization alone cannot explain their trace element signatures.
Finally, there is a complete absence of lamproite contemporaneous with potassic rocks in the eastern Lhasa block, which could only be explained with preferential arrest of the hypothetical parental lamproite at depth.
In summary, the ultrapotassic rocks of lamproitic character described here differ from the more widely studied potassic rocks of eastern Lhasa and Qiangtang blocks in major, trace element and Sr–Nd–Pb isotope systematics to an extent that a separate mantle source is supported.
Of particular significance in evaluating potential mantle sources is the very steep trend in the uranogenic Pb-isotope diagram, with a slope of 0· 768, corresponding to an age greater than that of the Earth (Fig. 6a and b). This is incontrovertible evidence for binary mixing. Lead isotopes cannot discriminate between mixing of two actual melts and mixing of melt sources or contamination of one melt upon ascent. However, trace element systematics, such as a positive covariation between Nb/Ta and Pb/Nd (Fig. 8e) suggest that the purest lamproites with the highest Nb/Ta were particularly enriched in Pb. The mixing hypothesis for the southern Tibetan lamproites is further corroborated by the covariations of Ba/Nb values with La/Nb values, Th/Ce values with Th/Sm and Th/U values with Cs/Rb (Fig. 8c, d and f). Together, these observations can be used to identify the nature of potential isotopic and chemical end-members.
Nature and origin of the enriched magma source component
Lamproites are known for their very unusual Sr–Nd–Pb isotopic and chemical characteristics, which, in many respects, require an incompatible element ‘enriched’ source (LREE, LILE, radiogenic Sr, high to extremely high 207Pb/204Pb) but in others require a ‘primitive’ source (e.g. high Mg-number, high compatible element content). Numerous experimental studies call for the presence of potassium-bearing minerals in the mantle source of lamproites: phlogopite and amphibole (see the review by Edgar & Vukadinovic, 1992). The incompatible trace element and Sr–Nd–Pb isotopic characteristics of most orogenic lamproites indicate that the metasomatic components in their magma source regions are principally derived from subducted lithosphere, including subducted sediments (Nelson, 1992).
Metasomatism of mantle by highly incompatible enriched fluids and melts offers the potential, first, to overwhelm the mantle's original isotopic signature and, second, to rapidly evolve radiogenic isotope ratios on account of extreme parent/daughter ratios. In the context of the studied rocks, the key question is which of these processes was responsible for the large isotopic divergence, particularly in Pb. For this purpose, we next compare lamproitic Pb isotope ratios with those of potential subducted sediment.
We recall that the Mibale, Xungba (Miller et al., 1999) and Wenbu lamproites (Ding et al., 2003) have low Nd values (–12· 18 to –16· 03), unradiogenic 206Pb/204Pb ratios (18· 45–18· 56), together with high 207Pb/204Pb (15· 679–15· 841), 208Pb/204Pb ratios (39· 557–40· 058) and 208Pb*/206Pb* (1· 075–1· 145). In the Pb isotope diagrams (Fig. 9), it can easily be seen that the lamproites are higher in 207Pb/204Pb and 208Pb/204Pb but lower in 206Pb/204Pb, compared with modern (granite-derived) feldspars from the Indus River system (Clift et al., 2002). They correspond much better to Indian Ocean turbidites (Hemming & McLennan, 2001). In the 207Pb/204Pb vs 208Pb/204Pb diagram (Fig. 9d), in particular, the lamproites and Indian Ocean turbidites overlap. Chazi (Pabbai Zong and Shiquanhe) lamproites, on the other hand, have relatively high 206Pb/204Pb, and lower 207Pb/204Pb and 208Pb/204Pb (Figs 6 and 9a, b). These lamproites overlap with the Indus River sediments and some Indian Ocean turbidites in Pb isotope space (Fig. 9). Hence the large variation in Pb isotope compositions of the two lamproite suite could be entirely accounted for by inheritance from Tethyan oceanic sediments.
|
For those samples of Mibale and Chazi lamproites where we have combined Pb and Nd isotope data there is also an overlap with the range of modern deep-sea turbidites (Hemming & McLennan, 2001
) in the 208Pb/204Pb vs Nd diagram (Fig. 10). Apart from those samples with Nd < –20, modern deep-sea turbidites show a negative correlation between 208Pb/204Pb ratios and Nd, clearly related to the age of the continental sediment sources (Hemming & McLennan, 2001). If the high 208Pb/204Pb ratios and low Nd values of the lamproites indeed indicate inheritance from a sediment-derived metasomatizing agent, then these sediments must have been derived from old continental crustal sources.
|
There is an extreme difference between the concentration of Pb in oceanic sediment (3· 2–131 ppm; Tonga clay in Fig. 8a; Plank & Langmuir, 1998
) and in primitive mantle (0· 175 ppm, Hofmann, 1988), such that the addition of a very small fraction of Pb from oceanic sediment significantly affects mantle Ce/Pb and Pb isotopic ratios, which essentially take on the Pb isotopic characteristics of the sediment (Alther et al., 2004). Metasomatism can be envisaged as the interaction between a LILE-enriched fluid with ambient (often depleted) mantle peridotite (Tatsumi et al., 1986). However, this mechanism cannot explain the trace element and REE characteristics of the lamproites from south Tibet. Although the lamproites have pronounced negative Nb–Ta anomalies, the high HFSE abundances (Nb 34–80 ppm, Ta 1· 89–4· 5 ppm, Zr 340-1100 ppm) of the lamproites appear to require addition of HFSE to a depleted mantle source prior to melting. HFSE addition via an aqueous fluid is generally considered unlikely because of their low solubilities in hydrous fluids (Stolz et al., 1995), hence an alternative mechanism is required. This proposal is supported by the low Sr/Nd ratios (3· 1–9· 7) of the south Tibetan lamproites. Experimental data indicate that Sr is more soluble in high P–T aqueous fluids than the REE (Stalder et al., 1998), and elevated Sr/Nd ratios (>17; Anders & Grevesse, 1989) could be indicative of Sr addition to the arc magma source via an aqueous fluid (Kelemen et al., 2003). The south Tibetan lamproites also have relatively low Ba/La ratios (9–21) despite high Ba abundances. This results in clear negative Ba anomalies in the primitive mantle-normalized trace element abundance pattern (Fig. 5). Elliott et al. (1997) proposed that high Ba/La ratios in primitive arc magmas reflect addition of an aqueous fluid component to the mantle source, whereas high Th/La ratios have been interpreted as a component derived from partial melting of subducted sediment (Elliott et al., 1997; Kelemen et al., 2003; Plank, 2005). The lamproites do have very high Th/La ratios (0· 6–1· 4) suggesting that the Th-enriched sedimentary component in the Tibetan lamproite source was probably introduced via small-degree partial melts of subducted sediment, rather than aqueous fluids.
It is now well established that those lamproites for which high-quality trace element data exist have higher Nb/Ta ratios than MORB or continental crust (see also Murphy et al., 2002). Our lamproites are no exception (Fig. 8e). Consensus is emerging that the low continental crust Nb/Ta ratio (c. 11–13) compared with chondrites and MORB is caused by processes in subduction zones, where Ta is preferentially transported by fluids (relative to Nb) to the melt source in which the parental continental magmas form (Stolz et al., 1995; Kamber & Collerson, 2000; Kamber et al., 2003). Metasomatism of refractory lithospheric mantle by infiltration of such fluids should, therefore, lead to a potential magma source with low Nb/Ta. In contrast, the residual dehydrated slab will have an elevated Nb/Ta as a result of preferential Ta loss. Stolz et al. (1996) interpreted the high Nb/Ta values (up to 33) of subduction-related potassic volcanic rocks as due to modification of the subarc mantle wedge by silicic melts derived from the subducting slab and its sediment veneer.
A similar situation exists with respect to Th/U. Most lamproites have unusually high Th/U ratios when compared with MORB or ocean island basalt (OIB). Lead isotope data and U-series systematics (Williams et al., 1992) clearly demonstrate that this is not caused by Th/U partitioning during partial melting but is a characteristic of the source itself. The solubility of U in fluids emanating from subducted slabs is redox dependent; however, on average, Th is considered less mobile than U. Metasomatism of the mantle wedge by subduction zone derived fluids should thus lead to low Th/U sources, whereas the dehydrated slab will have a high Th/U ratio. It is important to note here that the relatively high Th/U ratio of post-Archaean upper continental crust is not a feature inherited from subduction zone processes but is caused by preferential loss of U in the weathering cycle. As a result, mature weathered sediments and red clays can acquire Th/U in excess of 10 (Plank & Langmuir, 1998).
We observe an anti-correlation between Th/U and Cs/Rb, which is also inconsistent with source metasomatism by subduction-derived fluids (Fig. 8f). Caesium is the most strongly enriched element in the continental crust and, by inference, the element most preferentially lost from dehydrated slabs, such that slab fluids are predicted to have high Cs/Rb ratios. However, our observation is that the lamproite samples with the highest Th/U have the lowest Cs/Rb ratios (Fig. 8f). Similarly, residual, partly dehydrated slabs should have comparatively low Cs/Rb ratios.
Previous studies have shown that the subduction-related magmas with a strong imprint of a partial melt or bulk melt of subducted sediment or subducted sediment itself in their source region have higher Th contents and Th/Ce ratios (Hawkesworth et al., 1997a, 1997b; Elburg et al., 2002; Guo et al., 2006) than those related to fluid metasomatism of their mantle source. The lamproites from southern Tibet have high Th/Ce and Th/Sm ratios (Fig. 8d) as well as a high Th content. The south Tibet lamproites show positive correlations both between Th/Ce and Th/Sm (Fig. 8d) and between Ba/Nb and La/Nb (Fig. 8c). These trends could suggest simple two-component mixing. One enriched end-member in the binary mixing system could be derived from subducted sediments. The average composition of GLOSS (Plank & Langmuir, 1998) plots within the mixing trends (Fig. 8c and d). However, we note that most of the lamproite samples have higher Th/Ce, Th/Sm, Ba/Nb and La/Nb, and that they plot beyond the range of the GLOSS and Indian MORB-source mantle (Fig. 8c and d). This could either argue for a local sediment composition different from GLOSS (as inidicated by the Pb isotope compositions) and/or fractionation (i.e. amplification) of the Th/Ce and Th/Sm ratios during partial melting of subducted sediment (Guo et al., 2006).
In summary, we propose that the enriched component of the Lhasa terrane lamproites can be derived from an oceanic sediment source. The trace element characteristics of the lamproites suggest that the enriched source component most probably was a partial melt derived from subducted sediment that had experienced dehydration-related preferential loss of the more fluid-mobile elements. In this model, the radiogenic Pb isotope ratios cannot be used to infer a long history of separation of the lamproite source per se. Rather, the complete lack of correlation between Th/U and 208Pb/204Pb (r2 = 0· 0089) argues for a young age for the source, in which the isotopic characteristics were inherited from ancient (weathered) continental crustal sources stored in sediment and transported into the mantle in a subduction zone.
Depleted source component and mantle metasomatism
The suggestion of deriving lamproites from melting subducted sediment meets with the apparent problem of their high MgO and compatible trace element contents. In the case of the Lhasa terrane lamproites, both Pb isotope and trace element systematics indicate, however, that a second end-member component, a separate melt or a contaminant was involved in their petrogenesis. In conventional Pb isotope space this component should have a much lower 208Pb/204Pb and 207Pb/204Pb than the ‘enriched’ end-member (Fig. 6a and b), possibly as unradiogenic as normal MORB (N-MORB). In the Dazi and Bangdaco areas, Na-rich basalts (Ding et al., 2003; Gao et al., 2006) are found with less radiogenic Pb isotope compositions. These were proposed to be derived from an asthenospheric mantle source beneath the Qiangtang and Lhasa terranes (Chung et al., 2005; Gao et al., 2006). The positively correlated array in 208Pb/204Pb vs 207Pb/204Pb (Fig. 9d) also projects to the range of composition of Indian ocean MORB (Rehkämper & Hofmann, 1997). It is therefore possible that the second (compatible element enriched) component is the depleted upper mantle.
This proposal is in agreement with trace-element ratios of similarly incompatible pairs, such as Nb/Ta, Pb/Nd, Th/Ce, Th/Sm, Ba/Nb and La/Nb, shown in Fig. 8c–e. In these diagrams, the lamproites have clear positive trends between Nb/Ta vs Pb/Nd, Th/Ce vs Th/Sm, Ba/Nb vs La/Nb. The Indian MORB (Rehkämper & Hofmann, 1997) and Bangdaco Na-rich basalts (Ding et al., 2003) have the required low Nb/Ta, Pb/Nd, Th/Ce, Th/Sm, Ba/Nb and La/Nb ratios (Fig. 8c–e).
We note that Guo et al. (2006) recently proposed a similar origin for post-collisional, ultrapotassic and potassic rocks of the Qiangtang terrane. Their trace element modelling indicates that the mantle source of the most primitive magmas in the North Qiangtang terrane contained a high amount of subducted sediment (0· 5–10%). Using a non-modal batch partial melting model (Wilson, 1989), Guo et al. (2006) determined that the most important factors that influence melt compositions in the envisaged environment are the degree of partial melting, the amount of subducted sediment melt added to the mantle source and the nature of residual phases in the source (e.g. phlogopite, rutile, titanite, apatite). The geochemical and mineral characteristics of the southern lamproites described here are generally consistent with the constraints of trace element modelling by Guo et al. (2006). According to their calculations, the proportion of subducted sediment-derived partial melt added to the mantle source ranges from 2% to around 10% for the primitive lamproite magmas studied here (Fig. 11).
|
|
GEODYNAMIC IMPLICATIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL METHODSPETROGRAPHY AND MINERAL...GEOCHEMISTRYDISCUSSIONGEODYNAMIC IMPLICATIONSREFERENCES |
|---|
If, as we have reasoned above, the geochemical and isotopic characteristics of the Lhasa terrane lamproites originated from partial melts of subducted sediment that metasomatized the overlying depleted upper mantle (either in the lithosphere or asthenosphere), there are geodynamic implications for the post-collision tectonic evolution of the Tibetan plateau. Metasomatism by partial melts of subducted sediments is significant, even if parts of the sub-continental lithospheric mantle were enriched by aqueous fluids from slab dehydration as inferred for the source of the associated potassic magmatism (Coulon et al., 1986
; Turner et al., 1993, 1996; Miller et al., 1999; Williams et al., 2001, 2004; Ding et al., 2003). The separate mantle sources of the post-collisional ultrapotassic and potassic rocks (Miller et al., 1999) reflect different styles of mantle source enrichment. Thus, the Precambrian Nd and Pb model ages inferred for the Tibetan mantle lithosphere by Miller et al. (1999) may not necessarily reflect the time when the lithosphere became isolated from asthenospheric convection, but rather the inheritance of isotopic signatures from melts derived from subducted oceanic sediments.
An important observation is that the lamproites appear to be restricted to the western part of the Lhasa block, whereas post-collisional potassic lavas occur throughout the Lhasa terrane (Nomade et al., 2004). Our results indicate a coexistence of distinct domains for the post-collisional ultrapotassic–potassic volcanism in south Tibet. The along-strike variation in the post-collisional ultrapotassic–potassic volcanism in the Lhasa block, in our view, is related to a westward increase in sediment input and the oblique subduction of the Neotethyan crust in the western Lhasa terrane. This inference was also made on the basis of a study of widespread adakite-like porphyries in southern Tibet (Gao et al., 2007). Compared with adakite-like porphyries from the eastern Lhasa terrane, the mantle source of the adakite-like porphyries in the western Lhasa terrane, where lamproites occur as well, also required higher amounts of subducted sediment (Gao et al., 2007).
In the convective thinning model (Turner et al., 1996; Williams et al., 2001, 2004), a substantial part of the lower sub-continental lithospheric mantle must have been removed before the initiation of the ultrapotassic–potassic volcanism in south Tibet. In fact, the eruption age (25–18 Ma) of the lamproites is the oldest of a period of renewed magmatic activity 10 Myr after the end of the subduction-related magmatism in southern Tibet (Mahéo et al., 2002), and the Lhasa terrane lamproitic volcanism may represent the initiation of the process of convection removal of sub-continental lithospheric mantle. If this inference is correct, removal of a substantial part of the lower sub-continental lithospheric mantle is not necessary at the onset of the lamproitic magmatism (25 Ma), because depleted asthenosphere itself is a valid end-member of two-component mixing for the lamproite magmas.
|
ACKNOWLEDGEMENTS |
|---|
This project is financially supported by the National Key Basic Research Project (No. 2002 CB 412600) and National Natural Science Foundation of China (40672042). We thank Prelevic Dejian, Zhengfu Guo and an anonymous reviewer for their constructive reviews.
*Corresponding author. Telephone: 0086 311 87207857. Fax: 0086 311 85882537. E-mail: gaoyf{at}sjzue.edu.cn
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGANALYTICAL METHODSPETROGRAPHY AND MINERAL...GEOCHEMISTRYDISCUSSIONGEODYNAMIC IMPLICATIONSREFERENCES |
|---|
Altherr R, Meyer HP, Holl A, Volker F, Alibert C, McCulloch MT, Majer V. (2004) Geochemical and Sr–Nd–Pb isotopic characteristics of late Cenozoic laucite lamproites from the East European Alpine belt (Macedonia and Yugoslavia). Contributions to Mineralogy and Petrology 147:58–73.[CrossRef][Web of Science]
Anders E and Grevesse N. (1989) Abundances of the elements: meteoritic and solar. Geochimica et Cosmochimica Acta 53:197–214.[CrossRef][Web of Science]
Arnaud NO, Vidal P, Tapponnier P, Matte PH, Deng WM. (1992) The high K2O volcanism of northwestern Tibet: geochemistry and tectonic implications. Earth and Planetary Science Letters 111:351–367.[CrossRef][Web of Science]
Boynton WV. (1984) Cosmochemistry of the rare earth elements: meteorite studies. In Henderson P (Ed.). Rare Earth Element Geochemistry(Elsevier, Amsterdam) pp. 63–114.
Chung S-L, Chu M-F, Zhang Y, Xie Y, Lo C-H, Lee T-Y, Lan C-Y, Li X, Zhang Q, Wang Y. (2005) Tibetan tectonic evolution inferred from spatial and temporal variations in post-collisional magmatism. Earth-Science Reviews 68:173–196.
Clift PD, Lee J, Hildebrand P, Shimizu N, Layne GD, Blusztajn J, Blum JD, Garzanti E, Khan AA. (2002) Nd and Pb isotope variability in the Indus River System: implications for sediment provenance and crustal heterogeneity in the Western Himalaya. Earth and Planetary Science Letters 200:91–106.[CrossRef][Web of Science]
Conticelli S and Peccerillo A. (1992) Petrology and geochemistry of potassic and ultrapotassic volcanism in central Italy: petrogenesis and inferences on the evolution of the mantle sources. Lithos 28:221–240.[CrossRef][Web of Science]
Contini S, Venturelli G, Toscani L, Capedri S, Barbieri M. (1993) Cr–Zr-armalcolite-bearing lamproites of Cancarix, SE Spain. Mineralogical Magazine 57:203–216.[CrossRef][Web of Science]
Coulon C, Maluski H, Bollinger C, Wang S. (1986) Mesozoic and Cenozoic volcanic rocks from central and southern Tibet: 39Ar/40Ar dating, petrological characteristics and geodynamic significance. Earth and Planetary Science Letters 79:281–302.[CrossRef][Web of Science]
DeCellers PG, Robinson DM, Zandt G. (2002) Implications of shortening in the Himalayan fold–thrust belt for uplift of the Tibetan plateau. Tectonics 21:1062 doi:10.1029/2001TC001322.
Ding L, Kapp P, Zhong DL, Deng WM. (2003) Cenozoic volcanism in Tibet: evidence for a transition from oceanic to continental subduction. Journal of Petrology 44:1835–1865.
Edgar AD and Vukadinovic D. (1992) Implications of experimental petrology to the evolution of ultrapotassic rocks. Lithos 28:205–220.[CrossRef][Web of Science]
Elburg MA, Bergen MV, Hoogewerff J, Foden J, Vroon P, Zulkarnain I, Nasution A. (2002) Geochemical trends across an arc–continent collision zone: magma sources and slab-wedge transfer processes below the Pantar Strait volcanoes, Indonesia. Geochimica et Cosmochimica Acta 66:2711–2789.
Elliott T, Plank T, Zindler A, White W, Bourdon B. (1997) Element transport from slab to volcanic front at the Mariana Arc. Journal of Geophysical Research 102:14991–15019.
Foley S. (1992a) Petrological characterization of the source components of potassic magmas: geochemical and experimental constraints. Lithos 28:187–204.[CrossRef][Web of Science]
Foley SF, Venturelli G, Green DH, Toscani L. (1987) The ultrapotassic rocks: characteristics, classification and constraints for petrogenetic models. Earth-Science Reviews 24:81–134.
Gao Y-F, Hou Z-Q, Wei R-H, Zhao R-S. (2003) Post-collisional adakitic porphyries in Tibet: geochemical and Sr–Nd–Pb isotopic constraints on partial melting of oceanic lithosphere and crust–mantle interaction. Acta Geology Sinica 77:194–203.
Gao Y-F, Hou Z-Q, Wei R-H, Meng X-J, Hu H-B. (2006) The geochemistry and Sr–Nd–Pb isotopes of basaltic subvolcanics from the Gangdese: contraints on depleted mantle source for post-collisional volcanisms in the Tibetan plateau. Acta Geologica Sinica 22:547–557 (in Chinese with English abstract).
Gao Y-F, Hou Z-Q, Kamber BS, Wei R-H, Meng X-J, Zhao R-S. (2007) Adakite-like porphyries from the southern Tibetan continental collision zones: evidence for slab melt metasomatism. Contributions to Mineralogy and Petrology 153:105–120.
Guo Z-F, Wilson M, Liu J-Q, Mao Q. (2006) Post-collisional, potassic and ultrapotassic magmatism of the Northern Tibetan plateau: constraints on characteristics of the mantle source, geodynamic setting and uplift mechanisms. Journal of Petrology 47:1177–1220.
Hawkesworth CJ, Turner SP, McDermott F, Peat DW, van Calsteren P. (1997a) U–Th isotopes in arc magmas: implications for element transfer from the subducted crust. Science 276:551–555.[CrossRef][Web of Science][Medline]
Hawkesworth CJ, Turner SP, McDermott F, Peat DW, van Calsteren P. (1997b) Element U and Th variations in island arc magmas: implications for U-series isotopes. Chemical Geology 139:207–221.[CrossRef][Web of Science]
Hemming SR and McLennan SM. (2001) Pb isotope compositions of modern deep sea turbidites. Earth and Planetary Science Letters 84:489–503.
Hofmann AW. (1988) Chemical differentiation of the Earth: the relationship between mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters 90:279–314.
Hofmann AW. (1997) Mantle geochemistry: the message from oceanic volcanism. Nature 385:219–229.[CrossRef]
Houseman GA, McKenzie DP, Molnar P. (1981) Convective thinning of a thickened boundary layer and its relevance for the thermal evolution of continental convergent belts. Journal of Geophysical Research 86:6115–6132.
Kamber BS and Collerson KD. (2000) Role of ‘hidden’ deeply subducted slabs in mantle depletion. Chemical Geology 166:241–254.[CrossRef][Web of Science]
Kamber BS, Greig A, Schoenberg R, Collerson KD. (2003) Refined solution to Earth's hidden niobium: implications for evolution of continental crust and depth of core formation. Precambrian Research 126:289–308.[CrossRef][Web of Science]
Kelemen PB, Yogodzinski GM, Scholl DW. (2003) Along-strike variation in the Aleutian island arc: genesis of high Mg-number andesite and implications for continental crust. In Eiler J (Ed.). The Subduction Factory. Geophysical Monograph, American Geophysical Union 138: pp. 223–246.
Kohn M and Parkinson CD. (2002) Petrologic case for Eocene slab breakoff during the Indo-Asian collision. Geology 30:591–594.
Kramers JD and Tolstikhin IN. (1997) Two terrestrial lead paradoxes, forward transport modeling, core formation and the history of the continental crust. Chemical Geology 139:75–110.[CrossRef][Web of Science]
Liao S-P, Chen Z-H, Luo X-C, Zhou A-J. (2002) Discovery of leucite phonolite in the Tangra Yumco area, Tibet, and its geological significance. Geological Bulletin of China 21:735–738 (in Chinese with English abstract).
Mahéo G, Guillot S, Blichert-Toft J, Rolland Y, Pêcher A. (2002) Slab breakoff model for the Neogene thermal evolution of South Karakorum and South Tibet. Earth and Planetary Science Letters 195:45–58.[CrossRef][Web of Science]
Meyer B, Tapponnier P, Bourjot LM, Etevier F, Gaudemer Y, Peltzer G, Guo S, Chen Z. (1998) Crustal thickening in Gansu–Qinghai, lithospheric mantle subduction, and oblique, strike-slip controlled growth of the Tibet plateau. Journal of Geophysical Research 135:1–47.
Miller C, Schuster R, Klotzli U, Mair V, Frank W, Purtscheller F. (1999) Post-collisional potassic and ultrapotassic magmatism in SW Tibet: geochemical, Sr–Nd–Pb-O isotopic constraints for mantle source characteristics and petrogenesis. Journal of Petrology 40:1399–1424.[CrossRef][Web of Science]
Mitchell RH and Edgar AD. (2002) Melting experiments on SiO2-rich lamproites to 6· 4 GPa and bearing on the sources of lamproites magmas. Mineralogy and Petrology 74:115–128.[CrossRef][Web of Science]
Molnar P and England P. (1978) Active tectonics of Tibet. Journal of Geophysical Research A: Space Physics 83:5361–5375.
Murphy DT, Kamber BS, Collerson KD. (2002) A refined solution to the first terrestrial Pb-isotope paradox. Journal of Petrology 44:39–53.
Nelson DR. (1992) Isotopic characteristics of potassic rocks: evidence for the involvement of subducted sediments in magma genesis. Lithos 28:403–420.[CrossRef][Web of Science]
Nixon PH, Thirwall MF, Buckley F, Davies CJ. (1984) Spanish and Western Australian lamproites: aspects of whole rock geochemistry. In Kornprobst J (Ed.). Kimberlites I: Kimberlites and Related Rocks(Amsterdam, Elsevier) pp. 285–286.
Nomade S, Renne PR, Mo X, Zhao Z, Zhou S. (2004) Miocene volcanism in the Lhasa block, Tibet: spatial trends and geodynamic implications. Earth and Planetary Science Letters 221:227–243.[CrossRef][Web of Science]
Peccerillo A. (1999) Multiple mantle metasomatism in central–southern Italy: geochemical effects, timing and geodynamic implications. Geology 27:315–318.
Plank T. (2005) Constraints from thorium/lanthanum on sediment recycling at subduction zones and the evolution of the continents. Journal of Petrology 46:921–944.
Plank T and Langmuir CH. (1998) The chemical composition of subducting sediment and its consequences for the crust and mantle. Chemical Geology 145:325–394.[CrossRef][Web of Science]
Rehkämper M and Hofmann AW. (1997) Recycled ocean crust and sediment in Indian Ocean MORB. Earth and Planetary Science Letters 147:93–106.[CrossRef][Web of Science]
Rogers NW, Hawkesworth CJ, Parker RJ, Marsh JS. (1985) The geochemistry of potassic lavas from Vulsini, central Italy, and implications for mantle enrichment processes beneath the Roman region. Contributions to Mineralogy and Petrology 90:244–257.[CrossRef][Web of Science]
Stalder R, Foley SF, Brey GP, Horn I. (1998) Mineral–aqueous fluid partitioning of trace elements at 900–1200°C and 3· 0–5· 7 GPa: new experimental data for garnet, clinopyroxene and rutile and implications for mantle metasomatism. Geochimica et Cosmochimica Acta 62:1781–1801.[CrossRef][Web of Science]
Stephen S and Taylor WR. (1992) Barium- and LREE-rich, olivine–mica-lamprophyres with affinities to lamproites, Mt. Bundey, Northern Territory, Australia. Lithos 28:303–325.[CrossRef][Web of Science]
Stolz AJ, Jochum KP, Hofmann AW, Stettel B. (1995) HFSE constraints on the nature of island arc and ocean island magma sources. Terra Nova 7:299–316.
Stolz AJ, Jochum KP, Stettel B, Hofmann AW. (1996) Fluid- and melt-related enrichment in the subarc mantle: evidence from Nb/Ta variations in island-arc basalts. Geology 24:587–590.
Sun SS and McDonough WF. (1989) In Saunders AD and Norry MJ (Eds.). Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. Magmatism in the Ocean Basins. Geological Society, London, Specical Publications 42:313–345.
Tapponnier P, Xu ZQ, Roger F, Meyer B, Arnaud N, Wittlinger G, Yang GS. (2001) Oblique stepwise rise and growth of the Tibet Plateau. Science 294:1671–1677.[CrossRef][Web of Science][Medline]
Tatsumi Y, Hamilton DL, Nesbitt RW. (1986) Chemical characteristics of fluid phase released from a subducted lithosphere and origin of arc magmas: evidence from high-pressure experiments and natural rocks. Journal of Volcanology and Geothermal Research 29:293–309.[CrossRef][Web of Science]
Taylor SR and McLennan S. (1995) The geochemical composition of the continental crust. Reviews of Geophysics 33:2241–265.[CrossRef][Web of Science]
Turner S, Hawkesworth C, Liu JQ, Rogers JQ, Kelley NS, van Calsteren P. (1993) Timing of Tibetan uplift constrained by analysis of volcanic rocks. Nature 364:50–54.[CrossRef]
Turner S, Arnaud N, Liu JQ, Rogers N, Hawkesworth C, Harris N, Kelley S, van Calsteren P, Deng WM. (1996) Post-collision, shoshonitic volcanism on the Tibetan plataeu: implications for convective thinning of the lithosphere and the source of ocean island basalts. Journal of Petrology 37:45–71.
Venturelli G, Capedri S, Di Battistini G, Crawford A, Kogarko LN, Celestini S. (1984) The ultrapotassic rocks of southeastern Spain. Lithos 7:37–54.
Williams HM, Turner S, Kelley S, Harris N. (2001) Age and composition of dikes in Southern Tibet: new constraints on the timing of east–west extension and its relationship to postcollisional volcanism. Geology 29:339–342.
Williams HM, Turner S, Pearce GA, Kelley SP, Harris N. (2004) Nature of the source regions for post-collisional, potassic magmatism in southern and northern Tibet from geochemical variations and inverse element modeling. Journal of Petrology 45:555–607.
Williams RW, Collerson KD, Gill JB, Denice C. (1992) High Th/U ratios in subcontinental lithospheric mantle–mass spectrometric measurement of Th isotopes in Gaussberg lamproites. Earth and Planetary Science Letters 111:257–268.[CrossRef][Web of Science]
Wilson M. (1989) Igneous Petrogenesis: a Global Tectonic Approach(Unwin Hyman, London) pp. 466.
Woolley AR, Bergman SC, Edgar AD, LeBas MJ, Mitchell RH, Rock NMS, Scott Smith BH. (1996) Classification of lamprophyres, lamproites, kimberlites, and the kalsilitic, melilitic and leucitic rocks. Canadian Mineralogist 34:175–186.
Yin A. (2000) Mode of Cenozoic east–west extension in Tibet suggesting a common origin of rifts in Asia during the Indo-Asia collision. Journal of Geophysical Research 105:21745–21759.
Xizang Bureau of Geology and Mineral Resources. (2002) Regional Geology of Xizang Autonomous Region(Geological Publishing House (in Chinese), Beijing).
Zhao Z-D, Mo X, Zhang S-Q, Guo T-Y, Zhou S, Dong G-C, Wang Y. (2001) Post-collisional magmatism in Wuyu basin, central Tibet: evidence for recycling of subducted Tethyan oceanic crust. Science in China (D) 44:Supplement, 27–34.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Maheo, J. Blichert-Toft, C. Pin, S. Guillot, and A. Pecher Partial Melting of Mantle and Crustal Sources beneath South Karakorum, Pakistan: Implications for the Miocene Geodynamic Evolution of the India-Asia Convergence Zone J. Petrology, March 1, 2009; 50(3): 427 - 449. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||