Geochemistry of Picritic and AssociatedBasalt Flows of the Western Emeishan FloodBasalt Province, China

ZHAOCHONG ZHANG1*, JOHN J. MAHONEY2, JINGWEN MAO1

AND FUSHENG WANG3

1STATE KEY LABORATORY OF GEOLOGICAL PROCESSES AND MINERAL RESOURCES, CHINA UNIVERSITY OF

GEOSCIENCES, BEIJING, 100083, P. R. CHINA

2SCHOOL OF OCEAN AND EARTH SCIENCE AND TECHNOLOGY, UNIVERSITY OF HAWAII,

HONOLULU, HI 96822, USA

3INSTITUTE OF GEOLOGY, CHINESE ACADEMY OF GEOLOGICAL SCIENCES, 100037, P.R. CHINA

RECEIVED DECEMBER 13, 2005; ACCEPTED JUNE 8, 2006;ADVANCE ACCESS PUBLICATION JULY 14, 2006

Picritic lava flows near Lijiang in the late Permian Emeishan flood

basalt province are associated with augite-phyric basalt, aphyric

basalt, and basaltic pyroclastic units. The dominant phenocryst in

the picritic flows is Mg-rich olivine (up to 91.6% forsterite

component) with high CaO contents (to 0.42 wt %) and glass

inclusions, indicating that the olivine crystallized from a melt.

Associated chromite has a high Cr-number (73–75). The estimated

MgO content of the primitive picritic liquids is about 22 wt %,

and initial melt temperature may have been as high as 1630–

1690�C. The basaltic lavas appear to be related to the picritic

ones principally by olivine and clinopyroxene fractionation. Age-

corrected Nd–Sr–Pb isotope ratios of the picritic and basaltic

lavas are indistinguishable and cover a relatively small range [e.g.

eNd(t) ¼ �1.3 to þ4.0]. The higher eNd(t) lavas are isotopicallysimilar to those of several modern oceanic hotspots, and have ocean-

island-like patterns of alteration-resistant incompatible elements.

Heavy rare earth element characteristics indicate an important role

for garnet during melting and that the lavas were formed by fairly

small degrees of partial melting. Rough correlations of isotope ratios

with ratios of alteration-resistant highly incompatible elements

(e.g. Nb/La) suggest modest amounts of contamination involving

continental material or a relatively low-eNd component in the source.Overall, our results are consistent with other evidence suggesting

some type of plume-head origin for the Emeishan province.

KEY WORDS: Emeishan; flood basalts; picrites; mantle plumes;

late Permian

INTRODUCTION

The Emeishan province of southwestern China (Fig. 1)is one of two continental flood basalt provinces thatformed near the end of the Permian in widely separatedlocations, the other being the Siberian Traps (e.g.Chung & Jahn, 1995). Eruption of a third flood basaltprovince, the Panjal Traps of northwestern India, mayhave overlapped these two provinces in time (e.g. Bhat &Zainuddin, 1981), and at least two oceanic plateauxformed in the ensuing �20 Myr (Lassiter et al., 1995;Genc, 2004). An even larger outburst of widespreadflood basalt and plateau volcanism occurred in theEarly Cretaceous, from about 145 to 110Ma (e.g. Renneet al., 1992; Mahoney et al., 1993, 2005; Stewart et al.,1996; Duncan, 2002; Kent et al., 2002; Tejada et al.,2002; Hoernle et al., 2004). Such periods have beenproposed to represent episodes of mantle overturn,which is postulated to trigger the formation and ascentof multiple plume heads (Stein & Hofmann, 1994). Theplume-head or starting-plume hypothesis (e.g. Richardset al., 1989; Griffiths & Campbell, 1991; Campbell,1998), in turn, has been the hypothesis invoked mostfrequently in recent years to explain individual con-tinental flood basalts and oceanic plateaux.However, evidence has mounted recently that several

major flood basalt provinces and plateaux lack keycharacteristics predicted by the plume-head model.For the late Permian flood basalts specifically,

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 PAGES 1997–2019 2006 doi:10.1093/petrology/egl034

*Corresponding author. E-mail: zcchang@cugb.edu.cn

� The Author 2006. Published by Oxford University Press. All

rights reserved. For Permissions, please e-mail: journals.permissions@

oxfordjournals.org

Czamanske et al. (1998) concluded that a plume-headorigin could be rejected for the Siberian Traps on thebasis of a lack of any pre- or syn-volcanic upliftassociated with the flood basalt event. More recentwork in the West Siberian Basin, however, has yieldedevidence of regional kilometer-scale uplift deemedentirely consistent with a plume-head origin (Saunderset al., 2005). Little is known about the Panjal Traps, ofwhich only a comparatively tiny outcrop area remains(most of it in a militarized zone). The Emeishanprovince, in contrast, has been the focus of severalrecent stratigraphic, geochronological, geochemical, andgeophysical studies. Evidence for >1 km of doming of theregional lithosphere shortly before volcanism has beendocumented, and together with associated variations incrustal thickness, upper-mantle and lower-crustal seismiccharacteristics, and basalt composition, has been arguedto strongly support a plume-head origin (He et al., 2003;Xu et al., 2004).Near-primary picritic rocks can provide estimates of

mantle source temperature; indications of higher thannormal temperature are commonly interpreted as petro-logical evidence for a plume source. As in other floodbasalt provinces, picritic flows [here we use the MgO >12 wt % definition of Le Bas (2000)] are rare in the

Emeishan. Picritic rocks reported previously (Chung& Jahn, 1995; Xu et al., 2001; Song et al., 2001) arenot lavas but intrusions in (or that can be traced into)Triassic limestone, and thus postdate the flood basaltevent (Zhang & Wang, 2002b; Zhang et al., 2004).However, we recently discovered several picriticlava flows in the Lijiang area (Fig. 2; Zhang & Wang,2002b). Here, we present chemical and isotopic data forthese flows and associated basalts, and discuss theimplications for mantle source composition and condi-tions of melting.

GEOLOGICAL BACKGROUND

General descriptions of the Emeishan province havebeen presented by Wang et al. (1993), Chung & Jahn(1995), Xu et al. (2001), Thompson et al. (2001), and Aliet al. (2005). Most of the province lies within the broadregion of Cenozoic uplift caused by the collision ofGreater India and Eurasia. As a consequence, the lavapile is deeply dissected. The principal remaining volcanicoutcrops (Fig. 1) cover an area of about 250 000 km2,although outliers and buried remnants are present over agreater area. The lava pile is thickest in the west, wherethe thickest sections, located near Lijiang and Binchuan,

100°E 105°E

ChengduChengdu

GuiyangGuiyang

LAOSLAOSBURM

A

BURMA

KunmingKunming

CHINA

(Ertan)(Ertan)

30°N

25°N

LIJIANGLIJIANG

PanzhihuaPanzhihua

YongshengYongshengBinchuanBinchuan

Fig. 1. Map showing principal outcrops of the Emeishan flood basalts in black (modified from Chung & Jahn, 1995, fig. 1). Square near Lijiangmarks the location of Fig. 2.

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 OCTOBER 2006

1998

exceed 5000 m. Total erupted volume is estimatedconservatively at about 300 000–500 000 km3, makingthe Emeishan a medium-size flood basalt province (Yinet al., 1992; Jin & Shang, 2000; Ali et al., 2005). The lavasequence is underlain by late Carboniferous–Permiansedimentary beds atop a Mesoproterozoic to latePaleoproterozoic metamorphic basement.In contrast to the Siberian Traps, which formed at

a relatively high northern latitude, emplacement ofthe Emeishan flood basalts occurred near the Equator(e.g. Enkin et al., 1992). Overall, the province appears tobe slightly older than the much better dated, �251 MaSiberian Traps (e.g. Kamo et al., 2003): recent U–Pbdating of zircons in the Xinjie and Panzhihua layeredmafic–ultramafic intrusions yielded ages of 259 ± 3Maand 263 ± 3Ma, respectively (Zhou et al., 2002a,2005), whereas 40Ar–39Ar ages of 254 ± 5 Ma

(Boven et al., 2002) and 251–255 Ma (Lo et al., 2002)have been reported for lava flows and two late-stageintrusions.Three volcanic stages are evident. Locally early

magmatism, consisting principally of alkalic basalt, isrecorded in the eastern part of the province. The majoreruptive stage, present throughout the province, com-prises augite-phyric, plagioclase-phyric, and aphyrictholeiitic basalt, along with corresponding basalticpyroclastic deposits. The central part of the provincepreserves late-stage products of bimodal basalt–trachyteand basalt–rhyolite volcanism. Several mafic–ultramaficlayered intrusions, which host the world’s largest V–Tideposits, are also exposed in the central region.Picritic flows that we discovered recently in the Lijiang

area are located in an 800 m thick lava section nearShiman and a 5500 m thick section at Daju, one of

T

T

Pz

Q

Q

QQ

Pz

Pz

J ins

ha R

i ve r

J ins

ha R

i ve r

Lijiang

Yulong Snow Mts.

Haba Snow Mts.

J insha R

i ve r

J i nsha R

i ve rDaju

PzPz

Pz

Pz

Pz

Pz

Pz

Pz

Pz

Mz

Mz

Mz

Mz

Mz

Mz

Mz

Mz

Mz

MzMz

Mz

Mz

Mz

Mz

Mz

Mz

Mz

Mz

Mz

Mz

Mz

MzMz

Mz

Mz

Mz

Mz

Q

Mz

Mz

Mz

MzMz

Zb

Kz

Mz

Kz

Kz

Kz Mz

Zb

Zb

Mz

KzKz

KzMz

QuaternaryQuaternary

TertiaryTertiary

Triassic basaltTriassic basalt

MesozoicMesozoic

Permian basaltPermian basalt

PaleozoicPaleozoic

DisconformityDisconformity

Q

KzKz

T

MzMz PzPz

Late ProterozoicLate Proterozoic

FaultFault

ZbZb

Kz

Sampling traverseSampling traverse

RiverRiverGeologicalGeologicalboundaryboundary

Mz

100° 00’ 101° 00’

27°20’

26°40’

Shiman

Ninglang

Mz

Mz

Fig. 2. Geological map of the Lijiang area (after an unpublished map of the Yunnan Bureau of Geology and Mineral Resources, 1978).

ZHANG et al. WESTERN EMEISHAN PICRITIC AND BASALT FLOWS

1999

the thickest in the entire Emeishan province (see Fig. 2).In both sections, the picritic flows are intercalatedwith augite-phyric basalt flows (Fig. 3). At Shiman, thelowermost and uppermost picritic flows are 3–5 m thick,whereas a middle picritic flow is 15–20 m thick. In theDaju section, the lowermost picritic flow varies from20 to 50 m thick and the others are 3–5 m thick. The twosections differ considerably in total thickness and it isunclear how, or whether, the picritic flows at Shimancorrelate with those at Daju. Massive aphyric basalt and,at Daju, amygdaloidal basalt dominate the middle toupper parts of the lava sections. Plagioclase-phyricbasalt, abundant in other parts of the province, is absent.Most of the picritic flows are highly porphyritic

(with 7–15 vol. % phenocrysts). They contain abundant

phenocrysts of forsteritic olivine, plus minor diopsidicclinopyroxene. The olivine phenocrysts are generallysubhedral to rounded, and rarely embayed or partlyresorbed (Fig. 4). Most range from 0.2 to 1mm across,although the largest reach 4mm. Olivine is generallyreplaced by serpentine, but some grains retain coresof unaltered olivine, many of which contain scatteredmelt inclusions. Strained, kink-banded crystals areabsent. Some olivine crystals host equant, euhedralto rounded Cr-spinel a few tens of microns across. Cr-spinel is also present as solitary grains in the groundmass.The groundmass consists principally of very fine-grained, probably originally glassy, mesostasis plus lesseramounts of olivine, anhedral diopside, and tiny crystalsof plagioclase. The groundmass olivine tends to be lessaltered than the olivine phenocrysts; some is nearlyequant, but elongated skeletal forms (as large as 0.5mm· 0.1mm) consisting of parallel sets in optical continuityare most common.The phyric basalts contain 5–15 vol. % of augite

phenocrysts 1–6mm across; some form clusters. Thegroundmass is fine grained, with intersertal or inter-granular texture, and consists predominantly of plagio-clase and augite with minor iron oxides. The aphyricbasalts contain a similar assemblage of minerals. Augitemicrophenocrysts, several tens of microns across, arepresent, and commonly form clots, producing a cumulo-phyric texture. The groundmass is composed of plagio-clase laths with interstitial subophitic augite and minoranhedral magnetite.

ANALYTICAL METHODS

Mineral compositions in selected samples were measuredby electron microprobe at the Chinese Academy ofGeological Sciences and the University of Cardiff,following Bloomer et al. (1982). Bulk-rock major andtrace element compositions were determined on agate-and alumina-ground powders at the Chinese Academyof Geological Sciences and the University of Hawaii.Major elements were determined by X-ray fluorescencespectroscopy using the methods of Norrish & Chappell(1977), and trace elements by inductively coupledplasma–mass spectrometry following Dulski (1994) andNeal (2001). Isotope ratios of Nd, Pb, and Sr andassociated isotope-dilution concentrations were mea-sured at the University of Hawaii (see Sheth et al., 2003)on small, handpicked chips of acid-cleaned rock to avoidalteration and olivine phenocrysts (e.g. Peng &Mahoney, 1995). Because such chips may not be strictlyrepresentative of bulk-rock mineralogy, particularly incoarser-grained or patchily altered samples, we use theisotope-dilution data here only for age-correcting isotoperatios.

DJ-36DJ-35DJ-34DJ-33DJ-26

DJ-31DJ-34-1DJ-30DJ-25DJ-20DJ-16DJ-14DJ-11

DJ-7

SM-17SM-14SM-15

SM-11SM-8SM-5

DJ-3DJ-2DJ-1

DJ-32

DJ-37

DJ-38

Daju

Shiman

Amygdaloidal basalt

Aphyric basalt

Basaltic pyroclasticrock

Pyroxene-phyric basalt

Triassic limestone

0

500

1000 m

Picritic lava

Fig. 3. Simplified stratigraphic columns of the Daju and Shimansections with sample locations.

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 OCTOBER 2006

2000

RESULTS

Electron microprobe data

Analyses of olivine, clinopyroxene, and Cr-spinel inpicritic samples are listed in Table 1. Mg-number [100 ·Mg/(Mg þ Fe), molar] in the olivine varies from 84.5to 91.6. The more Mg-rich olivine crystals are visuallyindistinguishable from those with Mg-number <90.Most olivine crystals contain chromite and melt inclu-sions, and all have significant amounts of NiO, CoO,Cr2O3, and CaO.Olivine with Mg-number >90 is a common feature of

komatiites (e.g. Arndt et al., 1977; Lesher, 1989). It alsohas been reported in picritic rocks from several conti-nental flood basalt provinces (e.g. Larsen & Pedersen,2000; Thompson & Gibson, 2000) but is found onlyrarely in ocean ridge basalts (Donaldson & Brown, 1977)and ocean island basalts. In Hawaiian lavas, Mg-numberin olivine phenocrysts does not exceed 91.3 (Garcia et al.,1995), and in Reunion lavas, 90.6 (Fretzdorff & Haase,2002).In the sample for which Cr-spinel was analyzed

(SM-14), Mg-number varies from 26 to 35, Cr-number[100 · Cr/(Cr þ Al), molar] from 73.1 to 75.2, TiO2

from 1.85 to 1.96 wt %, and total iron as FeO (FeO*)from 30.36 to 32.92 wt %. These compositions are

distinct from those of Cr-spinel in abyssal peridotites(Dick & Bullen, 1984) and picritic continental floodbasalts (e.g. West Greenland; Larsen & Pedersen, 2000)but are broadly similar to Cr-spinel in some Archaeankomatiites (Arndt et al., 1977; Nisbet et al., 1977; Lesher,1989; Barnes & Roeder, 2001).Clinopyroxene phenocrysts in picritic sample SM-14

have high TiO2 (as high as 2.52 wt %), CaO, and MgOcontents, with Mg-number between 72.1 and 80.5. Theyare classified as Ti-rich diopside. Those in phyric basaltsamples DJ-33 and DJ-34 are augitic. In comparison,basalts from other areas of the Emeishan province con-tain augite (Wang et al., 1993).

Bulk-rock major and trace elementcomposition

The MgO content of the picritic lavas ranges from 26.99to 12.25 wt % and their Mg-number varies from 81 to66 (Table 2). The associated basalt flows have Mg-number between 66 and 55 at MgO contents of 11.92–6.83 wt % (except for sample DJ-1, with an Mg-numberof 34 at MgO 4.17 wt %). These values are higher thanreported for many basalts from other areas of theEmeishan province (Cong, 1988; Xu et al., 2001;Zhang & Wang, 2002a). A general decrease in Ni

Fig. 4. Photomicrograph of picritic sample SM-14 showing unaltered cores in partly resorbed olivine phenocrysts. The groundmass consists ofolivine, ophitic clinopyroxene, plagioclase, and oxide-rich mesostasis (plane-polarized light).

ZHANG et al. WESTERN EMEISHAN PICRITIC AND BASALT FLOWS

2001

Table 1: Analyses of olivine, clinopyroxene, and Cr-spinel

Mineral SiO2 TiO2 Al2O3 FeO* MnO MgO CaO Na2O Cr2O3 NiO CoO V2O5 Sum Mg-no. Cr-no.

SM-14

Ol 40.23 8.32 0.10 50.87 0.37 0.07 0.42 0.06 100.44 91.6

Ol 39.97 8.47 0.09 50.46 0.35 0.06 0.41 0.07 99.61 91.4

Ol 40.11 8.58 0.08 50.27 0.29 0.05 0.40 0.21 100.00 91.3

Ol 40.04 8.49 0.15 50.22 0.31 0.09 0.43 0.01 99.73 91.3

Ol 40.19 8.75 0.11 50.02 0.29 0.06 0.37 0.02 99.80 91.1

Ol 40.12 8.77 0.06 50.02 0.30 0.17 0.37 0.21 100.01 91.0

Ol 40.18 9.88 0.01 48.98 0.28 0.01 0.36 0.13 99.83 89.8

Ol 39.70 10.07 0.28 48.77 0.42 0.08 0.28 0.21 99.81 89.6

Ol 39.94 10.49 0.21 48.54 0.22 0.10 0.27 0.20 99.98 89.2

Ol 39.88 10.79 0.16 48.23 0.30 0.10 0.33 99.78 88.8

Ol 39.21 12.21 0.24 47.40 0.33 0.09 0.40 99.88 87.4

Ol 39.02 12.70 0.24 46.95 0.39 0.03 0.36 0.05 99.75 86.8

DJ-31

Ol 37.78 0.03 12.45 0.13 48.91 0.25 0.06 99.61 87.5

Ol 38.12 0.05 12.86 0.15 47.58 0.32 0.05 99.13 86.8

Ol 37.46 0.01 0.04 13.82 0.15 47.08 0.38 0.02 0.06 99.02 85.8

DJ-26

Ol 37.89 0.03 12.09 0.14 49 0.27 0.06 99.48 87.8

Ol 37.97 0.01 0.04 13.3 0.14 47.39 0.29 0.03 0.06 99.23 86.3

Ol 37.86 0.02 0.03 14.51 0.15 46.03 0.45 0.02 0.04 99.11 85.0

DJ-34-1

Ol 38.45 0.01 0.04 10.89 0.11 50.12 0.23 0.05 0.08 99.98 89.1

Ol 37.71 0.02 0.08 14.22 0.14 46.25 0.41 0.06 98.9 85.2

Ol 37.43 0.01 0.04 8.99 0.09 52.46 0.22 0.14 99.38 91.2

Ol 37.92 0.05 10.34 0.12 50.79 0.3 0.11 99.63 89.7

SM-14

Cpx 50.03 1.20 3.41 6.93 0.18 15.898 21.50 0.23 0.24 0.10 0.18 0.226 100.12 80.3

Cpx 50.16 1.05 3.32 6.96 0.20 16.10 21.58 0.04 0.16 0.08 0.20 0.22 100.06 80.5

Cpx 50.01 1.08 3.54 7.17 0.23 15.87 21.42 0.07 0.26 0.04 0.06 0.17 99.90 79.8

Cpx 48.06 1.68 4.38 9.12 0.28 14.70 20.66 0.44 0.41 0.02 0.09 0.10 99.93 74.2

Cpx 46.98 2.52 5.34 9.92 0.14 14.41 20.02 0.23 0.24 0.06 0.11 0.26 100.23 72.1

DJ-33

Cpx 49.68 2.00 4.43 7.74 0.19 11.66 22.34 0.18 0.20 98.47 72.9

Cpx 50.46 1.41 4.60 6.35 0.07 13.26 22.51 0.27 0.37 99.34 78.8

Cpx 52.09 0.81 2.72 4.53 0.00 14.13 23.64 0.18 0.28 98.38 84.8

Cpx 47.68 2.85 7.16 7.45 0.09 10.80 22.49 0.12 0.20 98.84 72.1

Cpx 50.32 1.17 4.10 5.77 0.11 13.74 23.19 0.06 0.38 98.84 80.9

DJ-34

Cpx 52.87 1.17 4.89 4.35 0.00 14.43 19.93 0.27 0.20 98.14 85.5

Cpx 53.07 1.03 4.12 4.10 0.00 15.42 20.42 0.16 0.38 98.70 87.0

Cpx 52.64 1.31 5.36 4.27 0.00 14.57 20.07 0.31 0.33 98.86 85.9

Cpx 52.08 1.56 6.25 5.04 0.04 13.31 19.57 0.27 0.29 98.41 82.5

SM-14

Sp 0.07 1.94 10.26 30.36 0.54 8.95 46.39 0.29 0.30 0.25 99.36 34.5 75.2

Sp 0.08 1.96 10.41 31.39 0.45 8.42 46.82 0.18 0.10 0.18 99.98 32.3 75.1

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 OCTOBER 2006

2002

content with decreasing Mg-number implies removal ofsignificant amounts of olivine between picritic andbasaltic compositions.Trends of increasing SiO2, Al2O3, TiO2, and Na2O

with decreasing Mg-number (Fig. 5) are also consistentwith olivine fractionation, whereas the steady increase inAl2O3 and lack of pronounced FeO* enrichmentindicate little or no removal of plagioclase, consistentwith the absence of plagioclase phenocrysts in theselavas. Clinopyroxene fractionation appears to havecommenced around Mg-number of 62–64 (�10 wt %MgO), where CaO values reach a maximum. With oneexception, samples with Mg-number >64 have ratherhigh CaO/Al2O3 (0.95–1.17), suggesting a relativelyhigh pressure of melting (Hirose & Kushiro, 1993; Baker& Stolper, 1994). On a total alkalis vs SiO2 diagram (notshown), values for many of the samples suggest asomewhat alkalic character.However, some care must be taken in interpreting

both the major and trace element data. The samples arevariably altered in thin section, and this is reflected in,for example, their LOI (weight loss on ignition to1000�C) values, which range from 1.49 to 7.51 wt %.Consistent with the general replacement of olivinephenocrysts by secondary phases, the most olivine-richsamples have the highest LOI values (e.g. note thecorrelation of LOI with Mg-number; Fig. 5). Althoughpost-eruptive alteration clearly has not destroyed theoverall relationships among most of the major elements,it is probably the reason for the lack of any correlationbetween K2O and Mg-number (or the other majorelement oxides) and, together with excess phenocrysts insome samples, probably has broadened several of theother arrays in Fig. 5.Among the trace elements, Rb and Ba display

considerable variability relative to alteration-resistantelements such as Th and Nb (Fig. 6a and b); as withK2O, this largely appears to be an alteration effect.Large troughs or, in a few cases, peaks at Sr also arepresent in many of the primitive-mantle-normalized

element patterns in Fig. 6. Strontium is compatible inplagioclase yet, as noted, plagioclase phenocrysts areabsent. Europium is also compatible in plagioclase, butthe lavas lack significant Eu anomalies. Unlike Eu, Srdoes not correlate with alteration-resistant elements (orwith isotope ratios). Thus, Sr concentrations appear tohave been affected considerably by alteration, as hasbeen documented in altered subaerial basalts elsewhere(e.g. Lindstrom & Haskin, 1981; Clague & Frey, 1982;Fleming et al., 1992). Peaks at Pb are present in most ofthe patterns. The peaks are of highly variable size, andthe extent to which they are primary features is unclear.Only weak correlations are seen between Pb and thealteration-resistant incompatible elements, suggestingthat alteration has modified Pb concentrations signifi-cantly. Some of the patterns in Fig. 6 show a small Ptrough and/or variability in U relative to Nb and Th,which may likewise be an alteration effect.Patterns of alteration-resistant incompatible elements

for the picritic and basaltic lavas are similar in overallshape (Fig. 6c and d), but element concentrations arehigher in the basaltic samples, consistent with theirmore evolved nature. All of the lavas are enriched inthe more highly incompatible elements relative tomoderately incompatible ones; for example, primitive-mantle-normalized (La/Yb)P varies from 5.2 to 18.7. Incomparison, values for most basalts from other parts ofthe Emeishan province range between 2.9 and 11.4 (Xuet al., 2001; Zhang & Wang, 2002a; Xiao et al., 2004).Unlike some other areas of the province, sizeable troughsat Nb and Ta relative to La and Th are not observedamong our samples; rather, many have Nb–Ta peaks.Overall, patterns of the alteration-resistant elements arevery similar to those of many oceanic island basalts.Xu et al. (2001) used a Ti/Y ratio of 500 to subdivide

their Emeishan samples into low- and high-Ti (or -Ti/Y)types. By this definition, all of our lavas, which haveTi/Y between 502 and 767, are of the high-Ti/Y type.However, the TiO2 contents of the samples (1.14–3.16 wt %, except for two with higher values) are lower

Table 1: Continued

Mineral SiO2 TiO2 Al2O3 FeO* MnO MgO CaO Na2O Cr2O3 NiO CoO V2O5 Sum Mg-no. Cr-no.

Sp 0.03 1.95 11.27 32.82 0.64 6.46 45.59 0.09 0.26 0.26 99.37 26.0 73.1

Sp 0.11 1.85 10.95 30.08 0.26 10.37 0.02 43.71 0.20 100.01 38.0 73.3

BCR-2(meas) 54.53 2.26 13.48 12.43 0.02 3.60 7.13 3.15 0.00

BCR-2 (rec) 54.86 2.29 13.69 12.59 0.02 3.64 7.22 3.20 0.00

Ol, olivine; Cpx, clinopyroxene; Sp, spinel. Oxide values are in wt %. Relative analytical uncertainty (2s) is <1% for themajor elements and <16% for the minor elements. Analyses were made with Cameca CAMEBAX electron microprobes andused Smithsonian mineral standards and oxides for calibration. Average measured (meas) values for standard BCR-2 glassanalyzed as an unknown at the Institute of Mineral Resources, Chinese Academy of Geological Sciences (n ¼ 3) andUniversity of Cardiff (n ¼ 5) are compared with recommended (rec) values (Wilson, 1997).

ZHANG et al. WESTERN EMEISHAN PICRITIC AND BASALT FLOWS

2003

Table 2: Major and trace element data for bulk rocks

Sample: DJ-1 DJ-2 DJ-3 DJ-7 DJ-11 DJ-14 DJ-16 DJ-20 DJ-25 DJ-34-1 DJ-31 DJ-32 DJ-26 DJ-33

pic pic pic pic pic pic

SiO2 48.62 44.67 48.74 49.53 50.06 49.63 49.39 48.79 49.08 43.33 45.80 48.61 42.19 46.15

TiO2 4.23 2.36 2.12 2.43 2.85 1.99 2.45 3.16 2.21 1.44 1.14 2.49 1.82 2.26

Al2O3 12.94 11.20 9.15 13.90 14.88 11.34 12.93 13.12 10.40 6.39 5.54 12.90 8.68 12.59

FeO* 16.93 13.42 11.63 11.93 11.93 11.42 12.54 13.07 11.69 12.91 13.70 12.25 17.57 12.95

MnO 0.21 0.19 0.15 0.18 0.16 0.17 0.17 0.20 0.16 0.19 0.14 0.16 0.22 0.18

MgO 4.17 14.85 15.58 7.30 6.83 10.16 7.19 8.61 12.25 26.99 26.68 10.16 19.04 10.74

CaO 7.00 11.43 10.56 9.89 8.25 11.68 9.17 8.73 12.03 7.47 6.16 9.00 8.92 10.64

Na2O 3.32 1.90 2.01 3.53 4.05 2.19 1.94 3.49 1.50 0.85 0.41 3.35 0.81 1.61

K2O 2.41 0.08 0.14 0.92 1.16 1.07 4.08 1.06 0.82 0.01 0.01 0.96 0.04 2.39

P2O5 0.57 0.34 0.21 0.33 0.39 0.19 0.30 0.35 0.28 0.15 0.10 0.39 0.20 0.42

Total 100.40 100.45 100.29 99.94 100.57 99.84 100.17 100.57 100.42 99.75 99.67 100.27 99.51 99.93

LOI 1.49 5.23 4.58 2.94 3.33 2.09 2.56 2.88 4.74 7.51 6.62 3.73 5.17 2.89

Mg-no. 34.0 69.9 73.7 56.2 54.5 65.1 54.6 58.0 68.7 81.4 80.3 63.5 69.4 63.5

Rb 45.8 5.0 3.6 31.5 24.3 40.2 63.2 20.5 8.4 5.5 0.3 32.8 1.8 64.3

Ba 523 34.4 32.5 225 541 1370 908 412 230 64.9 35.0 176 51.1 430

Th 3.19 2.95 1.37 3.00 3.25 3.35 3.49 1.43 3.35 1.26 0.81 1.52 1.83 2.18

U 0.50 0.69 0.37 0.79 0.64 0.76 0.93 0.15 0.58 0.28 0.24 0.45 0.41 0.48

Nb 45.3 31.5 16.7 31.5 35.4 20.0 34.1 30.5 28.6 11.8 8.0 28.4 18.0 27.1

Ta 2.80 1.86 1.06 1.61 2.04 1.29 2.00 1.95 1.69 0.79 0.54 1.71 1.19 1.65

La 42.8 28.4 19.4 39.7 36.7 20.4 38.6 24.2 33.8 10.3 6.64 23.3 13.9 15.6

Ce 88.5 60.5 44.8 77.5 77.9 47.9 76.8 58.8 66.9 25.8 17.3 51.2 32.9 39.5

Pr 11.4 7.60 5.91 9.42 9.79 6.37 9.28 8.21 7.97 3.67 2.50 6.55 4.46 5.45

Pb 2.45 2.44 1.95 3.41 3.14 2.52 3.77 2.64 3.43 2.21 2.08 2.73 3.17 5.99

Sr 195 95.5 22.6 339 501 2648 968 498 67.0 42.8 27.1 268 47.9 156

Nd 46.9 30.2 25.6 33.6 37.3 26.9 36.3 35.7 30.0 16.2 11.2 27.0 19.2 24.1

Hf 3.99 3.34 3.80 4.20 3.68 4.01 2.44 1.78 3.81 2.97 3.40

Zr 155 119 148 170 182 98.6 70.0 157 117 138

Sm 9.55 5.95 5.23 6.37 7.98 5.35 6.74 7.81 5.84 3.37 2.54 6.19 3.98 5.05

Eu 2.81 1.76 1.43 1.63 2.21 1.63 2.05 2.21 1.55 1.15 0.83 1.67 1.25 1.68

Gd 8.22 4.88 4.82 5.78 7.41 4.99 6.29 7.28 5.26 3.17 2.45 5.13 3.69 4.80

Tb 1.41 0.84 0.77 0.88 1.13 0.74 0.85 1.14 0.77 0.48 0.38 0.87 0.56 0.72

Dy 6.90 4.11 3.80 4.61 5.37 4.27 4.19 5.44 3.89 2.58 2.16 4.65 3.09 4.09

Y 35.1 23.0 17.3 29.1 29.3 21.8 25.3 27.7 22.6 13.2 10.9 26.9 16.3 21.5

Ho 1.25 0.79 0.63 0.85 0.91 0.80 0.80 0.91 0.73 0.48 0.41 0.87 0.59 0.78

Er 2.72 1.96 1.47 2.10 2.14 2.07 1.89 2.11 1.81 1.27 1.06 2.09 1.56 2.10

Tm 0.34 0.28 0.20 0.29 0.27 0.30 0.26 0.28 0.26 0.18 0.15 0.29 0.23 0.30

Yb 1.83 1.71 1.16 1.80 1.68 1.81 1.57 1.68 1.58 1.05 0.91 1.77 1.40 1.81

Lu 0.22 0.25 0.16 0.26 0.23 0.28 0.22 0.23 0.24 0.16 0.14 0.25 0.22 0.27

Sc 18.6 45.6 20.9 18.1 31.2 32.5

V 321 173 112 228 288

Cr 45.7 750 2262 2509 2122 688

Co 31.0 49.2 88.9 88.8 98.1 55.1

Ni 45.5 174 1147 1272 967 259

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 OCTOBER 2006

2004

Table 2: Continued

Sample: DJ-34 DJ-35 DJ-36 DJ-37 DJ-38 SM-15 SM-14 SM-5 SM-8 SM-17 SM-11 BHVO-1

pic pic pic pic meas. rec.

SiO2 45.46 48.44 49.05 52.00 48.89 46.15 45.64 48.46 47.60 47.86 45.42 50.09 50.57

TiO2 2.16 2.10 3.80 2.71 2.63 1.57 1.73 2.43 2.86 1.99 2.56 2.81 2.74

Al2O3 11.94 9.97 14.04 12.63 10.67 8.15 8.12 13.01 11.34 10.59 12.82 13.82 13.97

FeO* 12.64 14.02 11.31 11.81 13.40 12.70 12.94 11.81 14.84 12.80 13.07 11.29 11.14

MnO 0.23 0.18 0.12 0.22 0.19 0.18 0.20 0.22 0.24 0.17 0.22 0.16 0.17

MgO 11.92 13.02 6.99 7.44 11.20 23.01 22.44 9.85 10.07 15.80 10.21 7.26 7.32

CaO 11.84 10.85 7.08 9.51 10.21 7.75 7.85 10.89 9.77 8.04 12.90 11.53 11.54

Na2O 2.00 1.13 2.85 3.48 2.80 0.52 0.85 2.15 3.02 1.67 1.86 2.23 2.29

K2O 0.76 0.50 3.64 0.26 0.19 0.22 0.20 0.21 0.17 1.22 0.20 0.53 0.53

P2O5 0.39 0.24 0.50 0.28 0.25 0.16 0.18 0.37 0.28 0.25 0.38 0.28 0.27

Total 99.34 100.44 99.39 100.32 100.41 100.41 100.14 99.39 100.18 100.40 99.63 100.01 100.55

LOI 2.91 4.07 2.43 2.96 3.69 5.47 4.62 2.81 1.93 4.22 2.96

Mg-no. 66.4 66.0 56.4 56.9 63.7 79.1 78.4 63.6 58.7 72.1 62.1

Rb 15.7 12.6 103.8 5.7 3.5 7.9 7.3 3.4 2.5 33.1 2.8 9.6 11

Ba 601 79.6 297 111 93.2 73.2 87.1 204 148 442 205 133 139

Th 2.29 1.65 4.16 1.59 1.66 1.94 1.97 2.64 3.24 4.11 2.62 1.21 1.08

U 0.52 0.36 0.76 0.31 0.33 0.44 0.65 0.72 0.59 0.80 0.47 0.41 0.42

Nb 27.8 22.8 44.6 26.5 23.0 14.0 15.1 30.3 27.8 30.3 30.4 19.4 19

Ta 1.72 1.36 2.88 1.60 1.46 0.90 0.98 1.88 1.76 1.83 1.88 1.18 1.23

La 16.7 20.0 34.1 28.6 29.6 11.7 13.1 20.6 24.1 32.2 20.3 15.5 15.8

Ce 39.9 47.0 87.0 62.2 60.8 27.3 32.2 49.6 60.8 65.3 49.5 37.7 39

Pr 5.49 5.92 12.3 8.00 7.64 3.61 4.47 6.79 8.41 7.79 6.82 5.5 5.7

Pb 3.83 2.39 4.39 2.45 2.63 1.97 3.05 2.86 4.21 3.22 2.98 2.14 2.1

Sr 292 675 149 411 130 123 134 427 296 539 380 388 390

Nd 23.9 24.4 52.5 32.3 31.0 15.4 19.2 19.2 35.4 29.5 29.2 24.5 25.2

Hf 3.53 6.98 2.69 2.98 4.30 4.96 4.37 4.41 4.33 4.38

Zr 140 281 111 121 173 196 199 176 180 179

Sm 5.17 5.24 10.57 6.68 6.35 3.37 4.05 6.03 7.04 5.91 6.16 6.2 6.2

Eu 1.67 1.31 3.25 1.67 1.54 0.95 1.33 1.99 2.19 1.68 2.00 2.05 2.06

Gd 4.99 4.17 9.13 5.57 4.73 2.96 3.25 5.67 6.14 5.02 5.71 6.27 6.4

Tb 0.74 0.67 1.33 0.85 0.80 0.52 0.58 0.83 0.91 0.67 0.85 0.95 0.96

Dy 4.31 3.37 7.27 4.14 3.81 2.82 3.30 4.76 4.91 3.68 4.78 5.38 5.25

Y 21.8 18.0 38.8 21.1 20.5 15.4 17.0 25.8 25.4 21.4 25.9 25.6 27.6

Ho 0.81 0.58 1.34 0.66 0.64 0.53 0.62 0.91 0.92 0.66 0.92 1.0 0.99

Er 2.11 1.32 3.49 1.51 1.51 1.30 1.59 2.36 2.32 1.63 2.39 2.58 2.4

Tm 0.30 0.18 0.49 0.19 0.20 0.19 0.23 0.34 0.33 0.23 0.34 0.33 0.33

Yb 1.86 1.09 2.86 1.10 1.15 1.28 1.34 2.05 1.90 1.41 2.04 2.01 2.2

Lu 0.28 0.15 0.44 0.14 0.16 0.18 0.21 0.31 0.28 0.21 0.32 0.27 0.29

Sc 33.4 24.3 25.6 32.2 38.6 39.0 32.0 31.8

V 303 235 222 307 332 326 319 317

Cr 660 83.1 2259 529 893 854 285 289

Co 54.4 34.8 81.7 46.2 56.6 53.4 46.9 45

Ni 241 76.3 945 195 254 214 117 121

Picritic lavas are indicated by ‘pic’. Major and minor element oxide values are in wt %; trace element data are in ppm. FeO*is total iron as FeO. Mg-number ¼ [molar Mg/(Mg þ Fe2þ)] · 100, assuming 15% of total iron oxide is ferric. LOI, weightloss on ignition to 1000�C. Relative uncertainties (2s) on major and minor element oxides are �1%; for SiO2 �0.5%. Fortrace elements, the relative uncertainty is 3–7%. An indication of accuracy is given by measured (meas.; average of five) andrecommended (rec.; Govindaraju, 1994) values for standard BHVO-1. Major elements were measured on Siemens 303ASand 3080E spectrometers, trace elements on VG PQ-2 Turbo and PQ-2S instruments.

ZHANG et al. WESTERN EMEISHAN PICRITIC AND BASALT FLOWS

2005

than those of the high-Ti/Y basalts described by Xu et al.(2001) and Xiao et al. (2004), which have TiO2

>3.5 wt %. In part, this difference reflects the generallyless evolved nature of the rocks of the Lijiang area.High-Ti and low-Ti suites are fairly common in other

continental flood basalt provinces (e.g. Carlson, 1991;

Sharma, 1997; Arndt et al., 1998; Pik et al., 1998; Kiefferet al., 2004; Melluso et al., 2006). As with our samples,many of the high-Ti suites have broadly ocean-island-like incompatible-element characteristics. Other distinc-tive features of such high-Ti rocks tend to be a variablyalkalic character and higher FeO* contents than those of

Al2O3

6

8

10

12

14

CaO

6

8

10

12

FeO*

10

12

14

16

18

SiO2

42

44

46

48

50

52 TiO2

1

2

3

4

5

K2O

0

1

2

3

4

30 40 50 60 70 80 90

Na2O

0

1

2

3

4

LOI

2

4

6

40 50 60 70 80 90

Mg #

DajuShiman

Fig. 5. Mg-number vs major element oxides and LOI (in wt %) for lavas of the Daju and Shiman sections. Mg-number ¼ 100 · molar Mg/(MgþFe2þ), assuming 15% of total iron oxide is ferric. The anomalously high FeO* content of sample DJ-26 appears to reflect an abundant secondaryopaque phase (magnetite?) observed in thin section in this sample.

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 OCTOBER 2006

2006

ocean-ridge basalts at similar MgO. High-Ti, high-MgOrocks in the Siberian Traps (picritic basalt andmeimechite) and Ethiopian Traps (picritic basalt) exhibitsome similarities to the Lijiang area picritic lavas, butgenerally have higher TiO2 and FeO* for their MgOcontents (e.g. Arndt et al., 1998; Pik et al., 1998).

Nd, Sr, and Pb isotope ratios

The lavas define a relatively small range of age-correctedeNd(t) (t ¼ 250 Ma) from þ4.0 to –1.3 (Table 3; Fig. 7a).There is no difference between picritic and basalticsamples, as the range for both is the same within error(Fig. 8a). Values of (87Sr/86Sr)t vary from 0.70344 to0.70524. Given that both Rb and Sr concentrations havebeen affected by alteration, some of the age-corrected Srisotope ratios are no doubt too high or too low. InFig. 7a, the Lijiang data lie within or slightly to thelow-87Sr/86Sr side of the larger array defined by otherEmeishan basalts. Basalts from other parts of theprovince exhibit a total eNd(t) and (87Sr/86Sr)t range ofþ4.8 to –4.8 and 0.70393–0.70759, respectively. Thisrange is notably smaller than those for many othercontinental flood basalts, including the Karoo, DeccanTraps, Columbia River, Parana, and Siberian Traps,although it is larger than found for the intra-oceanicCaribbean or Ontong Java plateaux (see Hawkesworth

et al., 1984; Carlson & Hart, 1988; Mahoney, 1988; Kerret al., 1997; Peate, 1997; Sharma, 1997; Tejada et al.,2004, and references therein).The range of age-corrected Pb isotope ratios is also

relatively small compared with those of other continentalflood basalts: (206Pb/204Pb)t varies from 17.93 to 18.88,(207Pb/204Pb)t from 15.51 to 15.59, and (208Pb/204Pb)tfrom 37.93 to 38.86. In Fig. 7c and d, the data lie slightlyabove the estimated 250 Ma field for modern EastPacific Rise mantle, and are broadly similar in thisrespect to Indian Ocean hotspot and ridge basalts. Pre-viously published Pb isotope data for the Emeishanprovince are for seven samples from sections nearBinchuan and Yongsheng (Zhang & Wang, 2003) tothe south and east of the Lijiang area. With oneexception, these data overlap or lie close to our datapoints in Fig. 7.The high-eNd(t), high-(206Pb/204Pb)t end of the

Emeishan data array lies close to (Fig. 7a and b) oroverlaps (Fig. 7c and d) the estimated 250 Ma field forthe sources of several modern Indian Ocean hotspots(Re–Cr–Am field). At the highest eNd(t) values, itapproaches the field for the sources of some Pacific‘C-type’ (Hanan & Graham, 1996) hotspots, such asthose feeding the Easter Seamount Chain and NazcaRidge. Interestingly, the majority of data for the othermajor late Permian flood basalt province, the Siberian

1

10

100

1

10

100

1

10

100

1

10

100

Rb Ba Th U K Nb Ta La Ce Pr Pb Sr Nd P Zr Hf Sm Ti Eu Gd Tb Dy Y Ho Er Tm Yb Lu Th Nb Ta La Ce Pr Nd Zr Hf Sm Ti Eu Gd Tb Dy Y Ho Er Tm Yb Lu

Th Nb Ta La Ce Pr Nd Zr Hf Sm Ti Eu Gd Tb Dy Y Ho Er Tm Yb LuRb Ba Th U K Nb Ta La Ce Pr Pb Sr Nd P Zr Hf Sm Ti Eu Gd Tb Dy Y Ho Er Tm Yb Lu

Sam

ple

/ Est

. prim

itive

man

tle

basaltic lavas

picritic lavas

(a) (c)

(b) (d)

avg. OIBavg. OIB

Fig. 6. Incompatible-element patterns of Lijiang area picritic (a, c) and basaltic (b, d) lavas. (c) and (d) show alteration-resistant elements only.Gray shading in (b) and (d) indicates the range of variation among the picritic lavas. Estimated (Est.) primitive-mantle normalizing values andaverage ocean-island basalt (OIB) pattern are from Sun & McDonough (1989).

ZHANG et al. WESTERN EMEISHAN PICRITIC AND BASALT FLOWS

2007

Table3:Nd,Sr,andPbisotopicdata

Sam

ple:

SM-17

SM-15

DJ-36

DJ-35

DJ-34

DJ-26

DJ-32

DJ-31

DJ-34-1

DJ-25

DJ-20

DJ-16

DJ-14

DJ-11

DJ-3

DJ-2

(143Nd/1

44Nd) 0

0.512457

0.512682

0.512678

0.512607

0.512698

0.512718

0.512744

0.512773

0.512734

0.512512

0.512681

0.512469

0.512494

0.512532

0.512606

0.512517

e Nd(0)

�3.6

þ0.8

þ0.7

�0.6

þ1.1

þ1.6

þ2.0

þ2.6

þ1.9

�2.5

þ0.8

�3.3

�2.8

�2.3

�0.7

�2.4

(87Sr/

86Sr)0

0.70586

0.70465

0.71034

0.70515

0.70495

0.70540

0.70550

0.70515

0.70522

0.70595

0.70513

0.70597

0.70534

0.70549

0.70513

0.70537

(206Pb/204Pb) 0

19. 160

19. 093

18. 553

18. 397

19. 158

19. 379

18. 383

18. 703

18. 948

18. 792

18. 577

18. 823

19. 011

(207Pb/204Pb) 0

15. 586

15. 624

15. 557

15. 520

15. 603

15. 619

15. 539

15. 570

15. 619

15. 605

15. 607

15. 590

15. 620

(208Pb/204Pb) 0

40. 166

39. 392

39. 064

38. 707

39. 256

39. 401

38. 773

39. 233

39. 378

39. 292

39. 102

39. 156

39. 559

Nd,ppm

30. 70

18. 37

48. 32

26. 85

21. 11

16. 12

23. 68

9.560

13. 83

22. 34

30. 18

27. 22

19. 35

29. 11

14. 89

28. 53

Sm

5.918

4.178

9.726

5.573

4.800

3.860

5.509

2.400

3.240

4.897

6.723

5.697

4.703

5.929

3.392

5.381

Sr

555.6

168.6

181.0

655.7

290.1

51. 06

241.2

28. 75

42. 83

59. 20

423.7

477.9

2109

397.8

32. 60

89. 90

Rb

33. 3

7.33

109

9.69

14. 2

1.90

29. 3

4.80

5.49

8.26

22. 8

45. 1

21. 5

13. 3

0.997

1.84

Pb

3.241

1.900

3.832

6.626

3.239

2.334

3.237

2.123

3.220

2.517

4.692

1.374

1.904

U1.22*

0.440

0.423*

0.360

0.461*

0.449

0.579

0.151

0.927

0.671

0.635

0.372

0.714*

Th

5.99*

1.94

2.73*

1.65

1.70*

1.52

3.35

1.43

3.49

3.44

3.25

1.37

2.56*

(143Nd/1

44Nd) t

0.512266

0.512457

0.512479

0.512402

0.512473

0.512481

0.512514

0.512524

0.512502

0.512295

0.512461

0.512262

0.512254

0.512331

0.512381

0.512330

e Nd(t)

�1.0

þ2.7

þ3.1

þ1.6

þ3.0

þ3.2

þ3.8

þ4.0

þ3.6

�0.4

þ2.8

�1.1

�1.3

þ0.2

þ1.2

þ0.2

(87Sr/

86Sr)t

0.70524

0.70420

0.70416

0.70500

0.70445

0.70503

0.70425

0.70344

0.70390

0.70451

0.70458

0.70500

0.70523

0.70514

0.70482

0.70515

(206Pb/204Pb) t

18. 187

18. 499

18. 273

18. 260

18. 793

18. 883

17. 933

18. 522

18. 211

18. 112

18. 233

18. 133

18. 048

(207Pb/204Pb) t

15. 536

15. 594

15. 543

15. 513

15. 584

15. 594

15. 516

15. 561

15. 581

15. 570

15. 589

15. 555

15. 571

(208Pb/204Pb) t

38. 61

38. 54

38. 48

38. 50

38. 82

38. 85

37. 93

38. 67

38. 47

38. 16

38. 53

38. 33

38. 43

Measured

isotope

ratios(subscript0)

are

age-co

rrected

(t)to

250

Ma.

Fractionation

correctionsare

0.1194

for

86Sr/

86Sran

d0.242436

for

148NdO/1

44NdO.

Measuremen

tsweremad

eonaVG

Sectormulticollectorinstrumen

tin

dyn

amic

modeforNdan

dSran

dstatic

modeforPbisotopes.Ratiosarereported

relative

tomeasured

87Sr/

86Sr¼

0.710240

±0.000015

(2s,

n¼

30)forNBS987Sran

d143Nd/1

44Nd¼

0.511850

±0.000008

(0. 511843measured;2s

,n¼

25)forLaJolla

Nd.Pb

isotopes

weremeasuredwithadouble-spikemethod(G

aler,1999);meanvalues

obtained

for5–10

ngload

sofNBS981Pbin

thelast

3yearswere16. 938,15. 494,an

d36. 712

for206Pb/2

04Pb,207Pb/2

04Pb,an

d208Pb/2

04Pb(±

230ppm;2s

,n¼

33).Within-runerrors

ontheisotopic

dataab

ove

areless

than

oreq

ual

totheexternal

uncertainties

onstan

dards.

ConcentrationsofSm,Nd,Rb,Sr,Pban

d,wheremarkedwithan

asterisk,Uan

dThweredetermined

byisotopedilution;other

Uan

dTh

dataarefrom

Tab

le2.

Uncertainties

onisotope-dilutionco

ncentrationsareless

than

0.2%

forSm

andNd,0.4%

forSr,1%

forRb,0.5%

forPb,1%

forU,an

d2%

for

Th.Totalp

roceduralb

lanks

are<40

pgforPb,<35

pgforSr,an

d<12

pgforNd.Itshould

benotedthat

e Nd¼

0today

correspondsto

143Nd/1

44Nd¼

0.51264;

values

forolder

times

assumepresentBulk

Earth

147Sm/1

44Nd¼

0.1967.

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 OCTOBER 2006

2008

Traps, define a broadly similar array to that of theEmeishan at the high-eNd(t) end [though with fewrepresentatives at eNd(t) > þ2; Fig. 7a and b], and alsofall slightly above the estimated 250 Ma East Pacific Risesource field in Fig. 7c and d. However, the SiberianTraps data extend to much lower (206Pb/204Pb)t andeNd(t) and higher (87Sr/86Sr)t than yet found for theEmeishan province.Rough correlations are present between eNd(t) (and to

a lesser extent, Sr and Pb isotope ratios) and ratiosof alteration-resistant incompatible elements in whichat least one of the elements is highly incompatible (e.g.Fig. 8c–f). Lower values of eNd(t) are associated with

higher La/Yb, La/Sm, Th/Nd, etc. and with lower Nb/La; in other words, with greater relative enrichment inhighly incompatible elements and with decreasing Nb–Ta peak size in the patterns of Fig. 6. Most of ouranalyses are for the thick Daju section, and in this sectionthere is also a crude correlation with stratigraphicposition, in that the stratigraphically higher lavas tendto have higher eNd(t) (Fig. 8b), Nb/La, etc. In contrast,ratios of moderately incompatible elements, such asSm/Yb and Ti/Y, do not correlate with stratigraphicposition or with isotope ratios; the highest values ofSm/Yb, Ti/Y, etc. tend to be found in samples withintermediate eNd(t).

-12

-8

-4

0

+4

+8

+12

-8

-4

0

+4

+8

+12

0.702 0.703 0.704 0.705 0.706 0.707 0.708

15.30

15.35

15.40

15.45

15.50

15.55

15.60

15.65

15.70

15.75

17.0 17.5 18.0 18.5 19.0 19.5

17.0 17.5 18.0 18.5 19.0 19.5

36.5

37.0

37.5

38.0

38.5

39.0

39.5

40.0

17.0 17.5 18.0 18.5 19.0 19.5

(206Pb/204Pb)t (206Pb/204Pb)t

(206Pb/204Pb)t(87Sr/86Sr)t

(208 P

b/20

4 Pb)

t

(207 P

b/20

4 Pb)

tε N

d(t)

ε Nd(

t)

East Pacific RiseEast Pacific Rise

East Pacific RiseEast Pacific Rise

Nazca-EasterChain

Nazca-Easter Chain

Nazca-EasterChain

Nazca-EasterChain

Lijiang area

Other Emeishan

Siberian Traps

Re-Cr-Am

Re-Cr-Am

Re-Cr-Am Re-Cr-Am

(a) (b)

(c) (d)

Fig. 7. eNd(t) vs (87Sr/86Sr)t (a) and (206Pb/204Pb)t (b); (206Pb/204Pb)t vs (

207Pb/204Pb)t (c) and (208Pb/204Pb)t (d) for the Lijiang area lavas. OtherEmeishan data are from Xu et al. (2001), Zhang & Wang (2003), and Xiao et al. (2004). Siberian Traps data are from Sharma et al. (1992), Lightfootet al. (1993), and Wooden et al. (1993). Fields for several Indian Ocean hotspots (Re-Cr-Am: Reunion, Mauritius shield, Crozet, Amsterdam), theNazca Ridge–Easter Seamount Chain (east of Salas y Gomez) in the Pacific, and the East Pacific Rise are adjusted to estimated 250 Ma positionsassuming hotspot and rise mantle-source values, respectively, of 147Sm/144Nd ¼ 0.18 and 0.24, 87Rb/86Sr ¼ 0.10 and 0.02, 238U/204Pb ¼ 10 and5, and 232Th/238U ¼ 3.3 and 2.3 (see Zhang et al., 2005); data sources are as in Zhang et al. (2005), plus Wendt et al. (1999), Ray et al. (2003),Sheth et al. (2003), Doucet et al. (2004), and Nohda et al. (2005).

ZHANG et al. WESTERN EMEISHAN PICRITIC AND BASALT FLOWS

2009

DISCUSSION

Original liquid composition

The olivine phenocrysts with Mg-number >90 contain0.22–0.45 wt % CaO and 0.03–0.17 wt % Cr2O3.Olivine in mantle peridotites is characterized by muchlower Ca and Cr contents (e.g. Gurenko et al., 1996;Thompson & Gibson, 2000). Along with the presence ofmelt inclusions and lack of strain texture, these featuresindicate that the Mg-rich olivines in our picritic lavascrystallized from melt and are not accidental xenocrystsof mantle olivine.Under equilibrium conditions, olivine composition

reflects the composition of the magma from which theolivine crystallized. Thus, the composition of olivine inpicritic rocks can be used to estimate parental magmacomposition (e.g. Simkin & Smith, 1970; Larsen &

Pedersen, 2000; Revillon et al., 1999). In Fig. 9a, olivineMg-number is plotted vs bulk-rock MgO content. Alsoshown are curves representing the loci of equilibriumvalues for liquids containing 8–13 wt % FeO; thesevalues correspond to the FeO content of parentalmagmas, assuming 15% of the total magmatic ironoxide content is ferric. If values for the most Mg-richolivine compositions from a sample plot on an olivine–liquid equilibrium curve at approximately the same FeOcontent as present in the bulk rock, the MgO contentof the rock is close to that of the liquid from which theolivine crystallized. Data points falling below the curvesindicate excess (cumulus) olivine or olivine that formedlater as the magma was cooling.Both cases are represented by the Emeishan picritic

rocks. Data for samples DJ-26, DJ-31, and DJ-34–1 allfall below the equilibrium curves in Fig. 9a, indicating

Mg #-2

-1

0

+1

+2

+3

+4

+5

50 55 60 65 70 75 80 85 0 2 4 6 8 10 12 14 16

(Nb/La)p-2

-1

0

+1

+2

+3

+4

+5

+5

+4

+3

+2

+1

0

-1

-2

0.8 1.0 1.2 1.4 1.6

(La/Yb)p-2

-1

0

+1

+2

+3

+4

+5

2 4 6 8 10 12 14 16 18 20

(La/Yb)p0.703

0.704

0.705

2 4 6 8 10 12 14 16 18 20

(Th/Nd)p-2

-1

0

+1

+2

+3

+4

+5

0.5 1.0 1.5 2.0 2.5 3.0 3.5

ε Nd(

t)

(87S

r/86

Sr)

t

stratigraphic orderDaju section

(a) (b)

(c) (d)

(e) (f)Basaltic lavasPicritic lavas

Fig. 8. Variation of eNd(t) with Mg-number (a), stratigraphic order in the Daju section (b), primitive-mantle-normalized (La/Yb)P (c), (Nb/La)P (d),and (Th/Nd)P (e). (f) Variation of (87Sr/86Sr)t with (La/Yb)P.

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 OCTOBER 2006

2010

that these samples do not represent primitive composi-tions. (However, they are not simply basalts with excessolivine phenocrysts, because the basalts have no olivinephenocrysts, and contain augite rather than the diopsidefound in the picritic lavas.) In contrast, the mostmagnesian olivine values measured for sample SM-14lie within the region defined by the FeO curves. Thehighest Mg-number (91.6) lies near the 11 wt % FeOcurve. This rock, with 22.40 wt % MgO (normalized to amajor element total of 100 wt % on a volatile-free basis),has an appropriate FeO* content (12.92 wt %), againassuming 15% of total iron oxide is ferric (see Figure 9b).We conclude that sample SM-14 is close to anunfractionated melt of the mantle source. Sample SM-15 has a very similar bulk composition.

An independent estimate of initial magmatic MgOcontent may be made from the Ni content of olivinein SM-14 (3300 ppm). Assuming olivine in the sourcemantle has 3500 ppm Ni, a likely upper limit (Korenaga& Kelemen, 2000), implies a melt with only marginallygreater MgO than that of SM-14. For aphyric samplesthat have lost only olivine, Korenaga & Kelemen’s(2000) Ni-in-olivine method of estimating initial mag-matic MgO content mathematically adds equilibriumolivine to the composition of a sample until an olivineNi content of 3500 ppm is reached. Our highest-MgOaphyric sample, DJ-36, has only 7 wt % MgO; as notedabove, clinopyroxene crystallization generally appearsto have begun around 10 wt % MgO in the Lijiang areamagmas. Nevertheless, application of this approach to

95

DJ-31DJ-26

DJ-34-1SM-14

90

85

80

7510 15 20 25 30 35

8

10

12

14

16

18

10 20 30 40 50

8

13

12

11

9

10

Oliv

ine

Mg#

MgO (wt%)

FeO

* (w

t%)

Fo 91.6

Fo 85

BasaltPicritic lavaOlivine

(a)

(b)

Fig. 9. (a) Bulk-rock MgO vs Mg-number of olivine in picritic samples. The curves indicate the relationship between equilibrium olivine Mg-number and liquid MgO content for liquids with FeO values between 8 and 13 wt % (labeled), assuming a Mg–Fe olivine–liquid partitioncoefficient of 0.31 (Roeder & Emslie, 1970). (b) MgO vs FeO* for olivine phenocrysts and bulk rocks of the Lijiang area. The line indicatesliquid compositions in equilibrium with olivine of Mg-number 91.6 (Fo 91.6), assuming 15% of total iron oxide in the liquid is ferric. Data forsamples SM-14 and SM-15 lie near this line.

ZHANG et al. WESTERN EMEISHAN PICRITIC AND BASALT FLOWS

2011

sample DJ-36 yields an estimated initial magmatic MgOcontent of 21.2 wt %, reasonably close to that of SM-14.

Melting conditions

We use sample SM-14 to estimate the conditions underwhich the primary picritic magmas were generated.From the equation T(�C) ¼ 17.86 · MgO(wt%) þ 1061(Nisbet et al., 1993), an estimated eruption (1 atm)temperature of about 1460�C is obtained. Our sampleslack hydrous primary minerals (amphibole, mica, etc.),so the magmatic water content was probably relativelylow. For this 1-atm temperature, a recent model ofadiabatic pressure–temperature paths for primary ultra-mafic magmas (Herzberg & O’Hara, 2002) suggests aninitial melting temperature of about 1630–1680�C andpressure of 4.2–5.0 GPa (Fig. 10). Magma formation atrelatively high pressures is consistent with the highCaO/Al2O3 (see above) and FeO* (12.70, 12.94 wt %),and relatively low SiO2 (45.64, 46.15 wt %) and Al2O3

(8.12, 8.15 wt %) of SM-14 and SM-15 (Walter, 1998).Arndt (2000) noted that the temperature estimated inthis manner may commonly have an error of as much as100�C (with correspondingly large errors in estimatedpressure), because of uncertainties in water content andoxygen fugacity. Additional sources of error are thedepth of magma segregation from the source (Nisbetet al., 1993), the proportion of crystals in the ascending

magma, and the nature of the melting process itself.Another empirical estimate of source potential tem-perature and pressure can be made from, respectively,T(�C) ¼ 2000[MgO/(SiO2 þ MgO)] þ 969 andln[10P](GPa) ¼ 0.00252T – 0.12SiO2 þ 5.027(Albarede, 1992), yielding a temperature of about1630�C and pressure of 3.9 GPa, in broad agreementwith the values above. The olivine saturation-surfacemodel of Putirka (2005, equations A and B) suggests asimilar initial temperature of 1650–1690�C. A hightemperature is also supported by the high Cr content ofthe Cr-spinel; a value of 1673�C is obtained from theequations of Fabries (1979) and Engi & Evans (1980),using the analysis with the highest Cr-number in Table 1.Recent estimates of mantle potential temperature for

the sources of Hawaii and Iceland are in the same rangeas above, whereas those for normal asthenosphere are150–230�C lower (Putirka, 2005). Thus, the sourcetemperature estimated for the Lijiang area picriticflows appears consistent with some type of mantlethermal plume, whether ‘standard-type’ (e.g. Griffiths &Campbell, 1991), non-Newtonian (e.g. Larsen & Yuen,1997), or thermochemical (e.g. Davaille et al., 2005),as the magma source. Interestingly, from inversionof lanthanide rare earth element (REE) data, Xu et al.(2001) also estimated a rather high potential temperature(>1550�C) for the source of their (low Mg-number) low-Ti/Y Emeishan basalts.

01000

1200

1400

1600

1800

2000

2

0 100 200

4 6 8Pressure (GPa)

Tem

pera

ture

(°C

)

Depth (km)

Fertile Peridotite

MORB

Solidu

s

Hawaii IcelandW. Greenland

Gorgona

Gorgona K.

All Liquid

10

20

30

38

Fig. 10. Adiabatic temperature–pressure paths for primary magmas produced by melting of fertile peridotite, from Herzberg & O’Hara (2002).*, estimated 1 atm temperature of SM-14; &, a rough estimate of temperature and pressure at the start of melting. Numbers on paths to the left ofthe solidus indicate wt % MgO of liquids. Ranges of estimated mantle pressure and temperature for Gorgona komatiites (K) and picritic basalts,West Greenland picritic lavas, modern Hawaii and Iceland, and MORB (mid-ocean ridge basalt) are from Herzberg & O’Hara (2002).

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 OCTOBER 2006

2012

A pressure of 4 GPa corresponds to a depth of about130 km, within the garnet stability field (e.g. Walter,1998). Indeed, the variation in the middle to heavy REEamong the Lijiang area lavas is relatively large,comparable with that in the light REE, and indicativeof an important role for garnet during melting (unlike thelight REE, the heavy REE are variably compatible ingarnet). For example, primitive-mantle-normalized (Sm/Yb)P and (La/Sm)P both vary by slightly more than afactor of two, in the range of 2.92–6.74 and 1.69–3.74,respectively.In a diagram of (Tb/Yb)P vs (Yb/Sm)P, melting of

garnet peridotite produces a markedly different trajec-tory from melting of spinel peridotite (Fig. 11a). TheLijiang area data lie closer to a melt path for garnetperidotite than to one for spinel peridotite. With theassumptions used to construct Fig. 11a, the data areconsistent with a 60–100% contribution to the melt frommelting in the presence of garnet, and with rather smallamounts of partial melting, �2–7%. We caution thatthese values should not be taken literally, because thepositions of the curves in Fig. 11a can vary with adifferent choice of melting model, distribution coeffi-cients, source composition, and/or mineral proportions.Moreover, given the range in eNd(t) among our samples,the assumption of a single model source compositionis not necessarily valid (see below). However, the

conclusion that much of the melting occurred ingarnet-facies mantle and that the amount of partialmelting was rather modest is difficult to avoid. It seemsclear that compared with an intra-oceanic plateau suchas the Ontong Java, for which mean melt percentages of25–30% have been estimated (Fitton & Godard, 2004;Herzberg, 2004), the Lijiang area magmas formedthrough much smaller amounts of partial melting atrather great depth. High-Ti, high-MgO lavas in theSiberian, Ethiopian, and Deccan Traps also have beeninterpreted to represent rather small percentages ofmelting involving substantial contributions from garnetperidotite (e.g. Arndt et al., 1998; Pik et al., 1998; Mellusoet al., 2006). In contrast, low-Ti/Y lavas at Binchuan andYongsheng appear to reflect a greater role for melting atshallower depths in the spinel stability field (Xu et al.,2001; Zhang & Wang, 2002a), whereas rare low-Ti/Y,high-MgO (komatiitic) lavas in Vietnam, which mightalso be related to the Emeishan event, are interpreted asmelts from a significantly more depleted and garnet-freesource (Hanski et al., 2004) (see Fig. 11b).

Mantle source and contamination

The high-eNd(t), high-(206Pb/204Pb)t end of the arraydefined by the Lijiang area data in Fig. 7 points to amantle source isotopically similar to that of several

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.0 1.5 2.0 2.5 3.0 3.5 4.0

(Tb/Yb)P

(Yb/

Sm

) P

Model source

1%5%

10%

15% Gar 100Gar 0

Gar 50

0

0.4

0.2

0.6

1.2

1.0

0.8

1.4

1.6

1.8

1.0 2.0 3.0 4.0

(Tb/Yb)P

(a) (b)

Basaltic lavasPicritic lavas

Binchuan & YongshengVietnam

Lijiang area

Fig. 11. (a) (Tb/Yb)P vs (Yb/Sm)P for Lijiang area lavas. The grid indicates the range of model melt compositions produced by 1%, 5%, 10%, and15% of aggregated fractional melting (Shaw, 1970) of peridotite in which the amount of melting that occurs in the presence of garnet varies from 0to 100%. The light lines indicate the percentage of melt contribution from garnet-facies mantle (Gar); e.g. Gar 0 corresponds to melt from spinelperidotite. Curves of constant melt fraction are shown by bold lines. The curves are for a source consisting of a 1:1 mix of estimated averagedepleted mantle (Workman & Hart, 2005) and model enriched mantle peridotite (Ito & Mahoney, 2005); such a source is consistent with the Ndisotope values of our highest-eNd samples. Partition coefficients are taken or interpolated from Salters & Stracke (2004). The unmelted peridotite isassumed to be 53% olivine, 30% orthopyroxene, 10% clinopyroxene, and 7% garnet or spinel, and melting of these minerals is assumed to occur inproportions of 10%, 10%, 40%, and 40%, respectively, after Janney et al. (2000). (b) Data for Binchuan and Yongsheng basalts (Xu et al., 2001;Zhang & Wang, 2002a; Xiao et al., 2004) and for komatiitic and basaltic lavas of Vietnam (Hanski et al., 2004) are compared with a field for theLijiang area data.

ZHANG et al. WESTERN EMEISHAN PICRITIC AND BASALT FLOWS

2013

present-day oceanic hotspots (allowing for isotopicevolution via radioactive decay of parent nuclides inmantle reservoirs since 250 Ma). Likewise, patterns ofalteration-resistant incompatible elements are ocean-island-like (Fig. 6c and d). These characteristics suggest asource in the convecting mantle or in lithospheric mantlenot long isolated from it. In addition to various versionsof the plume-head model, some non-plume hypotheses(e.g. Alt et al., 1988; Sheth, 2005) may accommodatesuch signatures, and we therefore cannot rule out thepossibility of a non-plume mechanism. However, thecombination of hotspot-type geochemical characteristicswith the evidence for a high mantle-source temperaturediscussed above and the apparently very strong evi-dence for a large amount of regional lithospheric upliftimmediately preceding flood volcanism (He et al., 2003;Xu et al., 2004) qualitatively appears consistent withsome type of plume-head origin.The Lijiang-area data points trending in Fig. 7 toward

lower eNd(t) and (206Pb/204Pb)t and higher (87Sr/86Sr)tcould indicate low-eNd material in the source mantleor variable contamination of magmas with Mesopro-terozoic or late Paleoproterozoic lithosphere (crust ormantle) overlying the high-eNd source region (recentdating of a granulite in the vicinity gives an age of 1140Ma; Xu et al., 2002). The low-eNd(t) material wasrelatively enriched in highly incompatible elements andhad lower Nb/La than the high-eNd source mantle(Fig. 8c–e). Many continental crustal rocks and siliceousmarine sediments have low Nb/La and incompatible-element patterns with pronounced troughs at Nb–Ta,together with strongly negative eNd values (e.g. Rudnick& Fountain, 1995; Plank & Langmuir, 1998). Non-picritic basalts with pronounced Nb–Ta troughs andmore negative eNd(t) than any of the Lijiang area lavasare present elsewhere in the Emeishan province, andhave been interpreted to represent moderate to sub-stantial lithospheric inputs (Xu et al., 2001; Zhang &Wang, 2002a; Xiao et al., 2004). In contrast, our lowest-eNd(t) samples barely reach negative eNd territory, andlack significant Nb–Ta troughs in their incompatible-element patterns. Thus, the contribution of low-Nb–Ta continental lithospheric or subducted-sediment-derived material in the mantle source of the Lijiangarea lavas appears likely to have been comparativelyminor. However, we cannot rule out the possibility of arole for a continental lithospheric end-member withrelatively high (though negative) eNd and Nb/La.Granitic gneisses from a Neoproterozoic (�800 Ma)phase of magmatic-arc activity are present in the region;Nd isotope data are not available, but these rocks havelarge Nb–Ta troughs in their incompatible-elementpatterns (e.g. Zhou et al., 2002b). It is possible thatother rocks formed in the same event had relativelyhigh eNd at 250 Ma; however, for the most part such

arc-related crust would also be expected to possessincompatible-element patterns with marked Nb–Tatroughs. In any case, no correlation between isotoperatios and SiO2 is present among the Lijiang area lavas,implying that bulk assimilation of high-SiO2 continentalcrust was comparatively minor. The absence of plagio-clase phenocrysts suggests the magmas spent little time atcrustal levels. Consistent with this inference, applicationof Putirka et al.’s (1996) geobarometer to the clinopyr-oxene data in Table 1 suggests the clinopyroxenephenocrysts equilibrated at depths between about 30and 70 km.

Temporal evolution

In the thick Daju section, for which we have the mostisotopic data, the stratigraphically higher lavas tend tohave higher eNd(t) (Fig. 8b), indicating that the contribu-tion of low-eNd material waned overall with time. Thiscould represent the gradual depletion of easily meltedlithospheric wall-rock in the magmatic conduit system,or the melting out of a more fusible component in themantle source itself. The variation in garnet-sensitiveratios such as (Yb/Sm)P and (Tb/Yb)P (Fig. 11a) testifiesto melting over a range of depths in the mantle source.More fusible, low-eNd material in the source shouldordinarily begin melting at greater depths than morerefractory components (e.g. Ito & Mahoney, 2005),which in turn could lead to correlations of garnet-sensitive element ratios with eNd(t) in the melts. Thelack of such correlations may favor a lithosphericinfluence, because melting in the source could belargely decoupled from assimilation of low-eNd litho-spheric wall-rock.In any case, lithological evidence suggests that the

Lijiang area lavas represent an early stage in the floodbasalt episode. In particular, plagioclase-phyric basaltsare absent in the olivine- and clinopyroxene-dominatedDaju and Shiman sections, whereas to the south and east(Binchuan and Yongsheng, respectively), plagioclase-phyric lavas overlie a sequence of clinopyroxene-phyricbasalts (Zhang et al., 2004). Still farther east, quartztholeiites overlie a succession of plagioclase-phyric flows(Hou et al., 1999). If these regional differences corre-spond to a temporal sequence, an early transition fromrelatively low-eNd(t) to higher-eNd(t) (� þ4, ±1) composi-tions was repeated at least once during the Emeishanepisode, because the stratigraphically higher lavas in theBinchuan and Ertan areas tend to have the highesteNd(t), as in the Lijiang area (Xu et al., 2001; Zhang &Wang, 2003; Xiao et al., 2004). At present, it is not clearhow to interpret these results in terms of lithosphericthinning with time, the location of magma sourcesrelative to any postulated plume axis, development ofseparate magma-plumbing systems, etc. For example, in

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 OCTOBER 2006

2014

addition to temporal changes in magma and/or mantlesource composition, peak volcanism may have migratedwith time in response to plate motion over a hotspot.Thick piles of low-Ti/Y tholeiites make up the bulk

of sections in the Binchuan and Ertan areas, fromwhich Xu et al. (2001, 2004) and Xiao et al. (2004)concluded that the western Emeishan province iscomposed predominantly of low-Ti/Y basalts. Thesebasalts have flatter REE patterns than the high-Ti/YLijiang area lavas. Compositions are too evolved to makemajor-element-based estimates of initial temperature,but application of McKenzie & O’Nions’ (1991) REEinversion led Xu et al. (2001) to conclude that the low-Ti/Y basalts represent comparatively high-degree partialmelts formed at high potential temperature (>1550�C);the setting was inferred to be above the axis of a plumehead. Minor high-Ti/Y basalts were found in theuppermost parts of the Binchuan and Ertan sections.They have broadly ocean-island-like incompatible-element signatures rather similar to those of the Lijiangarea lavas, but concentrations of incompatible elementsare significantly higher. These high-Ti/Y basalts wereinterpreted, again on the basis of REE inversion, to bevery small-degree partial melts (<1.5%) formed atgreater mantle depths and lower potential tempera-tures (<1500�C) during the waning stages of volcanism,probably from a cooling plume mantle (Xu et al., 2001).Sections in the eastern and southern parts of theprovince are dominated by high-Ti/Y basalts, whichXu et al. (2004) attributed to small-degree melting nearthe periphery of the plume head, where the lithosphericlid was thicker. Somewhat similar secular variationshave been observed in a portion of the �55 Ma EastGreenland flood basalt pile, and are also interpreted interms of cooling of a plume and variation in lithosphericthickness (Tegner et al., 1998). In the Maymecha Riverbasin of the Siberian Traps, alkalic, high-Ti picriticbasalts and meimechites lie above a sequence of low-Titholeiitic lavas. This late high-Ti group is interpreted tohave formed by small amounts of partial melting ofgarnet-facies mantle at even greater depths (Arndt et al.,1998) than those we estimate for the Lijiang area lavas.High-Ti lavas, some of which are picritic, are also

present in the lower portions of the Siberian Trapssequence (e.g. Lightfoot et al., 1993; Sharma, 1997). Ourwork shows that a thick sequence of high-Ti/Y lavaswith ocean-island-like source characteristics is present inthe western part of the Emeishan province. As notedabove, these lavas are likely to be earlier in the volcanicsuccession than the low-Ti/Y basalts at Binchuan andErtan. Formation of the Lijiang area magmas by modestamounts of partial melting, largely in garnet peridotitemantle, can be accommodated straightforwardly in thecontext of a plume-head model if melting occurredbeneath a thick lithospheric lid before plume-related or

other stress on the overlying lithosphere led to majorlithospheric thinning (e.g. see Campbell, 1998). Litho-spheric thinning might have been rapid in the case of aplume with non-Newtonian rheology (Larsen & Yuen,1997).

CONCLUSIONS

Olivine phenocrysts with Mg-number as high as 91.6 inpicritic lavas of the Lijiang area crystallized from liquids.We estimate that the picritic lavas formed from meltswith about 22 wt % MgO generated by relatively smallamounts of partial melting, much of which appears tohave occurred in the presence of residual garnet. Initialsource temperature is estimated to have been as high as1630–1690�C. The high temperature is consistent with,although not proof of, a plume-head origin, whereasthe importance of garnet during melting implies a ratherthick lithospheric lid. As such, the Lijiang area lavasmay represent an early stage in the Emeishan floodbasalt episode, before major lithospheric thinning hadoccurred.The basalt flows that make up the bulk of the Lijiang

area sequence are isotopically indistinguishable fromthe picritic lavas, and appear to be derived frompicritic magmas by removal of olivine and clinopyrox-ene. Isotopic and incompatible-element characteristicsof the lavas indicate an eNd(t) � þ4 mantle sourcesimilar to those of several present-day oceanic hotspots.An isotopically broadly similar source may have beeninvolved in the generation of the nearly contempora-neous Siberian Traps, although they were formed in alocation far from the Emeishan flood basalts. Theabsence of plagioclase phenocrysts in both the picriticand basaltic lavas implies that the Lijiang area magmasspent little time at shallow lithospheric levels prior toeruption, which may help explain the lack of evi-dence for significant contamination. However, correla-tions between isotope ratios and ratios of highlyincompatible elements suggest that some interaction ofascending magmas with continental crust or lithosphericmantle occurred. Alternatively, relatively low-eNd, low-Nb/La material may have been present in the mantlesource.

ACKNOWLEDGEMENTSWe are grateful to D. Pyle, D. VonderHaar, andN. Hulbirt for assistance with the work at SOEST, andto K. Johnson for discussions on primary magmas. Wethank reviewers N. Arndt, D. Peate and A. Saunders,and editor W. Bohrson for their thoughtful and con-structive comments. This research was supportedby Program for New Century Excellent Talents in

ZHANG et al. WESTERN EMEISHAN PICRITIC AND BASALT FLOWS

2015

University (NCET-04-0728), NSF-China grant(40273020), and US NSF grant EAR98-0531.

REFERENCES

Albarede, F. (1992). How deep do common basaltic magmas form and

differentiate? Journal of Geophysical Research 97, 10997–11009.

Ali, J. R., Thompson, G. M., Zhou, M. & Song, Y. (2005). Emeishan

large igneous province, SW China. Lithos 79, 475–489.

Alt, D., Sears, J. M. & Hyndman, D. W. (1988). Terrestrial maria: the

origins of large basalt plateaus, hotspot tracks and spreading ridges.

Journal of Geology 96, 647–662.

Arndt, N. T. (2000). Hot heads and cold tails. Nature 407, 458–461.

Arndt, N. T., Naldrett, A. J. & Pyke, D. R. (1977). Komatiitic and iron-

rich tholeiitic lavas of Munro Township, northeast Ontario. Journal of

Petrology 18, 319–369.

Arndt, N. T., Chauvel, C., Fedorenko, V. & Czamanske, G. (1998).

Two mantle sources, two plumbing systems: tholeiitic and alkaline

magmatism of the Maymecha River basin, Siberian flood volcanic

province. Contributions to Mineralogy and Petrology 133, 297–313.

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.

Barnes, S. J. & Roeder, P. L. (2001). The range of spinel compositions

in terrestrial mafic and ultramafic rocks. Journal of Petrology 42,

2279–2302.

Bhat, I. & Zainuddin, S. M. (1981). The Panjal Trap chemistry; witness

to the birth of Tethys. Geological Magazine 118, 355–365.

Bloomer, S., Melchior, J., Evans, C. & Francis, R. D. (1982). Techniques

for the chemical analysis of igneous rocks at the Scripps Institution of

Oceanography. La Jolla, CA: Scripps Institution of Oceanography

Report, Reference 82-7, 510, 68 pp.

Boven, A., Pasteels, P., Punzalan, L. E., Liu, J., Luo, X., Zhang, W.,

Guo, Z. & Hertogen, J. (2002). 40Ar/39Ar geochronological

constraints on the age and evolution of the Permo-Triassic Emeishan

volcanic province, southwest China. Journal of Asian Earth Science 20,

157–175.

Campbell, I. H. (1998). The mantle’s chemical structure: insights from

the melting products of mantle plumes. In: Jackson, I. (ed.) The

Earth’s Mantle: Composition, Structure, and Evolution. Cambridge:

Cambridge University Press, pp. 259–310.

Carlson, R. W. (1991). Physical and chemical evidence on the cause

and source characteristics of flood basalt volcanism. Australian Journal

of Earth Sciences 38, 525–544.

Carlson, R. W. & Hart, W. K. (1988). Flood basalt volcanism in the

northwestern United States. In: Macdougall, J. D. (ed.) Continental

Flood Basalts. Dordrecht: Kluwer Academic, pp. 35–62.

Chung, S. L. & Jahn, B. M. (1995). Plume–lithosphere interaction in

generation of the Emeishan flood basalts at the Permian–Triassic

boundary. Geology 23, 889–892.

Clague, D. A. & Frey, F. A. (1982). Petrology and trace element

geochemistry of the Honolulu Volcanics, Oahu: implications for the

oceanic mantle below Hawaii. Journal of Petrology 23, 447–504.

Cong, B. (1988). Formation and Development of the Panxi Ancient Rift. Beijing:

Scientia, 424 pp. (in Chinese).

Czamanske, G. K., Gurevitch, A. B., Fedorenko, V. & Simonov, O.

(1998). Demise of the Siberian plume, paleogeographic and paleotec-

tonic reconstruction from the prevolcanic and volcanic record, north–

central Siberia. International Geological Review 40, 95–115.

Davaille, A., Stutzmann, E., Silveira, G., Besse, J. & Courtillot, V.

(2005). Convective patterns under the Indo-Atlantic box. Earth and

Planetary Science Letters 239, 233–252.

Dick, H. J. B. & Bullen, T. (1984). Chromian spinel as a petrogenetic

indicator in abyssal and alpine-type peridotites and spatially

associated lavas. Contributions to Mineralogy and Petrology 86, 54–76.

Donaldson, C. H. & Brown, R. W. (1977). Refractory megacrysts and

magnesium-rich melt inclusions within spinel in oceanic tholeiites;

indicators of magma mixing and parental magma composition. Earth

and Planetary Science Letters 37, 81–89.

Doucet, S., Weis, D., Scoates, J. S., Debaille, V. & Giret, A. (2004).

Geochemical and Hf–Pb–Sr–Nd isotopic constraints on the origin of

the Amsterdam–St. Paul (Indian Ocean) hotspot basalts. Earth and

Planetary Science Letters 218, 179–195.

Dulski, P. (1994). Interferences of oxide, hydroxide and chloride species

in the determination of rare earth elements in geological samples by

inductively coupled plasma–mass spectrometry. Fresenius’s Journal of

Analytical Chemistry 350, 194–203.

Duncan, R. A. (2002). A time frame for construction of the Kerguelen

Plateau and Broken Ridge. Journal of Petrology 43, 1109–1119.

Engi, M. & Evans, B. W. (1980). Reevaluation of the olivine–spinel

geothermometer: discussion. Contributions to Mineralogy and Petrology 74,

201–203.

Enkin, R. J., Courtillot, V., Leloup, Ph., Yang, Z., Xing, L., Zhang, J.

& Zhuang, Z. (1992). The paleomagnetic record of Uppermost

Permian, Lower Triassic rocks from the South China Block.

Geophysical Research Letters 19, 2147–2150.

Fabries, J. (1979). Spinel–olivine geothermometry in peridotites

from ultramafic complex. Contributions to Mineralogy and Petrology 69,

329–336.

Fitton, J. G. & Godard, M. (2004). Origin and evolution of magmas

on the Ontong Java Plateau. In: Fitton, J. G., Mahoney, J. J.,

Wallace, P. J. & Saunders, A. D. (eds) Origin and Evolution of the Ontong

Java Plateau. Geological Society, London, Special Publications, 229,

151–178.

Fleming, T. H., Elliot, D. H., Jones, L. M., Bowman, J. R. &

Siders, M. A. (1992). Chemical and isotopic variations in an iron-

rich lava flow from the Kirkpatrick Basalt, North Victoria Land,

Antarctica: implications for low-temperature alteration. Contributions

to Mineralogy and Petrology 111, 440–457.

Fretzdorff, S. & Haase, K. M. (2002). Geochemistry and petrology of

lavas from the submarine flanks of Reunion Island (western Indian

Ocean): implications for magma genesis and the mantle source.

Mineralogy and Petrology 75, 153–184.

Galer, S. J. G. (1999). Optimal double and triple spiking for high

precision lead isotopic measurement. Chemical Geology 157, 255–274.

Garcia, M. O., Hulsebosch, T. P. & Rhodes, J. M. (1995). Olivine-rich

submarine basalts from the Southwest Rift Zone of Mauna Loa

Volcano: implications for magmatic processes and geochemical

evolution. In: Rhodes, J. M. & Lockwood, J. P. (eds) Mauna Loa

Revealed: Structure, Composition, History, and Hazards. Geophysical Mono-

graphs, American Geophysical Union 92, 219–239.

Genc, S. C. (2004). A Triassic large igneous province in the Pontides,

northern Turkey: geochemical data for its tectonic setting. Journal of

Asian Earth Science 22, 503–516.

Govindaraju, K. (1994). Compilation of working values and sample

descriptions for 383 geostandards. Geostandards Newsletter, Special Issue

18, 1–158.

Griffiths, R. W. & Campbell, I. H. (1991). Interaction of mantle plume

heads with the Earth’s surface and onset of small-scale convection.

Journal of Geophysical Research 96, 18295–18310.

Gurenko, A. A., Hansteen, T. H. & Schmincke, H. U. (1996).

Evolution of parental magmas of Miocene shield basalts of Gran

Canaria (Canary Islands): constraints from crystal, melt and fluid

inclusions in minerals. Contributions to Mineralogy and Petrology 89,

422–435.

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 OCTOBER 2006

2016

Hanan, B. B. & Graham, D. W. (1996). Lead and helium isotope

evidence from oceanic basalts for a common deep source of mantle

plumes. Science 272, 991–995.

Hanski, E., Walker, R. J., Huhma, H., Polyakov, G. V., Balykin, P. A.,

Hoa, T. T. & Phuong, N. T. (2004). Origin of Permian–Triassic

komatiites, northwestern Vietnam. Contributions to Mineralogy and

Petrology 147, 453–469.

Hawkesworth, C. J., Marsh, J. S., Duncan, A. R., Erlank, A. J. &

Norry, M. J. (1984). The role of continental lithosphere in the

generation of the Karoo volcanic rocks: evidence from combined

Nd- and Sr-isotope studies.. In: Erlank, A. J (ed.) Petrogenesis of the

Volcanic Rocks of the Karoo Province. Geological Society of South Africa, Special

Publications 13, 341–354.

He, B., Xu, Y., Xiao, L., Chung, S. & Wang, Y. (2003). Sedimentary

evidence for a rapid, kilometer-scale crustal doming prior to the

eruption of the Emeishan flood basalts. Earth and Planetary Science

Letters 213, 391–405.

Herzberg, C. (2004). Partial melting of lavas below the Ontong Java

Plateau. In: Fitton, J. G., Mahoney, J. J., Wallace, P. J. &

Saunders, A. D. (eds) Origin and Evolution of the Ontong Java Plateau.

Geological Society, London, Special Publications 229, 179–184.

Herzberg, C. & O’Hara, M. J. (2002). Plume-associated ultramafic

magmas of Phanerozoic age. Journal of Petrology 43, 1857–1883.

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.

Hoernle, K., Hauff, F., Werner, R. & Mortimer, N. (2004). New

insights into the origin and evolution of the Hikurangi oceanic

plateau. EOS Transactions, American Geophysical Union 85, 401–402.

Hou, Z., Lu, J. & Wang, Y. (1999). Emeishan large igneous province:

structure, genesis and feature. Geological Review 45, 885–891 (in

Chinese with English abstract).

Ito, G. & Mahoney, J. J. (2005). Flow and melting of a heterogeneous

mantle: 1. Importance to the geochemistry of ocean island and mid-

ocean ridge basalts. Earth and Planetary Science Letters 230, 29–46.

Janney, P. E., Macdougall, J. D., Natland, J. H. & Lynch, M. A. (2000).

Geochemical evidence from the Pukapuka volcanic ridge system

for a shallow enriched mantle domain beneath the South Pacific

superswell. Earth and Planetary Science Letters 181, 47–60.

Jin, Y. & Shang, J. (2000). The Permian of China and its interregional

correlation. In: Yin, H. (ed.) Permian–Triassic Evolution of Tethys and

Western Circum-Pacific. Developments in Palaeontology and Stratigraphy 18,

71–98.

Kamo, S. L., Czamanske, G. K., Amelin, Y., Fedorenko, V. A.,

Davis, D. W. & Trofimov, V. R. (2003). Rapid eruption of Siberian

flood-volcanic rocks, coincident with the Permian–Triassic boundary

and mass extinction at 251 Ma. Earth and Planetary Science Letters 214,

75–91.

Kent, R. W., Pringle, M. S., Mueller, R. D., Saunders, A. D. &

Ghose, N. C. (2002). 40Ar/39Ar geochronology of the Rajmahal

basalts, India, and their relationship to the Kerguelen Plateau.

Journal of Petrology 43, 1141–1153.

Kerr, A. C., Tarney, J., Marriner, G. F., Nivia, A. & Saunders, A. D.

(1997). The Caribbean–Colombian Cretaceous igneous province:

the internal anatomy of an oceanic plateau. In: Mahoney, J. J. &

Coffin, M. F. (eds) Large Igneous Provinces: Continental, Oceanic and

Planetary Flood Volcanism. Geophysical Monographs, American Geophysical

Union 100, 123–144.

Kieffer, B., Arndt, N., LaPierre, H., Bastien, F., Bosch, D., Pecher, A.,

Yirgu, G., Ayalew, D., Weis, D., Jerram, D. A., Keller, F. &

Meugniot, C. (2004). The transition from plateau to shield volcanism

in Ethiopia: a petrological and geochemical study. Journal of Petrology

45, 793–834.

Korenaga, J. & Kelemen, P. B. (2000). Major element heterogeneity in

the mantle source of the North Atlantic igneous province. Earth and

Planetary Science Letters 184, 251–268.

Larsen, L. M. & Pedersen, A. K. (2000). Processes in high-Mg, high-T

magmas: evidence from olivine, chromite and glass in

Palaeogene picrites from West Greenland. Journal of Petrology 41,

1071–1098.

Larsen, T. B. & Yuen, D. A. (1997). Ultrafast upwelling bursting

through the upper mantle. Earth and Planetary Science Letters 146,

393–399.

Lassiter, J. C., DePaolo, D. J. & Mahoney, J. J. (1995). Geochemistry of

the Wrangellia flood basalt province; implications for the role of

continental and oceanic lithosphere in flood basalt genesis. Journal of

Petrology 36, 983–1009.

Le Bas, M. J. (2000). IUGS reclassification of the high-Mg and picritic

volcanic rocks. Journal of Petrology 41, 1467–1470.

Lesher, C. M. (1989). Komatiite-associated nickel sulfide deposits.

In: Whitney, J. A. & Naldrett, A. J. (eds) Ore Deposition Associated with

Magmas. Reviews in Economic Geology 4, 45–101.

Lightfoot, P. C., Hawkesworth, C. J., Hergt, J., Naldrett, A. J.,

Gorbachev, N. S., Fedorenko, V. A. & Doherty, W. (1993).

Remobilisation of the continental lithosphere by a mantle plume:

major-, trace element and Sr-, Nd-, and Pb-isotope evidence from

picritic and tholeiitic lavas of the Noril’sk District, Siberian Trap,

Russia. Contributions to Mineralogy and Petrology 114, 171–188.

Lindstrom, M. M. & Haskin, L. A. (1981). Compositional heterogen-

eities in a single Icelandic tholeiite flow. Geochimica et Cosmochimica Acta

45, 15–31.

Lo, C., Chung, S., Lee, T. & Wu, G. (2002). Age of the Emeishan flood

magmatism and relations to Permian–Triassic boundary events.

Earth and Planetary Science Letters 198, 449–458.

Mahoney, J. J. (1988). Deccan Traps. In: Macdougall, J. D. (ed.)

Continental Flood Basalts Kluwer Academic Publishers,

Dordrecht, 151–194.

Mahoney, J. J., Storey, M., Duncan, R. A., Spencer, K. J. &

Pringle, M. (1993). Geochemistry and age of the Ontong Java

Plateau. In: Pringle, M., Sager, W., Sliter, W. & Stein, S. (eds) The

Mesozoic Pacific: Geology, Tectonics, and Volcanism. Geophysical Monographs,

American Geophysical Union 77, 233–261.

Mahoney, J. J., Duncan, R. A., Tejada, M. L. G., Sager, W. W. &

Bralower, T. J. (2005). Jurassic–Cretaceous boundary age and

MORB-type mantle source for Shatsky Rise. Geology 33, 185–188.

McKenzie, D. P. & O’Nions, R. K. (1991). Partial melt distributions

from inversion of rare earth element concentrations. Journal of

Petrology 32, 1021–1091.

Melluso, L., Mahoney, J. J. & Dallai, L. (2006). Mantle sources and

crustal input as recorded in high-Mg Deccan Traps basalts of

Gujarat (India). Lithos 89, 259–274.

Neal, C. R. (2001). The interior of the Moon: the presence of garnet in

the primitive, deep lunar mantle. Journal of Geophysical Research 106,

27865–27885.

Nisbet, E. G., Bickle, M. J. & Martin, A. (1977). The mafic and

ultramafic lavas of the Belingwe greenstone belt. Rhodesia. Journal of

Petrology 18, 521–566.

Nisbet, E. G., Cheadle, M. J., Arndt, N. T. & Bickle, M. J. (1993).

Constraining the potential temperature of the Archaean mantle: a

review of the evidence from komatiites. Lithos 30, 291–307.

Nohda, S., Kaneoka, I., Hanyu, T., Xu, S. & Uto, K. (2005).

Systematic variation of Sr-, Nd- and Pb-isotopes with time in lavas of

Mauritius, Reunion hotspot. Journal of Petrology 46, 505–522.

ZHANG et al. WESTERN EMEISHAN PICRITIC AND BASALT FLOWS

2017

Norrish, K. & Chappell, B. W. (1977). X-ray fluorescence spectro-

metry. In: Zussman, J. (ed.) Physical Methods in Determinative Mineralogy,

2nd edition. Academic Press, New York, pp. 201–272.

Peate, D. W. (1997). The Parana–Etendeka province.

In: Mahoney, J. J. & Coffin, M. F. (eds) Large Igneous Provinces:

Continental, Oceanic and Planetary Flood Volcanism. Geophysical Monographs,

American Geophysical Union 100, 217–245.

Peng, Z. Z. & Mahoney, J. J. (1995). Drill-hole lavas from the

northwestern Deccan Traps, and the evolution of Reunion hotspot

mantle. Earth and Planetary Science Letters 134, 169–185.

Pik, R., Deniel, C., Coulon, C., Yirgu, G., Hoffmann, C. & Ayalew, D.

(1998). The northwestern Ethiopian Plateau flood basalts: classifica-

tion and spatial distribution of magma types. Journal of Volcanology and

Geothermal Research 81, 91–111.

Plank, T. & Langmuir, C. H. (1998). The chemical composition of

subducting sediment and its consequences for the crust and mantle.

Chemical Geology 145, 325–394.

Putirka, K. D. (2005). Mantle potential temperatures at Hawaii,

Iceland, and the mid-ocean ridge system, as inferred from olivine

phenocrysts: evidence for thermally driven mantle plumes. Geochem-

istry, Geophysics, Geosystems 6, Q05L08, doi:10.1029/2005GC000915.

Putirka, K. D., Johnson, M. & Kinzler, R. (1996). Thermobarometry

of mafic igneous rocks based on clinopyroxene–liquid equilibrium,

0–30 kbar. Contributions to Mineralogy and Petrology 123, 92–108.

Ray, J. S., Mahoney, J. J., Johnson, K. T. M., Pyle, D. G., Naar, D. &

Wessel, P. (2003). Geochemistry of volcanism along the Nazca Ridge

and Easter Seamount Chain. EGS–AGU–EUG Joint Assembly, Nice,

France, Abstract EAE03-A-03352.

Renne, P. R., Ernesto, M., Pacca, I. G., Coe, R. S., Glen, J. M.,

Prevot, M. & Perein, M. (1992). The age of Parana flood volcanism,

rifting of Gondwanaland, and the Jurassic–Cretaceous boundary.

Science 258, 975–979.

Revillon, S., Arndt, N. T., Hallot, E., Kerr, A. C. & Tarney, J. S.

(1999). Petrogenesis of picrites from the Caribbean Plateau and the

North Atlantic magmatic province. Lithos 49, 1–21.

Richards, M. A., Duncan, R. A. & Courtillot, V. E. (1989). Flood-

basalts and hot spot tracks: plume heads and tails. Science 246,

103–107.

Roeder, P. L. & Emslie, R. F. (1970). Olivine–liquid equilibrium.

Contributions to Mineralogy and Petrology 29, 275–289.

Rudnick, R. L. & Fountain, D. M. (1995). Nature and composition of

the continental crust: a lower crustal perspective. Reviews in Geophysics

33, 267–309.

Salters, V. J. M. & Stracke, A. (2004). Composition of the depleted

mantle. Geochemistry, Geophysics, Geosystems 5, Q05004,

doi:05010.01029/02003GC000597.

Saunders, A. D., England, R. W., Reichow, M. K. & White, R. V.

(2005). A mantle plume origin for the Siberian Traps: uplift and

extension in the West Siberian Basin, Russia. Lithos 79, 407–424.

Sharma, M. A. (1997). Siberian Traps. In: Mahoney, J. J. &

Coffin, M. F. (eds) Large Igneous Provinces: Continental, Oceanic, and

Planetary Flood Volcanism. Geophysical Monographs, American Geophysical

Union 100, 273–295.

Sharma, M. A., Basu, R. & Nesterenko, G. V. (1992). Temporal Sr-,

Nd-, and Pb-isotopic variations in the Siberian flood basalts:

implications for the plume-source characteristics. Earth and Planetary

Science Letters 113, 365–381.

Shaw, D. M. (1970). Trace element fractionation during anatexis.

Geochimica et Cosmochimica Acta 34, 237–243.

Sheth, H. C. (2005). Were the Deccan flood basalts derived in part

from ancient oceanic crust within the Indian continental lithosphere?

Gondwana Research 8, 109–127.

Sheth, H. C., Mahoney, J. J. & Baxter, A. N. (2003). Geochemistry of

lavas from Mauritius: mantle source and petrogenesis. International

Geological Review 45, 780–797.

Simkin, T. & Smith, J. V. (1970). Minor-element distribution in olivine.

Journal of Geology 78, 304–325.

Song, X., Zhou, M., Hou, Z., Cao, Z., Wang, W. & Li, Y. (2001).

Geochemical constraints on the mantle source of the upper Permian

Emeishan continental flood basalts, southern China. International

Geological Review 43, 213–225.

Stein, M. & Hofmann, A. W. (1994). Mantle plumes and episodic

crustal growth. Nature 372, 63–68.

Stewart, K., Turner, S., Kelley, S., Hawkeworth, C., Kirstein, L. &

Mantovani, M. (1996). 40Ar/39Ar geochronology in the Parana

continental flood basalt province. Earth and Planetary Science Letters

143, 95–109.

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

the Ocean Basins. Geological Society, London, Special Publications 42,

313–345.

Tegner, C., Lesher, C. E., Larsen, L. M. & Watt, W. S. (1998).

Evidence from the rare-earth record of mantle melting for cooling of

the Tertiary Iceland plume. Nature 395, 591–595.

Tejada M. L. G., Mahoney, J. J., Neal, C. R., Duncan, R. A. &

Petterson, M. G. (2002). Basement geochemistry and geochronology

of Central Malaita, Solomon Islands, with implications for the origin

and evolution of the Ontong Java Plateau. Journal of Petrology 43,

449–484.

Tejada, M. L. G., Mahoney, J. J., Castillo, P. R., Ingle, S. P.,

Sheth, H. C. & Weis, D. (2004). Pin-pricking the elephant: evidence

on the origin of the Ontong Java Plateau from Pb–Sr–Hf–Nd

isotopic characteristics of ODP Leg 192 basalts. In: Fitton, J. G.,

Mahoney, J. J., Wallace, P. J. & Saunders, A. D. (eds) Origin and

Evolution of the Ontong Java Plateau. Geological Society, London, Special

Publications 229, 133–150.

Thompson, G., Ali, J. & Song, X. (2001). Emeishan basalts, SW China:

reappraisal of the formation’s type area stratigraphy and a discussion

of the significance as a large igneous province. Journal of the Geological

Society, London 158, 593–599.

Thompson, R. N. & Gibson, S. A. (2000). Transient high temperatures

in mantle plume heads inferred from magnesian olivines in

Phanerozoic picrites. Nature 407, 502–506.

Walter, M. J. (1998). Melting of garnet peridotite and the origin of

komatiite and depleted lithosphere. Journal of Petrology 39, 29–60.

Wang, Y., Li, J., Zhou, R., Wang, W., Tong, C. & Xiong, X. (1993).

Principle on Trace Element Geochemistry of Igneous Rocks and its

Applications—and Demonstration of the Petrogenesis of Emeishan Basalts:

Chengdu University of Science and Technology, (in Chinese).

Wendt, J. I., Regelous, M., Niu, Y., Hekinian, R. & Collerson, K. D.

(1999). Geochemistry of lavas from the Garrett transform fault:

insights into mantle heterogeneity beneath the eastern Pacific. Earth

and Planetary Science Letters 173, 271–284.

Wilson, S. A. (1997). The collection, preparation, and testing of USGS

reference material BCR-2, Columbia River Basalt. US Geological Survey

Open-File Report. http://minerals.cr.usgs.gov/geo_chem_stand/

basaltbcr2.html.

Wooden, J. L., Czamanske, G. K., Fedorenko, V. A., Arndt, N. T.,

Chauvel, C., Bouse, R. M., King, B. S. W., Knight, R. J. &

Siems, D. F. (1993). Isotopic and trace-element constraints on

mantle and crustal contributions to Siberian continental flood

basalts, Noril’sk area, Siberia. Geochimica et Cosmochimica Acta 57,

3766–3704.

JOURNAL OF PETROLOGY VOLUME 47 NUMBER 10 OCTOBER 2006

2018

Workman, R. K. & Hart, S. R. (2005). Major and trace element

composition of the depleted MORB mantle (DMM). Earth and

Planetary Science Letters 231, 53–72.

Xiao, L., Xu, Y. G., Mei, H. J., Zheng, Y. F., He, B. & Pirajno, F.

(2004). Distinct mantle sources of low-Ti and high-Ti basalts

from the western Emeishan large igneous province, SW

China: implications for plume–lithosphere interaction. Earth and

Planetary Science Letters 228, 525–546.

Xu, S. J., Yu, H. B. & Wang, R. C. (2002). Sm–Nd and Rb–Sr isotope

dating of the Shaba granulite of western Sichuan and its geological

significance. Geological Journal of China Universities 8, 634–648 (in

Chinese with English abstract).

Xu, Y., Chung, S., Jahn, B. & Wu, G. (2001). Petrologic and

geochemical constraints on the petrogenesis of Permian–Triassic

Emeishan flood basalts in southern China. Lithos 58, 145–168.

Xu, Y., He, B., Chung, S. L., Menzies, M. A. & Frey, F. A. (2004).

Geologic, geochemical, and geophysical consequences of plume

involvement in the Emeishan flood-basalt province. Geology 32,

917–920.

Yin, H. H., Huang, S., Zhang, K., Hansen, H. J., Yang, F. Q.,

Ding, M. & Bie, X. (1992). The effects of volcanism on the

Permo-Triassic mass extinction in South China. In: Sweet, W. C.

(ed.) Permo-Triassic Events in Eastern Tethys. Cambridge: Cambridge

University Press, pp. 146–157.

Zhang, S.-Q., Mahoney, J. J., Mo, X.-X., Ghazi, A. M., Milani, L.,

Crawford, A. J., Guo, T.-Y. & Zhao, Z.-D. (2005). Evidence for a

widespread Tethyan upper mantle with Indian-Ocean-type isotopic

characteristics. Journal of Petrology 46, 828–858.

Zhang, Z. & Wang, F. (2002a). Geochemistry of the two types of basalts

of the Emeishan basaltic province: evidence for mantle plume–

lithosphere interaction. Acta Geologica Sinica 76, 229–238 (in Chinese

with English abstract).

Zhang, Z. & Wang, F. (2002b). Discovery of Permian picritic lavas in the

Emeishan large igneous province. Geological Review 48, 448 (in Chinese).

Zhang, Z. & Wang, F. (2003). Sr-, Nd- and Pb-isotopic characteristics

of the Emeishan basalt province and discussion on their source

region. Journal of China University of Geosciences 28, 431–439 (in Chinese

with English abstract).

Zhang, Z., Wang, F., Hao, Y. & Mahoney, J. J. (2004). Geochemistry of

the picrites and basalts from the Emeishan large igneous basalt

province and constraints on their source region. Acta Geologica Sinica

78, 171–180 (in Chinese with English abstract).

Zhou, M.-F., Malpas, J., Song, X. Y., Kennedy, A. K.,

Robinson, P. T., Sun, M., Lesher, C. M. & Keays, R. R. (2002a).

A temporal link between the Emeishan large igneous province (SW

China) and the end-Guadalupian mass extinction. Earth and Planetary

Science Letters 196, 113–122.

Zhou, M.-F., Yan, D. P., Kennedy, A. K., Li, Y. Q. & Ding, J. (2002b).

SHRIMPU–Pb zircon geochronological and geochemical evidence for

Neoproterozoic arc-magmatism along the western margin of the

Yangtze Block, South China.Earth and Planetary Science Letters 196, 51–67.

Zhou, M.-F., Robinson, P. T., Lesher, C. M., Keays, R. R.,

Zhang, C.-J. & Malpas, J. (2005). Geochemistry, petrogenesis and

metallogenesis of the Panzhihua gabbroic layered intrusion and

associated Fe–Ti–V oxide deposits, Sichuan Province, SW China.

Journal of Petrology 46, 2253–2280.

ZHANG et al. WESTERN EMEISHAN PICRITIC AND BASALT FLOWS

2019