Geochemistry of late Miocene-Pleistocene basalts in the Phu Quy island area (East Vietnam Sea): Implication for mantle source feature and melt generation

herein). The objectives of this study are to use the major and trace element geochemical compo- sitions of basalts of Phu Quy island, in com- bination with geochemical data of basalts in the nearby areas and field survey data, in or- der to (1) classify the basalt series based on their geochemical and petrological character- istics, and possibility of crustal contamination of the melt; (2) to clarify the relationship be- tween the distribution characteristics of the basalts and local fault systems; and (3) to cal- culate the melting pressures of the basalts. 2. Geomorphological and tectonic charac- teristics of Phu Quy island Phu Quy island belongs to the south- Central continental shelf area from 10°29' to 10°33'N, 108°55'30" to 108°58'32.5"E of Phu Quy island district, Binh Thuan province. The island covers an area of 16.4 km2, surrounded by islets of Hon Tranh (2.8 km2), Hon Den, Hon Giua, Hon Do and Da Ty. The population is estimated to be 35,218 people in 2015, a density of 2,147 people/km2, with fishing as the major source of livelihood. According to the fieldwork data combined with the published literature (Koloskov et al., 1989, 1999, 2016), seven volcanic cones were identified in the Phu Quy island area including submarine volcanoes (Figure 1). The largest volcano is located in the northern part of the island with a height of 90 m, having a crater diameter of about 120 m and base diameter of about 1300 m. The remaining volcanoes are smaller in size, with crater diameters ranging between 80 and 100 m and base diameters from about 600 m to 1000 m. Volcanic activi- ty in the Vietnamese continental shelf is be- lieved to have occurred some 12 million years ago, the latest eruption having been recorded in 1923 forming the Ile des Cendres subma- Vietnam Journal of Earth Sciences, 39(3), 270-288 272 rine volcanic group (Lee et et al., 1998; Ko- loskov et al., 2009, 2016). The early eruption phase in Phu Quy island occurred about 2.6 Ma forming pyroclastic layers with a total thickness up to 60 m (Figure 1). The later cen- tral cone-type eruptions forming massive lava flows occurred about 1 Ma (Koloskov et al., 2016; Le Duc Anh et al., 2017). Figure 1. Location of Phu Quy island in the south-Central offshore of Vietnam (a) and a blow-up (b); shown are Cenozoic basaltic centers (shaded) in southern Vietnam and Western Highlands; also shown are (1) sampling sites and sample names, (2) iso-line contour, (3) central-type volcano, (4) supposed (marine) volcano. Distribution scheme is modified after Hoang et al. (1996); locations of volcanoes after field survey data Field survey records show that the geolog- ical structure of the northern and eastern areas of Phu Quy island was strongly deformed and crushed, forming two structurally frac- tured zones oriented northeast-southwest and sub-meridian, respectively (Figure 2a). The NE-SW oriented destruction zone is thought to be an ancient fault and is clearly docu- mented in the oil and gas seismic literature (Figure 2b) (Fyhn et al., 2009). This fracture zone is about 500 m wide with a 60° direc- tion forming a series of parallel fracture trends (Figure 2b). Although sub-meridian faults, according to seismic data, were rec- orded only in the vicinity of Phu Quy island (Figure 2b), the actual survey data at Phu Quy island, clearly record these faults also. These are smaller in size than the NE-SW faults forming depression zones, cutting young basalt layers (Figure 3b). Central cone volcanic eruptions occurred along the destruction zone (Figure 2a), suggesting vol- canic activity may reactivate old fault sys- tems (Rangin et al., 1995). Le Duc Anh, et al./Vietnam Journal of Earth Sciences 39 (2017) 273 Figure 2. (a) Geological scheme of Phu Quy island simplified from Koloskov et al. (1989, 1999, 2016) and field sur- vey data, (b) large NE-SW regional fault systems observed in the Phu Quy island area (after Fyhn et al., 2009); shown are basalt (1), welded pyroclastic deposits (2), soft, volcanic weathering products (3), volcanoes (4), supposed volcanoes (5), iso-line contour (6), defined faults (7), supposed faults (8). Age data are from Le and Hoang (2017) 3. Sampling and sample processing The basalt samples were collected at 16 locations on Phu Quy island (Figure 3). Thin sections of the samples were prepared for microscopic study at the Institute of Geology and Geophysics. Figure 3. (a, c) pyroclastic layers including volcanic bomb, ash, breccia and tuff products, (b) outcrops of massive basaltic blocks (far back) and pillow basaltic lavas, (d) extension faults running in NE-SW direction cutting through basaltic outcrops in Phu Quy island Vietnam Journal of Earth Sciences, 39(3), 270-288 274 For the chemical analyses, the samples were crushed into small chips about 1 cm by 1 cm, repeatedly rinsed with clean water be- fore being heated overnight in an oven at 100°C to reduce the effect of seawater, fol- lowed by rinsing with purified water before being dried in an oven. About 50g of the pro- cessed sample were crushed to <2 mm, about 25g of which were extracted to be ground in an agate mill. The major and trace element compositions were obtained from glass beads and pressed pellets using a Bruker Pioneer X-ray fluores- cence analyzer at the Institute of Geological Sciences, Vietnam Academy of Science and Technology (VAST). During the measure- ment process Geological Survey of Japan’s geochemical standards were used to estimate the analytical precision and accuracy of the XRF analyzer. Five representative samples were re-analyzed at the Geological Survey of Japan, using a Panalytical XRF to evaluate the analytical precision and accuracy. In general, the analytical 1 errors are better than ±0.5wt% for the major oxides and the trace elements (ppm) are better than ±6%. The data are shown in Table 1. Table 1. Major and trace element compositions, and mineralogical normative (CIPW) contents of the Phu Quy late Miocene - Pleistocene volcanic rocks Sample PQ01 PQ02 PQ03 PQ04 PQ05 PQ06 PQ07 PQ08 Rock type alkali basalt olivine-basalt olivine- basalt olivine- basalt olivine- basalt olivine- basalt olivine- basalt olivine- basalt SiO2 49.22 49.43 49 49.87 51.12 51.2 51.13 50.47 TiO2 2.29 2.4 2.22 2.39 1.88 1.85 1.78 2.1 Al2O3 14.47 14.32 13.88 14.12 14.88 14.7 14.82 13.77 FeO* 11.32 11.21 11.94 11.44 11.27 11.39 11.37 11.88 MnO 0.15 0.14 0.17 0.14 0.17 0.18 0.16 0.12 MgO 7.45 7.23 7.3 5.94 6.95 6.99 6.99 6.5 CaO 8.01 7.94 8.23 7.2 8.21 8.28 8.24 7.62 Na2O 3.17 3.41 3.18 3.25 3.29 3.42 3.45 3.33 K2O 2.4 1.64 1.9 2.62 1.28 1.2 1.23 1.4 P2O5 0.56 0.53 0.48 0.56 0.3 0.31 0.33 0.35 LOI 0.79 1.44 1.23 1.96 0.25 0.21 0.23 2.03 Total 98.9 98.1 98.1 97.6 99.3 99.4 99.3 97.4 CIPW Di 14.675 13.931 16.193 13.088 13.745 14.672 14.317 14.045 Hy 0 4.531 1.031 4.769 13.883 12.178 11.149 15.203 Ol 18.298 14.584 17.549 12.662 8.042 9.214 10.132 6.486 Ne 0.85 0 0 0 0 0 0 0 Q 0 0 0 0 0 0 0 0 Trace element (ppm) Rb 54 14 38 67 28 26 25 30 Sr 679 648 620 721 454 457 455 510 Y 20 28 23 27 26 31 44 31 Zr 220 270 180 270 136 135 135 160 Nb 44 47 41 53 24 24 24 33 Ba 597 434 534 583 402 427 335 352 Cr 267 272 345 262 300 306 305 314 Ni 211 189 240 212 210 226 221 212 Zn 111 113 119 128 114 123 116 132 Le Duc Anh, et al./Vietnam Journal of Earth Sciences 39 (2017) 275 Sample PQ10 PQ12 PQ13 PQ16 PQ17 PQ19 PQ20 PQ21 Rock type olivine- basalt olivine- basalt olivine- basalt olivine- basalt olivine- basalt olivine- basalt olivine- basalt olivine- basalt SiO2 48.44 50.96 50.79 49.15 49.12 51.45 48.46 49.11 TiO2 2.75 1.91 1.89 2.46 2.6 1.84 2.37 2.15 Al2O3 13.49 14.43 14.55 14.73 13.69 14.53 14.66 14.45 FeO* 13.22 11.54 11.56 11.34 12.19 11.40 12.14 12.77 MnO 0.13 0.16 0.15 0.14 0.14 0.15 0.15 0.15 MgO 6.48 6.93 7.15 7.62 6.87 6.98 6.66 6.09 CaO 7.18 8.25 8.23 7.57 7.84 8.19 7.42 7.5 Na2O 3 3.42 3.2 3.16 3.04 3.3 2.98 3.14 K2O 2.3 1.26 1.26 2.26 2.21 1.24 2.22 1.4 P2O5 0.49 0.31 0.27 0.47 0.48 0.33 0.55 0.37 LOI 2.39 0.49 0.32 0.88 1.71 0.09 2.15 2.62 Total 97.3 99.2 99.2 98.8 98.1 99.2 97.6 96.9 CIPW Di 13.12 15.308 14.347 12.395 15.285 14.233 10.839 11.269 Hy 5.611 11.068 13.823 0.482 4.229 15.925 5.168 14.186 Ol 14.565 9.756 8.516 18.863 14.531 6.589 15.315 8.472 Ne 0 0 0 0 0 0 0 0 Q 0 0 0 0 0 0 0 0 Trace element (ppm) Rb 64 29 30 43 57 28 35 33 Sr 757 459 464 667 713 457 718 547 Y 34 52 24 26 29 35 40 40 Zr 229 139 135 233 224 139 230 156 Nb 55 25 25 48 50 25 51 36 Ba 628 346 405 553 572 366 571 410 Cr 310 305 308 262 290 300 290 352 Ni 217 254 246 212 238 233 249 202 Zn 143 124 119 122 131 115 132 134 4. Analytical result 4.1. Petrography The Phu Quy island basalts are mostly massive, porphyritic with phenocrysts of oli- vine (5 to 12 vol%). The olivine phenocrysts are euhedral or subhedral, tablet or broken fragments with sizes ranging from 0.1 × 0.5 mm to 1 ×2 mm. Some olivine crystals are altered, being replaced partly or completely by iddingsite. The groundmass is intersertal, micro-doleritic, containing microcrysts of oli- vine, clinopyroxene, plagioclase, ore minerals and volcanic glass (Figure 4). 4.2. Major element compositions According to the CIPW normative compo- sitions (Table 1), except for PQ01 sample containing nepheline (Ne)-normative, which is an alkaline basalt, the remaining samples are quartz (Qz)- bearing free (subalkaline) ol- ivine basalt (Figure 5a, b). The basalts define two geochemical groups. One is high-SiO2 (50-52wt%), low-MgO, -TiO2 (<2.2wt%) - P2O5 (<0.4wt%) and low-K2O (<1.5wt%), termed as ‘low alkaline - high silica’ group, comprising samples PQ05, PQ06, PQ07, PQ08, PQ12, PQ13, PQ19 and PQ21. These were collected along the coastline, produced by early volcanic phase (Figures. 1b, 5b). The other is “high alkaline - low silica” group, having low- SiO2 (<50wt%), high-MgO, - TiO2 (2.2-2.8wt%) -P2O5 (>0.45-0.55wt%) and high-K2O (1.5 - 2.5wt%), comprising samples PQ01, PQ02, PQ03, PQ04, PQ10, PQ16 and PQ17. These were sampled near Vietnam Journal of Earth Sciences, 39(3), 270-288 276 volcanic craters, belonging to the later erup- tive phase. The high- and low- SiO2 contents correlate negatively with the corresponding MgO, from 6 to 7wt% and from 6.5 to 7.6wt%, respectively (Figure 6). The other ox- ides, however, do not show clear correlations with MgO. The SiO2 concentrations in melts depend mainly on melting pressure, the higher the pressure, the lower the SiO2 content; while MgO is dependent on both pressure and partial melting temperature, the higher the pressure and/or temperature, the higher the MgO con- tent (Takahashi and Kushiro, 1983; Hirose and Kushiro, 1993). Other major elements such as K2O, TiO2 and P2O5 behave like trace elements, whose contents are higher if the melting source is comparably more enriched and/or if the degree of partial melting is lower (Hirose and Kushiro, 1993; Turner and Hawkesworth, 1995; Kushiro, 1996, 1998; Kogiso et al., 1998).The above observations suggest that the basaltic melt of early eruptive phase may have originated from a relatively depleted and/or refractory source, under rela- tively low melting pressure and high degree of partial melting. In contrast, the later eruptive phase could have been derived from a relative- ly enriched and/or fertile source, under higher melting pressure temperatures and lower de- gree of partial melting. Figure 4. Phu Quy sub-alkaline (olivine) phyric basalt with intersertal (upper row) and doleritic (lower row) texture, showing phenocrysts of olivine partially replaced by iddingsite. Abbreviations Ol: olivine, Idd: iddingsite, Pl: plagioclase Le Duc Anh, et al./Vietnam Journal of Earth Sciences 39 (2017) 277 Figure 5. Classification diagram after CIPW normative compositions (after Cross et al., 1903) (a) and TAS (total al- kalis versus SiO2) (after La Bas et al., 1986) of the Phu Quy basalts; younger lavas (circles) have higher alkalis and more olivine content (closer to nepheline apex). Arrow shows evolutional trend. Ne: nepheline, Di: diopside, Qz: quartz, Ol: olivine, Hy: hypersthene Figure 6. Plots of MgO (wt.%) against major oxides of Phu Quy basalts; except for SiO2 which shows two broadly negative correlation trends with MgO for early and later eruptive lavas, the other oxides such as K2O, TiO2 and P2O5 de- fine higher and lower fields, respectively, for late (circle) and early (diamond) lavas. Plotted for comparison are syn- opening East Vietnam Sea (cross, after Li et al., 2014) showing even lower in MgO, TiO2 and K2O than Phu Quy basalts Vietnam Journal of Earth Sciences, 39(3), 270-288 278 4.2. Trace element compositions The primitive mantle normalized trace el- ement distribution patterns (after Sun and McDonough, 1989) have typical intraplate basalt-like signatures with a smooth decrease from highly incompatible to relatively immo- bile elements (Figure 7). However, the differ- ence in trace element concentrations between the two basaltic groups is significant. For ex- ample, Rb content varies from 35 to 56 ppm as compared to <35 ppm; Sr from 650 to 760 ppm compared to 450-510 ppm; Ba from 430 to 630 ppm (most >550 ppm) as compared to 335 to 410 ppm; and Zr>220 ppm as com- pared to 135 ppm, respectively, in the late ‘high alkaline - low silica’ and early ‘low al- kaline - high silica’ basaltic groups (Table 1). The trace element compositions of late basal- tic series are closely comparable to Pleiku youngest (<1 - 0.3 Ma) basalts in the Western Highlands, while those of Phu Quy early bas- alts are quite similar to Pleiku second young- est (5-2 Ma) basalts (Figure 7; Hoang, 2005a; Hoang et al., 2013) and are more enriched as compared to Pleiku >6 Ma basalts. The latter is believed to be derived from the relatively refractory and depleted lithospheric mantle, in contrast to the later 5-2 Ma and <1 - 0. Ma ba- saltic rocks which could have been derived from more fertile and enriched source under very high melting pressures, possibly in the asthenosphere (Hoang, 2005a, b; Hoang et al., 2013). 5. Discussion 5.1. Mantle geochemical characteristics 5.1.1. Crustal material interaction Basaltic magma is less viscous as com- pared to highly silicic magma, often pouring on the surface rapidly as lava flows or as ex- plosive pyroclastic deposits under the effect of high pressure. Therefore, the probability of crustal contamination in many cases is low. However, we should take into consideration the large distance from the melt segregation point to the surface, averaging by as much as 45 km to 90 km, for the tholeiitic and basaltic magmas, respectively (Kushiro, 1990, 1996), or long residence time and magma fractional crystallization at intermediate magma cham- bers in the crust that can interact with the wall-rock material. In addition, mantle sources can be contaminated with crustal ma- terial being introduced into the mantle by sub ducting slabs. The prevalence of crustal contamination is reflected in the ratios of the elements consid- ered to be rich in crustal material such as Ba, Rb, K, Si, Al,... and poor in the high field strength elements (HFSE) such as Zr, Y, Ti, and Nb,... (Taylor and McLennan, 1991; Fon- taine and Rudnik, 1995). Crustal material con- tamination is therefore characterized by posi- tive linear correlations between, for example, Ba and Ba/Nb with SiO2. However, the nega- tive correlation between Ba (ppm) and SiO2 and positive correlation between Ba/Nb and SiO2 as shown in Figure 8 may rule out crustal contamination. With the trace element distri- bution pattern viewed as typical for the in- traplate basalt (Figure 7), the correlation be- tween Ba and Ba/Nb with SiO2 reflects the en- richment or depletion nature of source domain in combination with the effect of melting pressures and temperatures, and degrees of partial melting (Takahashi and Kushiro, 1983; McKenzie and O'Nion, 1991; Scarrow and Cox, 1995; Kogiso et al., 1998; Kamenetski et al., 2012). 5.1.2. Mantle geochemistry Although accounting for just under 1 % by weight of the total rock mass, the trace ele- ments fully reflect the source characteristics, melting mechanisms, fractional crystallization processes from the time they are generated until the moment they appear on the surface (Hofmann, 1988; Sun and McDonough, 1989; Kogiso et al., 1998). Le Duc Anh, et al./Vietnam Journal of Earth Sciences 39 (2017) 279 Figure 7. Primitive mantle normalized trace element distribution diagram for early and late Phu Quy basalts, shown for comparison are Pleiku basalts erupted at 8.4 - 4.6 Ma, 4.6- 3.4 Ma and <2 - 0.24 Ma (after Hoang et al., 2013). The trace element patterns of the Phu Quy basalts are closely similar to enriched Pleiku basalts younger than 4.6 Ma Figure 8. Negative and positive correlation between SiO2 and, respectively, Ba (a) and Ba/Nb (b) for early (diamond) and late (circle) Phu Quy basalts; also plotted are syn-opening East Vietnam Sea (after Li et al., 2014). Higher Ba accompanying lower SiO2 and lower Ba/Nb (e.g. higher Nb) accompanying lower SiO2 may suggest effect of high melting pressure and/or low degrees of partial melting. Arrow shows temporal evolution trend The positive correlation between the Nb/Y and Zr/Y ratios of the two basalt groups distributed along the limit field of ΔNb (ΔNb = 1.74 + log (Nb/Y) -1.92 Log (Zr/Y)) is an index reflecting the melting of mantle sources, the nature of source enrichment or depletion Vietnam Journal of Earth Sciences, 39(3), 270-288 280 and the possibility of crustal contamination (Fitton et al., 1997; Hoang et al., 2016).The “low alkaline - high silica” basaltic group is characterized by low Nb/Y and low Zr/Y, dis- tributed near the mid-ocean ridge basalt field, both depleted (N-MORB) and enriched man- tle (E-MORB) (Kogiso et al., 1998; Relegous et al., 1999) (Figure 9), which is quite similar to the Dak Mil Neogene basalt (Dak Nong province, Western Highlands) (Hoang et al., 2013), that is interpreted to have been derived from a relatively depleted and refractory litho- sphere mantle or from sources that have un- dergone previous melting. The Phu Quy “high alkaline - low silica” basaltic group is distrib- uted near the Hawaii and North Arch oceanic island basalt fields (after Norman and Garcia, 1999; Frey et al., 2000), and plotted in Pleiku Pleistocene field (Hoang, 2005a, Hoang et al., 2013) (Figure 9). This distribution field is usually explained by the magma being gener- ated from enriched and fertile mantle sources (Hirose and Kushiro, 1993; Scarrow and Cox, 1995; Kushiro, 1996, 1998). Figure 9. Plots of Nb/Y versus Zr/Y for Phu Quy basalts; shown are fields of Pleiku 4-2 Ma lavas, Dak Nong >7Ma (after Hoang et al., 2013), North Arch alkaline basalts (Frey et al., 2000), Hawaiian picrites (Norman and Garcia, 1999), Depleted Mid-Ocean Ridge Basalts (N-MORB) (Regelous et al., 1999), Arc basalt, for comparison. The Phu Quy lavas are relatively enriched and plotted in the field of intraplate volcanic geochemistry The high field strength elements such as Nb, Zr and Y are not affected by the weather- ing process and considered to reflect the geo- chemistry of the source and the tectonic set- ting of magma occurrence. Of the above ele- ments, the distribution coefficient between the solid and the liquid phase (Kd) of Nb is much smaller than Zr, the latter is smaller than Y (Hofmann, 1988; Sun and McDonough, 1989). Therefore, the Nb/Y and Zr/Y ratios may vary depending on the degree of partial melting. Note that the clearly negative linear relationship is found only between SiO2 and MgO (Figure 6), while the other oxides do not Le Duc Anh, et al./Vietnam Journal of Earth Sciences 39 (2017) 281 exhibit a clear linear relationship with MgO. For basaltic rocks that occur in a small area having a similarity in eruption age, the unclear linearity in the geochemical relationship can only be explained by the difference in degrees of partial melting, melting pressures and/or temperatures, producing different eruption ep- isodes, regardless of their having been derived from the same mantle source. Thus, the linear relationship between Nb/Y and Zr/Y may be due to the difference in the melting degree and fractional crystallization processes of different melt portions generated from a single source mantle (e.g. McKenzie and Bickle, 1988; McKenzie and O'Nion, 1991). 5.2. Mantle dynamic model The formation and development of Phu Quy island was based on the relationship be- tween mantle dynamics (temperature and pressure) and the regional tectonic context re- sulting in melt generation and volcanic erup- tion (e.g. Latin and White, 1990). Based on the chemical compositions of the volcanic rocks, it is possible to determine the composi- tion of the primitive melt, indirectly determin- ing the mantle temperature and pressure (McKenzie and Bickle, 1988). 5.2.1. Decompression melting: mantle dynam- ics versus lithospheric extension Volcanic activities in the continental shelf of Vietnam is considered to be related to lith- ospheric extension processes triggered by deep-seated fault systems (Li et al., 2013; Franke et al., 2013; Sava et al., Carter et al., 2000). The processes were believed to com- bine with extruding mantle flows followed the Neo-Tethys closure as a result of the Indo- Eurasian collision (Flower et al., 1998, 2001; Hoang and Trinh, 2009; Yan et al., 2014). The mantle melting occurs normally as a result of extensional tectonic activity occur- ring in a large area and having a profound ef- fect on the lithosphere mantle. Typical de- compression melting producing basaltic melt occurs beneath the oceanic rift, that serves as a diverging boundary of two plates moving away from each other under the influence of two subduction zones and the combined effect of pushing force by basaltic lava having been formed as a result of the rifting process. Lith- ospheric extension occurs in two general forms, namely, uniform stretching and simple shear stretching (Latin and White, 1990). An extensional value is defined as  factor, the ratio between the original to extended litho- spheric thickness. Under normal mantle ther- mal condition, e.g. 1280°C, melting under the effect of uniform stretching is only possible if > 2.8, while there is no melting under pure shear stretching (McKenzie and Bickle, 1988; Latin and White, 1990). However, in higher mantle thermal conditions, for example, 1480°C, a uniform stretching can cause man- tle melting with a factor as small as 1.5 (compared to 4 of simple shear stretching). Seismic and deep borehole data acquired along and near the fossil spreading axis in the NE, Central and SW sub-basins of East Vietnam Sea revealed basalt and pyroclastic deposits interbedded with early Miocene (>16 Ma) sediments (Li et al., 2014, 2015). However, the rate and extent of the syn- spreading volcanism were modest, corre- sponding to slower to intermediate spreading rate (20 km to 35 km/Myr); although the spreading rate sometimes reached 70-80 km/Myr (Li et al., 2014; IODP 349 Sci- entific report, 2014). Detailed studies of the seismic and magmatic cross-sections in the southwestern sub-basin of East Vietnam Sea, revealed numerous submarine volcanoes that mostly appeared after 5 Ma, about 10 million years after the cessation of the spreading of the East Vietnam Sea (Li et al., 2013; 2014; 2015; Ding et al., 2016). In particular, these eruptions do not appear to be associated with any major tectonic phase in the area (Li et al., 2013). Also, when studying the regional ex- tension faults, these authors concluded that Vietnam Journal of Earth Sciences, 39(3), 270-288 282 the extension factor () hardly surpassed 2.8 to trigger melting of the mantle having a nor- mal thermal state (e.g. 1280°C); therefore, Li et al. (2013) proposed that upper mantle dy- namics rather than lithospheric tectonics caused mantle melting and magmatism in the SW sub-basin (Li et al., 2013; 2014). Neogene - Quaternary basaltic volcanism is widespread in many areas in the Western Highlands, mainly from about 7 million years (Hoang, 2005a; Hoang and Flower, 1998; Ho- ang et al., 1996; 2013), following the uplift of the territory. The basaltic layers cover an area of thousands of square kilometers, up to sev- eral hundred meters thick, distributed at inter- sections, or along the regional extension fault zones, with sub-meridian being the major di- rection (Rangin et al., 1995). There are no comprehensive studies evaluating the extent of lithospheric extension and accompanying extension factor values of the fault systems in the Western Highlands of Vietnam. But stud- ies of extension fault systems between longi- tudes 109° and 110°E, running from south of the Gulf of Bac Bo (Northern Gulf) southward to Con Co and Ly Son volcanic islands, to Phu Quy island then Ile des Cendres subma- rine volcanic group (Koloskov et al., 2016 and references therein), with a Moho plane at depth of about 28 km, shows a factor of about 1.3 (Nguyen et al., 2004; Nguyen and Nguyen, 2013). This extension value is much too small to trigger mantle melting under the normal thermal state (e.g. 1280°C, after Latin and White, 1990; see Carter et al., 2000). Tamaki (1995), followed by Flower et al. (1998; 2001), on the basis of deep seismic tomography data, proposed that the collision of Indian subcontinent with Eurasia in the late Eocene led to the closure of the Neo-Tethys situated in between the plates, which not only deformed the lithosphere but also generated the east-west directed asthenospheric flows (Tamaki, 1995; Yang and Liu, 2009, and ref- erences therein). These asthenospheric flows not only caused mantle melting and magma- tism from the Paleogene to the present time in East and Southeast Asia, but also acted as a major driver in the opening of the Western Pacific marginal seas, including East Vietnam Sea (Flower et al., 1998; 2001; 2013; Hoang and Flower, 1998; Hoang et al., 2013; 2014). While the extension factor in the East Vietnam (South China) Sea region and the Western Highlands of Vietnam is not large enough to cause decompression melting, a hotter-than-normal mantle (i.e. 1380°C) caused by the extrusion of asthenospheric flows is sufficient to trigger melting and magmatism, and also to generate various min- eral resources in the SW sub-basin (Li et al., 2013; 2014; 2015; Ding et al., 2016). 5.2.2. Primitive melts The geochemical composition of melt formed by partial melting of mantle peridotite is strongly correlated with the melting pres- sure and temperature, the degree of partial melting and the source geochemistry (McKenzie and Bickle, 1988; McKenzie and O'Nion, 1991). For example, the SiO2 content depends on the melting pressure, the lower the SiO2 content is, the higher the melting pres- sure. The FeO content is proportional to the melting temperature and the source chemistry; the more fertile the source is, the higher the FeO. The MgO content is proven to be pro- portional to the degree of partial melting (Takahashi and Kushiro, 1983; Hirose and Kushiro, 1993; Kushiro, 1996; 1998). Two trends of negative correlations between SiO2 and MgO observed in the major elemental di- agrams (Figure 6) represent two high and low melting pressure ranges, forming, respective- ly, low and high SiO2 groups. Elements like K, Na and Ti depend on the degree of partial melting, the lower is the degree, the higher is their content (Kushiro, 1996; 1998). There- fore, the low SiO2 content accompanied by high concentrations of Na2O, K2O and TiO2 Le Duc Anh, et al./Vietnam Journal of Earth Sciences 39 (2017) 283 (and vice versa) suggests that the melt genera- tion below Phu Quy island occurred in at least two major phases. The early phase was con- trolled by low melting pressure and high melt- ing degree (low alkalinity - high silica); while the later stage operated under high pressure and low melting degree (high alkalinity - low silica). The continuity between high- low melting pressures and low- high melting de- grees is also illustrated by the negative corre- lation between SiO2 and Ba (and positive cor- relation with Ba/Nb) (Figure 8), and also the positive relationship between Zr/Y and Nb/Y. (Figure 9), which may reflect the mixing pro- cess of mantle sources through column melt- ing (Kamenetski et al., 2012). In order to re-establish the primitive basalt composition of Phu Quy island, we apply a mathematical model based on the principle of olivine addition to basalt compositions having MgO content higher than 6wt% (Scarrow and Cox, 1995; Turner and Hawkesworth, 1995; Hoang and Flower, 1998, 2013, 2016). Oli- vine is compensated incrementally 0.1% (after Yamashita et al., 1996), using the Fe-Mg dis- tribution coefficient (Kd(Fe/Mg)) between oli- vine and the melt of about 0.30 (Roeder and Emslie, 1970; Takahashi and Kushiro, 1983) and forsterite (Mg/(Mg+ Fe2+) of olivine in the residue is approximately 0.90 (Hirose and Kushiro, 1993). The rationale for this calcula- tion is, first, that olivine with Fo83-89 is the ma- jor phenocryst observed in the Phu Quy basalt having MgO>6%; second, basalt with MgO>6% is generally less affected by pyrox- ene and plagioclase fractionation (Turner and Hawkesworth, 1995; Hoang and Flower, 1998; Hoang et al., 2013). The computed compositions of primitive melts are presented in Table 2. Table 2. Computed primitive melts of the Phu Quy basalts Sample PQ05 PQ06 PQ07 PQ08 PQ12 PQ13 PQ19 PQ21 SiO2 49.42 49.41 49.36 49.42 49.34 49.26 49.66 48.30TiO2 1.59 1.56 1.50 1.77 1.62 1.60 1.55 1.80Al2O3 12.51 12.33 12.44 11.56 12.15 12.27 12.20 12.06FeO* 11.95 12.04 12.02 12.88 12.24 12.24 12.08 13.97 MnO 0.17 0.18 0.16 0.13 0.16 0.15 0.15 0.16 MgO 13.12 13.13 13.15 13.31 13.07 13.29 13.12 13.06 CaO 6.93 6.97 6.94 6.42 6.97 6.96 6.91 6.29 Na2O 2.76 2.87 2.89 2.79 2.88 2.69 2.77 2.62K2O 1.07 1.01 1.03 1.17 1.06 1.06 1.04 1.17P2O5 0.25 0.26 0.28 0.29 0.26 0.23 0.28 0.31Total 99.78 99.76 99.77 99.74 99.75 99.76 99.76 99.72 Sample PQ01 PQ02 PQ03 PQ04 PQ10 PQ16 PQ17 PQ20 SiO2 48.14 48.57 48.14 48.83 47.69 48.14 48.19 47.78TiO2 1.98 2.07 1.91 2.00 2.32 2.13 2.22 2.02Al2O3 12.44 12.29 11.90 11.74 11.33 12.69 11.64 12.42FeO* 11.88 11.91 12.63 12.49 14.23 11.89 12.99 13.05 MnO 0.15 0.14 0.17 0.15 0.14 0.14 0.14 0.16 MgO 13.04 13.20 13.15 13.15 13.13 13.21 13.01 13.14 CaO 6.91 6.84 7.08 6.01 6.05 6.54 6.69 6.31 Na2O 2.72 2.92 2.72 2.70 2.52 2.72 2.58 2.52K2O 2.06 1.41 1.63 2.17 1.93 1.94 1.88 1.88P2O5 0.48 0.45 0.41 0.46 0.41 0.40 0.41 0.47Total 99.81 99.81 99.76 99.71 99.74 99.82 99.76 99.74 Using CMAS (CaO-MgO-Al2O3-SiO2) tet- rahedral diagrams of Walker et al. (1979), and the sub-solidus curves constructed from exper- imental results (Hirose and Kushiro, 1993; Ku- shiro, 1996, 1998), the primitive melt composi- tions projected from the diopside apex on to the Vietnam Journal of Earth Sciences, 39(3), 270-288 284 Ol-Pl-Q plane show melting pressures ranging from 18-20kb to 25kb (Figure 10). As shown above, Phu Quy island’s basalts, being quite primitive (with MgO>6wt%), may be formed from the same fertile and enriched source (Figure 9). The relatively close melting pres- sure range may suggest melts of the early and later phases may be produced by melting of mantle peridotites at different pressures (depths) in the same mantle source. Figure 10. Projections of computed primitive melt compositions on to Pl -Ol-Sil plane (after Walker et al., 1979), showing melting pressures of Phu Quy lavas ranging between about 18 to 25 Kb. Sub-solidus lines are constructed using experimental results of Hirose and Kushiro (1993), Kushiro (1996, 1998), (after Yamashita et al., 1996) 6. Summary and Conclusions From the above descriptions we come to the following conclusions: The late Miocene-Pleistocene volcanic rocks of Phu Quy island form two geochemi- cal groups. One is low-alkaline high-silica ol- ivine basalt, forming the base of the volcanic island. The other is high alkaline-low silica alkaline basalt, characterized by stratovolca- no-type eruptions. The trace element ratios such as Ba/Nb, Ba/Zr and Ba/Y of the early basalts are much higher compared to the later eruptive rocks, possibly due to the higher degree of partial melting. The Nb/Y, Zr/Y ratios and accompa- nying MgO, K2O, TiO2 and P2O5 contents are relatively low, suggesting that they may be derived from a relatively depleted and refrac- tory source, possibly in the lithospheric man- tle. The later (younger) alkaline basalts show- Le Duc Anh, et al./Vietnam Journal of Earth Sciences 39 (2017) 285 ing geochemical features contrasting with the early basalts such as higher TiO2, P2O5, and MgO, and lower SiO2 may be derived from a deeper and more enriched source, for example, asthenosphere. However, the continuity in the negative correlation between ratios such as Ba/Nb and Nb/Y versus SiO2 (or positive with MgO) suggests that both series of Phu Quy basalt may also be produced from the same mantle source due to the effect of variable melting pressures. Neogene-Quaternary volcanic activity in the south- Central Vietnam continental shelf and elsewhere in Vietnam was unrelated to an important regionally tectonic phase. The infiltration of asthenospheric flows, the consequence of Neo-Tethys closure following the collision of India and Eurasia, not only readily raises the mantle temperature to melt, but may also appear as an important driving force in the marginal sea openings, including the East Vietnam Sea. The melting pressures calculated based on the computed primitive melt compositions suggest that the early phase basaltic melt may be formed from 18 to 20 kb (ca. 55-60 km) and the later phase may be generated between 20 and 25 kb (about 60 to 75 km). The decompression polybaric melting of a single mantle source may have facilitated mixing among different melt portions to form linear geochemical relationships observed between two basaltic groups. Acknowledgements This paper was conducted under a project within the framework of prioritized scientific and technological research projects of the Vietnam Academy of Science and Technology ‘Study of geochemistry and petrology of basalts in the coastal and offshore areas in south- Central Vietnam with implications to mantle and lithospheric dynamics’ coded VAST.06.04/17-18. Proofreading and valuable suggestions by Rolando Pena (UP at Quezon) are gratefully acknowledged. References Baker M.B., Hirschmann M.M., Ghiorso M.S., Stolper E.M., 1995. Compositions of near-solidus peridotite melts from experiments and thermodynamic calculations. Nature 375, 308-311. Baker M.B., Stolper E.M., 1994. Determining the composition of high pressure mantle melts using diamond aggregates. Geochimica et Cosmochimica Acta 58, 2811-2827. Carter A., Roques D., Bristow C.S., 2000. Denudation history of onshore central Vietnam: constraints on the Cenozoic evolution of the western margin of the South China Sea. Tectonophysics 322, 265-277. Ding W., Li J., Clift P.D., & Expedition I.O.D.P., 2016. Spreading dynamics and sedimentary process of the Southwest Sub-basin, South China Sea: constraints from multi-channel seismic data and IODP Expedition 349. Journal of Asian Earth Sciences, 115, 97-113. Fitton J.G., Saunders A.D., Norry M.J., Hardarson B.S., Taylor R.N., 1997. Thermal and chemical structure of the Iceland plume. Earth and Planetary Science Letters 153, 197-208. Flower M.F.J., Zhang M., Tu K., Xie G.H., Chen C.Y., 1992. Magmatism in the South China Basin 2.Post- spreading Quaternary basalts from Hainan Island, South China. Chemical Geology 97, 65-87. Flower M., Tamaki K., Hoang N., 1998. Mantle Extrusion: A model for Dispersed Volcanism and DUPAL-like Asthenosphere in East Asia and the Western Pacific. In: Mantle dynamics and plate Interactions in East Asia, edited by: Flower, M et al., Geodynamic 27, 67-88. Flower M.F.J., Russo R.M., Tamaki K., Hoang Nguyen, 2001. Mantle contamination and the Izu-Bonin- Mariana (IBM) ‘high-tide mark’, evidence for mantle extrusion caused by Tethyan closure. Tectonophysics 333, 9-34. Franke D., Savva D., Pubellier M., Steuer S., Mouly B., Auxietre J.L., Meresse F., Chamot L.R., 2013. The final rifting evolution in the South China Sea. Marine and Petroleum Geology xxx, 1-17. Vietnam Journal of Earth Sciences, 39(3), 270-288 286 Fyhn M.B.W., Boldreel L.O., Nielsen L.H., 2009. Geological development of the Central and South Vietnamese margin: implications for the establishment of the South China Sea, Indochinese escape tectonics and Cenozoic volcanism. J. Tectonophysics, 478 (3-4), 184-214. Hirose K., Kushiro I., 1993. Partial melting of dry peridotites at high pressures: determination of compositions of melts segregated from peridotite using aggregates of diamond. Earth and Planetary Science Letters 114, 477-489. Hofmann A.W., 1988. Chemical differentiation of the Earth: the relationship between mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters 90, 297-314. Kamenetski V.S., Chung S.-L., Kamenetski M.B., Kuzmin D.V., 2012. Picrites from the Emeishan large igneous province, SW China: a compositional continuum in primitive magmas and their respective mantle sources. Journal of Petrology 53 (N10), 2095-2113. Kogiso T., Hirose K., Takahashi E., 1998. Melting experiments on homogeneous mixtures of peridotite and basalt: Application to the genesis of ocean island basalts. Earth and Planetary Science Letters 162, 45-61. Koloskov A.V., 1999. Ultrabasic inclusions and volcanics as a self-regulated geologic system. Nauchnyi Mir, Moscow (in Russian). Koloskov A.V., Fedorov P.I., Rashidov V.A., 2016. New data on products composition of the Quaternary volcanic activity in the shelf zone of NW margins of the South China Sea and the problem of asthenospheric diapirism. Koloskov A.V., Flerov G.B., Sharas’kin A.Y., 1989. Rift-Related Volcanism in the System of Eastern Asian Volcanic Belts. In Magmatism of Rifts: Petrology, Evolution, Geodynamics, Ed. by O. A. Bogatikov, Nauka, Moscow, 139-144 (in Russian). Koloskov A.V., Rashidov V.A., Gatinskii Y.G., et al., 2003. Late Cenozoic volcanism in the continental shelf zones of Vietnam. In Proceedings of the Annual Conference Devoted to the Day of Volcanologists. Naukadlya Kamchatki, Petropavlovsk Kamchatskii, 9-15 (in Russian). Kudrass H. R., Hiedicke M., Cepek P., Kreuzer H., and Müller P., 1986. Mesozoic and Cenozoic rocks dredged from the South China Sea (Reed Bank area) and Sulu Sea and their significance for plate tectonic reconstructions. Mar. Pet. Geol.3, 19-30. Kushiro I., 1990. Partial melting of mantle wedge and evolution of island arc crust. Journal of Geophysical Research 95, 15929-15939. Kushiro I., 1996. Partial melting of a fertile mantle peridotite at high pressures: An experimental study using aggregates of diamond. In Basu A., and Hart S., eds. Earth processes: Reading the isotopic clock: American Geophysical Union Geophysical Monograph 95, 109-122. Kushiro I., 1998. Compositions of partial melts formed in mantle peridotites at high pressures and their relation to those of primitive MORB. Physics of the Earth and Planetary Interiors 107, 103-110. Latin D., White N., 1990. Generating melt during lithospheric extension: uniform vs. simple shear. Geology 18, 327-331. Le Bas M.J., Le Maitre R.W., Streckeisen A., Zanettin B., 1986. A chemical classification of volcanic rocks based on the total alkali-silica diagram. Journal of Petrology 27, 745-750. Le Duc Anh, Nguyen Hoang, 2017. Geochemistry of Cenozoic basalts in the south-Central Vietnam coastline region: implications for regional mantle - lithosphere interaction. Vietnam Journal of Earth Sciences, in preparation. Lee T.Y., Lo C-H., Chung S-L., Chen C-Y., Wang L., Lin W-P., Nguyen Hoang, Cung Thuong Chi, Nguyen Trong, Yem, 1998. 40Ar/39Ar dating result of Neogene basalts in Vietnam and its tectonic implication. In: Flower, M.F.J., et al. (Eds.), Mantle Dynamics and Plate Interactions in East Asia. Geodynamics Series, vol. 27, American Geophysics Union, 317-330. Li C.-F., et al., 2014. Ages and magnetic structures of the South China Sea constrained by deep tow magnetic surveys and IODP Expedition 349, Geochem. Geophys. Geosys., 15, 4958-4983. Li C-F., et al., 2015. Seismic stratigraphy of the central South China Sea basin and implications for neotectonics. J. Geophys. Res. Solid Earth, 120, 1377-1399. Doi:10.1002/2014JB011686. Le Duc Anh, et al./Vietnam Journal of Earth Sciences 39 (2017) 287 Li C.-F., Lin J., Kulhanek D.K., 2013. South China Sea tectonics: Opening of the South China Sea and its implications for southeast Asian tectonics, climates, and deep mantle processes since the late Mesozoic, IODP SCI. Prosp., 349. Li L., Clift P., Nguyen The Hung, 2013. The sedimentary, magmatic and tectonic evolution of the southwestern South China Sea revealed by seismic stratigraphic analysis. Marine Geophysical Research 34, 341-365. Li L., Clift P.D., Stephenson R., Nguyen T.H., 2014. Non-uniform hyper-extension in advance of seafloor spreading on the Vietnam continental margin and the SW South China Sea. Basin Research 26, 106-134. Malinovsky A.I., Rashidov V.A., 2015. Compositional characteristics of sedimentary and volcano- sedimentary rocks of Phu Quy-Catwickisland group in the continental shelf of Vietnam. Bulletin of Kamchatka Regional Association of ‘Educational - Scientific’ Center, Earth Sciences 27 (3), 12-34 (in Russian with English summary). McKenzie D., Bickle M.J., 1988. The volume and composition of melt generated by extension of the lithosphere. Journal of Petrology 26, 625-679. McKenzie D., O’Nions R.K., 1991. Partial melt distribution from inversion of rare earth element concentrations. Journal of Petrology 32, 1021-1091. Nguyen Hoang, 2005a. Source characteristics and melting conditions of Cenozoic basalts from Pleiku, south-central Vietnam. Journal of Geology Series A 286, 15-22 (in Vietnamese with English abstract). Nguyen Hoang, 2005b. Lithospheric mantle beneath Pleiku: evidence from mantle xenolith. Journal of Geology Series A 287, 8-19 (in Vietnamese with English abstract). Nguyen Hoang, Flower M.F.J., 1998. Petrogenesis of Cenozoic basalts from Vietnam: implication for origins of a diffuse igneous province. J. Petrol. 39, 369-395. Nguyen Hoang, Flower M.F.J., Carlson R.W., 1996. Major, trace element, and isotopic compositions of Vietnamese basalts: interaction of enriched mobile asthenosphere with the continental lithosphere? Geochimica et Cosmochimica Acta 60, 4329-4351. Nguyen Hoang, Uto K., 2003. Geochemistry of Cenozoic basalts in the Fukuoka district (northern Kyushu, Japan), implications for asthenosphere and lithospheric mantle interaction. Chemical Geology 198, 249-268. Nguyen Hoang, Phan Trong Trinh, 2009. Overview of petrologic and geochemical characteristics of Neogene-Quaternary basalts in the East Vietnam Sea and neighboring regions and their related mantle dynamics. Journal of Geology, 312A, 39-57 (in Vietnamese with English abstract). Nguyen Hoang, Flower M.F.J., Cung Thuong Chi, Pham Tich Xuan, Hoang Van Quy, Tran Thanh Son, 2013. Collision-induced basalt eruptions at Pleiku and Buon Me Thuot, south-central Viet Nam. Journal of Geodynamics 69, 65-83. Nguyen Hoang, Ogasawara M., Tran Thi Huong, Phan Van Hung, Nguyen Thi Thu, Cu Sy Thang, Pham Thanh Dang, Pham Tich Xuan, 2014. Geochemistry of Neogene basalts in the Nghia Dan district, western Nghe An. Vietnam Journal of Earth Sciences 36(4), 403-412. DOI: 10.15625/0866- 7187/36/4/6428. Nguyen Hoang, Tran Thi Huong, Ogasawara M., Le Duc Anh, Nguyen Thi Mai, Nguyen Thi Thu, Cu Sy Thang, Le Thi Phuong Dung, 2016. Petrography and geochemistry of Permian basalts of the Cam Thuy formation and their relation to Song Da and Emeishan magmatic rocks. Vietnam Journal of Earth Sciences 38, 372-392. DOI: 10.15625/0866- 7187/38/4/8993. Nguyen Nhu Trung, Lee S.M., Bui Cong Que, 2004. Satellite gravity anomalies and their correlation with the major tectonic features in the South China Sea. Gondwana Research 7(2), 407-424. Nguyen Nhu Trung, Nguyen Thi Thanh Huong, 2013. Topography of the Moho and Earth crust structure beneath the East Vietnam Sea from 3D inversion of granite field data. Acta Physica 61(2), 357-384. Norman M.D, Garcia M.O., 1999. Primitive magmas and source characteristics of the Hawaiian plume: petrology and geochemistry of shield picrites. Earth and Planetary Science Letters 168, 27-44. Pearce J.A., Norry M.J., 1979. Petrogenetic implications of Ti, Zr, Y, and Nb variations in volcanic rocks. Contributions to Mineralogy and Petrology 69, 33-47. Vietnam Journal of Earth Sciences, 39(3), 270-288 288 Rangin C. Huchon, Le Pichon X., Bellon H., Lepvrier C., Roques D., Hoe N.D., Quynh V., 1995. Cenozoic deformation of central and south Vietnam. Tectonophysics 251, 179-196. Regelous M., Niu Y., Wendt J.I., Batiza R., Greig A., Collerson K.D., 1999. Variations in the geochemistry of magmatism on the East Pacific Rise at 10°30’N since 800 ka. Earth Planet. Sci. Lett. 168, 45-63. Roeder P.L., Emslie R.F., 1970. Olivine-liquid Equilibrium. Contributions to Mineralogy and Petrology 29, 275-289. Ru K., Pigott J.D., 1986. Episodic rifting and subsidence in the South China Sea. AAPG Bull 70(9), 1136-1155. Rudnik R.L., Fontaine D.M., 1995. Nature and composition of the continental crust: a lower crustal perspective. Review Geophysics, 33, 267-309. Sava D., Pubellier M., Franke D., Chamot-Rooke N., Meresse F., Steuer S., Auxietre J.L., 2014. Different expressions of rifting on the South China Sea margins. Marine and Petroleum Geology 58, 579-598. Scarrow, J.H., Cox, K.G., 1995. Basalts generated by decompressive adiabatic melting of a mantle plume: a case study from the Isle of Skye, NW Scotland. Journal of Petrology 36, 3-22. Sun S.S., McDonough W.F., 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: Saunders, A.D., Norry, M.J. Eds. Magmatism in Ocean Basins. Geol. Soc. Spec. Publ., London, 313-345. Takahashi E., Kushiro I., 1983. Melting of a dry peridotite at high pressures and basalt magma genesis. American Mineralogist 68(9-10), 859-879. Tamaki K., 1995. Upper mantle extrusion tectonics of southeast Asia and formation of western Pacific backarc basins. In: International Workshop: Cenozoic Evolution of the Indochina Peninsula, Hanoi/Do Son, April, p.89 (Abstract with Programs). Taylor S.R., McLennan S.M., 1985. The continental crust: Its composition and evolution. Blackwell Scientific Publishers, Oxford. Tran Duc Luong, Nguyen Xuan Bao, 1981. Geological map 1:500.000 of Vietnam. Tu K., Flower M.F.J., Carlson R.W., Zhang M., Xie G.- H., 1991. Sr, Nd, and Pb isotopic compositions of Hainan basalts (South China), implications for a subcontinental lithosphere DUPAL source. Geology 19(6), 567-569. Tu K., Flower M.F.J., Carlson R.W., Zhang M., Xie G.H., 1992. Magmatism in the South China Basin, 1.Isotopic and trace element evidence for an endogenous Dupal mantle component. Chemical Geology 97, 47-63. Turner S., Hawkesworth C., 1995. The nature of the sub- continental mantle: constraints from the major element composition of continental flood basalts. Chemical Geology 120, 295-314. Walker D., Shibata T., DeLong D.E., 1979. Abyssal tholeiites from the Oceanographer Fracture Zone III. Phase equilibria and mixing. Contributions to Mineralogy and Petrology 70, 111-125. Yamashita S., Tatsumi Y., Nohda S., 1996. Temporal variation in primary magma compositions in the northeast Japan Arc, The Island Arc5, 276-288. Yan Q., Shi X., Castillo R.B., 2014. The late Mesozoic- Cenozoic tectonic evolution of the South China Sea: A petrologic perspective. Journal of Asian Earth Sciences 85, 178-201. Yang Y., Liu M., 2009. Crustal thickening and lateral extrusion during the Indo-Asian collision: a 3D viscous flow model. Tectonophysics 465, 128-135. Zhou P., Mukasa S., 1997. Nd-Sr-Pb isotopic, and major- and trace-element geochemistry of Cenozoic lavas from the Khorat Plateau, Thailand, sources and petrogenesis. Chemical Geology 137, 175-193.