UBC Faculty Research and Publications

Flood basalts from Mt. Capitole in the central Kerguelen Archipelago: insights into the growth of the.. Weis, Dominique; Scoates, James S. 2007

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
[if-you-see-this-DO-NOT-CLICK]
Weis_AGU_2007GC001608.pdf [ 2.53MB ]
[if-you-see-this-DO-NOT-CLICK]
[if-you-see-this-DO-NOT-CLICK]
Metadata
JSON: 1.0075972.json
JSON-LD: 1.0075972+ld.json
RDF/XML (Pretty): 1.0075972.xml
RDF/JSON: 1.0075972+rdf.json
Turtle: 1.0075972+rdf-turtle.txt
N-Triples: 1.0075972+rdf-ntriples.txt
Original Record: 1.0075972 +original-record.json
Full Text
1.0075972.txt
Citation
1.0075972.ris

Full Text

Geochemistry Geophysics Geosystems  3  G  AN ELECTRONIC JOURNAL OF THE EARTH SCIENCES Published by AGU and the Geochemical Society  Article Volume 8, Number 6 12 June 2007 Q06007, doi:10.1029/2007GC001608 ISSN: 1525-2027  Flood basalts from Mt. Capitole in the central Kerguelen Archipelago: Insights into the growth of the archipelago and source components contributing to plume-related volcanism Guangping Xu and Frederick A. Frey Department of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA (gpxu@mit.edu)  Dominique Weis and James S. Scoates Department of Earth and Ocean Sciences, University of British Columbia, 6339 Stores Road, Vancouver, British Columbia, Canada V6T 1Z4  Andre´ Giret Laboratoire de Ge´ologie-Petrologie, Universite´ Jean Monnet, CNRS-UMR 6524, 23 Rue du Docteur Paul Michelon, F-42023 Saint-E´tienne, France  [1] The Kerguelen Archipelago, constructed on the submarine Northern Kerguelen Plateau, is attributed to Cenozoic volcanism arising from the Kerguelen hot spot. Geochemical studies of 325 to 1000 m thick lava sections of the $30 to 25 Ma flood basalt forming the bulk of the archipelago show a temporal change from older tholeiitic basalt to younger slightly alkalic basalt. This compositional transition is expressed in a 630 m lava section at Mt. Capitole where the lava sequence is lowermost tholeiitic basalt overlain by slightly alkalic basalt overlain by plagioclase-rich cumulates that are mixtures of plagioclase-phyric basalt and more evolved magmas. During growth of the archipelago, magma supply from the hot spot was variable and at times sufficiently low to enable extensive crystal fractionation; e.g., at Mt. Capitole and nearby Mt. Tourmente only 10 of 120 lava flows have >6 wt% MgO. On the basis of this study and previous isotopic data for the $34 Ma submarine lavas erupted on the Northern Kerguelen Plateau, other flood basalt sections in the Kerguelen Archipelago, and younger lavas erupted in the archipelago and at Heard Island, there is significant Sr, Nd, Hf, and Pb isotopic heterogeneity that can be explained by two stages of mixing. The first mixing event, best shown by the submarine lavas, is between components that are related to Indian Ocean mid-ocean ridge basalt (MORB) and the Kerguelen hot spot. From $34 Ma to <1 Ma, on average the proportion of the MORB-related component decreased. Subsequently, a second mixing process involved addition of a component with relatively high 87Sr/86Sr (>0.7060) and low 143 Nd/144Nd (<0.5125) and 176Hf/177Hf (<0.2827) and nonradiogenic Pb isotope ratios (<17.9 for 206 Pb/204Pb). We infer that this component was lower continental crust. Components: 21,781 words, 17 figures, 10 tables. Keywords: Kerguelen mantle plume; Kerguelen Archipelago; Mt. Capitole; lower continental crust; Sr; Nd; Hf; Pb isotopic ratios. Index Terms: 1037 Geochemistry: Magma genesis and partial melting (3619); 1038 Geochemistry: Mantle processes (3621); 1065 Geochemistry: Major and trace element geochemistry. Received 13 February 2007; Accepted 15 March 2007; Published 12 June 2007.  Copyright 2007 by the American Geophysical Union  1 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Xu, G., F. A. Frey, D. Weis, J. S. Scoates, and A. Giret (2007), Flood basalts from Mt. Capitole in the central Kerguelen Archipelago: Insights into the growth of the archipelago and source components contributing to plume-related volcanism, Geochem. Geophys. Geosyst., 8, Q06007, doi:10.1029/2007GC001608.  1. Introduction [2] The Kerguelen hot spot has produced 15 to 24 Â 106 km3 of basaltic magma over $120 My [Coffin and Eldholm, 1994; Coffin et al., 2002]. This long volcanic record includes a large igneous province (Kerguelen Plateau-Broken Ridge), a hot spot track (the >5000 km long $82–38 Ma Ninetyeast Ridge), and the recently active islands (Kerguelen Archipelago, McDonald and Heard Islands) [e.g., Wallace et al., 2002]. Determination of spatial and temporal variations in geochemical characteristics of the basalt forming the Kerguelen Plateau, Ninetyeast Ridge and Kerguelen Archipelago are essential for understanding the history of the Kerguelen hot spot. The early, dominantly Cretaceous, volcanic activity of the Kerguelen hot spot is recorded in basalt recovered from the Kerguelen Plateau and Broken-Ridge by the Ocean Drilling Program (Legs 119, 120 and 183). Studies of these drill cores show a complex record of varying magma production rates [Coffin et al., 2002] and changes in the relative proportions of magma source components, including mantle plume, midocean ridge basalt (MORB) and continental-related components [e.g., Mahoney et al., 1995; Frey et al., 2002b; Ingle et al., 2002; Kieffer et al., 2002; Neal et al., 2002; Weis and Frey, 2002; Frey et al., 2003]. [ 3 ] The Cenozoic Kerguelen Archipelago (6500 km2) formed on the Northern Kerguelen Plateau (Figure 1). The archipelago has a history of volcanism from $30 to 0.1 Ma that is interpreted as magmatism resulting from the stem of the Kerguelen mantle plume [e.g., Weis et al., 1993; Nicolaysen et al., 2000]. Unlike the submarine Kerguelen Plateau and Ninetyeast Ridge, the Kerguelen Archipelago is currently a subaerial expression of the Kerguelen hot spot that can be studied in detail. The archipelago is largely, 85% of the surface, formed of flood basalt ranging from 28–29 Ma tholeiitic basalt in the northwest (Mts des Ruches, Fontaine, Bureau and Rabouille`re) to 24–26 Ma alkalic basalt in the east (Mt. Crozier and sections at Ravin Jaune and du Charbon) (Figure 1). A transition from tholeiitic to alkalic volcanism occurs in flood basalt sections from the Plateau Central. For example, at Mt. Tourmente (Figure 1), a 597 m section of lava flows ranges from $26 Ma transitional basalt (i.e., near the tholeiitic-  alkalic boundary line on a total alkalis versus SiO2 plot) in the lower 80% of the section to overlying $25.3 Ma alkalic basalt in the upper 20% of the section. In contrast, at Mt. Marion Dufresne, also in the Plateau Central (Figure 1), the lowermost lavas in a 700 m section are alkalic basalt and the lavas become less alkaline upward in the section [Annell et al., 2007]. If tholeiitic basalt reflects higher magma flux than alkalic basalt, as commonly inferred, the temporal variations in magma flux were different at Mts Tourmente and Marion Dufresne. [4] With the objective of understanding fluctuations in magma flux arising from the Kerguelen hot spot, we studied a 630 m lava section from Mt. Capitole at an intermediate location on the Plateau Central (Figure 1). We find an upward, i.e., decreasing age, change from slightly tholeiitic to slightly alkalic basalt in the Mt. Capitole section, but the uppermost plagioclase-phyric lavas reflect a plagioclase accumulation process similar to that forming plagioclase-phyric to -ultraphyric basalt at Mt. Marion Dufresne [Annell et al., 2007]. The accumulation of plagioclase phenocrysts in subgroups of lavas at Mts Capitole and Marion Dufresne provide further evidence for periods of reduced basaltic magma flux from the hot spot. [5] An important result is that isotopic data for Sr, Nd, Hf and Pb for Mt. Capitole lavas combined with previously published isotopic data for other archipelago lavas can be explained by mixing between three components. First mixing between a component, such as mid-ocean ridge basalt or its source, with relatively low 87 Sr/ 86 Sr, high 143 Nd/144Nd and 176Hf/177Hf and intermediate 206 Pb/204Pb, with a plume-related component with i n t er m edi a te 8 7 Sr/ 8 6 Sr, 1 4 3 Nd/ 1 4 4 Nd and 176 Hf/177Hf and high 206Pb/204Pb, $18.5, followed by addition of a component with high 87Sr/86Sr, low 143Nd/144Nd and 176Hf/177Hf and quite low 206 Pb/204Pb (<18). This last component is isotopically similar to some lower continental crust.  2. Geology [6] Mt. Capitole in the central part of the Kerguelen Archipelago, near the eastern edge of the Cook 2 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 1. Map of the Kerguelen Archipelago [after Yang et al., 1998] showing the major geologic units, the location of studied stratigraphic sections of flood basalt, and Mt. Ross, which is the youngest volcanic edifice in the archipelago. Mt. Capitole (red dot) is in the Plateau Central. Ages for these sections are from Weis et al. [1993, 1998], Nicolaysen et al. [2000], Doucet et al. [2002], and Annell et al. [2004]. Inset is a map showing the Southeast Indian Ocean Ridge (SEIR), the Kerguelen Plateau, forming a Cretaceous large igneous province, and the Cenozoic Kerguelen Archipelago and Heard Islands located on the Northern and Central Kerguelen Plateau, respectively. Filled stars show Kerguelen Plateau drill sites discussed in the text (Site 738, Mahoney et al. [1995]; Site 747, Frey et al. [2002b]; Site 1137, Ingle et al. [2002]; Site 1140, Weis and Frey [2002]).  ice cap (Figure 1), has a NE–SW orientation and is asymmetric with average slopes of 32° for the western flank and 18° for the eastern flank. It is cut by basaltic dikes with east–west orientation. In this region of the glaciated plateau, it is not possible to identify individual volcanic centers. [7] Fifty-five samples from distinct basalt flows were collected on a westward traverse from the summit (sample 93–459) with an altitude of 860 m to the Valle´e des Merveilles, an altitude of 230 m (sample 93–514); intercalated within the basalt flows are sedimentary breccias and conglomerates which indicate temporal breaks in eruption (Figure 2). For example, there is a 4 m thick breccia with angular pebbles of basalt located at 700 m (between samples 93–473 and 93–474), a 0.2 m thick  red bed consisting of basaltic pebbles in a red matrix located at 670 m (between samples 93– 477 and 93–478), and a 1.5 m thick breccia at 565 m (between samples 93–485 and 93–486). No age information is available but we assume that the Mt. Capitole section formed at $25 Ma, i.e., similar to the age of lavas from Mts Tourmente [Nicolaysen et al., 2000] and Marion Dufresne [Annell et al., 2004].  3. Analytical Techniques [8] Ten samples, mostly plagioclase-phyric, were chosen for analyses of phenocrysts, xenocrysts and amphibole inclusions within plagioclase (Table 1). Olivine, plagioclase, clinopyroxene and amphibole 3 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 2. Location of studied samples (black horizons with sample numbers) in the Mt. Capitole section. The base is at 69°1705100E and 49°1903200S, and the summit is at 69°1900000E and 49°1905100S. The vertical exaggeration is a factor of 5. The open regions indicate no outcrop or extremely weathered rocks. Samples with elevation greater than 690 m form the Upper Transitional Group, which is defined on the basis of petrography (Table 1) and lava compositions. Sample 93– 491, lower in the section, has the characteristics of this group. The Low-Silica Group lavas from 660 m to 560 m have relatively low SiO2/Fe2O3* (Fe2O3* is total iron). All other lavas belong to the Lower Transitional Group. Also shown are layers of sedimentary breccias and conglomerates, such as a 4 m thick breccia with angular pebbles of basalt at 700 m, a 0.2 m thick red matrix containing basaltic pebbles at 670 m, and a 1.5 m thick breccia at 565 m.  were analyzed with the 4-spectrometer JEOL 733 microprobe at Massachusetts Institute of Technology, using 15 kV accelerating voltage, 10 nA beam current and a beam size of 1 mm (10 mm for plagioclase). The counting time was 40 seconds for all elements except for Ca and Al (30 seconds) and Na (5 seconds) in plagioclase; Na was counted for 15 seconds for pyroxene and amphibole. Analyses of plagioclase, pyroxene, olivine and amphibole are in Tables 2a–2d. [9] For whole rock analyses, samples were abraded with sand-paper to remove surficial alteration features and contaminants introduced by sawing. Then they were coarse-crushed in a hydraulic piston crusher and reduced to powder in an agate shatterbox. Major element and some trace element (such  as Cr, Ni and V) concentrations were determined by X-ray fluorescence spectrometry at the University of Massachusetts, Amherst (Tables 3 and 4). Major element compositions are reported as the mean of duplicate analyses and loss on ignition (LOI) is the weight loss after heating 10 min at 1020°C using Pt-Au crucibles. Estimates of accuracy and precision were discussed by Rhodes [1996]. Most trace element abundances (Table 4) were determined at MIT by inductively coupled plasma mass spectrometry using a Fisons VG Plasmaquad 2 + S with both internal and external drift monitors. The relative standard deviation for all trace elements determined in BHVO-2 (15 analyses, Table 4) is less than 3% [Huang and Frey, 2003]. Scandium was determined by instrumental neutron activation analysis in 21 4 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 1. Petrographic Characteristics of Mt. Capitole Samples Group  Sample  Height, ma  Phenocryst/Xenocryst,b volume%  Upper Transitional Group (UTG)d  93 – 459c 93 – 460 93 – 461 93 – 462 93 – 463 93 – 464 93 – 465 93 – 467 93 – 468 93 – 469 93 – 470 93 – 471 93 – 472 93 – 473 93 – 474  860 840 840 825 815 810 800 780 760 750 740 735 730 715 690  40% plagioclase 35% plagioclase 25% plagioclase 20% plagioclase and <1% augite 3% plagioclase <1% plagioclase 25% plagioclase none 2% plagioclase 5% plagioclase 15% plagioclase 10% plagioclase 2% olivine, 1% augite and 5% plagioclase 40% plagioclase 15% plagioclase  Lower Transitional Groupd  93 – 475 93 – 476 93 – 477  690 685 680  2% plagioclase 2% plagioclase 10% plagioclase  Low-Silica Groupd  93 – 478 93 – 479 93 – 480 93 – 481 93 – 482 93 – 483 93 – 484 93 – 485 93 – 486  660 640 630 610 600 590 580 570 560  3% plagioclase None None <1% plagioclase <1% plagioclase <1% plagioclase 12% plagioclase and 3% augite  Lower Transitional Groupd  93 – 487 93 – 488 93 – 489 93 – 490  560 550 540 540  <1% plagioclase 10% plagioclase 15% plagioclase <1% plagioclase and augite  UTGd  93 – 491  540  25% plagioclase and 15% augite  Lower Transitional Groupd  93 – 492 93 – 493 93 – 494 93 – 495 93 – 496 93 – 497 93 – 498 93 – 499 93 – 500 93 – 501 93 – 502 93 – 503 93 – 504 93 – 505 93 – 506 93 – 507 93 – 508 93 – 509 93 – 510 93 – 511 93 – 512 93 – 513 93 – 514  540 530 520 510 505 490 480 470 465 455 440 435 430 420 410 400 390 380 350 310 270 250 230  15% plagioclase <1% plagioclase <1% plagioclase <1% plagioclase <1% plagioclase <1% plagioclase <1% plagioclase <1% plagioclase <1% plagioclase <1% plagioclase none <1% plagioclase <1% plagioclase <1% plagioclase <1% plagioclase <1% plagioclase none none <1% plagioclase <1% plagioclase none none <1% plagioclase  18% plagioclase and 2% augite  and augite and augite  and augite and augite  and augite  5 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  samples, following the procedures of Ila and Frey [2000] (Table 4). [10] Eighteen relatively fresh samples with minimum alteration were chosen for Sr, Nd, Hf and Pb isotopic analyses at the Pacific Centre for Isotopic and Geochemical Research at the University of British Columbia (UBC). Prior to isotopic analysis of Sr, Nd and Pb the samples were leached repeatedly in an ultrasonic bath with 6N HCl following the procedure described by Weis et al. [2005]. Analysis of leached and unleached aliquots for sample 93–465 shows that leaching resulted in residues with slightly higher 143Nd/144Nd and distinctly lower 87Sr/86Sr and Pb isotopic ratios (Table 5). At UBC Sr and Nd isotopic ratios were determined using a thermal ionization mass spectrometer (Triton) and Pb isotopic ratios were determined using a multiple-collector ICP-MS (Nu021) [Weis et al., 2005, 2006]. Normalization procedures and data for standards are in the footnotes for Table 5. [11] About 200 mg of unleached rock powder was dissolved for Hf isotopic analyses, following the procedure of Blichert-Toft et al. [1997]. The Hf isotopic compositions were measured by MC-ICPMS (Nu021) at UBC. The 176Hf/177Hf ratios are normalized to the Hf JMC 475 in-house standard value of 0.282160 [Blichert-Toft et al., 1997]. External reproducibility based on three duplicates is within in-run uncertainties, i.e., <6 Â 10À6 (Table 5). [12] For Sr and Pb isotopic analyses, plagioclase grains with relatively few inclusions were picked from two samples, 93–459 and 93–471, using a binocular microscope. Leaching procedures followed those of Housh and Bowring [1991]: grains were leached using 7N HNO3 for 30 min on a hotplate ($125°C); the residue was rinsed with Milli-Q H2O, leached by 6N HCl on a hotplate for 30 min and rinsed with Milli-Q H2O; this residue was leached with 5% HF + 0.5N HBr (8:1) for 10 min on a hotplate stirring every 2 min followed by rinsing twice with Milli-Q H2O. This last step was repeated until the sample was white with no visible black inclusions. The final residue was dissolved by concentrated HF and 7N HNO3. An  aliquot was taken for ICP-MS analyses to determine the parent/daughter abundance ratios (Table 6). The remaining aliquots were passed through 120 mL Pb and 50 mL Sr columns and analyzed by a thermal ionization multicollector mass spectrometer (Micromass Isoprobe-T) at MIT using dynamic mode for Sr and static mode for Pb (Table 5).  4. Results 4.1. Petrography [13] The textures of Mt. Capitole lavas range from aphyric to moderately phyric (Table 1), typically with a fine-grained groundmass of plagioclase, clinopyroxene, olivine, opaque minerals and devitrified, altered brown glass. Sample 93–472 is an exception; it has an intergranular texture with a coarse-grained groundmass of plagioclase and clinopyroxene and is altered (loss on ignition = 4.4 wt%, Table 3). Most samples (38) contain less than 5 vol% phenocrysts (>0.7 mm), and 8 samples are aphyric. Most of these aphyric to slightly phyric lavas are found in the lower part of the section. In contrast, 16 samples contain abundant phenocrysts or xenocrysts (!10 vol% and up to 40 vol%), dominantly plagioclase with sparse clinopyroxene; olivine phenocrysts occur only in sample 93–472; nine of these 16 samples are from the uppermost 170 m (Table 1). Most of these plagioclase grains are 0.7–3 mm in width, but a few laths are up to 7 mm in length; some grains are resorbed (Figure 3a). [14] The phenocryst assemblages in each of the three studied sections in the Plateau Central (Figure 1) are quite different. Lavas from Mt. Tourmente are largely aphyric; i.e., 62 of 64 samples have less than 5 vol% phenocrysts [Frey et al., 2002a]. Lavas from the lowermost 300 m of the Mt. Marion Dufresne section are also dominantly aphyric but with decreasing age plagioclasephyric, up to 60 vol%, lavas are abundant [Annell et al., 2007]. This upward succession from aphyric to plagioclase-phyric occurs at Mt. Capitole and Mt. Marion Dufresne. However, the latter section is unique in that olivine-phyric, up to 20 vol%, lavas  Notes to Table 1: a Meters above sea level. b Phenocrysts/xenocrysts are crystals with size ! 0.7 mm. Volume proportions estimated from observation of thin sections using a polarizing microscope. c Sample numbers in bold indicate samples with mineral analyses determined by electron microprobe. d On the basis of petrography and whole rock composition, the Mt. Capitole lavas are divided into three groups that correlate with their stratigraphic positions. See the text for details.  6 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 2a. Plagioclase Compositions of Mt. Capitole Lavasa Type 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460  plag1 plag1 plag2 plag2 plag3 plag3 plag4 plag4 plag5 plag5 plag5 plag5 plag6 plag6 plag7 plag7 plag7 plag7 plag7 plag8 plag8 plag8 plag9 plag9 plag10 plag11 plag11 plag12 plag13 plag1 plag1 plag1 plag1 plag1 plag1 plag1 plag1 plag2 plag2 plag2 plag2 plag2 plag2 plag3 plag3 plag3 plag4 plag4 plag4 plag4 plag5 plag5 plag5 plag5 plag6 plag6 plag6 plag7 plag7 plag8 plag8  core rim core rim core rim core rim core mid-core rim rim core rim core core core rim rim core rim rim core rim core core rim groundmass groundmass core core mid-core rim rim rim rim rim core core core core rim rim core core rim core core rim rim core core rim rim core rim rim core rim core rim  SiO2  Al2O3 FeO MgO  CaO  Na2O K2O Total  An  Ab  Or  Upper 50.66 49.36 49.65 52.05 49.22 48.90 48.04 48.66 51.16 49.35 49.21 49.28 49.31 49.02 48.89 48.81 48.22 48.86 48.66 48.60 54.91 48.28 48.87 49.31 49.10 48.81 49.01 55.52 52.69 48.00 49.08 47.93 52.52 48.52 48.00 52.10 47.50 50.77 48.67 49.38 48.92 53.07 49.13 50.58 49.37 52.39 50.05 49.28 49.56 52.72 50.60 50.24 48.92 49.60 50.24 48.19 49.31 48.99 49.14 47.44 49.07  Transitional Group 31.12 0.59 0.18 32.03 0.66 0.15 31.65 0.57 0.14 30.13 0.76 0.09 32.36 0.57 0.15 32.50 0.69 0.08 33.02 0.56 0.13 31.58 0.63 0.14 30.56 0.59 0.13 32.03 0.59 0.15 32.23 0.59 0.11 31.93 0.57 0.13 31.87 0.54 0.15 31.67 0.58 0.14 31.90 0.56 0.16 32.45 0.59 0.17 32.62 0.53 0.16 32.35 0.59 0.13 32.22 0.60 0.10 32.55 0.52 0.15 27.88 0.96 0.09 32.94 0.62 0.13 32.61 0.56 0.16 32.02 0.82 0.15 31.79 0.52 0.15 32.51 0.57 0.13 32.06 0.68 0.11 27.84 0.84 0.05 29.70 1.09 0.16 33.22 0.58 0.13 32.88 0.63 0.17 33.58 0.54 0.14 30.08 0.61 0.15 33.00 0.52 0.13 33.62 0.53 0.14 30.51 0.67 0.14 33.66 0.51 0.14 31.55 0.53 0.16 32.99 0.58 0.14 32.50 0.60 0.14 32.68 0.61 0.15 30.02 0.66 0.13 32.71 0.59 0.11 31.41 0.61 0.13 32.68 0.60 0.14 30.36 0.62 0.16 32.20 0.58 0.17 32.80 0.61 0.15 32.69 0.64 0.12 29.98 0.74 0.13 31.85 0.65 0.14 31.85 0.55 0.15 32.90 0.67 0.15 32.32 0.55 0.13 31.44 0.55 0.23 33.54 0.63 0.14 32.53 0.60 0.18 32.59 0.61 0.17 32.75 0.53 0.17 33.96 0.48 0.14 33.04 0.52 0.14  14.30 15.03 14.88 12.84 15.36 15.55 15.18 14.87 13.69 14.98 15.16 15.07 14.84 15.09 15.13 15.52 15.97 15.26 15.37 15.65 10.34 15.61 15.78 15.03 15.62 15.46 15.16 10.22 12.40 16.37 15.86 16.64 13.10 15.92 16.50 13.34 16.61 14.42 16.24 15.45 15.66 12.74 15.69 14.35 15.36 13.38 15.00 15.55 15.69 12.75 14.46 14.64 15.94 15.23 14.83 16.50 15.28 15.75 15.90 17.44 15.99  3.32 2.57 2.78 3.85 2.68 2.35 2.34 2.87 3.64 2.63 2.56 2.66 2.82 2.83 2.90 2.57 2.37 2.48 2.66 2.61 5.52 2.46 2.67 3.08 2.58 2.64 2.85 5.53 4.14 2.03 2.42 1.67 3.75 2.37 2.07 3.64 2.15 3.26 2.08 2.57 2.45 3.92 2.56 3.46 2.99 3.82 3.02 2.52 2.49 3.91 3.23 2.89 2.42 2.67 3.00 2.01 2.72 2.27 2.30 1.72 2.17  69.9 75.9 74.3 64.0 75.6 78.1 77.8 73.7 66.8 75.4 76.1 75.2 73.9 74.1 73.8 76.5 78.5 76.7 75.6 76.4 49.6 77.5 76.1 72.5 76.6 76.0 74.2 49.1 61.4 81.3 77.9 84.2 65.1 78.4 81.1 66.3 80.7 70.5 80.8 76.4 77.5 63.4 76.8 69.1 73.5 65.3 72.8 76.9 77.3 63.4 70.6 73.1 78.0 75.3 72.6 81.5 75.2 78.8 78.8 84.5 79.8  29.4 23.5 25.1 34.7 23.9 21.4 21.7 25.8 32.1 24.0 23.3 24.0 25.4 25.2 25.6 23.0 21.1 22.6 23.7 23.0 47.9 22.1 23.3 26.8 22.9 23.5 25.2 48.1 37.1 18.2 21.5 15.3 33.7 21.1 18.4 32.7 18.9 28.8 18.7 23.0 21.9 35.3 22.7 30.1 25.9 33.8 26.5 22.5 22.2 35.2 28.6 26.2 21.4 23.9 26.6 18.0 24.2 20.6 20.6 15.1 19.6  0.8 0.7 0.6 1.3 0.5 0.6 0.5 0.5 1.0 0.7 0.6 0.8 0.7 0.7 0.7 0.5 0.5 0.8 0.7 0.6 2.5 0.5 0.5 0.7 0.5 0.5 0.6 2.8 1.5 0.5 0.6 0.4 1.2 0.4 0.5 1.0 0.4 0.7 0.5 0.6 0.5 1.3 0.5 0.7 0.6 1.0 0.7 0.6 0.5 1.4 0.8 0.7 0.6 0.7 0.8 0.5 0.6 0.6 0.7 0.4 0.6  0.13 0.11 0.10 0.22 0.09 0.09 0.09 0.09 0.17 0.11 0.11 0.13 0.12 0.12 0.11 0.09 0.08 0.13 0.11 0.11 0.43 0.08 0.09 0.12 0.08 0.09 0.11 0.48 0.25 0.08 0.09 0.07 0.20 0.07 0.08 0.18 0.06 0.11 0.08 0.09 0.09 0.22 0.09 0.13 0.10 0.17 0.11 0.10 0.09 0.23 0.14 0.12 0.10 0.12 0.14 0.08 0.10 0.11 0.11 0.06 0.10  100.3 99.9 99.8 99.9 100.4 100.2 99.3 98.9 100.0 99.8 100.0 99.8 99.7 99.5 99.7 100.2 99.9 99.8 99.7 100.2 100.1 100.1 100.7 100.5 99.8 100.2 100.0 100.5 100.4 100.4 101.1 100.6 100.4 100.5 100.9 100.6 100.6 100.8 100.8 100.8 100.6 100.8 100.9 100.7 101.2 100.9 101.1 101.0 101.3 100.5 101.1 100.4 101.1 100.6 100.4 101.1 100.7 100.5 100.9 101.3 101.0  7 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 2a. (continued)  93 – 460 93 – 460 93 – 462 93 – 462 93 – 462 93 – 462 93 – 462 93 – 462 93 – 462 93 – 462 93 – 462 93 – 462 93 – 462 93 – 462 93 – 462 93 – 462 93 – 462 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465  plag8 plag9 plag1 plag1 plag2 plag2 plag2 plag3 plag3 plag4 plag4 plag5 plag5 plag6 plag6 plag7 plag7 plag1 plag1 plag1 plag2 plag2 plag2 plag3 plag3 plag4 plag4 plag4 plag5 plag5 plag5 plag6 plag6 plag6 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag7 plag8 plag8 plag9 plag9 plag9  Type  SiO2  Al2O3 FeO MgO  CaO  Na2O K2O Total  An  Ab  rim groundmass core rim core rim rim core rim core rim core rim core rim core rim core rim rim core rim rim core rim core rim rim core rim rim core rim rim core core rim around olivine inclusion core core core core core core rim rim rim around amphibole inclusion rim around amphibole inclusion rim around amphibole inclusion rim around amphibole inclusion rim around amphibole inclusion rim around amphibole inclusion rim around olivine inclusion rim around pyroxene inclusion rim around pyroxene inclusion rim around pyroxene inclusion rim around pyroxene inclusion rim around pyroxene inclusion core rim core rim rim  52.73 52.65 49.14 48.27 49.10 49.07 54.70 49.12 53.73 48.26 55.69 48.67 49.83 50.59 58.41 53.10 62.35 47.65 48.77 48.71 48.45 48.78 48.62 48.32 48.35 48.99 47.98 48.58 48.46 48.32 48.71 47.66 53.13 47.57 48.13 48.33 54.43 48.50 65.53 58.76 49.41 49.50 48.91 47.86 49.11 53.91 56.49 57.53 57.06 52.80 56.15 56.23 66.95 59.19 54.70 56.89 58.11 48.62 53.10 49.85 48.54 47.90  29.78 30.08 32.01 32.54 32.14 33.10 28.25 32.37 29.06 32.90 27.01 32.45 31.75 31.78 26.02 29.65 22.62 33.00 32.14 32.11 32.56 32.28 32.60 33.00 32.61 31.64 32.92 32.48 32.49 32.38 32.29 32.72 29.25 32.64 32.48 33.22 28.64 32.93 20.90 25.91 32.15 32.73 33.33 32.90 31.78 29.04 27.42 26.68 27.45 30.07 27.69 27.82 20.16 26.03 28.18 27.70 27.00 33.14 29.38 31.39 32.35 32.43  12.63 12.85 15.28 16.02 15.42 15.79 11.05 15.53 11.82 16.05 9.64 15.53 15.25 14.65 7.96 12.46 4.09 16.41 15.40 15.49 15.99 15.50 15.95 16.19 16.12 15.09 15.94 15.68 15.63 15.79 15.57 16.16 12.19 16.01 15.85 16.18 11.37 15.90 2.10 7.93 15.12 15.66 16.22 16.29 15.18 12.01 9.95 9.16 9.66 13.23 10.22 10.25 1.08 7.48 10.47 9.56 8.92 15.80 12.37 14.61 15.56 15.75  3.78 3.97 2.67 2.55 2.73 2.35 4.99 2.44 5.18 2.08 5.54 2.60 2.89 3.12 7.13 4.29 7.76 2.17 2.82 2.75 2.16 2.41 2.20 2.16 2.31 2.73 2.34 2.43 2.88 2.34 2.78 2.27 4.40 2.37 2.58 2.16 5.12 2.13 8.25 6.96 2.65 2.38 2.03 2.23 2.69 4.34 5.60 5.93 6.03 4.11 5.58 5.17 7.58 6.80 5.00 5.44 5.97 2.51 4.65 3.24 2.58 2.58  63.9 63.3 75.6 77.4 75.4 78.4 53.9 77.4 54.8 80.6 47.7 76.3 74.0 71.7 36.8 60.8 20.2 80.4 74.6 75.1 80.0 77.6 79.7 80.3 79.1 74.9 78.7 77.7 74.7 78.6 75.2 79.5 59.7 78.6 77.0 80.2 54.2 80.1 9.8 37.3 75.3 77.8 81.0 79.9 75.4 59.5 48.2 44.2 45.5 62.9 48.9 51.1 5.3 36.2 52.4 47.8 43.8 77.0 58.5 71.1 76.6 76.9  34.6 1.6 35.4 1.2 23.9 0.5 22.3 0.3 24.1 0.5 21.1 0.5 44.1 2.0 22.0 0.6 43.5 1.7 18.9 0.5 49.5 2.8 23.1 0.6 25.4 0.7 27.6 0.6 59.6 3.6 37.9 1.3 69.5 10.3 19.2 0.4 24.7 0.7 24.1 0.8 19.6 0.4 21.9 0.5 19.9 0.4 19.4 0.4 20.5 0.3 24.6 0.5 20.9 0.4 21.8 0.5 24.9 0.4 21.1 0.3 24.3 0.5 20.2 0.3 39.0 1.4 21.1 0.4 22.6 0.4 19.3 0.5 44.1 1.7 19.4 0.5 70.0 20.2 59.2 3.6 23.9 0.7 21.4 0.7 18.3 0.6 19.8 0.3 24.1 0.5 38.8 1.7 49.1 2.6 51.8 4.0 51.4 3.2 35.4 1.7 48.3 2.9 46.7 2.2 67.0 27.7 59.6 4.2 45.3 2.2 49.3 2.9 53.0 3.2 22.2 0.8 39.8 1.7 28.5 0.4 23.0 0.5 22.8 0.4  0.68 0.77 0.59 0.64 0.60 0.60 0.64 0.53 0.76 0.62 1.56 0.58 0.63 0.63 0.79 0.96 1.34 0.55 0.71 0.75 0.58 0.62 0.65 0.59 0.65 0.58 0.62 0.62 0.62 0.59 0.66 0.51 0.73 0.70 0.59 0.72 1.00 0.74 0.46 0.49 0.70 0.54 0.64 0.58 0.60 0.96 0.55 0.39 0.46 0.64 0.45 0.83 0.47 0.60 1.14 0.69 0.61 0.62 0.98 0.57 0.56 0.57  0.15 0.16 0.15 0.13 0.13 0.13 0.10 0.14 0.09 0.13 0.33 0.14 0.17 0.15 0.03 0.16 0.29 0.17 0.12 0.11 0.15 0.13 0.14 0.14 0.13 0.17 0.12 0.12 0.16 0.17 0.12 0.13 0.11 0.12 0.16 0.08 0.12 0.06 0.02 0.04 0.11 0.16 0.15 0.16 0.16 0.28 0.15 0.17 0.07 0.22 0.16 0.14 0.03 0.09 0.24 0.11 0.10 0.20 0.13 0.17 0.16 0.15  0.26 0.21 0.08 0.05 0.08 0.08 0.34 0.10 0.30 0.08 0.47 0.10 0.12 0.11 0.66 0.23 1.74 0.07 0.13 0.14 0.07 0.09 0.06 0.06 0.06 0.09 0.06 0.09 0.08 0.05 0.09 0.05 0.23 0.06 0.07 0.08 0.29 0.08 3.62 0.64 0.12 0.13 0.11 0.06 0.08 0.29 0.45 0.69 0.56 0.29 0.50 0.37 4.76 0.73 0.38 0.49 0.55 0.14 0.29 0.08 0.08 0.06  100.0 100.7 99.9 100.2 100.2 101.1 100.1 100.2 100.9 100.1 100.2 100.1 100.6 101.0 101.0 100.9 100.2 100.0 100.1 100.1 100.0 99.8 100.2 100.5 100.2 99.3 100.0 100.0 100.3 99.6 100.2 99.5 100.0 99.5 99.9 100.8 101.0 100.3 100.9 100.7 100.3 101.1 101.4 100.1 99.6 100.8 100.6 100.6 101.3 101.4 100.8 100.8 101.1 100.9 100.1 100.9 101.3 101.0 100.9 99.9 99.8 99.5  Or  8 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 2a. (continued) Type 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 472 93 – 472 93 – 472 93 – 473 93 – 473 93 – 473 93 – 473 93 – 473 93 – 473 93 – 473 93 – 474 93 – 474 93 – 474 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491  plag9 plag10 plag10 plag11 plag11 plag12 plag12 plag12 plag12 plag12 plag13 plag13 plag14 plag1 plag1 plag1 plag1 plag1 plag1 plag1 plag1 plag1 plag1 plag1 plag1 plag1 plag1 plag1 plag2 plag3 plag3 plag4 plag5 plag5 plag6 plag7 plag7 plag7 plag7 plag7 plag7 plag8 plag9 plag10 plag1 plag1 plag2 plag1 plag1 plag2 plag2 plag3 plag4 plag4 plag1 plag1 plag2 plag1 plag1 plag1 plag2 plag2  rim core rim core rim core rim rim rim rim core rim groundmass core core core core core core core rim rim rim rim rim rim rim rim core core rim core core rim rim core core core rim rim rim core core groundmass core rim groundmass core rim core rim core core rim core rim core core rim rim core rim  SiO2  Al2O3 FeO MgO  CaO  Na2O K2O Total  An  Ab  Or  48.00 47.91 47.60 47.88 48.02 48.72 48.38 53.51 47.79 54.97 48.77 48.90 59.49 50.62 50.02 49.89 50.49 48.34 49.72 48.83 48.86 49.27 48.45 48.43 48.41 48.67 48.82 49.20 48.14 49.30 47.98 48.34 48.63 48.72 54.43 49.41 49.10 48.51 48.53 50.94 49.17 49.12 49.49 53.07 50.25 50.78 51.10 50.37 50.85 50.18 48.84 51.01 48.44 49.80 51.13 51.83 53.92 50.19 54.69 52.85 51.68 53.23  32.75 33.04 32.62 32.87 32.66 32.71 32.68 29.40 33.03 28.86 33.06 33.33 24.67 31.65 32.10 32.09 31.66 33.14 32.20 32.71 32.77 32.52 33.26 33.14 33.22 32.86 32.93 32.91 33.36 32.32 33.28 33.44 32.78 32.96 29.09 32.38 32.57 33.20 33.09 31.42 32.80 32.72 32.78 29.20 31.38 31.25 30.67 31.83 31.42 31.70 32.62 30.97 32.91 31.82 31.08 30.55 28.91 31.17 28.20 29.51 30.42 29.34  15.97 16.05 16.00 16.02 16.04 15.96 16.06 12.07 16.22 11.30 16.29 16.11 7.02 14.96 15.06 15.24 14.68 16.42 15.30 15.92 15.76 15.79 16.26 16.15 16.47 16.02 15.82 15.82 16.50 15.58 16.37 16.27 16.02 15.86 11.58 15.61 15.46 16.22 15.91 14.40 15.54 15.49 15.58 12.15 14.32 14.26 13.82 14.86 14.19 14.63 15.75 14.07 16.05 14.88 13.87 13.65 11.98 14.29 10.92 12.62 13.44 12.14  2.37 2.27 2.55 2.32 2.25 2.73 2.23 4.52 1.97 4.96 2.39 2.19 6.92 2.80 2.83 2.56 2.96 1.84 2.76 2.49 2.37 2.54 2.26 2.10 2.06 2.34 2.33 2.51 1.94 2.58 1.78 1.92 2.15 2.15 4.66 2.49 2.63 2.27 2.23 3.22 2.44 2.61 2.54 4.14 3.23 3.34 3.53 2.94 3.59 2.97 2.73 3.66 2.39 3.17 3.46 3.98 4.60 2.94 4.98 4.20 3.72 4.19  78.5 79.4 77.3 79.0 79.4 76.0 79.6 58.7 81.8 54.7 78.8 80.0 33.7 74.2 74.2 76.2 72.9 82.8 75.0 77.6 78.3 77.1 79.6 80.6 81.2 78.8 78.6 77.3 82.2 76.6 83.3 82.1 80.2 80.0 57.1 77.2 76.1 79.5 79.4 70.7 77.4 76.3 76.9 60.9 70.6 69.8 67.9 73.1 68.1 72.6 75.7 67.5 78.5 71.7 68.3 64.7 58.1 72.4 53.8 61.8 66.1 60.7  21.1 20.3 22.3 20.7 20.2 23.5 20.0 39.8 18.0 43.5 20.9 19.7 60.2 25.2 25.2 23.2 26.6 16.8 24.5 22.0 21.3 22.5 20.1 19.0 18.4 20.8 21.0 22.2 17.5 22.9 16.4 17.6 19.4 19.6 41.6 22.3 23.5 20.1 20.1 28.6 22.0 23.2 22.7 37.5 28.8 29.6 31.4 26.2 31.2 26.7 23.7 31.8 21.1 27.7 30.8 34.1 40.4 27.0 44.4 37.2 33.0 38.0  0.4 0.3 0.4 0.3 0.4 0.4 0.5 1.4 0.2 1.8 0.3 0.3 6.1 0.6 0.6 0.5 0.6 0.5 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.5 0.4 0.5 0.3 0.3 0.4 0.4 1.3 0.5 0.4 0.3 0.5 0.6 0.5 0.5 0.4 1.6 0.5 0.6 0.7 0.7 0.8 0.7 0.5 0.7 0.4 0.6 0.9 1.2 1.5 0.6 1.9 1.0 0.9 1.3  0.62 0.54 0.52 0.52 0.74 0.54 0.52 0.69 0.54 0.65 0.52 0.53 0.79 0.58 0.62 0.57 0.56 0.48 0.60 0.54 0.53 0.50 0.55 0.56 0.57 0.60 0.59 0.58 0.53 0.56 0.59 0.58 0.54 0.61 0.54 0.61 0.53 0.54 0.52 0.55 0.58 0.58 0.49 0.90 0.45 0.50 0.47 0.51 0.54 0.52 0.53 0.52 0.46 0.55 0.62 0.74 0.62 0.67 0.73 0.58 0.62 0.58  0.12 0.16 0.13 0.15 0.11 0.15 0.13 0.12 0.01 0.00 0.00 0.03 0.13 0.18 0.18 0.17 0.18 0.16 0.22 0.19 0.12 0.17 0.17 0.15 0.16 0.15 0.17 0.18 0.17 0.17 0.12 0.14 0.16 0.14 0.11 0.21 0.20 0.21 0.23 0.23 0.21 0.24 0.20 0.07 0.17 0.14 0.20 0.16 0.16 0.17 0.17 0.19 0.13 0.14 0.20 0.16 0.19 0.19 0.11 0.17 0.20 0.14  0.07 0.05 0.07 0.05 0.07 0.08 0.08 0.25 0.03 0.31 0.05 0.05 1.06 0.10 0.10 0.09 0.10 0.08 0.08 0.08 0.07 0.07 0.06 0.07 0.07 0.07 0.07 0.08 0.06 0.08 0.05 0.05 0.06 0.07 0.22 0.09 0.07 0.06 0.08 0.11 0.09 0.08 0.07 0.27 0.09 0.10 0.12 0.12 0.13 0.11 0.09 0.13 0.07 0.10 0.16 0.22 0.26 0.11 0.32 0.18 0.15 0.22  99.9 100.0 99.5 99.8 99.9 100.9 100.1 100.6 99.6 101.1 101.1 101.1 100.1 100.9 100.9 100.6 100.6 100.4 100.9 100.8 100.5 100.9 101.0 100.6 101.0 100.7 100.7 101.3 100.7 100.6 100.2 100.7 100.4 100.5 100.6 100.8 100.6 101.0 100.6 100.9 100.8 100.8 101.2 99.8 99.9 100.4 99.9 100.8 100.9 100.3 100.7 100.5 100.5 100.5 100.5 101.1 100.5 99.6 100.0 100.1 100.2 99.8  9 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 2a. (continued) Type  SiO2  Al2O3 FeO MgO  CaO  Na2O K2O Total  An  Ab  Or  31.07 28.81 31.33 30.07 29.46 28.83 28.69 29.60 29.42 30.92 30.53 31.14 29.62 29.41 29.77 29.94 32.50 30.43 29.48  93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491  plag3 plag3 plag4 plag4 plag4 plag5 plag5 plag6 plag6 plag7 plag7 plag8 plag8 plag9 plag10 plag11 plag12 plag12 plag12  core rim core rim rim core rim rim core core rim core rim core core plag inclusion in plag11 core rim rim around pyroxene inclusion  50.54 54.12 50.54 52.52 53.18 54.15 53.86 52.66 53.07 51.13 51.19 50.44 53.34 53.34 53.07 52.94 48.86 52.10 53.38  93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486  plag1 plag1 plag1 plag1 plag1 plag2 plag2 plag2 plag2 plag2 plag2 plag2 plag3  core core core core rim core core rim rim rim rim rim core  Low-Silica 53.50 29.20 53.30 29.39 53.56 29.38 52.98 29.49 55.42 27.82 53.70 28.77 50.04 31.71 51.93 30.32 56.01 27.86 52.80 29.89 51.98 30.48 54.59 28.25 51.19 30.85  0.66 0.66 0.64 0.66 0.62 0.58 0.86 0.63 0.58 0.64 0.70 0.67 0.55 0.61 0.63 0.71 0.59 0.56 0.68  0.17 0.12 0.16 0.16 0.16 0.14 0.31 0.16 0.17 0.16 0.18 0.16 0.16 0.14 0.16 0.14 0.13 0.19 0.19  14.19 11.47 14.36 12.90 12.27 11.56 11.53 12.53 12.34 13.88 13.75 14.38 12.27 12.32 12.49 12.75 15.57 13.26 12.29  3.16 4.84 3.16 4.21 4.27 4.76 4.59 4.26 4.14 3.33 3.60 3.04 4.46 4.51 4.46 3.98 2.51 3.88 4.20  0.15 0.28 0.10 0.16 0.21 0.25 0.24 0.21 0.20 0.13 0.14 0.10 0.21 0.19 0.20 0.20 0.08 0.15 0.32  99.9 100.3 100.3 100.7 100.2 100.3 100.1 100.0 99.9 100.2 100.1 99.9 100.6 100.5 100.8 100.7 100.3 100.6 100.6  70.7 55.8 71.1 62.3 60.6 56.5 57.3 61.1 61.5 69.2 67.3 71.9 59.6 59.5 60.0 63.2 77.0 64.8 60.6  28.5 42.6 28.3 36.8 38.2 42.1 41.3 37.7 37.3 30.0 31.9 27.5 39.2 39.4 38.8 35.7 22.5 34.4 37.5  0.9 1.6 0.6 0.9 1.2 1.4 1.4 1.2 1.2 0.8 0.8 0.6 1.2 1.1 1.1 1.2 0.5 0.8 1.9  Group 0.54 0.50 0.63 0.50 0.76 0.55 0.57 0.61 0.68 0.70 0.62 0.67 0.56  0.13 0.12 0.11 0.13 0.10 0.17 0.14 0.12 0.09 0.12 0.14 0.10 0.13  11.92 12.34 12.05 12.33 10.55 11.78 14.79 13.09 10.42 12.62 13.22 11.10 13.74  4.56 4.20 4.61 4.21 4.99 4.81 2.95 3.68 5.41 4.19 4.12 5.34 3.48  0.22 0.20 0.20 0.22 0.37 0.22 0.11 0.16 0.35 0.21 0.17 0.31 0.14  100.1 100.1 100.5 99.9 100.0 100.0 100.3 99.9 100.8 100.5 100.7 100.4 100.1  58.4 61.2 58.4 61.0 52.7 56.8 73.0 65.7 50.5 61.7 63.3 52.5 68.0  40.4 37.7 40.5 37.7 45.1 42.0 26.3 33.4 47.4 37.1 35.7 45.7 31.2  1.3 1.2 1.1 1.3 2.2 1.2 0.7 1.0 2.0 1.2 1.0 1.7 0.8  a  Compositions are in wt% and were determined by electron microprobe at MIT.  dominate the uppermost 400 m [Annell et al., 2007]. In contrast, only one olivine-phyric ($2 vol%) lava occurs in the Mt. Capitole section (Table 1).  4.2. Mineral Compositions 4.2.1. Plagioclase [15] Plagioclase is the most abundant phenocryst/ xenocryst in Mt. Capitole basalt, especially in the uppermost 170 m of the section which we refer to as the Upper Transitional Group (Table 1). Within this group plagioclase cores range from An85 to An56 (Table 2a and Figure 4); a similarly wide range of plagioclase core compositions, An85 to An63, occurs in the plagioclase-phyric lavas exposed further south on the Plateau Central at Mt. Marion Dufresne (Figure 4). Annell et al. [2007] concluded that the wide range of plagioclase core compositions in individual samples, including cores  >An80, reflect mixing of magmas with distinct population of plagioclase phenocrysts. Consistent with this conclusion, the plagioclase phenocrysts/ xenocrysts in the Upper Transitional Group are texturally distinct from those in the other groups; they are commonly resorbed (Figure 3a) and contain inclusions of olivine, pyroxene, amphibole, Fe-Ti oxides and rare apatite (Figures 3b, 3c, 3d, 3e, and 3f). Rims of large plagioclase grains span a wide compositional range, from sodic-rich compositions (An47) to An83 (Tables 2a–2d). Even more sodic plagioclase (An30 –An61) occurs as inclusions (Figures 3c, 3d, 3e, and 4) and as partial rims surrounding inclusions of olivine, clinopyroxene, amphibole and Fe-Ti oxide (Figures 3c, 3d, and 3e). The large compositional variation of the plagioclase rims surrounding inclusion minerals (Figures 3d and 3e), ranging from labradorite to anorthoclase within less than 50 mm, reflects nonequilibrium crystallization. The large plagioclase 10 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 2b. Pyroxene Compositions of Mt. Capitole Lavasa  93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 460 93 – 462 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 471 93 – 472 93 – 472 93 – 472 93 – 472 93 – 472 93 – 473 93 – 473 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 486 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491  cpx inclusion in plag5 cpx in groundmass cpx in groundmass cpx in groundmass cpx in groundmass cpx in groundmass cpx1 core cpx1 rim cpx1 rim cpx2 core cpx2 core cpx2 rim cpx3 core cpx in groundmass pigeonite inclusion1 in plag7 pigeonite inclusion2 in plag7 pigeonite inclusion3 in plag7 cpx inclusion1 core in plag7 cpx inclusion1 rim in plag7 cpx inclusion2 core in plag7 cpx inclusion2 rim1 in plag7 cpx inclusion2 rim2 in plag7 cpx inclusion3 in plag7 cpx inclusion4 in plag7 cpx inclusion5 in plag7 cpx inclusion6 in plag7 cpx inclusion in plag12 cpx in groundmass cpx in groundmass cpx in groundmass cpx in groundmass cpx in groundmass cpx inclusion1 in plag1 cpx inclusion2 in plag1 cpx inclusion3 in plag1 cpx inclusion1 in plag5 cpx inclusion2 in plag5 cpx in groundmass cpx1 core cpx1 rim cpx2 core cpx3 core cpx3 rim cpx inclusion core in plag2 cpx inclusion rim in plag2 cpx1 core cpx1 rim cpx2 core cpx2 rim cpx2 rim cpx3 core cpx3 core cpx3 rim cpx3 rim cpx3 rim cpx1 core cpx1 rim cpx2 core cpx2 rim cpx2 rim cpx3 core cpx3 rim  SiO2  TiO2  Al2O3  Cr2O3  FeO  MnO  MgO  CaO  Na2O  Total  Mg#  50.08 49.07 48.07 48.48 48.88 50.98 49.92 51.12 52.16 51.97 50.43 51.57 52.29 52.00 51.25 50.59 49.85 51.43 50.47 51.42 52.43 51.16 52.40 51.33 51.19 51.04 51.05 48.07 49.17 52.16 51.55 49.69 50.66 50.63 52.38 52.26 51.07 50.73 50.07 49.32 50.48 52.13 50.64 49.00 50.15 51.54 51.35 51.04 51.15 51.61 52.28 51.66 51.40 51.39 52.03 51.68 51.38 51.41 51.63 51.56 51.48 51.82  1.19 1.95 2.36 2.19 2.01 1.20 1.69 1.36 1.14 1.00 1.31 1.11 0.82 1.02 0.32 0.48 0.47 1.13 1.46 1.17 1.22 1.47 1.28 1.10 1.43 1.45 1.51 1.27 1.67 0.82 1.06 1.49 1.39 1.53 0.73 0.94 1.31 1.21 1.79 2.08 1.10 1.03 1.00 2.15 1.86 1.21 1.18 1.13 1.09 1.20 0.90 1.10 1.03 1.07 0.98 0.93 0.95 0.81 0.80 1.20 0.86 0.93  2.00 4.14 5.05 3.79 4.79 2.17 3.84 1.72 1.85 2.56 3.99 2.06 2.17 1.76 0.10 0.50 0.16 2.32 3.22 1.70 1.62 2.14 0.97 1.18 1.82 2.22 2.64 4.86 4.47 1.94 2.14 4.16 3.19 2.78 1.17 2.10 2.75 2.49 3.70 3.58 1.42 1.94 1.29 3.81 2.27 2.49 2.50 2.91 1.55 2.29 2.35 2.03 1.47 1.73 2.35 2.76 2.56 2.63 2.46 1.44 2.27 2.51  0.02 0.04 0.06 0.00 0.31 0.00 0.09 0.00 0.00 0.06 0.20 0.02 0.08 0.00 0.02 0.02 0.12 0.04 0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.26 0.38 0.21 0.08 0.30 0.07 0.04 0.01 0.06 0.05 0.09 0.17 0.03 0.04 0.20 0.06 0.11 0.00 0.19 0.15 0.23 0.01 0.08 0.12 0.08 0.01 0.02 0.14 0.34 0.28 0.40 0.52 0.08 0.24 0.26  10.75 10.93 10.65 12.60 8.47 10.84 9.24 12.19 10.20 9.08 8.05 9.69 8.17 11.35 33.31 32.67 33.16 8.84 9.33 10.47 12.07 11.54 15.83 14.25 14.39 10.68 13.21 12.45 10.54 8.54 9.90 9.00 9.23 10.66 14.78 9.58 14.18 10.93 9.31 11.02 15.09 8.50 16.19 11.56 14.32 9.79 9.47 8.66 14.80 9.20 8.28 9.35 12.64 13.14 8.90 7.57 7.94 7.55 6.98 11.65 7.75 7.77  0.23 0.20 0.20 0.30 0.16 0.26 0.20 0.27 0.27 0.21 0.19 0.24 0.23 0.16 0.72 0.63 0.62 0.21 0.17 0.17 0.26 0.24 0.40 0.40 0.38 0.20 0.12 0.29 0.26 0.24 0.26 0.23 0.17 0.23 0.33 0.20 0.33 0.24 0.19 0.21 0.29 0.19 0.38 0.23 0.36 0.23 0.23 0.22 0.37 0.20 0.17 0.25 0.35 0.30 0.23 0.17 0.21 0.18 0.15 0.23 0.17 0.21  14.34 13.73 13.49 12.80 14.18 14.10 14.15 13.11 14.56 15.80 14.76 14.40 16.05 14.32 12.99 12.41 14.13 14.62 14.51 14.24 14.12 13.66 12.77 13.05 11.72 13.62 14.15 12.97 13.86 16.84 15.18 14.84 15.24 14.33 14.97 16.18 17.01 14.43 14.28 13.43 12.44 15.76 12.24 13.36 13.35 15.31 15.43 15.52 13.10 15.60 15.92 15.39 14.42 14.47 15.72 16.23 16.19 16.03 16.28 14.98 16.07 15.95  20.61 19.85 20.06 19.23 20.96 19.94 21.21 20.02 20.67 19.82 21.22 20.68 20.24 19.72 3.08 3.16 2.89 20.82 20.37 20.35 20.18 20.51 18.33 18.66 18.56 20.73 18.61 17.60 19.99 18.63 19.64 20.34 20.50 20.56 16.40 19.05 13.91 20.18 20.00 19.77 18.36 20.15 17.67 19.88 18.04 19.92 19.86 20.05 17.81 19.79 20.36 19.40 18.38 17.96 20.09 20.21 20.47 20.39 20.43 18.58 20.27 20.31  0.34 0.33 0.37 0.49 0.37 0.43 0.30 0.28 0.32 0.28 0.36 0.33 0.27 0.20 0.12 0.12 0.00 0.38 0.36 0.40 0.31 0.44 0.39 0.30 0.31 0.38 0.46 0.52 0.42 0.39 0.37 0.35 0.32 0.37 0.31 0.32 0.17 0.34 0.28 0.35 0.22 0.26 0.23 0.29 0.26 0.31 0.29 0.21 0.27 0.27 0.26 0.32 0.31 0.26 0.26 0.26 0.33 0.25 0.27 0.24 0.25 0.23  99.6 100.2 100.3 99.9 100.1 99.9 100.6 100.1 101.2 100.8 100.5 100.1 100.3 100.5 101.9 100.6 101.4 99.8 100.1 99.9 102.2 101.2 102.4 100.3 99.8 100.3 101.8 98.3 100.8 99.8 100.2 100.4 100.8 101.2 101.1 100.7 100.8 100.6 99.8 99.8 99.5 100.2 99.7 100.4 100.6 101.0 100.5 100.0 100.2 100.2 100.6 99.6 100.0 100.3 100.7 100.2 100.3 99.6 99.5 100.0 99.4 100.0  70.4 69.1 69.3 64.4 74.9 69.9 73.2 65.7 71.8 75.6 76.6 72.6 77.8 69.2 41.0 40.4 43.2 74.7 73.5 70.8 67.6 67.9 59.0 62.0 59.2 69.5 65.6 65.0 70.1 77.9 73.2 74.6 74.6 70.5 64.4 75.1 68.1 70.2 73.2 68.5 59.5 76.8 57.4 67.3 62.4 73.6 74.4 76.2 61.2 75.1 77.4 74.6 67.0 66.3 75.9 79.3 78.4 79.1 80.6 69.6 78.7 78.5  11 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 2b. (continued) 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 93 – 491 a  cpx4 core cpx4 rim cpx4 rim cpx4 rim cpx5 core cpx5 rim cpx6 core cpx6 rim cpx7 core cpx7 rim cpx8 core cpx8 rim cpx9 core cpx inclusion1 in plag11 cpx inclusion2 in plag11  SiO2  TiO2  Al2O3  Cr2O3  FeO  MnO  MgO  CaO  Na2O  Total  Mg#  51.75 52.08 51.43 52.22 52.11 51.78 49.34 52.37 50.54 52.21 50.30 51.43 51.42 49.00 49.94  0.96 0.81 0.82 0.88 0.86 0.91 1.46 0.83 1.17 0.97 1.31 1.28 1.10 1.57 1.62  2.53 2.20 1.14 1.83 1.92 2.38 4.71 1.61 3.58 1.39 3.93 2.18 2.82 3.19 2.11  0.23 0.30 0.05 0.30 0.22 0.23 0.51 0.40 0.39 0.15 0.27 0.12 0.25 0.00 0.03  8.08 7.23 15.95 9.29 8.40 8.35 9.42 8.74 8.78 11.26 9.32 11.52 8.13 17.13 17.66  0.17 0.20 0.46 0.21 0.23 0.22 0.25 0.27 0.21 0.23 0.14 0.23 0.16 0.29 0.37  15.92 16.40 14.08 16.22 16.30 16.02 16.29 17.05 15.50 16.51 15.97 15.27 16.06 13.26 11.20  20.58 20.44 16.39 18.90 19.96 19.79 16.74 18.38 19.78 17.19 18.31 18.21 20.49 14.72 18.39  0.22 0.26 0.18 0.27 0.24 0.28 0.35 0.22 0.30 0.21 0.32 0.23 0.30 0.15 0.26  100.5 99.9 100.5 100.1 100.2 100.0 99.1 99.9 100.3 100.1 99.9 100.5 100.7 99.3 101.6  77.8 80.2 61.1 75.7 77.6 77.4 75.5 77.7 75.9 72.3 75.3 70.3 77.9 58.0 53.1  Compositions are in wt% and were determined by electron microprobe at MIT.  grains are sieve-textured plagioclase [e.g., Nelson and Montana, 1992].  4.2.2. Pyroxene [16] Clinopyroxene is a more common phenocryst in Mt. Capitole lavas than olivine, but it rarely exceeds 3 vol%; sample 93–491 with $15 vol% clinopyroxene is an exception (Table 1). The Mg# of clinopyroxene phenocrysts ranges from 57 to 81 (Table 2b). Plagioclase phenocrysts in the Upper Transitional Group contain abundant inclusions of clinopyroxene with a similar range in Mg#, but rare pigeonite inclusions with low Mg# (40 to 43) also occur (Figure 3d). [17] High Al2O3 clinopyroxene (5 – 8.6 wt%) occurs in a section of mildly alkaline lavas from  Mt. Crozier in the eastern archipelago (Figure 1). Damasceno et al. [2002] concluded that highpressure (up to 12 kbar) fractionation of highAl2O3 clinopyroxene was an important process for these alkalic basalts. Such aluminous clinopyroxene phenocrysts are not present in Mt. Capitole lavas; they range from 1.29–4.71 wt% Al2O3, and crystallization pressures inferred from clinopyroxene/melt thermobarometers are 1 to 2.7 kbar at 1130°C [Putirka et al., 2003].  4.2.3. Olivine [18] Olivine phenocrysts, $2 vol%, occur only in sample 93–472 (Table 1) which has the highest MgO content (8.0 wt%) among Mt. Capitole lavas. These olivines are normally zoned, ranging from cores with Fo76 – 82 to rims with Fo74 – 77 (Table 2c).  Table 2c. Olivine Compositions of Mt. Capitole Lavasa  93 – 465 93 – 465 93 – 465 93 – 465 93 – 472 93 – 472 93 – 472 93 – 472 93 – 472 93 – 472 93 – 472 93 – 472 93 – 472 93 – 472 93 – 472 93 – 472 a  olivine inclusion1 core in plag7 olivine inclusion1 rim in plag7 olivine inclusion2 in plag7 olivine in groundmass near plag7 olivine1 core olivine1 rim olivine1 rim olivine2 core olivine2 rim olivine3 core olivine4 core olivine5 core olivine6 core olivine6 rim olivine7 core olivine7 rim  SiO2  Cr2O3  FeO  MnO  MgO  CaO  NiO  Total  Mg#  35.07 34.67 35.72 34.65 39.38 38.27 38.84 38.41 38.65 38.89 38.24 38.26 39.43 38.23 38.35 38.06  0.01 0.06 0.02 0.07 0.01 0.03 0.03 0.02 0.04 0.05 0.03 0.03 0.01 0.02 0.01 0.01  37.12 42.64 32.04 43.39 17.06 20.86 22.43 19.18 21.57 19.36 22.27 21.21 16.99 22.72 20.47 23.06  0.55 0.61 0.40 0.64 0.24 0.27 0.25 0.26 0.32 0.31 0.32 0.31 0.18 0.27 0.29 0.31  26.16 20.29 30.98 23.39 41.94 39.45 37.60 40.44 38.42 40.85 38.42 39.26 42.61 37.48 39.52 37.62  0.38 0.27 0.34 0.36 0.30 0.30 0.35 0.31 0.36 0.28 0.40 0.38 0.27 0.35 0.36 0.38  0.06 0.07 0.08  99.4 99.0 99.6 102.6 99.1 99.3 99.7 98.8 99.6 100.0 99.8 99.6 99.7 99.3 99.2 99.6  55.7 45.9 63.3 49.0 81.4 77.1 74.9 79.0 76.0 79.0 75.5 76.7 81.7 74.6 77.5 74.4  0.18 0.12 0.15 0.17 0.22 0.24 0.12 0.10 0.11 0.13 0.11 0.09  Compositions are in wt% and were determined by electron microprobe at MIT.  12 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 2d. Compositions of Amphibole Inclusions in Plagioclase Phenocrysts/Xenocrysts in Mt. Capitole Lavasa  93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 459 93 – 460 93 – 462 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 465 93 – 471 93 – 491 93 – 491 a  amph amph amph amph amph amph amph amph amph amph amph amph amph amph amph amph amph amph amph amph amph  inclusion in plag4 inclusion in plag4 inclusion in plag7 inclusion in plag7 inclusion in plag7 inclusion in plag8 inclusion in plag8 inclusion in plag3 inclusion in plag2 inclusion1 rim1 in plag7 inclusion1 core in plag7 inclusion1 rim2 in plag7 inclusion1 rim3 in plag7 inclusion1 rim4 in plag7 inclusion1 rim5 in plag7 inclusion9 in plag7 inclusion11 in plag8 inclusion12 in plag9 inclusion in plag7 inclusion in plag7 inclusion3 in plag11  SiO2  TiO2  Al2O3  Cr2O3  FeO  MnO  MgO  CaO  Na2O  Total  Mg#  43.82 43.27 42.61 43.83 43.98 44.55 43.82 44.26 43.91 48.42 45.14 44.62 45.35 43.70 44.30 45.64 42.65 45.87 45.00 41.16 43.32  4.91 4.66 5.41 5.09 5.37 4.25 4.79 5.05 4.79 2.15 5.07 4.64 4.80 4.69 4.14 5.10 4.52 4.16 4.35 5.32 5.26  3.12 2.77 2.35 3.11 3.22 5.47 2.90 2.70 3.95 2.02 3.08 4.66 3.92 2.89 2.47 2.97 3.24 4.28 3.37 4.20 2.64  0.04 0.04 0.05 0.01 0.08 0.04 0.10 0.02 0.05 0.03 0.00 0.05 0.08 0.04 0.05 0.07 0.01 0.07 0.06 0.09 0.05  21.90 22.34 23.49 23.78 22.99 22.65 21.08 17.15 18.38 20.86 20.55 23.89 22.47 22.56 22.56 18.54 21.29 19.17 21.41 24.62 23.81  0.42 0.47 0.36 0.31 0.33 0.29 0.38 0.25 0.46 0.43 0.29 0.48 0.39 0.36 0.40 0.34 0.37 0.48 0.43 0.64 0.60  11.93 11.61 12.05 11.49 11.17 12.08 12.83 13.21 12.19 14.06 15.91 10.86 12.84 12.81 13.91 15.35 12.71 12.82 13.36 9.82 10.71  12.67 12.22 13.19 11.89 12.14 11.56 12.72 16.07 13.15 10.93 8.86 10.02 9.85 10.70 9.96 11.52 12.81 11.21 12.05 10.79 11.17  0.50 0.53 0.38 0.55 0.59 1.13 0.84 0.80 0.64 0.28 0.26 0.62 0.66 0.39 0.58 0.51 0.36 0.46 0.57 0.40 0.31  99.3 97.9 99.9 100.1 99.9 102.0 99.5 99.5 97.5 99.2 99.2 99.8 100.4 98.1 98.4 100.0 98.0 98.5 100.6 97.0 97.9  49.3 48.1 47.8 46.3 46.4 48.7 52.0 57.9 54.2 54.6 58.0 44.8 50.4 50.3 52.4 59.6 51.6 54.4 52.6 41.5 44.5  Compositions are in wt% and were determined by electron microscope at MIT.  Highly evolved olivine (Fo46 – 63) also occurs as inclusions in the abundant plagioclase phenocrysts/ xenocrysts that characterize the Upper Transitional Group (Tables 1 and 2c and Figure 3d).  4.2.4. Amphibole [19] Amphibole phenocrysts/microphenocrysts occur in alkaline Kerguelen Archipelago lavas [Giret et al., 1980; Damasceno et al., 2002; Gagnevin et al., 2003]. Amphibole crystals in the Mt. Capitole section are present in the groundmass and as inclusions in plagioclase xenocrysts (Figures 3e and 3f). They are calcic-amphibole ranging from (titano-) magnesiohornblende to tschermakite according to the classification of Leake et al. [1997] (Table 2d). Amphibole inclusions in plagioclase phenocrysts/xenocrysts are commonly enclosed by Na-rich plagioclase rims that vary in composition along the elongated amphibole inclusion (Figure 3e).  4.3. Whole-Rock Compositions 4.3.1. Major Elements [20] Like other sections of flood basalt from the northern and central part of the Kerguelen archipelago, lavas from the Mt. Capitole section are dominantly tholeiitic to transitional basalt on the  basis of a silica-total alkalis diagram (Figure 5). They are evolved basalts with $46 to 53 wt% SiO2 and 3.3 to 8 wt% MgO (Table 3). Although there are no simple geochemical variations with relative eruption age, i.e., stratigraphic height in Figure 6, the lava compositions can be divided into three groups that correlate with stratigraphic position. The first group is the uppermost 15 tholeiitic/ transitional lavas from above 690 m. These lavas are dominantly plagioclase-phyric (Table 1) and are characterized by relatively high Al2O3 coupled with relatively low TiO2 and Fe2O3 (as total iron) (Figure 6). They are designated as the Upper Transitional Group. Sample 93–491, lower in the section at 540 m, is compositionally and petrographically similar to this group (Table 1 and Figure 6). For this group, abundances of SiO2, Al2O3, Na2O and K2O are negatively correlated with MgO, whereas CaO shows a slight positive correlation (Figure 7). The negative Al2O3 – MgO trend of this group contrasts with the positive trend of other Mt. Capitole and Mt. Tourmente lavas (Figure 7). Neither TiO2 nor P2O5 is inversely correlated with MgO in this group (Figure 7). [21] The second group of Mt. Capitole lavas are samples 93–478 to 93–486 from 660 m to 560 m. They have relatively low SiO2 contents and SiO2/ Fe2O3* ratios, high TiO2 and Fe2O3* contents and are alkalic or very close to the tholeiitic-alkalic 13 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 3. Major Element Compositions in Basalt From Mt. Capitolea CaO  Na2O  K2O  P2O5  Total  LOIc  18.05 16.80 15.35 16.22 15.71 14.55 16.20 14.29 14.94 14.54 15.92 15.91 15.10 15.89 15.00 14.49  Upper Transitional Group Lavas 10.97 0.17 4.78 11.57 12.44 0.19 4.67 10.52 13.35 0.20 4.93 10.56 12.41 0.19 5.46 11.00 12.92 0.19 6.47 10.76 13.89 0.20 5.81 10.44 12.17 0.18 6.00 11.10 13.39 0.21 6.54 10.91 13.19 0.18 6.42 11.00 13.05 0.19 6.27 10.90 12.25 0.18 6.30 10.75 12.50 0.19 6.24 11.13 12.90 0.19 8.00 11.15 13.30 0.20 5.93 10.37 13.08 0.19 5.48 10.80 12.96 0.19 7.12 11.01  2.74 2.91 3.01 2.65 2.43 2.88 2.55 2.66 2.04 2.46 2.93 2.46 2.27 2.88 2.79 2.12  0.67 0.88 0.78 0.64 0.66 0.70 0.69 0.48 0.51 0.56 0.54 0.59 0.46 0.55 0.57 0.20  0.26 0.34 0.34 0.29 0.28 0.32 0.26 0.32 0.30 0.30 0.27 0.28 0.29 0.31 0.37 0.29  99.96 99.83 99.89 99.68 99.88 99.73 99.92 99.91 99.56 99.65 99.44 100.20 99.87 99.82 99.58 99.48  1.36 2.99 0.57 5.20 1.62 2.20 2.29 1.94 4.42 1.91 2.47 2.94 4.39 4.23 1.18 3.07  3.26 3.23 3.17  13.31 13.20 14.25  Lower Transitional Group Lavas 14.65 0.20 4.55 10.19 14.80 0.20 5.06 9.74 14.31 0.18 5.22 10.08  2.86 2.90 2.85  0.77 1.00 0.76  0.45 0.45 0.41  99.75 99.70 99.73  1.95 2.20 1.49  47.88 49.33 48.55 47.12 46.17 47.69 47.46 47.76 47.33  3.87 3.74 3.89 4.20 4.47 4.02 3.78 3.75 3.83  12.95 13.18 12.59 13.27 13.30 13.34 13.52 13.38 14.18  9.76 9.42 9.38 9.83 9.13 10.32 10.47 10.21 10.11  2.73 2.85 2.76 2.95 2.72 2.60 2.40 2.39 2.92  0.47 0.97 0.80 0.71 0.98 0.75 0.41 0.60 0.42  0.50 0.50 0.48 0.52 0.55 0.47 0.43 0.44 0.45  99.43 100.22 99.69 100.13 99.48 99.83 99.62 99.78 100.08  1.66 2.04 0.77 1.40 1.17 2.18 1.91 1.14 1.68  48.75 49.25 48.58 51.66 47.84 47.57 49.66 49.88 50.29 49.10 49.86 49.48 49.73 49.39 48.52 48.53 48.25 51.43 49.76 50.18 50.22 50.54 52.88  3.56 3.48 3.29 3.82 3.33 3.35 3.34 3.32 3.50 3.55 3.63 3.23 3.57 3.14 3.00 3.31 3.90 3.93 3.46 3.44 3.72 3.51 3.50  12.93 13.16 13.60 12.87 13.70 13.84 12.97 13.20 13.39 13.32 13.30 13.23 13.24 13.93 14.26 13.62 13.33 13.31 13.80 13.71 12.91 13.15 13.41  Lower Transitional Group Lavas 15.70 0.22 4.95 9.75 15.12 0.23 5.05 10.08 14.74 0.22 5.39 10.40 14.51 0.18 3.77 8.60 15.00 0.25 5.82 10.31 15.31 0.19 5.84 10.41 15.01 0.22 5.01 9.53 14.44 0.19 4.94 9.42 14.41 0.24 4.56 8.66 15.30 0.25 5.10 9.25 15.36 0.20 4.69 9.28 14.31 0.23 5.43 10.08 14.74 0.26 5.34 9.55 14.35 0.22 5.17 10.26 13.60 0.22 5.96 10.72 14.82 0.21 5.58 10.49 15.86 0.30 5.05 9.19 14.15 0.18 3.95 8.54 14.67 0.27 4.73 8.69 14.95 0.21 4.37 8.63 15.17 0.20 4.37 8.34 14.38 0.21 4.39 8.67 13.48 0.18 3.29 7.14  2.36 2.34 2.53 2.80 2.53 2.29 2.78 2.80 2.95 2.80 2.74 2.74 2.86 2.98 2.66 2.38 2.17 2.69 2.97 3.00 2.97 3.23 3.30  0.47 0.44 0.35 1.02 0.40 0.23 0.74 0.83 1.02 0.57 0.50 0.45 0.63 0.49 0.51 0.33 0.90 0.98 0.89 1.08 1.19 0.99 1.62  0.45 0.42 0.40 0.48 0.41 0.37 0.40 0.40 0.44 0.44 0.44 0.39 0.43 0.37 0.33 0.39 0.47 0.50 0.49 0.49 0.48 0.47 0.69  99.14 99.57 99.50 99.71 99.59 99.40 99.66 99.42 99.46 99.68 100.00 99.57 100.35 100.30 99.78 99.66 99.42 99.66 99.73 100.06 99.57 99.54 99.49  0.92 0.89 1.53 0.94 1.54 2.98 0.33 0.40 0.75 0.93 1.61 1.21 1.04 1.20 1.31 1.29 6.31 2.26 1.84 1.27 1.16 0.61 0.57  Sample  Height, m  SiO2  TiO2  Al2O3  93 – 459 93 – 460 93 – 461 93 – 462 93 – 463 93 – 464 93 – 465 93 – 467 93 – 468 93 – 469 93 – 470 93 – 471 93 – 472 93 – 473 93 – 474 93 – 491  860 840 840 825 815 810 800 780 760 750 740 735 730 715 690 540  48.71 48.69 48.77 48.49 48.13 48.18 48.50 48.50 48.45 48.89 47.95 48.52 47.24 47.89 48.57 48.58  2.04 2.39 2.60 2.33 2.33 2.76 2.27 2.61 2.53 2.49 2.35 2.38 2.27 2.50 2.73 2.52  93 – 475 93 – 476 93 – 477  690 685 680  49.51 49.12 48.50  93 – 478 93 – 479 93 – 480 93 – 481 93 – 482 93 – 483 93 – 484 93 – 485 93 – 486  660 640 630 610 600 590 580 570 560  93 – 487 93 – 488 93 – 489 93 – 490 93 – 492 93 – 493 93 – 494 93 – 495 93 – 496 93 – 497 93 – 498 93 – 499 93 – 500 93 – 501 93 – 502 93 – 503 93 – 504 93 – 505 93 – 506 93 – 507 93 – 508 93 – 509 93 – 510  560 550 540 540 540 530 520 510 505 490 480 470 465 455 440 435 430 420 410 400 390 380 350  Fe2O3*b  MnO  Low-Silica Group 16.39 0.26 15.84 0.20 16.42 0.24 16.47 0.24 17.31 0.26 15.72 0.27 15.98 0.24 15.85 0.22 16.01 0.22  MgO  Lavas 4.62 4.19 4.58 4.82 4.59 4.65 4.93 5.18 4.61  14 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 3. (continued) Sample  Height, m  SiO2  TiO2  Al2O3  Fe2O3*b  MnO  MgO  CaO  Na2O  K2O  P2O5  Total  LOIc  93 – 511 93 – 512 93 – 513 93 – 514  310 270 250 230  51.39 50.93 50.68 50.14  3.68 2.90 2.83 3.30  13.60 13.50 13.88 13.42  14.37 13.71 13.62 14.45  0.23 0.20 0.21 0.19  3.78 5.12 5.35 4.74  8.37 9.53 9.76 9.04  2.81 2.83 2.48 3.13  1.01 0.65 0.48 0.71  0.60 0.37 0.36 0.41  99.84 99.74 99.65 99.53  1.93 0.43 0.88 1.24  a  Major oxide abundances (wt.%) were determined by X-ray fluorescence (XRF) at the University of Massachusetts following the procedures of Rhodes [1996]. b Fe2O3* indicates all iron reported as Fe2O3. c Loss on ignition (LOI) indicates weight loss after heating to 1020°C for 30 min.  boundary (Figures 5, 6, and 7). They are designated as the Low-Silica Group. This group does not vary widely in MgO (4.1 to 5.2 wt%); in general its compositional range overlaps with the uppermost group of slightly alkalic lavas in the Mt. Tourmente section (Figure 7). [22] All other samples from Mt. Capitole, 30 lavas from 680–690 m and 230–560 m (Figure 6) form the third group designated as Lower Transitional Group. The major element compositions of this group largely overlap the transitional lavas that occur in the lower 80% of the Mt. Tourmente section (Figure 7). [23] In summary, as at Mt. Tourmente, at Mt. Capitole there is a transition from tholeiitic to alkalic basalt with decreasing age; however, plagioclase-rich lavas are abundant in the upper part of the Mt. Capitole section. Plagioclase-phyric lavas are absent at Mt. Tourmente [Frey et al., 2002a], but they also occur at Mt. Marion Dufresne [Annell et al., 2007].  4.3.2. Trace Elements [24] Abundances of Th, Nb, Pb, Zr and Yb are highly correlated in Mt. Capitole lavas; in contrast abundances of K, Rb, Sr and Ba are poorly correlated with Th abundance (Figure 8). The ranges in Rb and K contents (factors of 33 and 8, respectively) are much greater than those for relatively immobile incompatible elements, such as Nb, Zr and Th (factors of 2 to 4). The ranges for Ba and Sr (factors of 3.8 and 3.1, respectively) are comparable to those for immobile incompatible elements. We infer that Mt. Capitole samples experienced post-magmatic alteration and that Rb and K were mobile during the alteration, but that Ba and Sr were less mobile. Despite their relatively low Th content the plagioclase-rich Upper Transitional Group lavas have K, Rb, Sr and Ba contents similar to lavas in the other groups (Figure 8). Note that there is a relative Sr depletion, i.e., relatively  low Sr/Ce and Sr/Nd ratios, in the Lower Transitional Group and Low-Silica Group lavas but not in the plagioclase-rich Upper Transitional Group (Figure 9). [25] All Mt. Capitole samples are enriched in incompatible elements relative to primitive mantle (Figure 9). The highest incompatible element contents are in the low MgO (3.3 to 3.8 wt%) lavas of the Lower Transitional Group; the lowest contents are in the plagioclase-rich Upper Transitional Group; incompatible element contents in the Low-Silica Group overlap with those of alkalic lavas at Mt. Tourmente (Figure 9). At a given MgO content, incompatible element contents increase in the order: Upper Transitional Group < Lower Transitional Group < Low-Silica Group. [26] All Mt. Capitole lavas contain relatively low and variable abundances of transition elements (Ni = 34–128 ppm; Cr = 2–272 ppm, Table 4) that are positively correlated with MgO. Abundance of Sc ranges from 28 to 32 ppm for lavas with MgO greater than 5.5 wt%, but ranges to lower Sc ($24 ppm) with decreasing MgO content (Table 4). Like TiO2, the Upper Transitional Group samples have the lowest V contents while Low-Silica Group lavas have relatively high V abundances (Table 4).  4.4. Sr, Nd, Hf, and Pb Isotopes [27] Although there are no long-term systematic temporal variations of Sr, Nd, Hf and Pb isotopic ratios with stratigraphic height, i.e., inferred eruption age, in the Mt. Capitole section, samples of the Upper Transitional and Low-Silica Groups define trends of increasing 87Sr/86Sr and decreasing 143 Nd/144Nd, 176Hf/177Hf with decreasing height; in contrast the lower Transitional Group lavas show no systematic variations of isotope ratios with height (Figure 10). [28] Most of the Mt. Capitole lavas define an inverse correlation of 87Sr/86Sr and 143Nd144Nd, 15 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 4. (Representative Sample). Trace Element Abundances in Basalt From Mt. Capitolea [The full Table 4 is available in the HTML version of this article at http://www.g-cubed.org.] Rb  Ba  Th  U  93 – 459 93 – 460 93 – 461 93 – 462 93 – 463 93 – 464 93 – 465 93 – 467 93 – 468 93 – 469 93 – 470 93 – 471 93 – 472 93 – 473 93 – 474 93 – 491  7.97 14.4 15.2 8.04 8.47 7.57 9.85 3.05 6.02 5.13 6.65 8.60 6.17 7.23 3.56 1.17  168 231 186 152 150 188 170 182 235 159 176 164 114 147 198 96  1.83 2.31 2.00 1.74 1.59 2.14 1.53 1.92 1.75 1.76 1.69 1.55 1.34 1.83 2.58 1.89  0.321 0.360 0.444 0.322 0.298 0.381 0.315 0.389 0.229 0.353 0.275 0.284 0.171 0.222 0.500 0.340  93 – 475 93 – 476 93 – 477  6.33 17.1 7.05  232 224 201  3.02 2.93 2.52  0.648 0.620 0.441  93 – 478 93 – 479 93 – 480 93 – 481 93 – 482 93 – 483 93 – 484 93 – 485 93 – 486  8.86 21.8 7.79 5.35 14.1 21.1 3.15 9.70 1.97  210 214 218 202 219 177 166 164 200  2.97 2.92 2.97 2.83 3.27 2.56 2.51 2.67 3.00  0.555 0.654 0.637 0.526 0.661 0.564 0.516 0.607 0.531  93 – 487 93 – 488 93 – 489 93 – 490 93 – 492 93 – 493 93 – 494 93 – 495 93 – 496 93 – 497 93 – 498 93 – 499 93 – 500 93 – 501 93 – 502 93 – 503 93 – 504 93 – 505 93 – 506 93 – 507 93 – 508  8.84 9.59 4.58 18.3 3.95 2.00 9.52 12.7 21.0 5.09 14.1 8.80 13.1 4.00 3.13 1.61 12.6 18.9 9.89 19.6 26.6  185 158 159 227 158 192 184 172 206 210 182 182 211 156 149 157 187 215 245 232 241  2.71 2.54 2.70 3.25 2.54 2.27 2.51 2.61 3.05 3.10 3.02 2.64 3.01 2.18 1.96 2.54 2.84 3.22 3.54 3.35 3.40  0.588 0.558 0.518 0.737 0.503 0.424 0.591 0.603 0.690 0.635 0.607 0.578 0.650 0.441 0.389 0.579 0.653 0.732 0.737 0.789 0.790  Nb  Ta  La  Pb  Pr  Nd  Sr  Zr  Upper Transitional Group Lavas 16.0 1.00 14.3 35.1 20.4 1.24 18.9 41.0 20.0 1.26 18.1 41.3 15.9 1.01 14.7 34.5 16.1 1.00 14.4 32.2 19.2 1.23 18.1 41.5 14.1 0.91 13.1 31.5 18.2 1.13 16.2 40.7 16.6 1.02 14.7 36.8 16.9 1.03 14.9 36.5 16.7 1.04 14.4 36.2 16.8 1.03 14.9 34.2 14.2 0.89 12.8 30.9 16.1 1.01 15.2 33.5 22.0 1.33 20.5 42.8 17.0 1.08 14.7 35.1  1.43 1.73 1.83 1.38 1.35 1.70 1.26 1.71 1.54 1.53 1.52 1.36 1.11 1.46 2.10 1.49  4.31 5.44 5.75 4.45 4.39 5.33 3.93 5.22 4.64 4.63 4.61 4.66 4.01 4.56 6.01 4.75  18.2 22.8 24.0 19.6 18.6 23.3 17.2 22.4 20.3 20.1 19.8 20.1 17.6 19.7 26.0 21.0  410 379 367 362 331 361 349 329 298 317 349 331 389 291 349 301  124 163 171 137 132 159 118 155 139 143 137 143 127 145 195 160  Lower Transitional Group Lavas 26.6 1.60 24.0 51.8 25.9 1.56 22.8 51.3 22.6 1.38 21.2 47.2  2.23 2.25 2.02  7.08 6.81 6.24  30.1 29.5 26.7  323 300 343  233 225 203  54.5 53.9 56.2 52.9 60.9 53.8 48.4 49.3 56.7  2.28 2.27 2.29 2.21 2.42 2.56 1.86 2.12 2.36  7.64 7.55 7.55 7.38 8.55 7.34 6.61 6.86 7.60  33.0 32.0 32.2 31.8 37.2 32.3 29.2 29.8 33.3  323 298 306 297 347 330 329 327 355  262 249 258 251 298 244 226 236 259  Lower Transitional Group Lavas 25.6 1.59 21.8 51.8 25.5 1.60 21.6 49.8 24.9 1.58 21.9 52.5 28.1 1.78 25.7 57.9 25.1 1.49 20.2 48.7 23.8 1.52 19.1 46.5 25.5 1.59 21.9 51.0 24.2 1.48 21.1 46.1 26.9 1.66 24.2 48.5 26.0 1.62 24.6 51.9 29.0 1.85 25.6 58.1 25.3 1.54 21.3 50.8 27.0 1.69 24.5 54.3 19.8 1.24 17.6 40.1 19.1 1.20 16.0 37.1 24.1 1.44 20.3 47.1 25.2 1.67 20.4 52.7 28.2 1.76 24.5 56.2 31.6 1.87 26.9 60.8 31.3 1.95 26.3 61.0 32.2 2.02 28.2 65.2  2.16 2.11 2.03 2.57 1.98 1.97 2.07 1.94 2.38 2.33 2.49 2.05 2.33 1.75 1.59 1.97 2.23 2.56 2.49 2.75 2.73  6.89 6.79 6.69 7.67 6.37 6.32 6.95 6.47 7.22 7.36 7.86 6.68 7.28 5.43 5.09 6.40 7.17 7.45 8.38 8.26 8.72  30.0 29.7 29.4 33.6 27.8 26.9 30.7 28.5 30.9 31.7 33.3 29.3 31.2 23.7 22.2 27.2 29.6 32.6 35.7 35.0 38.0  306 302 344 333 318 304 316 340 313 324 323 337 317 317 306 309 290 336 344 322 313  233 231 232 268 227 216 233 227 252 247 265 228 249 184 171 219 226 271 291 289 292  Low – Silica Group Lavas 29.1 1.72 24.4 29.1 1.71 23.7 28.5 1.73 24.5 29.4 1.74 23.2 34.0 2.08 26.5 27.5 1.69 23.3 24.9 1.53 21.4 25.0 1.53 22.2 29.0 1.83 23.6  Ce  a Trace element abundances are in ppm. Ni, Cr, and V were determined by XRF at the University of Massachusetts following the procedures of Rhodes [1996]. Sc was determined by INAA at MIT following the procedures of Ila and Frey [2000]. All others were determined by ICP-MS at MIT. The abundances for BHVO-2 are the average values of 15 analyses, with relative standard deviation of 3% [Huang and Frey, 2003].  16 of 34  0.512686 0.512694 0.512690 0.512722 0.512689 0.512706 0.512704 0.512717 0.512706  7 6 6 6 5 5 5 6 5  5 5 5  5  0.512704  0.512717 0.512728 0.512714  6 7 3 5 6  0.512697 0.512693 0.512697 0.512680 0.512705  0.51266 0.51267 0.51267 0.51270 0.51266 0.51268 0.51268 0.51269 0.51268  0.51269 0.51270 0.51269  0.51268  0.51268 0.51267  0.51267 0.51267 0.51267  0.51266 0.51265  0.28288  0.28288  0.28287 0.28288 0.28288  18.462 35 18.4718 26  18.465 23 18.3973 11 18.4193 8 18.4191 11 18.4236 8 18.4037 8  Transitional 0.28288 0.28287 0.28286 0.28289 0.28286 0.28289 0.28289 0.28288 0.28288  Group 18.3954 18.3769 18.3906 18.4618 18.3938 18.4200 18.4243 18.4382 18.4451 8 10 14 13 7 5 9 15 14  Low-Silica Group 0.28290 18.4721 8 0.28289 18.4562 13 18.4386 10  Lower 0.282884 5 0.282871 4 0.282869 3 0.282897 5 0.282869 4 0.282891 5 0.282894 5 0.282887 5 0.282881 6  0.282905 4 0.282897 4  0.282885 4  0.282884 4  0.282876 4 0.282886 4 0.282882 5  Upper Transitional Group 0.282863 4 0.28286 18.4418 13  18.325 18.304 18.318 18.382 18.321 18.347 18.331 18.367 18.374  18.399 18.386 18.382  18.384 18.387 18.45 18.341 18.356 18.355 18.360 18.351 18.337 18.45 18.391  15.5449 15.5514 15.5520 15.5553 15.5494 15.5407 15.5405 15.5421 15.5487  8 9 13 12 7 6 7 13 13  15.5505 7 15.5488 12 15.5520 9  15.640 29 15.5536 24  15.584 19 15.5398 9 15.5426 7 15.5424 9 15.5459 8 15.5421 7  15.5512 15  15.542 15.548 15.549 15.552 15.546 15.537 15.536 15.545 15.539  15.547 15.546 15.549  15.548 15.549 15.58 15.537 15.540 15.539 15.543 15.540 15.539 15.640 15.550 52 22 20 21 21 20  38.8748 38.8102 38.8299 38.8682 38.8425 38.8488 38.8534 38.8800 38.9004  22 26 32 34 18 15 19 35 33  38.9155 19 38.8762 28 38.8991 26  38.910 70 38.9414 57  38.894 38.8821 38.8937 38.8915 38.9124 38.8658  38.9699 36  38.765 38.704 38.724 38.754 38.736 38.746 38.721 38.770 38.790  38.807 38.762 38.814  38.861 38.926 38.88 38.783 38.792 38.789 38.810 38.770 38.814 38.90 38.840  a Notes: (1) Within each group, samples are in stratigraphic order. (2) Prior to isotopic analyses, all samples were acid-leached following the procedures of Weis et al. [2005]. The effects of acid leaching are shown by data for leached and unleached (UL) aliquots of sample 93 – 465. (3) Measured Sr isotopic ratios were normalized to 86Sr/88Sr = 0.1194, and Nd ratios were normalized to 146Nd/144Nd = 0.7219. Mean measured 87Sr/86Sr for SRM 987 at UBC during the course of study was 0.710260 ± 13 (2s, n = 42), and 143Nd/144Nd for La Jolla standard was 0.511858 ± 7 (2s, n = 18). The 87Sr/86Sr data for plagioclase phenocrysts were normalized to SRM 987 Sr standard of 0.710260 to avoid inter-lab bias. 176Hf/177Hf ratios reported were normalized to JMC475 Hf standard of 0.282160. Pb isotopic ratios were measured using Tl spiking (with a 205Tl/203Tl = 2.3885) for fractionation correction [Weis et al., 2005]. Mean measured 206Pb/204Pb, 207Pb/204Pb, and 208Pb/204Pb for SRM 981 Pb standard at UBC were 16.9418 ± 22 (2s, n = 94), 15.4979 ± 26 (2s, n = 94), and 36.7184 ± 63 (2s, n = 94), respectively. The Pb isotopic ratios of plagioclase were analyzed by TIMS at MIT using a fractionation correction of 0.12 ± 0.03%/amu, based on the values of Todt et al. [1996]. (4) Two sigma (2s) errors apply to last decimal place. The external reproducibilities for 87Sr/86Sr and 143Nd/144Nd based on three duplicates (93 – 465, 93 – 490, and 93 – 512) are better than 20 Â 10À6 and 11 Â 10À6, respectively, which is within or slightly larger than the machine in-run uncertainties. The external reproducibilities at UBC for 206Pb/204Pb, 207Pb/204Pb, and 208Pb/204Pb are better than 744 ppm, 428 ppm, and 525 ppm, respectively. (5) Subscript ‘‘i’’ (initial) measured ratios corrected to 25.7 Ma, the age of lavas from nearby Mt. Tourmente. Parent/daughter abundance ratios used for age corrections are data for unleached samples (Table 4) except for sample 472, which has a very high Nb/U ratio indicating U loss; therefore Nb/U = 40 was used to calculate U content, which was used to calculate the initial Pb isotope ratios. For samples 93 – 471 and 93 – 459, parent/daughter ratios are also available for leached samples; the calculated initial ratios for the two sets of parent/daughter ratios are within analytical uncertainties.  0.70485 0.70499 0.70498 0.70473 0.70500 0.70478 0.70482 0.70476 0.70477  0.704913 0.705046 0.705039 0.704770 0.705060 0.704841 0.704935 0.704790 0.704803  93 – 476 93 – 490 93 – 490 93 – 495 93 – 505 93 – 507 93 – 510 93 – 512 93 – 512  6 6 7 6 9 8 6 7 7  0.704783 7 0.70471 0.704716 8 0.70467 0.704776 9 0.70471  93 – 479 93 – 482 93 – 483  8  0.512685  G  3  0.70477 0.70478 0.70476 0.70476  0.70481 0.70482 0.70474 0.70481 0.70480 0.70478  Sr/86Sr 2s (87Sr/86Sr)i 143Nd/144Nd 2s (143Nd/144Nd)i 176Hf/177Hf 2s (176Hf/177Hf)i 206Pb/204Pb 2s (206Pb/204Pb)i 207Pb/204Pb 2s (207Pb/204Pb)i 208Pb/204Pb 2s (208Pb/204Pb)i  93 – 459 0.704835 10 93 – 459 93 – 459 plag 0.704740 10 93 – 463 0.704832 9 93 – 465 0.704828 9 93 – 465 0.704808 7 93 – 465UL 0.704875 7 93 – 471 0.704802 7 93 – 471 93 – 471 plag 0.704761 8 93 – 472 0.704772 6  87  Table 5. Sr, Nd, Hf, and Pb Isotope Compositions for Mt. Capitole Whole Rock Lavas and Plagioclase Phenocrystsa Geochemistry Geophysics Geosystems  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  17 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 6. Selected Trace Element Ratios for Whole Rocks and Plagioclase Phenocrysts (Sr/Nd)PM 93 – 459 93 – 459 93 – 459 93 – 471 93 – 471 93 – 471  plagioclase whole rock whole rock plagioclase whole rock whole rock  leacheda unleachedb leached unleached  28.5 2.74 1.45 22.1 2.44 1.05  87  Rb/86Sr  0.00157 0.054 0.056 0.00193 0.067 0.075  87  Sr/86Sr  0.704750 0.704835 0.704771 0.704802  147  Sm/144Nd 0.138 0.196 0.140 0.127 0.207 0.141  143  Nd/144Nd  0.512685 0.512705  238  U/204Pb 3.19 12.6 14.4 1.81 15.2 13.3  235  U/204Pb  0.0232 0.091 0.104 0.0132 0.111 0.097  232  Th/204Pb 9.6 34 86 7.5 41 76  a  Whole rock was leached repeatedly in 6 N HCl following the same procedures used in Sr, Nd, and Pb isotope analyses before dissolving for ICP-MS analyses. b Results from Table 4.  but two samples of the Lower Transitional Group are offset to higher 87Sr/86Sr (Figure 11a). The lowest 87Sr/86Sr ratios are in the lavas of the LowSilica Group and one of these samples (93–482) has the lowest 87Sr/86Sr and highest 143Nd/144Nd; the Low-Silica Group lavas overlap with the field for Mt. Tourmente but the other two groups include samples that range to higher 87Sr/86Sr and lower 143 Nd/144Nd (Figure 11a). [29] Mt. Capitole lavas define a positive trend in 143 Nd/144Nd versus 176Hf/177Hf overlapping with the field of Mt. Tourmente lavas but extend to lower 143Nd/144Nd and 176Hf/177Hf (Figure 11b); this trend is parallel to the slope of mantle-OIB array [Vervoort et al., 1999]. [30] In plots of 208Pb/204Pb and 207Pb/204Pb versus 206 Pb/204Pb, there is overlap among the three compositional groups of Mt. Capitole lavas (Figure 12a). Also the two samples of the Lower Transitional Group that are offset to high 87Sr/86Sr (Figure 11a) have anomalously high 207Pb/204Pb (Figure 12b); one of these samples (93–490) was analyzed in duplicate (Table 5). As with Sr, Nd and Hf isotopic ratios, Pb isotopic ratios in the LowSilica Group lavas overlap with the field for Mt. Tourmente lavas, but lavas from the Upper and Lower Transitional Group range to higher 87 Sr/86Sr, lower 143Nd/144Nd and 176Hf/177Hf and higher 208Pb/204Pb at a given 206Pb/204Pb (Figures 11 and 12). [31] The Upper Transitional Group lavas from Mt. Capitole show a correlation between (206Pb/204Pb)i and (Sr/Nd)PM (PM stands for primitive mantle value of Sun and McDonough [1989]) (Figure 13). Since high (Sr/Nd)PM is characteristic of plagioclase (Table 6), the plagioclase-rich component is inferred to have relatively high 206Pb/204Pb. Plagioclase grains from two Upper Transitional Group samples, 93 –459 and 93–471, which have the  extremes in Sr/Nd ratios among the five samples analyzed for radiogenic isotopes, were analyzed for Sr and Pb isotopes (Table 5). Plagioclase xenocrysts from these two samples have the same Sr and Pb isotope ratios within analytical uncertainties (Figure 12a and Table 5), indicating that these plagioclase xenocrysts were derived from the same source. The plagioclases and whole-rock have similar 87Sr/86Sr, but as expected from the correlations between Sr/Nd versus 206Pb/204Pb (Figure 13), the plagioclase xenocrysts have more radiogenic Pb isotope ratios than their whole rocks (Figure 12a and Table 5).  5. Discussion 5.1. Origin of Compositional Variations in Mt. Capitole Lavas 5.1.1. Role of Crystal Fractionation and Accumulation [32] Mt. Capitole lavas define two different Al2O3 versus MgO trends, a positive trend, similar to Mt. Tourmente lavas, for the Low-Silica and Lower Transitional Group lavas and a negative trend for the Upper Transitional Group (Figure 14). A negative Al2O3 versus MgO trend defined by Mt. Crozier lavas, in the northeast part of the archipelago (Figure 1), was inferred to reflect fractionation of a clinopyroxene-dominated assemblage at high pressure by Damasceno et al. [2002] and Scoates et al. [2006]. They inferred that lithospheric thickness increased as the archipelago evolved from a near-ridge setting at $40 Ma to its present intraplate location (see inset of Figure 1); therefore younger flood basalts, such as at Mt. Crozier, were likely to stagnate at higher pressure where the fractionating mineral assemblage has a high proportion of clinopyroxene. We favor a different interpretation, i.e., plagioclase accumulation, for 18 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 3. Backscattered electron images of polished thin sections of Upper Transitional Group samples. (a) Plagioclase xenocryst from sample 93– 471 showing irregular morphology that is interpreted as resorption. (b) Sieve-textured plagioclase xenocryst from 93– 465 showing abundant inclusions of clinopyroxene, olivine, amphibole, and Fe-Ti oxide. (c) Expanded scale of Figure 3b showing Na-rich plagioclase domains. (d) Expanded scale of Figure 3b showing olivine, clinopyroxene, pigeonite, amphibole, Fe-Ti oxide, and apatite inclusions. The inset with increased contrast shows the Na-rich plagioclase rims around the olivine, clinopyroxene, amphibole, and Fe-Ti oxide inclusions. These Na-rich plagioclase rims have variable compositions ranging from An5 to An54. (e) Expanded scale of Figure 3c showing the amphibole inclusion. The inset with increased contrast shows the Narich plagioclase rim partly surrounding the amphibole inclusion. Note that the plagioclase An composition decreases systematically along the elongated direction of the amphibole, which indicates nonequilibrium crystallization. (f) Another example of an amphibole inclusion in a plagioclase xenocryst from sample 93– 459. PLAG, plagioclase; CPX, clinopyroxene; PIG, pigeonite; OL, olivine; AMPH, amphibole; TMT, titanomagnetite.  19 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 4. Variation of plagioclase phenocryst and xenocryst compositions (An, in mol%) as a function of stratigraphic height (meters above sea level) in the Mt. Capitole and Mt. Marion Dufresne sections of the Plateau Central (this paper and Annell et al. [2007]). Plagioclase-phyric basalt dominantly occurs within the Upper Transitional Group (UTG) at Mt. Capitole (Table 1) and from near the base of the Mt. Marion Dufresne section. In both sections, plagioclase cores range widely in composition with >An80 cores occurring in the upper part of the Mt. Capitole section and the lower part of the Mt. Marion Dufresne section.  the negative Al2O3 versus MgO trend defined by the Upper Transitional Group at Mt. Capitole. The abundant plagioclase (Table 1) is obvious evidence for plagioclase accumulation. In addition, these lavas have the geochemical characteristics of plagioclase, that is, relatively high Al2O3 content, relatively low abundance of incompatible elements, (Sr/Nd)PM and Eu/Eu* >1, relatively high Ba/Th, and positive correlations of Sr/Nd and Eu/ Eu* with Al2O3/TiO2 (Figures 9, 14, and 15). In contrast, the Low-Silica Group and Lower Transitional Group lavas have (Sr/Nd)PM and Eu/Eu* <1 with (Sr/Nd)PM decreasing as MgO decreases (Figure 15). Such trends are consistent with cofractionation of plagioclase and a mafic phase, such as clinopyroxene, which decreased Al2O3 and MgO, respectively.  5.1.2. Role of Magma Mixing [33] Several characteristics of plagioclase in the Upper Transitional Group of Mt. Capitole indicate magma mixing: (1) many plagioclase grains are resorbed (Figure 3a); (2) the plagioclase grains have abundant olivine, pyroxene and amphibole inclusions with low Mg# and Na-rich plagioclase rims partly surrounding these inclusions (Figures 3d and 3e and Table 2), indicating that a plagioclaserich magma was invaded by a more evolved magma which reacted with the plagioclase crystals via interconnecting channels formed by dissolution; and (3) plagioclase is not in isotopic equilibrium with their whole rocks, i.e., plagioclase  xenocrysts have more radiogenic Pb isotope ratios (Figure 12a and Table 5).  5.1.3. Role of Variable Extents of Melting [34] In the lower 500 m of the Mt. Capitole section slightly alkaline lavas (Low-Silica Group) overlie tholeiitic lavas (Lower Transitional Group) (Figures 2 and 5). The nearby Mt. Tourmente section (Figure 1) records a similar compositional change. Moreover the Low-Silica Group at Mt. Capitole is similar in major and trace element compositions and isotopic ratios (Sr, Nd and Pb) to the upper alkalic lavas in the Mt. Tourmente section (Figures 7, 9, 11, and 14). Frey et al. [2002a] inferred that this temporal, tholeiitic to alkalic transition, reflects a decrease in extent of melting with decreasing eruption age.  5.2. Inferences From Flood Basalt Compositions at Three Locations in the Plateau Central [35] From northwest to southeast in the Kerguelen Archipelago, the exposed flood basalt changes from older, 29 to 26 Ma, tholeiitic and transitional basalt (Mts de Ruches, Fontaine, Bureau and Rabouille`re; Figure 1) to younger, 25 to 24 Ma, slightly alkalic basalt (Mt. Crozier, Ravin Jaune and Charbon; Figure 1) [Frey et al., 2000; Damasceno et al., 2002]. Frey et al. [2000] proposed that this change in composition reflects a decrease in melting extent of the Kerguelen mantle 20 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 5. Na2O + K2O versus SiO2 classification plot showing that the Mt. Capitole lavas straddle the alkalictholeiitic dividing line of Macdonald and Katsura [1964]. The filled squares indicate 15 samples from the uppermost 170 m of the section and 93– 491 from the lower section, designated as the ‘‘Upper Transitional Group.’’ The 9 filled circles indicate ‘‘Low-Silica Group’’ lavas; they have low SiO2/Fe2O3* ratios and are from the elevation range of 560 m to 660 m. The other 29 samples define the ‘‘Lower Transitional Group’’; labeled sample 93– 510 near the bottom of the section (Figure 2) is the most evolved lava with the lowest MgO and highest SiO2. Major element data were adjusted to a Fe2+/Fetotal ratio of 0.85. In general, Mt. Capitole lavas are less alkalic than flood basalts erupted in the Southeast Province [Frey et al., 2000] and at Mt. Crozier in the Courbet Peninsula [Damasceno et al., 2002]. They generally overlap with lavas from Mt. Tourmente [Frey et al., 2002a] and lavas erupted in the north-central (Mt. Bureau and Mt. Rabouille`re) and northwest (Mt. des Ruches and Mt. Fontaine) parts of the archipelago [Yang et al., 1998; Doucet et al., 2002].  plume as lithosphere thickness increased during the transition from a ridge-centered to intraplate setting in the Northern Kerguelen Plateau (see Figure 1 inset). Also the increasing proportion of highly evolved magmas with decreasing eruption age indicates a decrease in supply of basaltic magma to the crust [Weis et al., 1998; Frey et al., 2000]. The flood basalt sections in the Plateau Central are consistent with these interpretations. At Mt. Capitole alkalic basalt overlies tholeiitic basalt; the youngest lavas are plagioclase-phyric lavas that formed by mixing of plagioclase-rich magma with a highly evolved magma. At Mt. Tourmente [Frey et al., 2002a] alkalic basalt overlies tholeiitic to transitional basalt, most lavas are aphyric with low MgO contents (4.05 to 6.38 wt% in 64 lavas). At Mt. Marion Dufresne [Annell et al., 2007] the lower 300 m of alkalic lavas with <5.2 wt% MgO grades upward to plagioclase-phyric lavas overlain by 400 m of olivine-phyric, less alkalic lavas with 7 to 11 wt% MgO; within this upper interval there are three quartz-bearing basaltic andesites that reflect mixing of an evolved, quartz-bearing magma with basaltic magma. These characteristics of flood basalt in the Plateau Central show that these sections recorded a complex temporal transition from tholeiitic to alkaline volcanism and that the accompanying decrease in flux of basaltic magma provided time intervals for cooling and fractionation of basaltic magma.  5.3. Origin of Isotopic Variability in Kerguelen Archipelago Lavas [36] Basalt from the Cenozoic Northern Kerguelen Plateau, the Kerguelen Archipelago, and Heard  Figure 6. Abundance of TiO2, Fe2O3*, and Al2O3 (wt.%) and SiO2/Fe2O3* ratio versus stratigraphic height (meters) in the Mt. Capitole section. Fe2O3* is total iron as Fe2O3. Compared to the Lower Transitional Group, the Upper Transitional Group lavas (elevation greater than 690 m, except for 93– 491 at 540 m) have relatively low TiO2 and Fe2O3* and high Al2O3 and SiO2/Fe2O3*, whereas Low-Silica Group lavas (elevation between 560 m and 660 m) have low SiO2/Fe2O3* and high TiO2 and Fe2O3*. 21 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 7. TiO2, P2O5, CaO, Al2O3, K2O, Na2O, SiO2, and Fe2O3* abundance versus MgO content (all in wt%) for Mt. Capitole samples. The encircled fields shown for comparison are transitional lavas (open field) and alkalic lavas (gray field) from nearby Mt. Tourmente [Frey et al., 2002a]. Note that there is a negative Al2O3 – MgO trend for the Upper Transitional Group lavas that contrasts with other Mt. Capitole lavas and the Mt. Tourmente fields. In general, the Low-Silica Group lavas and Lower Transitional Group lavas from Mt. Capitole overlap the alkalic and transitional lavas from Mt. Tourmente, respectively.  Island define an inverse trend between 87Sr/86Sr and 143Nd/144Nd that ranges from the field of Southeast Indian Ridge (SEIR) MORB to 87Sr/86Sr of $0.7060 (Figure 11a). This trend is commonly inferred to reflect mixing of a plume-related component with relatively high 87Sr/86Sr and low 143 Nd/144Nd with a component similar to SEIR MORB [e.g., Gautier et al., 1990]. This conclusion is especially robust for the $34 Ma submarine Ocean Drilling Program (ODP) Site 1140 basalt recovered from the Northern Kerguelen Plateau,  which erupted within 50 km of the SEIR [Weis and Frey, 2002]. Among lavas forming the Kerguelen Archipelago, the MORB-like component is minimal in the youngest alkalic lavas (e.g., Mt. Ross and Southeast Province Upper Miocene Series) and some of the oldest tholeiitic lavas (Group P of Mts Bureau and Rabouille`re, where P indicates plumederived [Yang et al., 1998]) and most abundant in some of the older tholeiitic to transitional basalt (e.g., Group D lavas from Mt. Bureau, where D  22 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 8. Abundance of Rb, K2O, Sr, Ba, Nb, Pb, Zr, and Yb versus Th content (all in ppm, except K2O in weight percent) for Mt. Capitole samples. The 2s standard deviation indicated in each panel is ±3%. The highest Th and other incompatible element abundances are in two samples (93 – 510 and 93– 511) with the lowest MgO contents (Table 3). Sample 93 –483 has higher Pb abundance at a given Th content possibly due to Pb contamination. Abundances of K, Rb, Sr, and Ba do not vary systematically with Th content, but Rb and K abundances are much more variable than Sr and Ba abundances.  indicates relatively depleted [Yang et al., 1998]) (Figure 11a). [37] In contrast to the well-defined linear trend in Figure 11a, plots of 87 Sr/ 86 Sr, 143 Nd/ 144 Nd, 176 Hf/177Hf and 208Pb/204Pb versus 206Pb/204Pb show more complexity (Figure 16). As in Figure 11a, in Figure 16 Site 1140 basalts from the Northern Kerguelen Plateau extend from the SEIR N-MORB field toward the Kerguelen plume field ( 87 Sr/ 8 6 Sr $0.7052, 1 43 Nd/ 1 44 Nd $0.5126, 176 Hf/177Hf $0.2829 and 206Pb/204Pb $18.53); two-component mixing between Kerguelen plume and MORB-like components is inferred [Weis and  Frey, 2002]. However, lavas collected from some sections of the flood basalt forming the Kerguelen Archipelago define trends that are at high angles to the Site 1140 trend (Figure 16). Some trends, such as the Charbon/Jaune lavas from the Southeast Province, range from the plume field to higher 87 Sr/86Sr and lower 206Pb/204Pb; the Southeast Province UMS field has a similar slope but at higher 8 7 Sr/ 8 6 Sr, lower 1 4 3 Nd/ 1 4 4 Nd and 206 Pb/204Pb and higher 208Pb/204Pb at a given 206 Pb/204Pb; other groups, such as lavas from Mt. Capitole and Mts des Ruches and Fontaine, define trends subparallel to the trends of the Southeast Province lavas, but they originate from the plume23 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 9. Incompatible trace element abundance in Mt. Capitole lavas normalized to the primitive mantle estimates of Sun and McDonough [1989]. The field for alkalic lavas from Mt. Tourmente overlaps with the Low-Silica Group lavas from Mt. Capitole. Important features are the negative slopes from Nb to Yb with a pronounced relative depletion in Sr for the Low-Silica Group and Lower Transitional Group lavas. The Upper Transitional Group lavas are not depleted in Sr. Mt. Capitole lavas range to high Ba/Rb ratios as a result of Rb depletion.  SEIR MORB mixing trend. Other than Northern Kerguelen Province Site 1140 lavas, the largest proportion of a MORB-related component is in Group D lavas from Mt. Bureau (Figure 16). [38] We conclude that some lavas, such as Group P of Mt. Bureau and Mt. Rabouille`re, Charbon/Jaune and Upper Miocene Series from the Southeast  Province, and Heard Island (Big Ben Series) define isotopic fields consistent with mixing of a plume component with a component having higher 87 Sr/86Sr, and lower 143Nd/144Nd, 176Hf/177Hf and 206 Pb/ 204 Pb and high 208 Pb/ 204 Pb at a given 206 Pb/204Pb (Figure 16). However, other groups, such as Group D of Mt. Bureau and Mt. Rabouille`re, Mt. des Ruches, Mt. Fontaine, Mt. Tourmente and Mt. Capitole lavas, were created by two distinct mixing processes; the first process involving variable proportions of MORB-like and plumerelated components followed by variable addition of a component with high 87Sr/86Sr, and low 143 Nd/ 1 4 4 Nd, 1 7 6 Hf/ 1 7 7 Hf and 2 0 6 Pb/ 2 0 4 Pb (Figure 16). Evidence that such a component is present in the mantle below the archipelago is a metasomatized, clinopyroxene-bearing dunite xenolith found in a Upper Miocene Series basanite breccia; it has acid-leached whole-rock 206Pb/204Pb of 17.72 and 87Sr/86Sr of 0.7072 and an acidleached clinopyroxene separate has 87Sr/86Sr of 0.7056 [Mattielli et al., 1999] (see arrow in Figure 16a). The metasomatic component may be derived from the plume, perhaps originating as deeply recycled continental lithosphere [Barling et al., 1994; Doucet et al., 2005] or deeply recycled oceanic crust containing sediment. Alternatively as concluded by Mattielli et al. [1999] and consistent with the two stage mixing model presented here, this component may have been introduced relatively recently during ascent of plume-derived magma, perhaps by interaction with continental components in the underlying Cretaceous Kerguelen Plateau (e.g., ODP Site 747 in Figure 16e).  Figure 10. Initial (a) 87Sr/86Sr, (b) 143Nd/144Nd, and (c) 176Hf/177Hf versus stratigraphic height (meters) in the Mt. Capitole section calculated at 25.7 Ma. Although there is no long-term correlation, if grouped together, the Upper Transitional and Low Silica Groups define trends of increasing 87Sr/86Sr and decreasing 143Nd/144Nd and 176Hf/177Hf with decreasing eruption age. The 2 sigma errors shown are for analyses of SRM 987 (Sr), La Jolla (Nd), and JMC475 (Hf) standard (see Table 5). 24 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 11. (a) Initial (87Sr/86Sr)i versus (143Nd/144Nd)i showing that lavas from the Kerguelen Archipelago and Heard Island define a trend ranging from the field for Southeast Indian mid-ocean ridge basalt (SEIR N-MORB) to relatively high 87Sr/86Sr and low 143Nd/144Nd. ‘‘K. Plume’’ is the average Kerguelen plume composition from Table 3 of Weis and Frey [2002]. Red squares show data for Mt. Capitole lavas. The fields designate data for submarine basalt from ODP Site 1140 on the Northern Kerguelen Plateau [Weis and Frey, 2002], several stratigraphic sections from the 29-25 Ma flood basalt forming the Kerguelen Archipelago (i.e., the 28 –30 Ma northern sections of Group P (plume) and Group D (relatively depleted) lavas from Mts Bureau and Rabouille`re [Yang et al., 1998]), lavas from Mts Fontaine and des Ruches in the north [Doucet et al., 2002], 25– 26 Ma lavas from Mt. Capitole and Mt. Tourmente [Frey et al., 2002a] in the Plateau Central, $25 Ma lavas from Charbon/Jaune in the Southeast Province [Frey et al., 2000], two groups of younger (<10 Ma) and more alkalic lavas with MgO > 3 wt% (i.e., lavas from Mt. Ross [Weis et al., 1998] and basanites of the Upper Miocene Series (UMS) in the Southeast Province [Weis et al., 1993]), and Pleistocene/Holocene lavas (Big Ben Series) from Heard Island [Barling et al., 1994], a recently volcanically active island 440 km southeast of the archipelago (Figure 1 inset). A second group of Heard Island lavas, Laurens Peninsula Series, overlaps with the field for Mt. des Ruches and Fontaine. The 2s uncertainties are less than the size of the symbol. All the data are age-corrected to their eruption ages. Data sources are this study, the above references, and Mahoney et al. [2002] for SEIR N-MORB. (b) Expanded scale of Figure 11a showing data for the two sections sampling the Plateau Central, i.e., a field for Mt. Tourmente and data points for the 3 Mt. Capitole groups. (c) Initial (143Nd/144Nd)i versus (176Hf/177Hf)i for Kerguelen Archipelago lavas. The fields designate data for submarine basalt from ODP Site 1140 on the Northern Kerguelen Plateau [Weis and Frey, 2002], the 30-25 Ma flood basalt forming the Kerguelen Archipelago (lavas from Mts Bureau, Fontaine, Rabouille`re, des Ruches, and Tourmente), and two groups of younger (<10 Ma) and more alkalic lavas with MgO > 3 wt% from the archipelago (Mt. Ross [Weis et al., 1998] and basanites of the UMS from the Southeast Province [Weis et al., 1993]). ODP Site 747 lavas, age-corrected to 26 Ma, from the Central Kerguelen Plateau are shown as an example of inferred lower continental crust contamination in the Cretaceous basalt forming the Kerguelen Plateau [Frey et al., 2002b]. Mantle OIB array is taken from Vervoort et al. [1999]. (d) Expanded scale of Figure 11c showing data for the two lava sections from the Plateau Central. Data sources are the same as for Figure 11a plus Mattielli et al. [2002], Chauvel and Blichert-Toft [2001], Hanan et al. [2004], and Graham et al. [2006].  25 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 12. Initial (206Pb/204Pb)i versus (208Pb/204P)i and (207Pb/204Pb)i for Mt. Capitole lavas. (a) The data define a linear trend in 208Pb/204Pb versus 206Pb/204Pb, overlapping with one end of the measured field defined by Mt. Tourmente lavas. Plagioclase xenocrysts from Upper Transitional Group lavas have higher initial 206Pb/204Pb ratios than the whole rocks. (b) Samples 93– 490 and 93– 505, which are offset to higher (87Sr/86Sr)i at a given (143Nd/144Nd)i (Figure 11), have higher (207Pb/204Pb)i at a given (206Pb/204Pb)i. Plagioclase data are not shown in Figure 12b because of large uncertainties in 207Pb/204Pb ratios.  [39] What is the origin of the component with low 143 Nd/144Nd, 176Hf/177Hf, 206Pb/204Pb and high 87 Sr/86Sr ratios? Both ancient sediment and subcontinental lithosphere (lower continental crust and mantle) may have these characteristics [e.g., Huang et al., 1995; Rehka¨mper and Hofmann, 1997; Downes et al., 2001; Liu et al., 2004; Janney et al., 2005; Lustrino, 2005]. A difficulty with attributing low 206Pb/204Pb to recycled sediment is that sediment is likely to be accompanied by a much larger mass of altered igneous crust; this basaltic crust may mask the effects of sediment. For example, altered MORB can have very high 238U/204Pb, and this ratio is further increased by subduction zone processes [Kelley et al., 2005], which counteracts the effect of the low 238U/204Pb in sediment. Consequently, models favoring recycled sediments may assume extreme, perhaps unrealistic, values for sediment. As an example, a model for explaining the DUPAL anomaly of Indian Ocean MORB [Rehka¨ mper and Hofmann, 1997] used a 238 204 U/ Pb ratio of 2, whereas GLOSS (Global subducted sediment) has 238U/204Pb of 5.1 [Plank and Langmuir, 1998], and a Pb content of 55 ppm, whereas GLOSS has 20 ppm Pb [also see Zhang et al., 2005]. In contrast, lower continental crust has a relatively low 238U/204Pb ratio ($3 [Rudnick and Gao, 2004]), which will lead to a relatively low 206 Pb/204Pb ratio with increasing age. [40] There is evidence for subcontinental lithospheric mantle beneath the Kerguelen Archipelago; i.e., some harzburgite xenoliths in basanite dikes in the Courbet Peninsula (Figure 1) have the low 187 Os/188Os characteristic of subcontinental litho-  spheric mantle [Hassler and Shimizu, 1998]. However, basalts from the Kerguelen Archipelago [Yang et al., 1998; Weis et al., 2000] and Heard Island [Barling et al., 2003] are not characterized by such low Os isotopic ratios. Therefore it is unlikely that subcontinental lithospheric mantle was the major source component that led to the  Figure 13. (Sr/Nd)PM versus initial (206Pb/204Pb)i for Upper Transitional Group lavas from Mt. Capitole. (Sr/Nd)PM broadly increases with abundance of plagioclase phenocrysts; the exception, sample 93– 472, has abundant microphenocrysts of plagioclase. The correlation indicates that plagioclase with high Sr/Nd ratio (Table 6) has radiogenic Pb isotopic ratios. Two sigma errors for (Sr/Nd)PM and (206Pb/204Pb)i are ±3% and the in-run uncertainties, respectively. 26 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  is small. For example, the few d18O measurements of olivine phenocrysts from the Kerguelen and Heard Islands lavas are within the range of upper mantle peridotite and MORB sources [Eiler et al., 1997]. If lower continental crust has d 18O of $8.1% [Simon and Le´cuyer, 2005], the absence of anomalous d18O in Kerguelen Archipelago and Heard Island basalt limits lower continental crust to less than 14%; i.e., larger amounts of these components would result in d 18O greater than that found in upper mantle peridotite and MORB sources which range from 5.0 – 5.4% [Eiler et al., 1997].  Figure 14. Al2O3 versus MgO (wt.%) showing that lavas from the flood basalt sections in the Northern Kerguelen Archipelago define broad trends consistent with initial olivine fractionation (negative Al2O3 - MgO trend) followed by segregation of a plagioclase-rich assemblage (positive Al2O3 - MgO trend). In contrast, the younger, 24– 25 Ma flood basalt from the eastern archipelago (Mt. Crozier and Ravin Jaune and du Charbon) defines a steep inverse Al2O3 - MgO trend that dominantly reflects high-pressure clinopyroxene fractionation [Damasceno et al., 2002; Scoates et al., 2006]. Mt. Capitole lavas (symbols as in Figure 12) show two trends; the uppermost lavas, Upper Transitional Group, define a negative Al2O3 versus MgO trend, but in this case, the trend reflects plagioclase accumulation. In contrast, the Low-Silica and Lower Transitional Groups define a positive Al2O3 versus MgO trend that is consistent with plagioclase fractionation. Inset shows the fractionation/accumulation trends of different phase assemblages; using the measured plagioclase core and clinopyroxene compositions in Mt. Capitole lavas (Table 2), the vectors for plagioclase addition and clinopyroxene fractionation are similar. Data sources are the same as Figure 5.  low high  143 87  Nd/144Nd, 176Hf/177Hf, 206Pb/204Pb and Sr/86Sr in some archipelago lavas.  [41] Some lower continental crust, especially of Archean age, has very unradiogenic Pb isotopic ratios [e.g., Dickin, 1981; Huang et al., 1995]. Moreover, Archean cratons (India, South Africa, Antarctica and Australia) surround the Indian Ocean. Therefore we evaluate evidence for lower continental crust as a component that contributed to Kerguelen Archipelago lavas. On the basis of oxygen isotopic ratios, the proportion of lower continental crust in Kerguelen Archipelago lavas  [42] We have previously argued that the absence of relative depletion in Nb and Ta abundance is inconsistent with a continental component contributing to Kerguelen Archipelago lavas [e.g., Yang et al., 1998; Doucet et al., 2002; Frey et al., 2002a]. Specifically, Kerguelen Archipelago lavas lack the marked relative depletion in Nb, i.e., (La/Nb)PM and (Th/Nb)PM !1.5 (PM indicates primitive mantle from Sun and McDonough [1989]), found in Cretaceous basalt forming the Kerguelen Plateau at ODP Sites 738, 747 and 1137 (Figure 17a). Such plateau basalt is interpreted to be plume-derived basalt that assimilated continental crust [Mahoney et al., 1995; Weis et al., 2001; Ingle et al., 2002; Frey et al., 2003]. Mt. Capitole lavas in the Lower Transitional and Low-Silica Groups range from only $0.75 to $1 in (La/Nb)PM and (Th/Nb)PM, but these ratios are positively correlated (Figures 17a and 17b). Although low degree of melting (<6%) can change La/Nb and Th/Nb ratios, the melting trend leads to more variable La/Nb than Th/Nb (Figure 17b). Two samples (93–490 and 93–505) from the Lower Transitional Group which are offset to higher 8 7 Sr/ 8 6 Sr at a given 143 Nd/144Nd and offset to higher 207Pb/204Pb at a given 206Pb/204Pb have the lowest 206Pb/204Pb and relatively high Th/Nb ratios (Figures 11a, 12b, and 17b). These characteristics are consistent with the involvement of a continental component in these samples. Figure 17b shows mixing trends for two estimates of lower continental crust compositions. We note that these amounts of lower continental crust, 6–20%, are maximum values because the lower continental crust of stable, mature continents (i.e., Archean cratons) may be silicic, e.g., the Lewisian in Scotland [Rudnick and Gao, 2004; Willbold and Stracke, 2006]. Such lower continental crust is readily partially melted by basaltic magma; consequently lower proportions of incompatible element rich melt would be required. 27 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 15. Geochemical parameters controlled by plagioclase: (a) (Sr/Nd)PM versus Eu/Eu*, (b and c) (Sr/Nd)PM and (Ba/Th)PM versus MgO content (wt%), and (d and e) Th abundance and Al2O3/TiO2 versus (Sr/Nd)PM. Eu* is Eu abundance interpolated from chondrite-normalized abundances of Sm and Gd, and subscript ‘‘PM’’ designates normalized to primitive mantle estimate [Sun and McDonough, 1989]. Ten of 16 Upper Transitional Group lavas have more than (or equal to) 10 vol% plagioclase phenocrysts (Table 1), which is consistent with their (Sr/Nd)PM and Eu/Eu* greater than 1, and relatively high (Ba/Th)PM and Al2O3/TiO2. These are all characteristics of plagioclase accumulation. All other Mt. Capitole lavas define trends of decreasing (Sr/Nd)PM and Eu/Eu* with decreasing MgO and increasing Th. These characteristics reflect plagioclase fractionation. Dashed and solid lines in Figure 15d are plagioclase accumulation/fractionation trends starting from aphyric sample 93– 467 (An76 (solid line) and An52 (dashed line); tick marks are 5% intervals). Partition coefficients (Sr and Nd) for plagioclase are from Bindeman et al. [1998], and DTh = 0.05. For Upper Transitional Group, plagioclase accumulation is the major process, and for other Mt. Capitole lavas, plagioclase fractionation is required, but in detail clinopyroxene (±olivine) fractionation is also required.  [43] Compared to oceanic basalt, lower continental crust has distinctive incompatible trace element ratios that involve Nb and Pb (Table 7). For example, lower continental crust has Ce/Pb and Nb/U ratios of 5 and 25, respectively [Rudnick and Gao, 2004], whereas fresh ocean island basalt (OIB) has Ce/Pb and Nb/U ratio of 25 ± 5 and  47 ± 10, respectively [Hofmann et al., 1986]. Mt Capitole lavas have average Ce/Pb (24 ± 2.7) and Nb/U (45 ± 9.5 for lavas with LOI <2.5%). Although these averages for Mt. Capitole lavas overlap those of OIB, Mt. Capitole lavas define a weak correlation between Ce/Pb, Nb/U and (Th/ Nb)PM (e.g., Figure 17c). Mass balance calcula28 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  tions shows that addition of 18% lower continental crust of Rudnick and Gao [2004] or 6% lower continental crust of Shaw et al. [1994] decreases Ce/Pb from 23.5 to 18.5 and 21, and Nb/U from 43.5 to 42.3 and 41.3, respectively. [44] Lavas related to the Kerguelen hot spot that have high (La/Nb)PM also have distinctive radiogenic isotopic ratios. For example, Kerguelen Plateau lavas with high (La/Nb) PM have high 208 Pb/204Pb at a given 206Pb/204Pb [Frey et al., 2003, Figure 10]. In Figure 17d we show that as in Figure 16, Site 1140 lavas define a mixing line between SEIR MORB and the plume, whereas the Big Ben Series of Heard Island and the Upper Miocene Series from Southeast Province in the  Kerguelen Archipelago define a trend between the plume and lower continental crust.  6. Summary [45] Geochemical and petrographic characteristics define three distinct basalt types in the Mt. Capitole section. The Lower Transitional Group, tholeiitic/ transitional lavas, is compositionally distinct from the overlying Low-Silica Group, transitional to alkalic lavas. This upward transition from tholeiitic to alkalic composition is also observed at nearby Mt. Tourmente and is analogous to the $30 to 24 Ma compositional change of the flood basalt forming the bulk of the Kerguelen Archipelago. In  Figure 16 29 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  contrast the uppermost lavas, Upper Transitional Group, are distinguished by abundant plagioclase xenocrysts that show evidence for magma mixing. [46] Mt. Capitole lavas define trends in 206Pb/204Pb versus 87Sr/86Sr, 143Nd/144Nd, 176Hf/177Hf and 208 Pb/204Pb that do not extrapolate to the field of SEIR N-MORB (Figure 16). These trends cannot be explained by plume-MORB mixing. We propose a two-step mixing process for forming the $29–25 Ma flood basalt of the Kerguelen Archipelago; that is mixing of MORB-like and plumerelated components followed by variable addition of a continental-related component with high 87 Sr/86Sr and low 143Nd/144Nd, 176Hf/177Hf and  206  Pb/204Pb. This temporal sequence of events explains the slopes of arrays for the Mt. Capitole lavas and lavas from Mt. des Ruches and Fontaine (Figure 16). Our mixing scenario schematically illustrated in Figure 16e is similar to that proposed by Doucet et al. [2005]. Mixing trend 1 involves the Kerguelen plume- and MORB-like components (thick black curve in Figure 16e). Mixing trend 2 involves addition of a continental component, probably lower continental crust, to a Kerguelen plume-derived magma (thick red curve in Figure 16e) or to mixtures of the plume- and MORB-like components (thin red curves in Figure 16e). The first mixing event is best represented by NKP Site 1140 lavas and the second mixing event is consis-  Figure 16. Initial 87Sr/86Sr, 143Nd/144Nd, 176Hf/177Hf, and 208Pb/204Pb versus 206Pb/204Pb. All data are agecorrected except for Pb data for Mt. Tourmente and SE Charbon/Jaune lavas, which lack U and Pb abundance data. Red squares indicate Mt. Capitole data. The 2s uncertainties are less than the size of the symbol. (a) The field for SEIR N-MORB is at relatively low 87Sr/86Sr and 206Pb/204Pb, whereas the inferred ratios for the Kerguelen mantle plume are at relatively high 87Sr/86Sr and 206Pb/204Pb. The average (K. Plume) and radiogenic (rad. K. Plume in Figures 16d and 16e) Kerguelen plume compositions are from Table 3 of Weis and Frey [2002] for 87Sr/86Sr, 143 Nd/144Nd, and 206Pb/204Pb and from Mattielli et al. [2002] for 176Hf/177Hf. Other data fields are as in Figure 11. Note that samples 41 and 42 from Mt. des Ruches are distinct from other lavas in this section. Lavas from the Northern Kerguelen Plateau, Site 1140, are an example of binary mixing between plume and MORB-like components [Weis and Frey, 2002], but the elongated trends defined by the groups of Kerguelen and Heard basalt require components with relatively high 87Sr/86Sr and low 206Pb/204Pb. (b and c) 143Nd/144Nd and 176Hf/177Hf versus 206 Pb/204Pb. In contrast to the trends in Figure 16a, the slopes for archipelago groups are positive because 87Sr/86Sr is inversely correlated with 143Nd/144Nd and 176Hf/177Hf. Fields defined by data from the same references as in Figure 11 plus Chauvel and Blichert-Toft [2001], Hanan et al. [2004], and Graham et al. [2006]. (d) (206Pb/204Pb)i versus (208Pb/204Pb)i showing Mt. Capitole data and fields for various sections of the Kerguelen Archipelago and Heard Island lavas. Lavas from NKP Site 1140 and Group D lavas from Mt. Bureau and Rabouille`re define trends that extrapolate toward the SEIR N-MORB field; these trends were attributed to the mixing of Kerguelen plume and SEIR MORB – like components (thick black lines) [Yang et al., 1998; Doucet et al., 2002; Weis and Frey, 2002], but several sections of lavas from Kerguelen Archipelago (Mt. Capitole, Mt. des Ruches and Fontaine, Mt. Bureau and Rabouille`re (Group P), and SE Charbon/Jaune) and Big Ben Series lavas from Heard Island define trends toward higher 208Pb/204Pb at a given 206Pb/204Pb than the field for SEIR N-MORB. Also shown is a field for continentalrelated clasts in a conglomerate intercalated with basalt from ODP Site 1137 on the Kerguelen Plateau [Ingle et al., 2002]; none of the Kerguelen Archipelago or Heard Island fields extrapolate toward this field. (e) A schematic diagram showing two mixing events. Triangles are data for Site 1140 lavas. The green field schematically shows that although lower continental crust (LCC) is isotopically heterogeneous, a distinguishing characteristic of many LCC samples is unusually low 206Pb/204Pb and variable 87Sr/86Sr [e.g., Huang et al., 1995; Downes et al., 2001; Liu et al., 2004; Lustrino, 2005]. The field for ODP Site 747 lavas from the Central Kerguelen Plateau (CKP) is an example of inferred LCC contamination in the Cretaceous basalt forming the Kerguelen Plateau [Frey et al., 2002b]. Note that MORB-plume mixing could be either solid-solid mixing or mixing of melts. The MORB-plume mixing trajectory is for melt mixing, whereas the addition of LCC assumes bulk assimilation of LCC, i.e., a maximum estimate (see Table 7 for parameters used for mixing end-members). Mixing curves between Kerguelen plume and LCC are near linear because Sr/Pb ratios for Kerguelen plume and average LCC are similar. The isotopic variation of Mt. Capitole lavas can be explained by mixing of Kerguelen plume primary melt with 50% SEIR MORB followed by $6% LCC addition using the modeling parameters in Table 7. The ticks on the red line are a proportion of LCC at 1% intervals. The proportions for MORB are indicated next to the black line. Triangles stand for Site 1140 lavas. Two geographically separate Pleistocene/Holocene lava groups from Heard Island have been studied: Big Ben Series and Laurens Peninsula Series (LPS) [Barling et al., 1994]. Like some lavas from the Kerguelen Archipelago, Big Ben Series lavas (x) extend to low 206Pb/204Pb and high 87Sr/86Sr. In contrast, the LPS lavas (open circle) with high 3 He/4He (16.2– 18.3 R/Ra [Hilton et al., 1995]) have lower 87Sr/86Sr and higher 206Pb/204Pb than proposed for the Kerguelen plume, perhaps reflecting plume heterogeneity. 30 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Figure 17. (a) Abundance ratio of (Th/Nb)PM versus (La/Nb)PM showing the field for Kerguelen Archipelago lavas and Mt. Capitole data for the Low-Silica and Lower Transitional Group (red squares). Subscript PM indicates ratios normalized to primitive mantle ratios [Sun and McDonough, 1989]. Also shown is average lower continent crust (LCC) from Rudnick and Gao [2004]. Shown for comparison are data points for oceanic basalt inferred to contain a component derived from continental crust; i.e., Kerguelen Plateau Sites 738, 747, and 1137 [Mahoney et al., 1995; Frey et al., 2002b; Ingle et al., 2002] and Pitcairn Island [Eisele et al., 2002; Honda and Woodhead, 2005]. (b) Expanded scale of Figure 17a showing the positive trend for Mt. Capitole lavas in the Lower Transitional Group (red squares) and Low-Silica Group (blue circles). Upper Transitional Group lavas from Mt. Capitole which have accumulated plagioclase, not plotted in Figure 17a, are shown as a field because accumulation of plagioclase creates higher La/Nb ratios at a given Th/Nb [Bindeman et al., 1998]. Lower Transitional Group lavas 93– 490 and 93– 505 with relatively high 87Sr/86Sr and 207Pb/204Pb at a given 143Nd/144Nd and 206Pb/204Pb, respectively, have relatively high Th/Nb and La/Nb ratios. These characteristics are consistent with involvement of LCC and inconsistent with partial melting trend. Using the average lower continental crust composition of Rudnick and Gao [2004], $20% LCC is needed to explain the maximum variation of Th/Nb and La/Nb ratios in Mt. Capitole lavas. However, if the Shaw et al. [1994] estimate of lower continental crust is used, then only 6% LCC is needed. These are maximum values of LCC (see text). Since SEIR N-MORB has low Th/Nb but relatively high La/Nb ratios, the Kerguelen plume composition has to be slightly offset from the trend defined by Mt. Capitole lavas. We note that the average Heard Island LPS (filled large pink triangle) which may represent the extreme Kerguelen plume composition has such La/ Nb and Th/Nb ratios. Trace element compositions for Kerguelen plume, SEIR N-MORB, and LCC are in Table 7. The 2s uncertainties shown in Figure 17b are ±3%. (c) Ce/Pb versus (Th/Nb)PM for Mt. Capitole lavas (symbols as in Figure 5). Incorporation of LCC into oceanic basalt creates an inverse correlation. Error bars indicate ±3% 2s uncertainties. Sample 93– 483 and 93– 510 are outliers; 93– 483 is offset to high Pb in Figure 8, possibly because of Pb contamination, and 93– 510 is the most evolved sample (Figures 7 and 8). (d) Initial 87Sr/86Sr versus (La/Nb)PM showing that the Kerguelen Plateau Site 1140 data are consistent with mixing of MORB- and plume-related components, whereas the Heard Island Big Ben Series (BBS) and the Upper Miocene Series from the southeast Kerguelen Archipelago (SE UPMS) define a trend between plume and LCC-related components. The Mt. Capitole data define a trend emanating from the MORB-Plume mixing line toward a LCC component. 31 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Table 7. Trace Element Contents and Isotopic Ratios for Modelinga 87  Kerguelen plume (av.) Kerguelen plume (rad.) SEIR N-MORB LCC [Huang et al., 1995] LCC [Rudnick and Gao, 2004] LCC [Shaw et al., 1994]  Sr/86Sr  0.70523 0.703 0.707  206  Pb/204Pb  18.533 18.642 17.98 16.44  208  Pb/204Pb 39.2 39.38 37.8 36.85  Sr  Pb  Th  Nb  La  300  3  1.05  10.5  7.7  100  0.25  0.05  1  1.4  348 447  4 6  1.2 2.6  5 5.6  8 21  a  Isotopic compositions of average (av.) and radiogenic (rad.) Kerguelen plume are from Weis and Frey [2002]. SEIR N-MORB isotopic compositions are the averages of N-MORB data from Mahoney et al. [2002] and Kempton et al. [2002]. We estimated the trace element compositions of SEIR N-MORB using Mahoney et al. [2002] data based on Mg# - X plots. Lower continental crust isotopic compositions are from a xenolith composition from Huang et al. [1995]. The trace element compositions of lower continental crust are from Rudnick and Gao [2004] for average LCC and Shaw et al. [1994] for stable and mature LCC. We estimated Sr, Th, Nb, and La contents for the Kerguelen plume melt using MgO-X plots for the flood basalts. The slopes of such plots change abruptly at $6.5 wt% MgO, reflecting the onset of cotectic crystallization and fractionation of plagioclase and clinopyroxene. Lavas with >6.5 wt% define a nearly horizontal trend due to olivine fractionation, and we averaged the abundances of Sr, Th, Nb, and La of samples with >6.5 wt% MgO. Due to scarcity of Pb data for Kerguelen Archipelago lavas, we calculated a Pb content for Kerguelen plume melt based on the most primitive melt composition of Mt. Capitole lavas and the proportion of MORB which can be constrained by Sr contents and isotopic ratios. There are large uncertainties in estimating of trace element contents of primary melts for Kerguelen plume and SEIR N-MORB as well as choosing isotopic ratios and concentrations in LCC; therefore the modeling is schematic and qualitative rather than quantitative.  tent with the trend of Mt. Capitole lavas. Since Cretaceous Kerguelen Plateau may underlie the Cenozoic Kerguelen Archipelago, and some basalt forming the plateau has assimilated continental crust [e.g., Mahoney et al., 1995; Frey et al., 2002b], it is possible that the continental crust signature evident in some archipelago lavas was acquired by assimilation of plateau lavas that were contaminated by lower continental crust [Ingle et al., 2003].  Acknowledgments [47] G.X. thanks M. Lo Cascio, Stephanie Ingle, and Sonia Doucet for assistance during sample preparation in Brussels, B. Kieffer and J. Barling for help in acquisition of the whole rock isotopic data, F. Dudas and S. Bowring for advice and assistance during plagioclase separation and acquisition of isotopic data, B. Grant and R. Kayser for their assistance with ICP-MS analyses, and P. Ila for her assistance in obtaining INAA data. N. Chatterjee, T. L. Grove, E. Medard, and J. Barr are thanked for their assistance in obtaining mineral composition data. Each of these individuals plus M. Lustrino and T. Plank are thanked for helpful discussions. This study was supported by National Science Foundation grants to F.F. and NSERC Discovery grants to D.W. and J.S. We thank M. Regelous and A. Saunders for their review comments and the editor, V. Salters, for his comments and suggestions.  References Annell, H., D. Weis, J. Scoates, and A. Giret (2004), Evidence for a depleted mantle component in mildly alkalic high-MgO basalts from Mt. Marion Dufresne, Kerguelen Archipelago, southern Indian Ocean, Eos Trans. AGU, 85(47), Fall Meet. Suppl., Abstract V31B-1436. Annell, H., S. J. Scoates, D. Weis, and A. Giret (2007), Petrology and phenocryst mineral chemistry in flood basalts from  the tholeiitic-alkalic transition, Mt. Marion Dufresne, Kerguelen Archipelago, southern Indian Ocean, Can. Mineral, in press. Barling, J., S. L. Goldstein, and I. A. Nicholls (1994), Geochemistry of Heard-Island (southern Indian-Ocean): Characterization of an enriched mantle component and implications for enrichment of the sub-Indian Ocean mantle, J. Petrol., 35, 1017 – 1053. Barling, J., J. S. McBride, D. D. Lambert, and I. A. Nicholls (2003), Osmium isotope evidence for mantle vs. magma contamination in Heard Island lavas, paper presented at GAC/MAC Meeting, Geol. Assoc. of Can., Vancouver, B. C. Bindeman, I. N., A. M. Davis, and M. J. Drake (1998), Ion microprobe study of plagioclase-basalt partition experiments at natural concentration levels of trace elements, Geochim. Cosmochim. Acta, 62, 1175 – 1193. Blichert-Toft, J., C. Chauvel, and F. Albare`de (1997), Separation of Hf and Lu for high-precision isotope analysis of rock samples by magnetic sector multiple collector ICP-MS, Contrib. Mineral. Petrol., 127, 248 – 260. Chauvel, C., and J. Blichert-Toft (2001), A hafnium isotope and trace element perspective on melting of the depleted mantle, Earth Planet. Sci. Lett., 190, 137 – 151. Coffin, M. F., and O. Eldholm (1994), Large igneous provinces: Crustal structure, dimensions, and external consequences, Rev. Geophys., 32, 1 – 36. Coffin, M. F., M. S. Pringle, R. A. Duncan, T. P. Gladczenko, M. Storey, R. D. Muller, and L. A. Gahagan (2002), Kerguelen hotspot magma output since 130 Ma, J. Petrol., 43, 1121 – 1137. Damasceno, D., J. S. Scoates, D. Weis, F. A. Frey, and A. Giret (2002), Mineral chemistry of mildly alkalic basalts from the 25 Ma Mont Crozier section, Kerguelen Archipelago: Constraints on phenocryst crystallization environments, J. Petrol., 43, 1389 – 1413. Dickin, A. P. (1981), Isotope geochemistry of tertiary igneous rocks from the Isle of Skye, NW Scotland, J. Petrol., 22, 155 – 189. Doucet, S., D. Weis, J. S. Scoates, K. Nicolaysen, F. A. Frey, and A. Giret (2002), The depleted mantle component in Kerguelen Archipelago basalts: Petrogenesis of tholeiitictransitional basalts from the Loranchet Peninsula, J. Petrol., 43, 1341 – 1366. 32 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Doucet, S., J. S. Scoates, D. Weis, and A. Giret (2005), Constraining the components of the Kerguelen Mantle Plume: A Hf-Pb-Sr-Nd isotopic study of picrites and high-MgO basalts from the Kerguelen Archipelago, Geochem. Geophys. Geosyst., 6, Q04007, doi:10.1029/2004GC000806. Downes, H., A. J. W. Markwick, P. D. Kempton, and M. F. Thirlwall (2001), The lower crust beneath cratonic north-east Europe: Isotopic constraints from garnet granulite xenoliths, Terra Nova, 13, 395 – 400. Eiler, J. M., K. A. Farley, J. W. Valley, E. Hauri, H. Craig, S. R. Hart, and E. M. Stolper (1997), Oxygen isotope variations in ocean island basalt phenocrysts, Geochim. Cosmochim. Acta, 61, 2281 – 2293. Eisele, J., M. Sharma, S. J. G. Galer, J. Blichert-Toft, C. W. Devey, and A. W. Hofmann (2002), The role of sediment recycling in EM-1 inferred from Os, Pb, Hf, Nd, Sr isotope and trace element systematics of the Pitcairn hotspot, Earth Planet. Sci. Lett., 196, 197 – 212. Frey, F. A., D. Weis, H. J. Yang, K. Nicolaysen, H. Leyrit, and A. Giret (2000), Temporal geochemical trends in Kerguelen Archipelago basalts: Evidence for decreasing magma supply from the Kerguelen Plume, Chem. Geol., 164, 61 – 80. Frey, F. A., K. Nicolaysen, B. K. Kubit, D. Weis, and A. Giret (2002a), Flood basalt from Mont Tourmente in the Central Kerguelen Archipelago: The change from transitional to alkalic basalt at $25 Ma, J. Petrol., 43, 1367 – 1387. Frey, F. A., D. Weis, A. Y. Borisova, and G. Xu (2002b), Involvement of continental crust in the formation of the Cretaceous Kerguelen Plateau: New perspectives from ODP Leg 120 sites, J. Petrol., 43, 1207 – 1239. Frey, F. A., M. F. Coffin, P. J. Wallace, and D. Weis (2003), Leg 183 synthesis: Kerguelen Plateau-Broken Ridge—A large igneous province, Proc. Ocean Drill. Program Sci. Results, 183, 1 – 48. Gagnevin, D., et al. (2003), Open-system processes in the genesis of silica-oversaturated alkaline rocks of the Rallierdu-Baty Peninsula, Kerguelen Archipelago (Indian Ocean), J. Volcanol. Geotherm. Res., 123, 267 – 300. Gautier, I., D. Weis, J. P. Mennessier, P. Vidal, A. Giret, and M. Loubet (1990), Petrology and geochemistry of the Kerguelen Archipelago basalts (south Indian-Ocean)— Evolution of the mantle sources from ridge to intraplate position, Earth Planet. Sci. Lett., 100, 59 – 76. Giret, A., B. Bonin, and J. M. Le´ger (1980), Amphibole compositional trends in oversaturated and undersaturated alkaline plutonic ring complexes, Can. Mineral., 18, 481 – 495. Graham, D. W., J. Blichert-Toft, C. J. Russo, K. H. Rubin, and F. Albare`de (2006), Cryptic striations in the upper mantle revealed by hafnium isotopes in southeast Indian Ridge basalts, Nature, 440, 199 – 202. Hanan, B. B., J. Blichert-Toft, D. G. Pyle, and D. M. Christie (2004), Contrasting origins of the upper mantle revealed by hafnium and lead isotopes from the Southeast Indian Ridge, Nature, 432, 91 – 94. Hassler, D. R., and N. Shimizu (1998), Osmium isotopic evidence for ancient subcontinental lithospheric mantle beneath the Kerguelen Islands, southern Indian Ocean, Science, 280, 418 – 421. Hilton, D. R., J. Barling, and O. E. Wheller (1995), Effect of shallow-level contamination on the helium isotope systematics of ocean-island lavas, Nature, 373, 330 – 333. Hofmann, A. W., K. P. Jochum, M. Seufert, and W. M. White (1986), Nb and Pb in oceanic basalts: New constraints on mantle evolution, Earth Planet. Sci. Lett., 79, 33 – 45. Honda, M., and J. D. Woodhead (2005), A primordial solarneon enriched component in the source of EM-I-type ocean  island basalts from the Pitcairn Seamounts, Polynesia, Earth Planet. Sci. Lett., 236, 597 – 612. Housh, T., and S. A. Bowring (1991), Lead isotopic heterogeneities within alkali feldspars: Implications for the determination of initial lead isotopic compositions, Geochim. Cosmochim. Acta, 55, 2309 – 2316. Huang, S., and F. A. Frey (2003), Trace element abundances of Mauna Kea basalt from phase 2 of the Hawaii Scientific Drilling Project: Petrogenetic implications of correlations with major element content and isotopic ratios, Geochem. Geophys. Geosyst., 4(6), 8711, doi:10.1029/2002GC000322. Huang, Y. M., P. Vancalsteren, and C. J. Hawkesworth (1995), The evolution of the lithosphere in southern Africa: A perspective on the basic granulite xenoliths from kimberlites in South-Africa, Geochim. Cosmochim. Acta, 59, 4905 – 4920. Ila, P., and F. Frey (2000), Trace element analysis of USGS standards AGV2, BCR2, BHVO2, DTS2 and GSP2 by INAA, J. Radioanal. Nucl. Chem., 244, 599 – 602. Ingle, S., D. Weis, and F. A. Frey (2002), Indian continental crust recovered from Elan Bank, Kerguelen Plateau (ODP Leg 183, Site 1137), J. Petrol., 43, 1241 – 1257. Ingle, S., D. Weis, S. Doucet, and N. Mattielli (2003), Hf isotope constraints on mantle sources and shallow-level contaminants during Kerguelen hot spot activity since <120 Ma, Geochem. Geophys. Geosyst., 4(8), 1068, doi:10.1029/ 2002GC000482. Janney, P. E., A. P. Le Roex, and R. W. Carlson (2005), Hafnium isotope and trace element constraints on the nature of mantle heterogeneity beneath the central Southwest Indian Ridge (13°E to 47°E), J. Petrol., 46, 2427 – 2464. Kelley, K. A., T. Plank, L. Farr, J. Ludden, and H. Staudigel (2005), Subduction cycling of U, Th, and Pb, Earth Planet. Sci. Lett., 234, 369 – 383. Kieffer, B., N. T. Arndt, and D. Weis (2002), A bimodal alkalic shield volcano on Skiff Bank: Its place in the evolution of the Kerguelen Plateau, J. Petrol., 43, 1259 – 1286. Leake, B. E., et al. (1997), Nomenclature of amphiboles: Report of the subcommittee on amphiboles of the International Mineralogical Association Commission on New Minerals and Mineral Names, Eur. J. Mineral., 9, 623 – 651. Liu, Y., S. Gao, H. Yuan, L. Zhou, X. Liu, X. Wang, Z. Hu, and L. Wang (2004), U-Pb zircon ages and Nd, Sr, and Pb isotopes of lower crustal xenoliths from North China Craton: Insights on evolution of lower continental crust, Chem. Geol., 211, 87 – 109. Lustrino, M. (2005), How the delamination and detachment of lower crust can influence basaltic magmatism, Earth Sci. Rev., 72, 21 – 38. Macdonald, G. A., and T. Katsura (1964), Chemical compositional of Hawaiian lavas, J. Petrol., 5, 82 – 133. Mahoney, J. J., W. B. Jones, F. A. Frey, V. J. M. Salters, D. G. Pyle, and H. L. Davies (1995), Geochemical characteristics of lavas from Broken Ridge, the Naturaliste Plateau and southernmost Kerguelen Plateau: Cretaceous plateau volcanism in the southeast Indian-Ocean, Chem. Geol., 120, 315 – 345. Mahoney, J. J., D. W. Graham, D. M. Christie, K. T. M. Johnson, L. S. Hall, and D. L. Vonderhaar (2002), Between a hotspot and a cold spot: Isotopic variation in the Southeast Indian Ridge asthenosphere, 86°E – 118°E, J. Petrol., 43, 1155 – 1176. Mattielli, N., D. Weis, J. S. Scoates, N. Shimizu, J. P. Mennessier, M. Gre´goire, J. Y. Cottin, and A. Giret (1999), Evolution of heterogeneous lithospheric mantle in a plume environment beneath the Kerguelen Archipelago, J. Petrol., 40, 1721 – 1744. 33 of 34  Geochemistry Geophysics Geosystems  3  G  xu et al.: flood basalts in kerguelen archipelago 10.1029/2007GC001608  Mattielli, N., D. Weis, J. Blichert-Toft, and F. Albare`de (2002), Hf isotope evidence for a Miocene change in the Kerguelen Mantle Plume composition, J. Petrol., 43, 1327 – 1339. Neal, C. R., J. J. Mahoney, and W. J. Chazey III (2002), Mantle sources and the highly variable role of continental lithosphere in basalt petrogenesis of the Kerguelen Plateau and Broken Ridge LIP: Results from ODP Leg 183, J. Petrol., 43, 1177 – 1205. Nelson, S. T., and A. Montana (1992), Sieve-textured plagioclase in volcanic-rocks produced by rapid decompression, Am. Mineral., 77, 1242 – 1249. Nicolaysen, K., F. A. Frey, K. V. Hodges, D. Weis, and A. Giret (2000), 40Ar-39Ar geochronology of flood basalts from the Kerguelen Archipelago, southern Indian Ocean: Implications for Cenozoic eruption rates of the Kerguelen Plume, Earth Planet. Sci. Lett., 174, 313 – 328. Plank, T., and C. H. Langmuir (1998), The chemical composition of subducting sediment and its consequences for the crust and mantle, Chem. Geol., 145, 325 – 394. Putirka, K. D., H. Mikaelian, F. Ryerson, and H. Shaw (2003), New clinopyroxene-liquid thermobarometers for maric, evolved, and volatile-bearing lava compositions, with applications to lavas from Tibet and the Snake River Plain, Idaho, Am. Mineral., 88, 1542 – 1554. Rehka¨mper, M., and A. W. Hofmann (1997), Recycled ocean crust and sediment in Indian Ocean MORB, Earth Planet. Sci. Lett., 147, 93 – 106. Rhodes, J. M. (1996), Geochemical stratigraphy of lava flows sampled by the Hawaii Scientific Drilling Project, J. Geophys. Res., 101, 11,729 – 11,746. Rudnick, R. L., and S. Gao (2004), Composition of the continental crust, in Treatise on Geochemistry, edited by H. D. Holland and K. K. Turekian, pp. 1 – 64, Elsevier-Pergamon, Oxford, U. K. Scoates, J., M. Lo Cascio, D. Weis, and D. Lindsley (2006), Experimental constraints on the origin and evolution of mildly alkalic basalts from the Kerguelen Archipelago, southeast Indian Ocean, Contrib. Mineral. Petrol., 151, 582 – 599. Shaw, D. M., A. P. Dickin, H. Li, R. H. McNutt, H. P. Schwarcz, and M. G. Truscott (1994), Crustal geochemistry in the Wawa-Foleyet Region, Ontario, Can. J. Earth Sci., 31, 1104 – 1121. Simon, L., and C. Le´cuyer (2005), Continental recycling: The oxygen isotope point of view, Geochem. Geophys. Geosyst., 6, Q08004, doi:10.1029/2005GC000958. Sun, S.-S. and W. F. McDonough (1989), Chemical and isotopic systematics of oceanic basalts: Implications for mantle composition and processes, in Magmatism in the Ocean Basins, edited by A. D. Saunders and M. J. Norry, Geol. Soc. Spec. Publ., 42, 313 – 345. Todt, W., R. A. Cliff, A. Hanser, and A. W. Hofmann (1996), Evaluation of a 202Pb/205Pb double spike for high-precision  lead isotope analysis, in Earth Processes: Reading the Isotopic Code, Geophys. Monogr. Ser., vol. 95, edited by A. Basu, and S. R. Hart, pp. 429 – 437, AGU, Washington, D. C. Vervoort, J. D., P. J. Patchett, J. Blichert-Toft, and F. Albare`de (1999), Relationships between Lu-Hf and Sm-Nd isotopic systems in the global sedimentary system, Earth Planet. Sci. Lett., 168, 79 – 99. Wallace, P. J., F. A. Frey, D. Weis, and M. F. Coffin (2002), Origin and evolution of the Kerguelen Plateau, Broken Ridge and Kerguelen Archipelago: Editorial, J. Petrol., 43, 1105 – 1108. Weis, D., and F. A. Frey (2002), Submarine basalts of the northern Kerguelen Plateau: Interaction between the Kerguelen Plume and the Southeast Indian Ridge revealed at ODP Site 1140, J. Petrol., 43, 1287 – 1309. Weis, D., F. A. Frey, H. Leyrit, and I. Gautier (1993), Kerguelen Archipelago revisited: Geochemical and isotopic study of the southeast province lavas, Earth Planet. Sci. Lett., 118, 101 – 119. Weis, D., F. A. Frey, A. Giret, and J. M. Cantagrel (1998), Geochemical characteristics of the youngest volcano (Mount Ross) in the Kerguelen Archipelago: Inferences for magma flux, lithosphere assimilation and composition of the Kerguelen Plume, J. Petrol., 39, 973 – 994. Weis, D., S. B. Shirey, and F. A. Frey (2000), Re-Os systematics of Kerguelen Plume basalts: Enriched components and lower mantle source, Eos Trans. AGU, 81(48), Fall Meet. Suppl., Abstract V11C-09. Weis, D., S. Ingle, D. Damasceno, F. A. Frey, K. Nicolaysen, J. Barling, and L. S. S. Party (2001), Origin of continental components in Indian Ocean basalts: Evidence from Elan Bank (Kerguelen Plateau, ODP Leg 183, Site 1137), Geology, 29, 147 – 150. Weis, D., B. Kieffer, C. Maerschalk, W. Pretorius, and J. Barling (2005), High-precision Pb-Sr-Nd-Hf isotopic characterization of USGS BHVO-1 and BHVO-2 reference materials, Geochem. Geophys. Geosyst., 6, Q02002, doi:10.1029/ 2004GC000852. Weis, D., et al. (2006), High-precision isotopic characterization of USGS reference materials by TIMS and MC-ICP-MS, Geochem. Geophys. Geosyst., 7, Q08006, doi:10.1029/ 2006GC001283. Willbold, M., and A. Stracke (2006), Trace element composition of mantle end-members: Implications for recycling of oceanic and upper and lower continental crust, Geochem. Geophys. Geosyst., 7, Q04004, doi:10.1029/2005GC001005. Yang, H.-J., F. A. Frey, D. Weis, A. Giret, D. Pyle, and G. Michon (1998), Petrogenesis of the flood basalts forming the northern Kerguelen Archipelago: Implications for the Kerguelen Plume, J. Petrol., 39, 711 – 748. Zhang, S. Q., J. J. Mahoney, X. X. Mo, A. M. Ghazi, L. Milani, A. J. Crawford, T. Y. Guo, and Z. D. Zhao (2005), Evidence for a widespread tethyan upper mantle with Indian-Oceantype isotopic characteristics, J. Petrol., 46, 829 – 858.  34 of 34  

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 7 3
France 4 0
Japan 4 0
China 3 7
City Views Downloads
Ashburn 5 0
Tokyo 4 0
Unknown 4 5
Beijing 2 0
Shenzhen 1 6
Sunnyvale 1 0
University Park 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.38318.1-0075972/manifest

Comment

Related Items