UBC Faculty Research and Publications

Constraining the components of the Kerguelen mantle plume: A Hf-Pb-Sr-Nd isotopic study of picrites and… Scoates, James S.; Weis, Dominique 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-Weis_AGU_2004GC000806.pdf [ 1.4MB ]
Metadata
JSON: 52383-1.0075977.json
JSON-LD: 52383-1.0075977-ld.json
RDF/XML (Pretty): 52383-1.0075977-rdf.xml
RDF/JSON: 52383-1.0075977-rdf.json
Turtle: 52383-1.0075977-turtle.txt
N-Triples: 52383-1.0075977-rdf-ntriples.txt
Original Record: 52383-1.0075977-source.json
Full Text
52383-1.0075977-fulltext.txt
Citation
52383-1.0075977.ris

Full Text

Geochemistry Geophysics Geosystems  3  G  Article Volume 6, Number 4 19 April 2005 Q04007, doi:10.1029/2004GC000806  AN ELECTRONIC JOURNAL OF THE EARTH SCIENCES Published by AGU and the Geochemical Society  ISSN: 1525-2027  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 Sonia Doucet DSTE, Universite´ Libre de Bruxelles, Avenue F.D. Roosevelt, 50, CP 160/02, Brussels, Belgium (sonia.mo.doucet@wanadoo.fr)  James S. Scoates and Dominique Weis DSTE, Universite´ Libre de Bruxelles, Avenue F.D. Roosevelt, 50, CP 160/02, Brussels, Belgium Now at Pacific Centre for Isotopic and Geochemical Research, Department of Earth and Ocean Sciences, 6339 Stores Road, University of British Columbia, Vancouver, British Columbia, Canada V6T1Z4  Andre´ Giret Laboratoire Magmas et Volcans, CNRS-UMR 6524, Universite´ Jean Monnet, 23 Rue Paul Michelon, 42023 St. Etienne, France  [1] We report geochemical and Hf-Pb-Sr-Nd isotopic analyses of a suite of picrites and associated highMgO volcanic rocks (6–17 wt.% MgO) from the Kerguelen Archipelago, which are rare compared to other oceanic islands, to better constrain the nature and the origin of components present in the Kerguelen mantle plume source. The Sr and Nd isotopic compositions of the transitional to mildly alkalic picrites and highMgO basalts closely match those of the 24 Ma mildly alkalic basalts from the Courbet Peninsula, whose compositions are considered to reflect the present geochemical expression of the enriched component of the Kerguelen mantle plume. However, linear trends in Pb isotopic compositions in the studied samples reflect involvement of a component with lower 206Pb/204Pb and 208Pb/204Pb than that inferred for the enriched Kerguelen plume. Contamination of the MgO-rich magmas by the Kerguelen Plateau cannot account for the observed Hf-Pb-Sr-Nd isotopic variations. Isotopic systematics in the picrites and the highMgO basalts are inconsistent with simple binary mixing between two distinct end-members and indicate the presence of small-scale heterogeneities within the Kerguelen plume itself as has been observed in other hot spot environments such as Hawaii and Iceland. The 34 to 26 Ma Kerguelen plume-related basalts that formed when the archipelago was close to the ridge axis ($50 to 250 km) show geochemical evidence for significant involvement of a Southeast Indian Ridge (SEIR)-like source. In contrast, the 24–25 Ma mildly alkalic basalts from the eastern and southeastern parts of the archipelago, which erupted about 400 km away from the SEIR, and the picrites and high-MgO basalts from this study show little or no contribution from a SEIR-like component. Thus chemical interaction between the SEIR axis and the Kerguelen plume effectively ceased prior to 25 Ma. Components: 14,117 words, 15 figures, 4 tables. Keywords: Kerguelen Archipelago; picrites; Hf-Pb-Sr-Nd isotopes; mantle plume components. Index Terms: 1025 Geochemistry: Composition of the mantle; 1040 Geochemistry: Radiogenic isotope geochemistry; 1065 Geochemistry: Major and trace element geochemistry; 3640 Mineralogy and Petrology: Igneous petrology; 9340 Geographic Location: Indian Ocean. Received 21 July 2004; Revised 29 November 2004; Accepted 1 February 2005; Published 19 April 2005.  Copyright 2005 by the American Geophysical Union  1 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Doucet, S., J. S. Scoates, D. Weis, and A. Giret (2005), Constraining the components of the Kerguelen mantle plume: A HfPb-Sr-Nd isotopic study of picrites and high-MgO basalts from the Kerguelen Archipelago, Geochem. Geophys. Geosyst., 6, Q04007, doi:10.1029/2004GC000806.  1. Introduction [2] Ocean island basalts (OIB) are related to the activity of mantle plumes [e.g., Wilson, 1963; Morgan, 1972] and show large geochemical variations compared to mid-ocean ridge basalts (MORB) generated at ocean spreading ridges. The geochemical variability in oceanic island basalts usually requires the involvement of more than two components [e.g., Fitton et al., 1997, 2003; Kempton et al., 2000; Harpp and White, 2001; Abouchami et al., 2000]. Proposed origins for the geochemical variations in hot spot basalts are variable and a matter of debate. For example, these variations have been interpreted as reflecting the presence of recycled components from the oceanic and/or continental crust/lithosphere in the plume source [e.g., Zindler and Hart, 1986], or as resulting from the preservation of compositional heterogeneities from the thermal boundary layer where the plume was initiated [e.g., Farnetani et al., 2002]. Studies of geochemical variations in hot spot basalts are important for understanding the interactions between different geochemical reservoirs and the composition, origin and location of geochemical reservoirs within the Earth. [3] Kerguelen volcanic products are particularly interesting to study from a geochemical perspective as they represent the surface expression of a mantle plume with over 132 Myr of activity, involving several distinctive tectonic environments. The geochemical characteristics of Kerguelen basalts allow to provide constraints on interactions between source components as the tectonic setting evolved (Figure 1). Some of the continental flood basalts present from 132 to 114 Ma at Indian Ocean margins and the major volcanic features generated in the Indian Ocean from 119 Ma to the present are commonly ascribed to the activity of the Kerguelen mantle plume [e.g., Weis et al., 1991; Frey et al., 1996; 2000a; Coffin et al., 2002; Kent et al., 1997, 2002; Ingle et al., 2002a]. From 132 Ma to 118 Ma, when the Bunbury and Naturaliste Plateaus and the Rajmahal traps formed, the hot spot was located beneath or close to continental margins. The isotopic geochemistry of these basalts is consistent with mixing between the inferred  enriched component of the Kerguelen plume and continental lithospheric components [e.g., Frey et al., 1996, Mahoney et al., 1983, 1995; Kent et al., 1997, 2002]. The Southern Kerguelen Plateau and Elan Bank (119–107 Ma), the Central Kerguelen Plateau (94–95 Ma), plus its conjugate Broken Ridge ($95 Ma) (ages from Whitechurch et al. [1992]; Coffin et al. [2002]; Duncan [2002]), formed by eruption of large volumes of basalt during several magmatic episodes that occurred from 119 to 95 Ma. Subsequent volcanic activity was responsible for the formation of the Ninetyeast Ridge ($80 to $40 Ma), Skiff Bank (68 Ma), the Northern Kerguelen Plateau ($34 Ma), the Kerguelen Archipelago (29 Ma to present), and HeardMacDonald Islands [Duncan, 1978, 1991, 2002; Weis et al., 1993, 1998b; Barling et al., 1994; Yang et al., 1998; Frey et al., 2000b; 2002b; Nicolaysen et al., 2000; Doucet et al., 2002; Kieffer et al., 2002]. There is no evidence for continental crust contamination in the Ninetyeast Ridge and postCretaceous lavas. The geochemistry of basalts that formed at 34 Ma at ODP Site 1140 (Leg 183), when the ridge axis was $50 km away from the Kerguelen hot spot, shows strong evidence for simple binary mixing between the enriched component of the Kerguelen plume and shallow asthenosphere that is the source for MORB along the Southeast Indian Ridge (SEIR) [Weis and Frey, 2002]. Volcanism on the archipelago younger than 30 Ma, which occurred as the distance between the hot spot and the SEIR increased, was accompanied by a decreasing geochemical contribution of a SEIR-like component [e.g., Gautier et al., 1990; Weis et al., 1993, 1998b; Yang et al., 1998; Doucet et al., 2002]. [4] The Kerguelen Archipelago is an important laboratory for testing interactions between hot spot and ridge geochemical reservoirs. The archipelago is composed of lavas, mainly basalts, that formed after 40 Ma, during which time the hot spot-ridge distance increased from 0 to $1200 km. Therefore the hypothesis of shallow ambient asthenosphere as the source for the depleted component present in a plume context can be evaluated by studying basalts from different time slices that are exposed in stratigraphic sections across the archipelago. 2 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 1. Bathymetric map of the Indian Ocean from Smith and Sandwell [1997] with hot spot locations from Mu¨ller et al. [1993] and from Steinberger [2000]. The major volcanic features in the Indian Ocean are indicated. NKP, CKP, and SKP refer to the Northern, Central, and Southern Kerguelen Plateaus, respectively. EB refers to Elan Bank, a western salient of the Kerguelen Plateau. ANR and ASP refer to the Aphanasey Nikitin Rise and the Amsterdam-St. Paul Plateau, respectively. Numbers are locations of different drill holes from ODP Legs 119, 120, and 183 on the Kerguelen Plateau and are referred to in the text.  Intensive sampling on the Kerguelen Archipelago during the last 15 years as part of the French mapping program CartoKer has allowed for the characterization of two major isotopic endmembers contributing to the genesis of Kerguelen Archipelago basalts: (1) an enriched component of the Kerguelen plume identified in all basalts and (2) a SEIR-like MORB component that decreases in contribution during relative migration of the  SEIR away from the hot spot during the last 40 Myr [e.g., Gautier et al., 1990; Weis et al., 1993; Yang et al., 1998; Doucet et al., 2002; Frey et al., 2002b]. Volcanic and plutonic rocks younger than 10 Ma on the Kerguelen Archipelago, which represent limited surface exposure on the archipelago (<5%), indicate the involvement of an additional component that is not recorded in the 24 – 29 Ma basalts [Weis et al., 1993, 1998b; 3 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 2. Simplified geological map of the Kerguelen Archipelago (after Nougier [1971]) showing the distribution of flood basalts (85% of the surface area), plutonic complexes (5%), and Quaternary deposits (10%). The locations of basaltic sections that have been studied are labeled with solid black circles, and the sample sites for the picrites and high-MgO basalts in this study are indicated by the large filled circles. Samples from the Aubert de la Ru¨e area are shown in red, and all other samples are shown in blue. Ar-Ar ages that are indicated close to section names are from Nicolaysen et al. [2000].  Mattielli et al., 2002]. The origin of this component is unclear; it could either reflect contamination of basaltic melts from the plume by assimilation of Kerguelen Plateau basement, or small-scale source heterogeneities within the Kerguelen plume itself.  2. Basaltic Rocks of the Kerguelen Archipelago [ 5 ] The 7000 km 2 Kerguelen Archipelago (Figure 2) represents the surface expression of  the last 40 Myr of activity of the Kerguelen mantle plume. It is presently located 1200 km to the southwest of the SEIR on the submarine Northern Kerguelen Plateau. The Northern Plateau formed after separation of the Central Kerguelen Plateau from Broken Ridge by spreading along the SEIR that began around 40 Ma, as inferred from estimates of the crustal thickness of the Kerguelen Archipelago [Charvis et al., 1995], from estimates of eruption rates on the archipelago [Nicolaysen et al., 2000], and paleogeographic reconstructions [e.g., Royer and Sandwell, 1989]. Plateau flood 4 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  basalts are the dominant volcanic products on the archipelago, covering 85% of the surface. They are intruded by plutonic rocks (gabbros to granites or syenites representing 5% of the surface area) with mantle isotopic signatures [Watkins et al., 1974; Giret and Lameyre, 1983; Giret, 1990; Weis and Giret, 1994; Gagnevin et al., 2003]. Glacial erosion has exposed outcrops of thick basaltic sections (0.4–1 km). Flow contacts have a general slope of 5° toward the southeast throughout most of the archipelago [Giret et al., 1992], except where the topography has been locally perturbed by the occurrence of plutonic bodies. Basaltic flows on the Kerguelen Archipelago are mostly aphyric and 1–5 m thick, but may exceptionally reach up to 40–50 m in thickness. Ar-Ar dating of the upper and lower flows of different basaltic sections across the archipelago indicates that these basaltic sections were constructed within $1 Myr, and that the exposed basaltic flows erupted between 29 and 24 Ma [Nicolaysen et al., 2000; Yang et al., 1998; Doucet et al., 2002; Frey et al., 2002b]. Older ages (28–29 Ma) are recorded in the north-central and northwestern parts of the archipelago (Mt. Bureau, Mt. Rabouille`re, Mt. des Ruches and Mt. Fontaine sections). An intermediate age (26 Ma) is recorded in the central Kerguelen Archipelago (Mt. Tourmente section) and the youngest ages (24–25 Ma) are found in the eastern (Mt. Crozier section on the Courbet Peninsula) and southeastern parts of the archipelago (Charbon and Jaune sections in the Southeast Province). During this period of time, from 29 to 24 Ma, the hot spot-ridge distance increased about 200 km, with the ridge moving from 225 to 400 km away from the hot spot when considering opening rates of 35 mm.yÀ1 [e.g., Royer and Sandwell, 1989]. Intensive stratigraphic geochemical investigations (at the individual flow scale) of different sections were designed to obtain a detailed temporal geochemical record of the sources and processes involved in the origin of the Kerguelen Archipelago basalts as the SEIR migrated away from the hot spot. [6] The older basalts (>26 Ma) exposed in the northwestern part of the Kerguelen Archipelago (Figure 2) are tholeiitic to transitional in composition, whereas those younger than 26 Ma exposed in the east and southeast are mildly alkalic. Differentiated trachytic lavas occur in the Southeast Province. There is evidence for decreasing magma supply and decreasing degree of melting with decreasing eruption age of the flood basalts on the Kerguelen Archipelago from the northwest to the southeast [Weis et al., 1998a; Frey et al.,  10.1029/2004GC000806  2000b; Damasceno et al., 2002]. In addition, there is a temporal evolution of isotopic composition. Basalts from the 26–29 Ma have variable isotopic compositions with both low and high 87Sr/86Sr ratios reflecting mixture of the Kerguelen plume component with a SEIR-like component. In contrast, basalts from the 24 –25 Ma sections are characterized by the enriched Kerguelen plume composition (see references cited for the Kerguelen Archipelago). The composition of the enriched component present within the Kerguelen plume has been estimated on the basis of the highest 206 Pb/204Pb (18.6) on the archipelago, which occurs in the mildly alkalic basalts from the 24 Ma Mt. Crozier basaltic section on the Courbet Peninsula (Figure 2) [Weis et al., 1998a, 2002]). [7] The last field mission of the French CartoKer geological mapping program (Nov–Dec 1999) was dedicated to examining the central portion of the southeastern part of the archipelago (Figure 2). According to the general distribution of the ages on the archipelago, and the regional slope to the southeast of basaltic flows, the ages of the basalts in this part of the archipelago are expected to range between $24 and $26 Ma. During this investigation, we noted the systematic occurrence of small quantities of cobble-sized glacially eroded and rounded fragments of olivine-clinopyroxene ± plagioclase-phyric basalts (i.e., basalts with elevated MgO contents) within moraines across the region. MgO-rich volcanic rocks do not occur in the outcropping basaltic sections from this area and rocks with MgO >6 wt.% are not common on the Kerguelen Archipelago (61 of 258 analyzed samples; see Figure 3). Most of the basalts average 3– 5 wt.% MgO and basalts with higher MgO contents (6–13 wt.% MgO) occur in the northwestern part of the archipelago in the 28–29 Ma Mt. Bureau, Mt. Rabouille`re, Mt. des Ruches and Mt. Fontaine basaltic sections [Yang et al., 1998; Doucet et al., 2002] (Figure 2). [8] We collected olivine- and clinopyroxene-phyric samples in moraines from five different sites between the Courbet and Aubert de la Ru¨e Peninsulas; the location of the sample sites is indicated on Figure 2. At each sampling site, we chose samples to encompass the variety of petrographic types present in the phenocryst-rich cobbles. Although we do not have precise ages for these samples, on the basis of the local orientations of glacially carved valleys relative to each of the sample sites for this study, we can estimate the general directions from which material was derived and what 5 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 3. Histogram of MgO (wt%) for Kerguelen Archipelago basalts (total number of samples is 258). The vast majority of basalts from the Kerguelen Archipelago have 3– 5 wt% MgO. The picrites and high-MgO basalts from this study are highlighted in blue. Data references are from Gautier et al. [1990], Weis et al. [1993, 1998b], Frey et al. [2000b, 2002b], Yang et al. [1998], Doucet et al. [2002], and D. Weis (unpublished data, 2000). The inset shows a representative thin section of a picrite (AG99-34) that contains abundant rounded olivine phenocrysts and minor clinopyroxene phenocrysts.  age this material might be. The samples collected from Armor (AG99-41 to 43) were likely eroded from the central Kerguelen Archipelago or Plateau Central to the northwest (upslope) of the sample site (Figure 2), which correspond to an age between 24–26 Ma considering the general distribution of the ages on the archipelago. The samples from the Anse de Notothe´nia (AG99-121 to 127) would have been derived predominantly from the northwest as well, but in an area that has been extensively eroded near the north-central shore of the Kerguelen Archipelago (Figure 2). The trace element and isotopic geochemistry of these samples (see sections to follow) is similar to the younger 24–25 Ma mildly alkalic basalts from the Courbet Peninsula and not to the 29 Ma tholeiitic-transitional basalts from the Mt. Rabouil-  le`re section to the northwest, suggesting that the Anse de Notothe´nia samples may represent material of similar ages as those from the Courbet Peninsula. The sample from the Courbet Peninsula (AG99-182) was likely derived and transported from the north (upslope) relative to the sample site, from 24–25 Ma mildly alkalic basalts. The sample from the Southeast Province on the Presqu’ıˆle de Jeanne d’Arc was likely derived and transported from the southwest (upslope) relative to the sample site, from the nearby 25 Ma Ravin du Charbon section. Finally, the samples collected from the lower eastern flanks of the Aubert de la Ru¨e Peninsula (AG99-34 to 38) (Figure 2) were likely derived and transported from the west, from higher levels (currently exposed or completely eroded) on the peninsula, or perhaps 6 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  from the young Mt. Ross stratovolcano further to the west. We cannot exclude the possibility that some of these samples are dikes or sills that could be younger than the surrounding basaltic sections. However, dikes and sills of olivine- ± clinopyroxene-phyric basalt have not been documented during the systematic remapping of the Kerguelen Archipelago during the CartoKer missions. [9] We report here the first extensive study of olivine-clinopyroxene-phyric picrites (4 samples; 13–17 wt.% MgO) and moderate to high-MgO basalts (13 samples; 6–10 wt% MgO) from the central to southeast parts of the Kerguelen Archipelago. We use the geochemical characteristics of these high-MgO basalts and picrites in an effort (1) to better characterize the composition of the Kerguelen mantle plume and (2) to identify the different source components involved during formation of the Kerguelen Archipelago basalts as the setting evolved from a ridge-centered to an intraplate position over the last 40 Myr (i.e., as the setting evolved from a position comparable to Iceland today to a position comparable to Hawaii).  3. Petrography and Mineral Chemistry [10] Most of the samples are relatively unaltered in thin section, except for sample AG99-153, and to a much lesser extent, samples AG99-36, 121, 125, and 126. The picrites and high-MgO basalts are characterized by the presence of 5–20 vol.% of 1– 15 mm-sized olivine phenocrysts. Several samples also contain abundant (15–20 vol.%), 1–20 mm clinopyroxene phenocrysts (e.g., AG99-36 and 153) and several other samples also contain 5–15 vol.% of 2–10 mm plagioclase phenocrysts (e.g., AG99-38, 41, 42, 43). The majority of the olivine phenocrysts are euhedral and slightly iddingsitized and serpentinized. Rounded olivines are rare, although they are particularly abundant in sample AG99-34 (Figure 3). In AG99-36, olivine coronas surround each of the clinopyroxene phenocrysts. Alteration in sample AG99-153 is reflected by the abundance of vacuoles filled by zeolites, by brown altered and devitrified groundmass, and by brown altered phenocrysts. Zeolitization is also present, but significantly less (<1 vol.%), in samples AG9935, AG99-121, AG99-125, and AG99-126. In sample AG99-34, the presence of slightly brown altered and devitrified groundmass and serpentinized and iddingsitized olivine indicates minor alteration. [11] Mineral compositions were determined by electron probe microanalysis on a fully automated  10.1029/2004GC000806  Cameca SX-50 microprobe at the University of British Columbia, operating in wavelengthdispersion mode, with the following operating conditions: excitation voltage 15 kV; beam current 20 nA; beam diameter >10 mm; peak counttime 20 s; background count-time 10s. Data reduction was done with the ‘‘PAP’’ f(rZ) method sections [Pouchou and Pichoir, 1985] using natural standards. Representative compositions of olivine and clinopyroxene phenocryst cores and rims for some of the 17 samples are reported in Tables 1a and 1b. All analyses are also available online as auxiliary material1 (Tables A1 and A2). The average compositions of olivine cores and rims are Fo80 and Fo73, respectively, and the average clinopyroxene core and rim compositions are Wo44En47Fs09 and Wo45En42Fs13, respectively (Figure 4). The most Mg-rich olivine (Fo 87 ) is from sample AG99-34.  4. Geochemistry 4.1. Analytical Techniques [12] Seventeen samples (200 – 300 cm3) were analyzed for major and trace element compositions. Surface alteration was removed using a diamond-embedded saw. The cut surfaces were abraded with sandpaper to remove the saw traces and any remaining visible alteration. The samples were coarsely crushed between tungsten-carbide plates in a hydraulic piston crusher before being reduced to powder in an agate planetary mill. [13] Major element oxides and some trace element concentrations were determined by X-ray fluorescence at the University of Massachusetts at Amherst following the analytical procedure described by Rhodes [1996]. Estimates of accuracy for analysis are detailed by Rhodes [1988]. Major element compositions are the mean of duplicate analyses. Some of the trace elements, including rare earth element concentrations, were determined by ICP-MS at ACME Laboratory (Vancouver). Interlaboratory drift and accuracy was controlled by duplicate measurements of ODP Leg 183, Site 1140 (31R1-57-61 cm) basalt, which has been measured in several different laboratories [Weis and Frey, 2002; Doucet et al., 2002]. The major and trace element concentrations are presented in Table 2, including results obtained for the sample from Site 1140. 1 Auxiliary material is available at ftp://ftp.agu.org/apend/gc/ 2004GC000806.  7 of 28  Geochemistry Geophysics Geosystems  3  picrites, transitional high-MgO basalts, and mildly alkalic high-MgO basalts, respectively. Fa and Fo represent the proportion (%) of the Fayalite and Forsterite olivine end-members, respectively.  a Major element oxides in wt.%. b Rock types 1, 2, and 3 refer to c  39.59 0.03 15.41 0.20 43.88 0.22 0.27 99.60 16 84  40.04 0.02 12.90 0.16 45.17 0.36 0.27 98.91 14 86  38.55 0.01 20.60 0.26 38.90 0.25 0.30 98.87 23 77  39.89 37.51 38.18 0.01 0.05 0.00 13.68 26.18 23.99 0.15 0.34 0.23 45.05 36.18 38.34 0.34 0.10 0.12 0.26 0.22 0.23 99.37 100.58 101.09 15 29 26 85 71 74  36.18 38.30 37.49 39.86 38.80 39.70 37.94 39.64 38.10 39.46 0.02 0.01 0.04 0.07 0.04 0.07 0.02 0.02 0.00 0.01 31.62 21.27 26.18 13.38 19.08 14.07 24.55 15.25 21.42 14.98 0.57 0.32 0.38 0.17 0.24 0.18 0.34 0.25 0.27 0.22 31.12 40.01 36.03 46.64 41.65 45.63 37.46 45.01 39.77 44.86 0.05 0.13 0.14 0.28 0.13 0.22 0.17 0.24 0.13 0.25 0.34 0.19 0.26 0.28 0.30 0.31 0.37 0.32 0.33 0.32 99.90 100.23 100.52 100.67 100.23 100.17 100.85 100.73 100.03 100.09 36 23 29 14 20 15 27 16 23 16 64 77 71 86 80 85 73 84 77 84  35.14 37.74 39.25 38.45 0.00 0.01 0.09 0.00 36.75 23.31 16.42 20.98 0.60 0.25 0.19 0.25 26.71 38.55 43.77 40.71 0.08 0.24 0.15 0.13 0.36 0.24 0.28 0.19 99.65 100.35 100.14 100.71 44 25 17 22 56 75 83 78  doucet et al.: kerguelen mantle plume  SiO2 Cr2O3 FeO MnO MgO NiO CaO Total Fac Foc  Core Rim Core Rim Core Rim Core Rim Core Rim Core Rim Core Rim Core Rim Core Rim Core Rim  Notothe´nia AG99-126 1 Notothe´nia AG99-125 2 Notothe´nia AG99-123 1 Notothe´nia AG99-122 2 Armor AG99-41 3 A. Ru¨e AG99-38 3 A. Ru¨e AG99-36 1 A. Ru¨e AG99-34 1 Location Sample Rock typeb  Table 1a. Representative Olivine Compositions From Picrites and High-MgO Basalts From the Kerguelen Archipelagoa  Notothe´nia AG99-127 3  Courbet AG99-182 2  G  10.1029/2004GC000806  [14] From the least altered samples (all with LOI < 1.7 wt%, except AG99-126 at 2.4 wt%), 12 samples were selected for Pb, Sr, Nd and Hf isotopic analyses to encompass the entire range of major and trace element compositions. The chemical procedures used are based on those initially described by Weis and Frey [1991], with slight modifications as indicated below. For chemical extraction of Pb, Sr and Nd, samples were acidleached in 6N HCl (3 to 6 steps) to remove alteration phases and the weight lost by leaching was between 30 and 60%. Duplicates (leached from separate powders, using an identical chemical procedure) were analyzed by TIMS (VG sector 54) and MC-ICP-MS (Nu Plasma). The results for these duplicates are presented in Table 3 and they are all within analytical error. Total blank values were (1 ng for each isotopic system considered, which is negligible with respect to the abundance of the elements in the dissolved samples (>16000, 50, 7000, 300 ng on average for Sr, Pb, Nd, and Hf, respectively). All isotopic measurements were performed at the Universite´ Libre de Bruxelles. Sr and Nd isotopes were measured on a multicollector thermal ionization mass spectrometer (Micromass VG Elemental Sector 54) in multidynamic mode on a single Ta and triple Re-Ta filament, respectively. Sr and Nd isotopic ratios were normalized to 86Sr/88Sr = 0.1194 and 146Nd/144Nd = 0.7219, respectively. The average 87Sr/86Sr of the NBS 987 and 143Nd/144Nd of the Rennes Nd standard [Chauvel and Blichert-Toft, 2001] during the period of our analyses were 0.710271 ± 7 (2s on 12 measurements) and 0.511961 ± 11 (2s on 7 measurements), respectively. No correction was applied to the isotopic ratios. A value for the Rennes standard of 0.511961 corresponds to the La Jolla value of 0.511856 [Chauvel and BlichertToft, 2001]. Replicate (not separately dissolved) measurements of three basalts for Nd isotopic ratios determined by the TIMS method were analyzed on the Nu Plasma MC-ICP-MS (Nu 015) at the Universite´ Libre de Bruxelles and the results overlap within error (Figure 5a). For the Pb isotopic analyses, we used the MC-ICP-MS with Tl spiking (with a 205Tl/203Tl = 2.3885) for fractionation correction [White et al., 2000]. The precision obtained by this method is about ten times better than the precision obtained by TIMS, which allows for an improved resolution of mixing trends [e.g., Blichert-Toft et al., 2003]. Note that a comparison of the TIMS and MC-ICP-MS results is given as auxiliary material (Figure A and Table B). Analysis of the NBS 981 Pb standard on the Nu Plasma MC-ICP-MS during the course of this study gave: 8 of 28  51.94 0.54 2.92 0.80 4.96 0.15 17.51 19.38 0.39 98.59 40.9 51.4 7.7  Core  Rim 50.53 1.51 2.18 0.00 9.26 0.21 13.87 21.32 0.41 99.29 45.4 41.1 13.5  Rim 47.39 2.00 7.28 0.31 7.77 0.13 14.10 19.82 0.60 99.39 45.0 44.5 10.5  Core  A. Ru¨e AG99-38 3  48.19 2.41 4.28 0.01 8.53 0.18 13.63 21.67 0.46 99.37 47.4 41.5 11.0  Rim 48.20 1.85 5.89 0.23 6.96 0.13 14.39 21.43 0.40 99.48 47.1 44.1 8.8  Core  Armor AG99-41 3  50.01 1.51 2.48 0.05 9.82 0.20 14.15 20.77 0.38 99.38 44.4 42.0 13.6  Rim 51.34 0.46 3.95 0.96 5.22 0.08 18.67 18.37 0.36 99.41 38.8 54.8 6.4  Core  Notothe´nia AG99-122 2  50.28 1.35 2.13 0.02 10.28 0.25 14.26 20.18 0.31 99.06 42.9 42.2 14.9  Rim 49.77 0.75 3.68 0.66 6.18 0.10 15.85 20.92 0.28 98.20 45.0 47.4 7.6  Core  Notothe´nia AG99-123 1  49.86 1.39 1.80 0.04 12.43 0.27 13.89 18.71 0.27 98.67 39.8 41.1 19.1  Rim 49.84 0.89 4.27 0.71 6.27 0.08 15.83 21.07 0.30 99.26 45.2 47.2 7.6  Core  Notothe´nia AG99-125 2  picrites, transitional high-MgO basalts, and mildly alkalic high-MgO basalts, respectively. Wo, En, and Fs represent the proportion (%) of Wollastonite, Enstatite, and Ferrosillite clinopyroxene end-members, respectively.  48.16 1.18 6.59 0.67 7.12 0.17 14.31 19.94 0.49 98.63 44.9 44.9 10.2  Core  A. Ru¨e AG99-36 1  49.96 0.97 4.41 1.06 5.33 0.12 15.66 20.71 0.37 98.60 44.9 47.2 7.9  a Major element oxides in wt.%. b Rock types 1, 2, and 3 refer to c  48.74 1.98 3.73 0.02 9.18 0.20 13.49 20.82 0.37 98.54 45.4 41.0 13.6  Rim  A. Ru¨e AG99-34 1  50.92 1.09 2.27 0.05 8.37 0.16 15.53 19.98 0.25 98.63 42.0 45.4 12.5  Rim  49.47 1.04 3.83 0.63 6.01 0.11 15.51 21.72 0.28 98.60 46.7 46.4 6.9  Core  Notothe´nia AG99-126 1  49.52 1.91 3.04 0.07 8.44 0.16 14.22 21.47 0.36 99.18 46.0 42.4 11.6  Rim  48.70 1.64 4.97 0.35 7.68 0.13 14.71 20.80 0.34 99.32 45.3 44.6 10.2  Core  Notothe´nia AG99-127 3  48.97 2.22 4.21 0.10 8.23 0.15 14.39 20.49 0.33 99.09 44.3 43.3 12.5  Rim  48.97 1.37 4.74 0.66 6.16 0.15 15.23 21.25 0.35 98.87 46.3 46.1 7.6  Core  Courbet AG99-182 2  G  3  SiO2 TiO2 Al2O3 Cr2O3 FeO MnO MgO CaO Na2O Total Woc Enc Fsc  Location Sample Rock typeb  Table 1b. Representative Clinopyroxene Compositions From Picrites and High-MgO Basalts From the Kerguelen Archipelagoa  Geochemistry Geophysics Geosystems  doucet et al.: kerguelen mantle plume 10.1029/2004GC000806  9 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 4. Olivine and clinopyroxene compositions in the picrites and high-MgO basalts. (a) Histogram showing the distribution of olivine core and rim forsterite contents. Olivine core compositions typically range between Fo75 – 85 and the highest forsterite content for an olivine phenocryst determined in this study is Fo87 (AG99-34). (b) Clinopyroxene compositions projected into the Diopside (Di) – Hedenbergite (Hd) – Enstatite (En) – Ferrosillite (Fs) pyroxene quadrilateral. Core and rim compositions are shown as black circles and black crosses, respectively, and are compared to the field of clinopyroxene core compositions from the 24 Ma Mt. Crozier section [Damasceno et al., 2002].  206  Pb/204Pb = 16.9402 ± 32 (189 ppm, 2s on the mean), 207Pb/204Pb = 15.4965 ± 29 (186 ppm), 208 Pb/204Pb = 36.7140 ± 71 (193 ppm). These values are consistent within analytical uncertainties with values for the NBS 981 standard obtained by triple spike analysis (206Pb/204Pb = 16.9403 ± 22, 207 Pb/ 204 Pb = 15.4974 ± 20, 208 Pb/ 204 Pb = 36.7246 ± 58 [Regelous et al., 2003]). [15] For the Hf isotopic analyses, we dissolved about 250 to 300 mg of whole rock powder following the procedure described by Blichert-Toft  et al. [1997], with slight modifications as described by Ingle et al. [2003]. The Hf isotopic compositions were analyzed in the static mode on the Nu Plasma MC-ICP-MS. Both Lu and Yb beams were monitored during the course of analysis for interference corrections on mass 176; the Yb beam was negligible. The measured Hf isotopic ratios were corrected for isobaric interference with Lu at mass 176 by monitoring the isotope 175Lu; the 176Lu interference was subtracted using a value of 37.69969 for 176Lu/175Lu [Rosman and Taylor, 1998]. The results are presented in Table 3. During 10 of 28  SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2 O P2O5 Total LOIc Na2O + K2O Al2O3/CaO Mg numberd AIe V Cr Co (ICP-MS) Ni Zn Cu (ICP-MS) Ga Rb (ICP-MS) Sr Cs (ICP-MS) Ba Y Zr Nb Hf (ICP-MS) Ta (ICP-MS) Pb Th (ICP-MS) U (ICP-MS) La (ICP-MS) Ce (ICP-MS) Pr (ICP-MS) Nd (ICP-MS) Sm (ICP-MS) Eu (ICP-MS) Gd (ICP-MS) Tb (ICP-MS) Dy (ICP-MS) Ho (ICP-MS) Er (ICP-MS) Tm (ICP-MS) Yb Lu  46.16 1.86 9.51 13.23 0.19 17.42 9.06 1.74 0.74 0.24 100.14 1.69 2.48 1.05 0.75 À0.17 192 771 81.3 503 105 83 14 15.3 249 <0.1 165 16.6 123 18.4 3.10 1.30 2 1.90 0.40 15.5 35.6 4.14 18.4 4.10 1.43 3.76 0.58 3.56 0.70 1.90 0.22 1.47 0.20  46.51 1.87 9.32 12.58 0.18 16.67 9.97 1.86 0.80 0.23 99.98 1.71 2.66 0.93 0.76 À0.12 208 974 78.5 452 97 102 14 17.3 528 <0.1 169 16.2 123 17.9 3.30 1.30 2 1.90 0.50 15.8 36.5 4.37 18.7 4.00 1.48 4.03 0.61 3.69 0.70 1.89 0.22 1.43 0.19  46.94 1.86 10.73 13.56 0.19 14.52 9.70 1.91 0.63 0.20 100.25 2.44 2.55 1.11 0.71 À0.39 196 533 62.8 323 102 73 17 12.3 246 <0.1 157 16.6 114 14.2 2.70 0.90 1 1.40 0.30 11.6 27.3 3.35 15.1 3.50 1.29 3.34 0.55 3.25 0.63 1.69 0.21 1.30 0.18  46.73 1.95 11.18 13.31 0.18 13.19 9.90 2.19 0.64 0.22 99.47 1.12 2.82 1.13 0.70 À0.04 206 472 60.1 290 103 81 17 12 265 <0.1 167 18.0 126 15.3 2.90 1.30 2 1.50 0.30 12.9 29.5 3.68 16.1 3.90 1.33 3.58 0.57 3.49 0.67 1.79 0.22 1.39 0.18  47.54 2.12 12.53 13.28 0.19 10.51 10.45 2.32 0.66 0.22 99.82 1.24 2.98 1.20 0.65 À0.18 228 380 53.1 189 109 89 20 11.1 289 <0.1 155 19.3 129 15.8 3.00 1.00 1 1.60 0.30 13.1 30.4 3.74 17.0 4.20 1.47 3.90 0.63 3.76 0.76 2.00 0.24 1.52 0.20  48.46 2.38 13.12 12.08 0.16 10.03 9.28 2.99 1.15 0.37 100.01 1.49 4.14 1.41 0.66 0.64 204 431 53.1 221 109 53 20 27.5 406 0.7 269 19.3 182 26.2 4.00 1.90 2 3.20 0.70 25 49.2 6.27 26.3 5.60 1.99 5.19 0.78 4.05 0.78 1.87 0.23 1.35 0.17  46.98 2.79 13.92 13.24 0.22 9.87 9.39 2.24 1.06 0.36 100.06 4.31 3.29 1.48 0.64 0.34 224 453 47.2 238 111 73 20 19.4 389 0.2 224 22.2 170 27.7 3.90 1.90 2 2.50 0.60 21.4 44.8 5.48 23.2 5.00 1.79 4.57 0.71 4.10 0.78 2.12 0.24 1.60 0.21  47.34 2.49 14.05 12.93 0.18 9.59 9.67 2.71 1.04 0.33 100.31 0.85 3.74 1.45 0.63 0.66 229 401 47.5 194 110 72 20 20.9 359 0.3 228 22.3 170 25.7 4.10 1.80 2 2.40 0.60 20.6 45.7 5.49 23.2 5.20 1.77 4.70 0.74 4.42 0.87 2.30 0.28 1.80 0.24  46.03 3.04 13.92 13.14 0.17 9.34 9.87 2.93 1.05 0.60 100.07 0.74 3.98 1.41 0.62 1.38 222 285 61.9 181 110 75 20 39.2 732 0.2 395 19.8 182 31.0 4.90 2.60 2 3.60 0.90 32.3 68.1 8.16 34.2 7.10 2.71 6.14 0.88 5.07 0.93 2.33 0.26 1.61 0.21  46.06 3.27 14.01 13.58 0.18 9.11 9.45 2.99 0.97 0.55 100.16 1.57 3.96 1.48 0.61 1.35 240 204 54.3 167 122 58 21 17.3 602 0.4 410 24.6 265 42.0 6.30 2.90 3 4.70 1.00 37.6 79.3 9.13 37.4 7.60 2.47 6.66 0.94 5.42 1.00 2.71 0.31 1.99 0.26  47.11 2.68 14.60 12.20 0.16 9.07 9.51 2.88 1.39 0.51 100.11 0.91 4.27 1.54 0.63 1.27 205 278 48.6 163 96 60 20 27.6 669 0.4 354 18.8 205 31.8 4.80 2.10 2 3.70 0.80 28.8 61.8 7.16 29.1 6.00 2.04 5.22 0.73 4.26 0.79 2.02 0.23 1.43 0.18  47.11 2.67 14.72 12.13 0.17 8.84 9.48 3.12 1.36 0.50 100.08 À0.04 4.48 1.55 0.63 1.48 207 272 50.4 169 103 57 20 26.6 597 0.2 342 19.3 212 33.1 5.00 2.30 2 3.80 0.90 30.4 65.2 7.51 30.7 6.10 2.14 5.48 0.76 4.45 0.82 2.09 0.25 1.47 0.19  47.02 2.68 14.62 12.24 0.16 8.79 9.46 3.02 1.33 0.49 99.81 0.38 4.35 1.55 0.63 1.38 204 258 45.7 171 101 71 20 24.3 602 0.1 355 19.1 210 32.7 4.80 2.10 3 3.70 0.80 28.1 61.7 6.94 29.7 5.70 1.97 4.96 0.71 4.21 0.75 1.96 0.23 1.40 0.19  46.57 2.99 14.77 12.93 0.18 7.75 9.84 3.02 1.52 0.61 100.17 1.06 4.53 1.50 0.58 1.73 191 200 44.2 90 96 64 20 26.2 629 0.2 369 24.5 218 38.1 4.70 2.70 2 4.10 0.90 31 65.1 8.15 34.1 6.70 2.49 6.31 0.90 4.72 0.89 2.19 0.26 1.62 0.22  47.66 2.81 15.63 12.13 0.16 7.29 9.16 3.10 1.71 0.74 100.38 1.68 4.81 1.71 0.58 1.60 178 172 44.5 123 112 57 21 36.3 890 0.3 566 22.0 218 36.9 4.97 2.63 3 4.17 0.97 36.8 77.9 9.47 38.9 7.63 3.14 6.48 0.88 4.84 0.89 2.23 0.26 1.56 0.21  44.14 2.43 12.43 11.22 0.57 6.33 19.48 2.46 0.87 0.33 100.26 8.22 3.32 0.64 0.57 1.42 240 498 48.6 210 97 70 18 9.6 347 <0.1 212 21.4 144 21.3 3.50 1.50 2 2.10 0.40 21.3 42.8 5.45 23.7 5.10 1.73 4.75 0.69 3.68 0.75 1.89 0.26 1.53 0.22  45.79 3.65 15.38 14.25 0.19 5.86 10.01 2.90 1.46 0.51 100.00 1.51 4.37 1.54 0.49 1.85 290 41 43.6 56 121 77 22 26.3 549 0.2 332 28.6 256 39.2 5.80 2.60 2 4.20 0.90 31.9 71.8 8.28 35.7 7.30 2.61 6.65 1.00 5.81 1.12 2.97 0.36 2.21 0.31  1.6 ± 0.0 0.4 ± 0.0 13.4 ± 0.2 33.3 ± 1.2 4.39 ± 0.11 21.0 ± 0.2 5.6 ± 0.2 1.95 ± 0.00 5.76 ± 0.11 0.98 ± 0.01 6.28 ± 0.01 1.3 ± 0.0 3.67 ± 0.03 0.47 ± 0.01 3.14 ± 0.03 0.44 ± 0.01  4.4 ± 0.0 1.0 ± 0.0  G  3 0.1 ± 0.0  6.0 ± 0.2  Notothe´ Notothe´ Notothe´ Notothe´ Notothe´ Notothe´ Notothe´ Location A. Ru¨e A. Ru¨e nia nia nia Courbet nia nia A. Ru¨e A. Ru¨e Armor Armor Armor nia A. Ru¨e Charbon nia 1140Sample name AG99-34 AG99-36 AG99-126 AG99-123 AG99-125 AG99-182 AG99-121 AG99-122 AG99-37 AG99-35 AG99-42 AG99-41 AG99-43 AG99-127 AG99-38 AG99-153 AG99-124 31R1f b Rock type 1 1 1 1 2 2 2 2 3 3 3 3 3 3 3 3 3 57 – 61 cm  Table 2. Major and Trace Element Compositions for Picrites and High-MgO Basalts From the Kerguelen Archipelagoa Geochemistry Geophysics Geosystems  doucet et al.: kerguelen mantle plume 10.1029/2004GC000806  11 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  the course of our analyses, replicate measurements of the Hf JMC 475 in-house standard gave 176 Hf/177Hf = 0.282158 ± 8 (2sm on 13 measurements; Figure 5b), which is within the range of previously published values for this standard [e.g., Blichert-Toft et al., 1997; Kempton et al., 2000; Chauvel and Blichert-Toft, 2001; Mattielli et al., 2002; Goolaerts et al., 2004]. We controlled our laboratory techniques and reproducibility by separately dissolving one sample (duplicate); the measurement reproduced the same value within error. Where the Hf concentrations were sufficient, we systematically replicated the measurement at the end of the session in the same way as was done for standard replicates in order to add an internal control on the potential drift during the day of the analysis, although none was observed.  4.2. Major Element Chemistry [16] Low values of loss-on-ignition (LOI) for the picrites and the high-MgO basalts of $0 to 1.7 wt% is consistent with the relatively unaltered condition of the samples observed in thin section. The minor degree of alteration is also reflected by the small number of leaching steps required to obtain a clear supernatant during acid leaching (3 to 6 steps, compared to the 8 to 12 required for typical Kerguelen Plateau basalts). Alteration in sample AG99-153 is coherent with the high LOI value of 8 wt.% (Table 2). Samples AG99-35, AG99-121, AG99-125, and AG99-126 have LOI values from 1.2 to 4.3 wt%, which is consistent with petrographic evidence for zeolitization in these samples. [17] According to the revised classification for high-MgO basalts and picrites of Le Bas [2000], four of the samples are picrites (AG99-34, 36, 123 and 126: 13–17 wt% MgO) and thirteen of the samples are high-MgO basalts (6–10 wt% MgO). In a diagram of total alkalis versus SiO2, the phenocryst-rich picritic basalts plot within the tholeiitic to transitional field, whereas the highMgO basalts are either transitional to mildly alkalic (4 of 13) or mildly alkalic (9 of 13) (Figure 6). All of the samples from this study overlap with the field for transitional to mildly alkalic basalts from  10.1029/2004GC000806  the Kerguelen Archipelago that have the lowest SiO2 contents. [18] The highest MgO content recorded in volcanic rocks from the Kerguelen Archipelago is represented by a picrite from this study (17.5 wt.% MgO). All of the picrites and most of the highMgO basalts (except for AG99-122 and -125 with MgO $10 wt.%) have whole rock Mg numbers that are too high relative to the forsterite contents of their olivine phenocrysts to represent melt compositions (Figure 7a). A similar relationship is shown for samples AG99-123, 34, and 36, with respect to the Mg number of their clinopyroxene phenocrysts (Figure 7b). Accumulation of olivine and/or clinopyroxene will lead to unsupportedly high whole rock Mg numbers relative to the equilibrium fields defined for Fe/ Mg partitioning between olivine-liquid [Roeder and Emslie, 1970] and clinopyroxene-liquid [Grove and Bryan, 1983]. Olivine and clinopyroxene core compositions from AG99-122 and 125 plot close to or within the equilibrium fields, suggesting that they may closely approximate melt compositions with $10 wt.% MgO (although the olivine compositions in AG99-122 are slightly more forsteritic than the equilibrium field, which may indicate incorporation of olivine xenocrysts). Note that the olivine compositions that plot well below the equilibrium field are rim compositions. The MgO abundance obtained for the high-MgO rocks by removing the observed modal abundance of olivine and/or clinopyroxene phenocrysts yields equilibrium whole rock compositions with $10–12 wt.% MgO. A similar range for the most MgO-rich liquid inferred to have been involved in the formation of Kerguelen Archipelago basalts was found in the Mt. des Ruches and Mt. Fontaine sections in the northwestern part of the archipelago [Doucet et al., 2002]. [19] In diagrams of major elements versus MgO (Figure 8), except for AG99-153 (altered) and AG99-124, the samples from this study are characterized by generally decreasing TiO2, Al2O3, Na 2O, P2 O5, and Al2O3/CaO with increasing  Notes to Table 2:  a Major element oxides in wt.%; trace element concentrations in b Rock types 1, 2, and 3 refer to picrites, transitional high-MgO c  ppm. Data are by XRF, except where specified. basalts, and mildly alkalic high-MgO basalts, respectively.  LOI is weight loss on ignition after 30 min at 1020°C. Mg number is calculated on the basis of atomic fractions [Mg2+/(Mg2+ + Fe2+)]; FeO and Fe2O3 have been recalculated on the basis of Fe2+/ (Fe + Fe3+) = 0.85. e AI is the Alkalinity Index (AI = 14.43 + (Na2O + K2O) À 0.37*SiO2) and represents the distance of the sample from the alkalic-tholeiitic boundary defined for Hawaiian lavas [MacDonald and Katsura, 1964]. f Reproducibility of ICP-MS data is shown with values ±1s on the mean of duplicate analysis of sample 1140-31R1; see interlaboratory comparison for the same sample 1140-31R1 by Doucet et al. [2002]. d  2+  12 of 28  2 2 2  3 3 3  AG99-125 AG99-182 AG99-122 AG99-122 AG99-122 AG99-122  AG99-37 AG99-35 AG99-41 AG99-41 AG99-41 AG99-38 AG99-124  17.4 16.7 14.5 13.2  0.178 0.705217 0.095 0.705232 0.145 0.705222 0.131 0.705213  7 6 7 7  0.1347 0.1293 0.1401 0.1464  0.512606 0.512605 0.512630 0.512655 0.512658  Picrites 10 À0.62 12.8 0.093 63 9 À0.64 16.0 0.116 63 8 À0.16 19.1 0.138 92 13 0.33 9.5 0.069 49 8 18.4834c 18.4040c 18.2893c 18.3306c  NU TIMS/NU TIMS/NU NU NU replicate TIMS/NU TIMS/NU  7.3 5.9  9.3 9.1 8.8  Mildly Alkalic High-MgO Basalts 0.155 0.705247 8 0.1255 0.512586c 5c À1.01 28.8 0.209 119 18.4394c 0.083 0.705243 7 0.1229 0.512588 7 À0.98 21.2 0.154 103 18.3578c 0.129 0.705214 6 0.1201 0.512647 16 0.18 28.8 0.209 126 18.4269c 0.512616c 20c 0.512634c 15c 0.118 0.705307 7 0.1186 0.512581 11 À1.11 20.6 0.149 92 18.4090c 0.139 0.705090 7 0.1236 0.512656 8 0.35 29.0 0.210 140 18.6769c  15.5288c 15.5492c 15.5482c 15.5454c  15.5459c 15.5352c 15.5306c 15.5310c  31c 14c 28c 9c  18c 9c 26c 19c  Pb/ Pbm 2smb  207 204  38.7934c 38.9448c 39.0509c 39.0432c  39.1039c 38.9848c 38.7670c 38.8006c  75c 35c 90c 28c  47c 31c 71c 47c  Pb/ Pbm 2smb  208 204  Lu/ Hf  0.0095c 0.282845c 0.0060c 0.282859c 0.0083c 0.282822c 0.0083c 0.282819c  6c 5c 5c 6c  5c 5c 7c 6c  Hf/ Hfm 2smb  176 177  0.0092c 0.282820c 0.0082c 0.282811c 0.0095c 0.282830c 0.0088c 0.282851c  177  176  2.57c 3.06c 1.75c 1.66c  1.68c 1.38c 2.03c 2.79c  eHfm  mildly alkalic high-MgO basalts, respectively. isotopic values considered; 2sm represents two sigma on the mean of the within run error. Data obtained by the NU method. All other data were obtained by the TIMS method.  12c 15.5363c 11c 38.9443c 40c 0.0060c 0.282776c 5c 0.16c 13c 15.5694c 11c 39.2661c 32c 0.0076c 0.282858c 7c 3.05c  14c 15.5487c 13c 39.0336c 36c 0.0061c 0.282770c 6c À0.07c 7c 15.5472c 7c 38.8598c 19c 0.0059c 0.282803c 7c 1.09c 19c 15.5513c 15c 38.9935c 52c 0.0054c 0.282830c 6c 2.06c  33c 17c 23c 10c  21c 10c 26c 20c  Pb/ Pbm 2smb  206 204  Transitional High-MgO Basalts TIMS/NU 10.5 0.111 0.705216 9 0.1494 0.512646 9 0.16 19.1 0.138 105 18.3139c TIMS/NU 10.0 0.196 0.705233 6 0.1287 0.512653 9 0.29 22.4 0.162 106 18.4051c TIMS/NU 9.6 0.168 0.705124 6 0.1355 0.512628 7 À0.20 19.2 0.139 79 18.4617c duplicates 0.512611 10 18.4600c NU 0.512620c 13c NU replicate 0.512610c 10c  TIMS/NU TIMS/NU TIMS/NU TIMS/NU NU  238 Nd/ U/ 235U/ 232Th/ Ndm 2smb eNdm 204Pb 204Pb 204Pb  143 144  doucet et al.: kerguelen mantle plume  a Rock types 1, 2, and 3 refer to picrites, transitional high-MgO basalts and, b The 2sm values refer to the last significant digits given in the table for the c  3 3  1 1 1 1  147 Sr/ Sm/ Srm 2smb 144Nd  87 86  G  3  AG99-34 AG99-36 AG99-126 AG99-123 AG99-123  Sample Rock Measurement MgO, 87Rb/ Name Typea Method wt.% 86Sr  Table 3. Sr-Nd-Pb-Hf Isotopic Compositions for Picrites and High-MgO Basalts From the Kerguelen Archipelago  Geochemistry Geophysics Geosystems  10.1029/2004GC000806  13 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 5. Measured Nd and Hf isotopic compositions and data quality control. (a) Comparative diagram of analyses of three samples for 143Nd/144Nd measured by TIMS (VG-Sector, Universite´ Libre de Bruxelles) and MC-ICP-MS (Nu 015, Universite´ Libre de Bruxelles). TIMS values are indicated in blue, and ICP-MS values are indicated in red. (b) 176Hf/177Hf values for the JMC 475 standard run during the course of analyses in this study. A minimum of five analyses of the JMC 475 standard are run before starting measurements on the samples, and one standard was analyzed between every two samples. The horizontal line across the diagram corresponds to the mean of the JMC 475 analyses, while the orange field corresponds to the 2 standard deviation of the mean.  MgO contents. Vectors pointing toward the olivine and clinopyroxene compositions in Figure 8 show graphically that the high MgO contents in the picrites are consistent with accumulation of olivine and/or clinopyroxene from mildly alkalic magmas with $10 wt.% MgO.  4.3. Trace Element Chemistry [20] Abundances of elements such as Ba, Sr, Rb, and Ce, which are sensitive to secondary alteration, are positively correlated with Nb abundances (Figure 9). The vast majority of the picrites and 14 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 6. Total alkalis versus SiO2 diagram. The alkalic-tholeiitic boundary is from [MacDonald and Katsura, 1964]. The picrites are shown as circles, and the transitional and mildly alkalic high-MgO basalts are shown as squares and triangles, respectively. Samples located near the Mt. Ross stratovolcano are shown in red, and the other samples are shown in blue. Whole rocks were assumed to have FeO/(FeO + Fe2O3) = 0.85. The high-MgO basalts can be divided into two groups on the basis of their alkalinity; the high-MgO basalts with 9.5– 10.5 wt% MgO are transitional to mildly alkalic, whereas the high-MgO basalts with 6 – 9.5 wt% MgO are all mildly alkalic. Note that the composition of sample AG99-153 in the alkalis versus silica diagram is likely to have been strongly influenced by alteration. Data sources: 24– 25 Ma mildly alkalic basalts from the Kerguelen Archipelago [Weis et al., 1993; Frey et al., 2000b; D. Weis, unpublished data, 2000], 26– 29 Ma tholeiitic-transitional basalts from the Kerguelen Archipelago [Frey et al., 2002b; Doucet et al., 2002; Yang et al., 1998], Site 1140 basalts from the Northern Kerguelen Plateau [Weis and Frey, 2002], and SEIR N-MORB [Price et al., 1986; Michard et al., 1986].  the high-MgO basalts are distributed along trends that represent relatively constant ratios for Rb/ Nb$0.8, Sr/Nb$16, Ba/Nb$10, and Ce/Nb$1.9. Some samples deviate from these general trends, especially in diagrams of Rb versus Nb and Sr versus Nb, such as AG99-36, -37 and -38, which have higher Rb/Nb, Sr/Nb, and Ba/Nb than the other picrites and high-MgO basalts. These samples with higher Ce, Ba, Sr, Rb for a given Nb were all sampled  at the Aubert de la Ru¨e location (Figure 2). The most intensely altered sample (AG99-153) has lower Rb/ Nb, as does AG99-35. These deviations from the general trends may indicate that the samples originated from distinctive parental magma compositions. Alternatively, addition or loss of mobile elements, especially Rb, during postmagmatic processes could also explain such variations. Such an assumption, however, appears inconsistent with the 15 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 7. (a) Whole rock Mg # versus olivine core forsterite compositions and (b) whole rock Mg # versus clinopyroxene core Mg #, where Mg # = (Mg/(Mg + Fe2+))*100. The blue fields indicate the equilibrium fields calculated using a Fe/Mg exchange partition coefficient of 0.30 ± 0.03 between olivine and basaltic liquid [Roeder and Emslie, 1970] and of 0.25 ± 0.05 between clinopyroxene and basaltic liquid [Grove and Bryan, 1983]. See text for explanation. Legend as in Figure 6.  observation that variations in isotopic compositions (to be discussed below) are correlated with specific trace element compositions. [21] Primitive mantle-normalized trace element patterns in the picrites and high-MgO basalts show large variations between compositions inferred for the enriched Kerguelen plume component and the SEIR-MORB source (Figure 10). Similar to other Kerguelen Archipelago basalts, the absence of negative Nb-Ta anomalies indicates that continental crust material was not involved in the genesis of these basalts [Weis et al., 1993, 1998a; Yang et al., 1998; Doucet et al., 2002; Frey et al., 2000b,  2002b]. The picrites show limited trace element compositional variations and have the least trace element-enriched patterns compared to the other MgO-rich rocks. The samples from the Aubert de la Ru¨ e Peninsula have systematically more enriched trace element patterns (higher Sr, Rb, Eu) than samples to the north. [22] The Zr and Nb concentrations of the picrites and high-MgO basalts are positively correlated and overlap with the 24–25 Ma mildly alkalic Kerguelen Archipelago basalts (Figure 11). In general, distinct Nb/Zr ratios are observed for the 26–29 Ma and 24–25 Ma Kerguelen basalts as well as for the 16 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 8 17 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  SEIR basalts. The similarity in Nb/Zr of the picrites and high-MgO basalts with the southeastern 24– 25 Ma mildly alkalic basalts suggests that they may be coeval.  4.4. Isotope Geochemistry [23] In Figure 12, we have indicated the isotopic ratios at 24 Ma, assuming that the picrites and high-MgO basalts and the 24–25 Ma mildly alkalic basalts are coeval. This assumption of age does not affect the overall interpretation of the results. The 87 Sr/86Sr values of the picrites and high-MgO basalts correlate negatively with 143Nd/144Nd (Figure 12a) and plot within the mantle trend, which indicates that the leaching steps performed prior to sample dissolution efficiently removed the alteration phases. The studied samples have limited variations in 87Sr/86Sr (0.70521 ± 11; 2s) and 143 Nd/144Nd (0.51262 ± 5; 2s), which overlap with the isotopic compositions of the 24 Ma mildly alkalic basalts from Mont Crozier section on the Courbet Peninsula [Weis et al., 2002]. The samples from Aubert de la Ru¨e are indicated in red and all other samples in blue. Despite limited 87Sr/86Sr and 143 Nd/144Nd variations in the MgO-rich rocks, the Aubert de la Ru¨e samples have distinctly more radiogenic Sr and less radiogenic Nd isotopic ratios (Figure 12a). Note that sample AG99-124 has the least radiogenic Sr isotopic composition. [24] Except for sample AG99-124, which has the most radiogenic Pb isotopic composition and is characterized by lower 208Pb/204Pb for a given 206 Pb/204Pb compared to the other high-MgO basalts and picrites, all of the samples are positively correlated in diagrams of 208Pb/204Pb versus 206 Pb/204Pb and 207Pb/204Pb versus 206Pb/204Pb (Figures 12b and 12c). The Pb isotopic compositions of the samples appear to be independent of MgO contents as both the picrites and high-MgO basalts show a similar range of isotopic compositions (206Pb/204Pb = 18.3 – 18.7, 207Pb/204Pb = 15.53 – 15.57, 208 Pb/ 204 Pb = 38.7 – 39.3). The Aubert de la Ru¨e samples have more limited Pb isotopic variations than the rest of the sample suite.  10.1029/2004GC000806  [25] Hf isotopic variations for the high-MgO basalts and picrites are relatively restricted (176Hf/177Hf = 0.28282 ± 6; 2s) and they are positively correlated with Nd isotopic ratios (Figure 12d). In a plot of eNd versus eHf, they overlap with the field for the Mt. Crozier basalts [Mattielli et al., 2002] (Figure 13c). Two of the samples from the Aubert de la Ru¨e Peninsula (AG99-37 and -38) have significantly lower eHf and nearly overlap with the field for the Mt. Ross lavas and the initial isotopic composition for basalts from Site 1137 (Elan Bank) on the Kerguelen Plateau. There is no correlation between the MgO contents of the picrites and high-MgO basalts and their Hf isotopic compositions.  5. Discussion [26] Major, trace element, and Sr-Nd isotopic compositions of the studied picrites and highMgO basalts from the Kerguelen Archipelago indicate that they were formed from similar parental magmas, characterized by about 10 wt% MgO, which accumulated and/or segregated olivine ± clinopyroxene (±plagioclase in the lowest MgO rocks). The trace element and SrNd isotopic systematics of these volcanic rocks show limited variations that overlap estimates for the enriched component of the Kerguelen plume (Figure 13a). This suggests that their parental magmas formed by partial melting of the enriched component of the Kerguelen plume. This argument is corroborated by values of DNb for the picrites and high-MgO basalts that overlap those of basalts from the southeastern Kerguelen Archipelago area (DNb = À0.03 to 0.15), and are within the range of estimates for the enriched component of the Kerguelen plume (DNb = À0.3 to 0.3). These values of DNb are shown in Figure 14 as the deviation from the lines indicated in a log (Nb/Y) versus log (Zr/Y) diagram. The DNb value is constant during mantle melting and therefore variations in the DNb value are interpreted as reflecting differences in the geochemical composition of the  Figure 8. Major elements versus SiO2 in the picrites and high-MgO basalts from the Kerguelen Archipelago. The geochemical variations of the studied samples form a general trend characterized by decreasing TiO2, Al2O3, Na2O, P2O5, and Al2O3/CaO with increasing MgO together with nearly constant CaO abundances for the large variations of MgO observed. Sample AG99-153 is strongly altered, and sample AG99-124 has lower P2O5, Al2O3, Na2O, and Al2O3/CaO. The relatively high MgO contents of the picrites are due to accumulation of olivine and/or clinopyroxene from a parental magma with $9 – 10 wt% MgO, as shown by the vectors pointing toward average olivine and clinopyroxene compositions measured in the samples. Legend as in Figure 6. 18 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 9. Nb versus Ce, Ba, Rb, and Sr (all in ppm). The good correlation of Nb with elements such as Ba and Rb, which are sensitive to secondary alteration, suggests that the samples have trace element concentrations that have not been significantly disturbed by postmagmatic alteration, except for AG99-153 and -35, which have clearly lost Rb. Ba is well correlated with Nb except for AG99-38 and for other samples from the Aubert de la Ru¨e Peninsula (red), which also have higher Rb for a given Nb content than the other picrites and high-MgO basalts. These variations are discussed in the text. Legend as in Figure 6.  source that generates parental magmas of the basalts [Fitton et al., 1997]. [27] However, Pb isotopic systematics of the picrites and high-MgO basalts reflect involvement of a component having less radiogenic Pb isotopic compositions than estimated for the enriched Kerguelen plume component (206Pb/204Pb = 18.6 and 208Pb/204Pb = 39.3) (Figure 13b). Isotopic compositions from the Mt. Crozier basalts have been  inferred to correspond to the composition of the enriched component of the Kerguelen plume because of their limited Sr, Nd and Hf isotopic variations together with the most radiogenic 206 Pb/204Pb ratios on the archipelago, which cannot be explained by contamination or by interaction with either depleted mantle or the older Kerguelen Plateau [Weis et al., 1998a]. Linear relationships between 208Pb/204Pb and 206Pb/204Pb of the picrites and high MgO basalts suggest that mixing between 19 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 10. Primitive-mantle normalized trace element diagrams for the Kerguelen Archipelago picrites and highMgO basalts. Transitional picrites (blue or red circles), transitional to mildly alkalic high-MgO basalts (blue or red squares), and mildly alkalic high-MgO basalts (blue or red triangles) are compared to the composition inferred for the enriched component of the Kerguelen plume [Frey et al., 2000b] and the SEIR N-MORB average value [Mahoney et al., 2002]. Primitive mantle normalizing values are from Sun and McDonough [1989].  two end-members occurred. The eHf versus eNd relationships are generally linear, although samples from the Aubert de la Ru¨e Peninsula have systematically lower eHf and eNd. Two of these samples are characterized by significantly lower eHf values than those estimated for the enriched plume component (Figure 13c) and they overlap in eHf-eNd compositions with the field determined for 0.1 to 2 Ma samples from the Mt. Ross volcano. Lower eHf observed in these two samples are related to lower time-integrated Lu/Hf ratios. Contamination of the source of these samples in the past by material characterized by low Lu/Hf may explain such observations. These two samples are also characterized by higher Rb/Nb, Ba/Nb, Sr/Nb, and Ce/Nb ratios than the other picrites and high-MgO basalts (Figure 9). These observations suggest that they were formed from a different parental magma than the magmas that generated the other studied picrites and high-MgO basalts. There is no simple correlation or inverse correlation between the suite of studied samples in a diagram of 206Pb/204Pb versus  176  Hf/177Hf that would allow for the identification of two different end-members with specific Hf and Pb isotopic compositions (Figure 15a). Therefore the isotopic variations observed in the picrites and the high-MgO basalts, including the two samples having slightly lower 176Hf/177Hf, indicate the presence of discrete components having slightly different Hf-Pb isotopic compositions within the source of the picrites and the high-MgO basalts. The exact nature of such discrete components is difficult to assess as different end-members could interact with each other within the geological context of the formation of the Kerguelen Archipelago, including the ambient upper mantle source for SEIR-MORB or different components within the Kerguelen plume. In addition to mantle sources, contamination of parental magmas at shallow levels by the Kerguelen Plateau lithosphere in the late stages of magma ascent could contribute to the isotopic composition of the basalts. Below we examine the possible involvement of these different components. 20 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  Figure 11. Zr versus Nb (ppm) in picrites and highMgO basalts compared to other basalts from the Kerguelen Archipelago and the SEIR. Basaltic volcanic rocks from the Kerguelen Archipelago define two distinct fields. One field (Nb/Zr = 0.10) corresponds to the 26– 29 Ma tholeiitic-transitional basalts that formed when the archipelago was relatively close to the SEIR ridge axis. This field is intermediate between the SEIR-N-MORB field (Nb/Zr = 0.05) and the field of the 24– 25 Ma mildly alkalic basalts from the Courbet Peninsula and Southeast Province (Nb/Zr = 0.15), which erupted when the archipelago was more than 450 km away from the ridge axis. The picrites and high-MgO basalts from this study fall within the field for the younger mildly alkalic basalts. Symbols and references as in Figures 3 – 10.  [28] The lower 206Pb/204Pb component involved in the formation of the studied picrites and high-MgO basalts compared to other basalts from the Kerguelen Archipelago is clearly different from the SEIRlike component because low values of 206Pb/204Pb are not associated with low 87Sr/86Sr, which is expected from such mixing (Figure 13d). These basalts formed when the archipelago was more than 400 km away from the SEIR axis, when the area where they formed was probably too far away to directly sample and be influenced by the depleted SEIR-MORB reservoir. [29] On the Kerguelen Archipelago, no evidence of interaction between ascending basaltic magma and the Kerguelen Plateau has been demonstrated in the geochemistry of the 24 – 29 Ma flood basalts. Clinopyroxene-melt thermobarometry and the  10.1029/2004GC000806  compositions of mildly alkalic basalts from the 24 Ma Mont Crozier basaltic section indicate that magmas may have stalled in deep magma storage areas, near the crust-mantle boundary, and there is no evidence of a crustal component in these basalts [Damasceno et al., 2002]. However, the Upper Miocene series (6–10 Ma) in the Southeast Province and the young stratovolcano Mont Ross (<2 Ma), which represent the last eruptive events on the archipelago and correspond to significantly lower degrees of melting, have Pb isotopic compositions trending toward lower 206Pb/204Pb ratios (Figures 15a and 15b), characteristic of some of the Kerguelen Plateau sites (e.g., ODP Site 747, Site 1137). They also have lower eHf for a given eNd in comparison to the archipelago flood basalts (Figure 13c). This has been interpreted either as reflecting assimilation of the Kerguelen Plateau by plume-derived magmas or as reflecting sampling of a particular zone of the plume characterized by slightly different eHf composition, and thus reflecting heterogeneities within the plume itself [Weis et al., 1998b; Mattielli et al., 2002]. [30] The high-MgO basalts and picrites form a linear trend in a diagram of eHf versus eNd. This trend is comparable to the general trend formed by the !24 Ma basalts, but with slightly lower eHf and eNd. Two samples (AG99-37 and -38), which have significantly lower eHf than other values used as estimates for the enriched plume component, overlap with differentiated lavas from Mt. Ross and they reflect involvement of the different component that was also involved in the genesis of the Mt. Ross lavas. On the Kerguelen Plateau, basalts from Sites 1137 on Elan Bank and 747 on the Central Kerguelen Plateau have distinctly lower eHf for a given eNd compared to basalts from the Kerguelen Archipelago and the rest of the plateau [Ingle et al., 2003]. The lower eHf observed in samples AG99-37 and -38 that overlap with the compositions of Site 1137 basalts (Figure 13c) could be due to contamination of the parental magmas by the Kerguelen Plateau. In Figures 13a, 13c, and 13d, the picrites and the high-MgO basalts also point toward compositions that are found in some basalts from the submarine Cretaceous Kerguelen Plateau/Broken Ridge, especially toward the compositional fields of basalts from Sites 747 and 1137 from the Central Kerguelen Plateau. [31] This argument, however, does not appear to be valid when examining the composition of the Site 1137-like source at 25 Ma, when the picrites and high-MgO basalts likely formed (see auxil21 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 12. Sr-Nd-Hf-Pb Isotopic variations for the picrites and high-MgO basalts from the Kerguelen Archipelago. Two sets of data are used in the four figures: measured isotopic ratios (symbols as in Figure 6) and age-corrected ratios at 24 Ma (black circles). Fields for age-corrected (24 Ma) compositions are shown with dashed lines, and fields for measured isotopic composition are shown with solid lines. Note that sample AG99-124, which has a higher 206 Pb/204Pb ratio, is not included in the fields. (a) 143Nd/144Nd versus 87Sr/86Sr; samples located near the Mt. Ross (in red) are characterized by the highest 87Sr/86Sr and the lowest 143Nd/144Nd. (b) 208Pb/204Pb versus 206Pb/204Pb; the samples are distributed along a linear trend except for sample AG99-124, which has lower 208Pb/204Pb for a given 206 Pb/204Pb compared to the other samples. (c) 207Pb/204Pb versus 206Pb/204Pb; the samples are distributed along a linear trend. (d) 176Hf/177Hf versus 143Nd/144Nd; the samples located near Mt. Ross (in red) are characterized by the less radiogenic Hf and Nd isotopic compositions.  iary material Figure B). In Pb isotopic space (Figure 13b), contamination of the parental magmas of the picrites and high-MgO basalts by assimilation of plateau resembling the compositions observed at Site 1137 can be excluded because significantly more radiogenic Pb isotopic ratios are observed in the studied rocks. As shown on Figure 13, each site drilled on the Kerguelen Plateau is characterized by distinct isotopic compositions  [e.g., Frey et al., 2002a; Ingle et al., 2002a, 2002b, 2003; Weis and Frey, 2002] and there is no appropriate candidate in the Kerguelen Plateau that is able to explain all of the isotopic variations in the highMgO basalts and picrites. There is also no correlation between MgO content and isotopic ratios in the high-MgO basalts and picrites. Such a correlation would be expected if contamination of Kerguelen plume-related magmas occurred during fraction22 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 13. Sr-Nd-Hf-Pb isotopic variations for the picrites and high-MgO basalts compared to other basalts from the Kerguelen Archipelago, Kerguelen Plateau, and SEIR. The picrites and high-MgO basalts are shown as solid dots. Red dots represent the samples from the Aubert de la Ru¨e Peninsula, and blue dots represent the other samples from this study. References as in Figure 3, except for SEIR [Mahoney et al., 2002; Chauvel and Blichert-Toft, 2001], plus Kerguelen Plateau [Frey et al., 2002a; Ingle et al., 2002a, 2003; Weis and Frey, 2002] and Broken Ridge [Neal et al., 2002]. Note that the fields are defined for isotopic ratios corrected following the eruption ages determined for each site provided in the references cited above. (a) 143Nd/144Ndi versus 87Sr/86Sri (where subscript i means age-corrected). The high-MgO basalts and picrites show limited variations and overlap those of Mt. Crozier, considered to represent the enriched component of the Kerguelen plume. (b) 208Pb/204Pb versus 206Pb/204Pb. Except for one sample (AG99124), the picrites and high-MgO basalts form linear trends in Pb-Pb isotopic diagrams toward lower ratios (206Pb/204Pb = 18.3 – 18.5, 208Pb/204Pb = 38.7– 39.2) than those measured in Mt. Crozier basalts. (c) eHf versus eNd. The picrites and high-MgO basalts are distributed along a trend that is parallel to, but slightly offset (although nearly within analytical uncertainty) to, lower eHf for a given eNd compared to the 24– 25 Ma basalts from the Kerguelen Archipelago. Two samples (AG99-37 and -38) have significantly lower eHf than the Mt. Crozier samples. (d) 87Sr/86Sr versus 206Pb/204Pb. The picrites and high-MgO basalts are characterized by variable 206Pb/204Pb compositions for nearly constant 87Sr/86Sr ratios and do not follow a simple binary mixing trend that has been drawn between the enriched Kerguelen plume component and the SEIR-MORB-like source (end-members used for this calculation are detailed by Doucet et al. [2002]).  23 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 14. Diagram of log (Nb/Y) versus log (Zr/Y) for the picrites and high-MgO basalts compared to the fields defined for other basalts from the Kerguelen Archipelago and SEIR. DNb represents the deviation from a line in a plot of log (Nb/Y) versus log (Zr/Y), which represents the limit between basalts derived from a MORB source and basalts derived from an enriched source (the enriched Iceland plume) (DNb = (1.74 + log(Nb/Y) À 1.92*log(Zr/Y) [Fitton et al., 1997]. The DNb value is constant during mantle melting, and therefore variations in the DNb value are interpreted as reflecting differences in the geochemical composition of the source that generates parental magmas of the basalts. This index has been used to identify the presence of an intrinsic depleted component within the Iceland plume, because the Icelandic basalts have constant DNb despite variations in Sr-Nd isotopic ratios that could reflect involvement of a MORB-like source. The values of DNb for the Icelandic basalts are different from MORB basalts. We observe for the Kerguelen Archipelago that the basalts with lower enrichment in trace elements are located in between the fields for SEIR N-MORB and enriched Kerguelen plume field. The picrites and the high-MgO basalts overlap the field defined for the enriched Kerguelen plume (on the lower limit of the Iceland plume field).  ation and/or assimilation processes in the course of magma ascent through the lithosphere. We do not observe low Nb/La in the picrites and high-MgO basalts, which would reflect contamination by continental crust components and which is characteristic of the contaminated Cretaceous Kerguelen Plateau lavas. All of these observations appear inconsistent with the hypothesis that the parental magmas of the picrites and high-MgO basalts were contaminated by the Kerguelen Plateau.  6. Summary and Conclusions [32] We report the first detailed geochemical and isotopic study of a relatively large suite of high-  MgO volcanic rocks on the Kerguelen Archipelago, the emergent part of the larger Northern Kerguelen Plateau in the Southern Indian Ocean. The small variations in trace element, and isotopic compositions of the picrites and high-MgO basalts together with the significant Pb and Hf isotopic variations observed lead to the following conclusions: [33] (1) The source for the studied picrites and high-MgO basalts shares compositional characteristics of the enriched Kerguelen plume endmember. (2) The Kerguelen plume is relatively homogeneous in composition with respect to trace element and Sr-Nd isotopic compositions. This is highlighted by the nearly constant Sr-Nd compo24 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  10.1029/2004GC000806  Figure 15. 176Hf/177Hf versus 206Pb/204Pb for the picrites and high-MgO basalts compared to other Kerguelen Archipelago and Kerguelen hot spot related basalts. Fields and data points (a) for all Kerguelen-related basalts and for SEIR-MORB and (b) for Kerguelen Archipelago basalts. The picrites and high-MgO basalts are characterized by variations in 206Pb/204Pb for an almost constant 176Hf/177Hf, which is near the enriched plume component composition. In Figure 15b, two vectors are shown pointing from the enriched composition of the Kerguelen plume toward the SEIR-MORB end-member and toward a ‘‘third’’ end-member. Data for Hf isotopic compositions are from Ingle et al. [2003], Chauvel and Blichert-Toft [2001], and Mattielli et al. [2002]. Other references and symbols as in Figure 13.  sitions and trace element ratios in the picrites and high-MgO basalts and in the other 24 –25 Ma basalts collected in the same area, which also lack in any contribution of a SEIR-like component. (3) The Kerguelen plume is characterized by dispersed Pb-Hf isotopic heterogeneities. These heterogeneities are reflected in the variable 206 Pb/204Pb and lower eHf in the picrites and high-MgO basalts compared with those of other 24–25 Ma basalts located in the same area. These variations cannot be explained by contamination (e.g., SEIR-like contamination, overlying Kerguelen Plateau contamination) and indicate the presence of discrete heterogeneities within the Kerguelen plume itself. (4) No evidence of mixing of the enriched Kerguelen plume source with a depleted component from a SEIR-like source is found in the 24–25 Ma mildly alkalic basalts from the southeastern Kerguelen Archipelago, which erupted about 400 km away from the SEIR ridge axis, nor in the picrites and the high-MgO basalts from this study collected in the same area. [34] The geochemical variability in oceanic island basalts is generally not due to simple binary mixing between the ambient asthenosphere (source for MORB) and ‘‘primitive’’ lower mantle reservoirs, but typically requires involvement of multiple mantle components. For example, at least four mantle components are required to explain isotopic  variations in the Iceland plume-related basalts [e.g., Fitton et al., 1997, 2003; Kempton et al., 2000] and in Gala´pagos seamount volcanic rocks [e.g., Harpp and White, 2001]. At Hawaii, the presence of discrete geochemical heterogeneities within the Hawaiian mantle plume has been proposed to explain the distinctive parallel mixing lines defined by each volcano in Pb isotopic space [e.g., Abouchami et al., 2000]. The origin of these heterogeneities is still unclear. This study of picrites and high-MgO basalts from the Kerguelen Archipelago provides the first documentation for small-scale heterogeneities within the Kerguelen plume source.  Acknowledgments [35] We are grateful to the captains and crews of the Marion Dufresne II and La Curieuse, the IFRTP, and the TAAF for logistical support during the CartoKer 1999/2000 mapping program. P. Kempton and N. Arndt are thanked for discussions that allowed for clarification of many points of the first drafts of the paper. We are especially grateful for the very constructive work of the two reviewers, G. Fitton and F. Hauff, whose comments allowed for important clarifications of some key points. We also thank the G-cubed editors, C. Chauvel and W. White, for their additional comments on the manuscript. C. Maerschalk and J. de Jong are thanked for technical assistance with mass spectrometry at the Universite´ Libre de Bruxelles. We thank M. Rhodes and M. Vollinger from the University of 25 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  Massachusetts at Amherst for carrying out the XRF major and trace element analyses, H. Annell for carrying out the microprobe analyses, and M. Raudsepp for monitoring the microbeam facility at the University of British Columbia. This work was supported by an ARC grant to D. Weis (FNRS Research Director), J. Scoates and N. Mattielli (ARC 98/03-233; Actions de Recherches Concerte´es from the Communaute´ Franc¸aise de Belgique) and NSERC Discovery grants to J. Scoates and D. Weis. S. Doucet’s work was also supported by a grant from the Re´gion Rhoˆne-Alpes (Eurodoc).  References Abouchami, W., S. J. G. Galer, and A. W. Hofmann (2000), High precision lead isotope systematics of lavas from the Hawaiian Scientific Drilling Project, Chem. Geol., 169, 187 – 209. 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. 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. Blichert-Toft, J., D. Weis, C. Maerschalk, A. Agranier, and F. Albare`de (2003), Hawaiian hot spot dynamics as inferred from the Hf and Pb isotope evolution of Mauna Kea volcano, Geochem. Geophys. Geosyst., 4(2), 8704, doi:10.1029/ 2002GC000340. Charvis, P., M. Recq, S. Operto, and D. Brefort (1995), Deep structure of the northern Kerguelen Plateau and hotspotrelated activity, Geophys. J. Int., 122, 899 – 924. 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., M. S. Pringle, R. A. Duncan, T. P. Gladczenco, M. Storey, R. D. Mu¨ller, and L. A. Gahagan (2002), Kerguelen hotspot magma output since 130 Ma, J. Petrol., 43, 1121 – 1139. 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. Doucet, S., D. Weis, J. S. Scoates, K. Nicolaysen, F. A. Frey, and A. Giret (2002), Petrogenesis of high- and low-MgO transitional basalts from the Loranchet Peninsula (Mont des Ruches and Mont Fontaine), Kerguelen Archipelago, J. Petrol., 43, 1341 – 1366. Duncan, R. A. (1978), Geochronology of basalts from the Ninetyeast Ridge and continental dispersion in the Eastern Indian Ocean, J. Volcanol. Geotherm. Res., 4, 283 – 305. Duncan, R. A. (1991), The age distribution of volcanism along aseismic ridges in the eastern Indian Ocean, Proc. Ocean Drill. Program Sci. Results, 121, 507 – 517. Duncan, R. A. (2002), A timeframe for construction of the Kerguelen Plateau and Broken Ridge, J. Petrol., 43, 1109 – 1119. Farnetani, C. G., B. Legras, and P. J. Tackley (2002), Mixing and deformations in mantle plumes, Earth Planet. Sci. Lett., 196, 1 – 15.  10.1029/2004GC000806  Fitton, J. G., A. D. Saunders, M. J. Norry, B. S. Hardarson, and R. N. Taylor (1997), Thermal and chemical structure of the Iceland plume, Earth Planet. Sci. Lett., 153, 197 – 208. Fitton, J. G., A. D. Saunders, P. D. Kempton, and B. S. Hardarson (2003), Does depleted mantle form an intrinsic part of the Iceland plume?, Geochem. Geophys. Geosyst., 4(3), 1032, doi:10.1029/2002GC000424. Frey, F. A., N. J. McNaughton, D. R. Nelson, J. R. DeLaeter, and R. A. Duncan (1996), Petrogenesis of the Bunbury Basalts, Western Australia: Interaction between the Kerguelen plume and Gondwana lithosphere?, Earth Planet. Sci. Lett., 144, 163 – 183. Frey, F. A., et al. (2000a), Origin and evolution of a submarine large igneous province: The Kerguelen Plateau and Broken Ridge, southern Indian Ocean, Earth Planet. Sci. Lett., 176, 73 – 89. Frey, F. A., D. Weis, H.-J. Yang, K. Nicolaysen, H. Leyrit, and A. Giret (2000b), Temporal geochemical trends in the Kerguelen Archipelago basalts: Evidence for decreasing magma supply from the Kerguelen plume, Chem. Geol., 164, 61 – 80. Frey, F. A., D. Weis, A. Y. Borisova, and G. Xu (2002a), 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., K. Nicolaysen, B. K. Kubit, D. Weis, and A. Giret (2002b), Flood basalts from Mont Tourmente in the central Kerguelen Archipelago: The change from tholeiitic/transitional to alkali basalts at $25 Ma, J. Petrol., 43, 1367 – 1387. 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 Kerguelen basalts (South Indian Ocean): Evolution of the mantle sources from ridge to an intraplate position, Earth Planet. Sci. Lett., 100, 59 – 76. Giret, A. (1990), Typology, evolution, and origin of the Kerguelen Plutonic Series, Indian Ocean: A review, Geol. J., 25, 239 – 247. Giret, A., and J. Lameyre (1983), A study of Kerguelen plutonism: Petrology, geochronology and geological implications, in Antarctic Earth Science, pp. 646 – 651, Cambridge Univ. Press, New York. Giret, A., O. Verdier, and P. Nativel (1992), The zeolitisation model of Kerguelen Islands, southern Indian Ocean, in Recent Progress in Antarctic Earth Science, edited by Y. Yoshida, pp. 457 – 463, Terra Sci., Tokyo. Goolaerts, A., N. Mattielli, J. de Jong, D. Weis, and J. S. Scoates (2004), Hf and Lu isotopic reference values for the zircon standard 91500 by multiple collector inductively couple plasma mass spectrometry (MC-ICP-MS), Chem. Geol., 206, 1 – 9. Grove, T. L., and W. B. Bryan (1983), Fractionation of pyroxene-phyric MORB at low pressure: An experimental study, Contrib. Mineral. Petrol., 84, 293 – 309. Harpp, K. S., and W. M. White (2001), Tracing a mantle plume: Isotopic and trace element variations of Gala´pagos seamounts, Geochem. Geophys. Geosyst., 2(6), doi:10.1029/ 2000GC000137. Ingle, S., D. Weis, J. S. Scoates, and F. A. Frey (2002a), Relationship between the early Kerguelen plume and continental flood basalts of the paleo-eastern Gondwanan margins, Earth Planet. Sci. Lett., 197, 35 – 50. 26 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  Ingle, S., D. Weis, and F. A. Frey (2002b), 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. Kempton, P. D., J. G. Fitton, A. D. Saunders, G. M. Nowell, R. N. Taylor, B. S. Hardarson, and G. Pearson (2000), The Iceland plume in space and time: A Sr-Nd-Pb-Hf study of the North Atlantic rifted margins, Earth Planet. Sci. Lett., 177, 255 – 271. Kent, R. W., A. D. Saunders, P. D. Kempton, and N. C. Ghose (1997), Rajmahal basalts, Eastern India: Mantle sources and melt distribution at a volcanic rifted margin, in Large Igneous Provinces: Continental, Oceanic, and Planetary Flood Volcanism, Geophys. Monogr. Ser., vol. 100, edited by J. J. Mahoney and M. F. Coffin, pp. 145 – 182, AGU, Washington. Kent, R. W., M. P. Pringle, R. D. Mu¨ller, A. D. Saunders, and N. C. Ghose (2002), 40Ar/39Ar geochronology of the Rajmahal basalts, India, and their relationship to the Kerguelen Plateau, J. Petrol., 43, 1141 – 1153. 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. Le Bas, M. J. (2000), IUGS reclassification of the high-Mg and picritic volcanic rocks, J. Petrol., 41, 1467 – 1470. MacDonald, G. A., and T. Katsura (1964), Chemical compositions of Hawaiian lavas, J. Petrol., 5, 83 – 133. Mahoney, J. J., J. D. McDougall, G. W. Lugmair, and K. Gopalan (1983), Kerguelen hot spot source for the Rajmahal traps and Ninetyeast Ridge, Nature, 303, 385 – 389. 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. Blichert-Toft, and F. Albare`de (2002), Hf isotope evidence for a Miocene change of regime of the Kerguelen hot spot, J. Petrol., 43, 1327 – 1339. Michard, A., R. Montigny, and R. Schlich (1986), Geochemistry of the mantle below the Rodriquez Triple junction and the South-East Indian Ridge, Earth Planet. Sci. Lett., 78, 104 – 114. Morgan, W. J. (1972), Plate motions and deep mantle convection, Mem. Geol. Soc. Am., 132, 7 – 22. Mu¨ller, R. D., J.-Y. Royer, and L. A. Lawver (1993), Revised plate motions relative to the hotspots from combined Atlantic and Indian ocean hotspot tracks, Geology, 21, 275 – 278. 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. Nicolaysen, K., F. A. Frey, K. Hodges, D. Weis, and A. Giret (2000), 40Ar/39Ar geochronology of flood basalts from the Kerguelen Archipelago, southern Indian Ocean: Implications  10.1029/2004GC000806  for Cenozoic eruption rates of the Kerguelen plume, Earth Planet. Sci. Lett., 174, 313 – 328. Nougier, J. (1971), Carte ge´ologique de reconnaissance des ˆıles Kerguelen au 1/200000, Comite´ Natl. Fr. des Rech. Antarct., Paris. Pouchou, J.-L., and F. Pichoir (1985), PAP f(rZ) procedure for improved quantitative microanalysis, in Microbeam Analysis, edited by J. T. Armstron, pp. 104 – 106, San Francisco Press, San Francisco, Calif. Price, R. C., A. K. Kennedy, M. Riggs-Sneeringer, and F. A. Frey (1986), Geochemistry of basalts from the Indian Ocean triple junction: Implications for the generation and evolution of Indian Ocean ridge basalts, Earth Planet. Sci. Lett., 78, 379 – 396. Regelous, M., A. W. Hofmann, W. Abouchami, and S. J. G. Galer (2003), Geochemistry of lavas from the Emperor Seamounts, and the geochemical evolution of Hawaiian magmatism from 85 to 42 Ma, J. Petrol., 44, 113 – 140. Rhodes, J. M. (1988), Geochemistry of the 1984 Mauna Loa eruption: Implication for magma storage and supply, J. Geophys. Res., 93, 4453 – 4466. Rhodes, J. M. (1996), Geochemical stratigraphy of lava flows sampled by the Hawaii Scientific Drilling Project, J. Geophys. Res., 101, 11,729 – 11,746. Roeder, P. L., and R. F. Emslie (1970), Olivine-liquid equilibrium, Contrib. Mineral. Petrol., 29, 275 – 289. Rosman, J. R., and P. D. P. Taylor (1998), Isotopic compositions of the elements, Pure Appl. Chem., 70, 217 – 236. Royer, J.-Y., and D. T. Sandwell (1989), Evolution of the eastern Indian Ocean since the late Cretaceous: Constraints from Geosat altimetry, J. Geophys. Res., 94, 13,755 – 13,782. Smith, W. H. F., and D. Sandwell (1997), Global seafloor topography from satellite altimetry and ship depth soundings, Science, 277, 1956 – 1962. Steinberger, B. (2000), Plumes in a convecting mantle: Models and observations for individual hotspots, J. Geophys. Res., 105(B5), 11,127 – 11,152. 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. Watkins, N. D., B. M. Gunn, J. Nougier, and A. K. Baksi (1974), Kerguelen: Continental fragment or oceanic island, Geol. Soc. Am. Bull., 85, 201 – 212. Weis, D., and F. A. Frey (1991), Isotope geochemistry of Ninetyeast Ridge basalts: Sr, Nd, and Pb evidence for the involvement of the Kerguelen hot spot, Proc. Ocean Drill. Program Sci. Results, 121, 591 – 610. 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., and A. Giret (1994), Kerguelen plutonic complexes: Sr, Nd, Pb isotopic study and inferences about their sources, ages and geodynamic setting, Bull. Soc. Ge´ol. Fr., 166, 47 – 59. Weis, D., F. A. Frey, A. D. Saunders, and I. Gibson (1991), Ninetyeast Ridge (Indian Ocean): A 5000 km record of a Dupal mantle plume, Geology, 19, 99 – 102. Weis, D., F. A. Frey, H. Leyrit, and I. Gautier (1993), Kerguelen Archipelago revisited: Geochemical and isotopic study of the SE Province lavas, Earth Planet. Sci. Lett., 118, 101 – 119. 27 of 28  Geochemistry Geophysics Geosystems  3  G  doucet et al.: kerguelen mantle plume  Weis, D., D. Damasceno, F. A. Frey, K. Nicolaysen, and A. Giret (1998a), Temporal isotopic variations in the Kerguelen plume: Evidence from the Kerguelen Archipelago, Mineral. Mag., 62A, 1643 – 1644. Weis, D., F. A. Frey, A. Giret, and J. M. Cantagrel (1998b), Geochemical characteristics of the youngest volcano (Mount Ross) in the Kerguelen Archipelago: Inferences for magma flux and composition of the Kerguelen Plume, J. Petrol., 39, 973 – 994. Weis, D., F. A. Frey, R. Schlich, M. Schaming, R. Montigny, D. Damasceno, N. Mattielli, K. E. Nicolaysen, and J. S. Scoates (2002), Trace of the Kerguelen mantle plume: Evidence from seamounts between the Kerguelen Archipelago and Heard Island, Indian Ocean, Geochem. Geophys. Geosyst., 3(6), 1033, doi:10.1029/2001GC000251.  10.1029/2004GC000806  White, W. M., F. Albare`de, and P. Te´louk (2000), High-precision analysis of Pb isotope ratios by multi-collector ICP-MS, Chem. Geol., 167, 257 – 270. Whitechurch, H., R. Montigny, J. Sevigny, M. Storey, and V. Salters (1992), K-Ar and 40Ar/39Ar ages of central Kerguelen Plateau basalts, Proc. Ocean Drill. Program Sci. Results, 120, 71 – 78. Wilson, J. T. (1963), Evidence from islands on the spreading of the ocean floor, Can. J. Phys., 41, 863 – 868. Yang, Y. 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. Zindler, A., and S. Hart (1986), Chemical geodynamics, Annu. Rev. Earth Planet. Sci. Lett., 14, 493 – 571.  28 of 28  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

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.52383.1-0075977/manifest

Comment

Related Items