Chemical Geology 372 (2014) 62–71

Contents lists available at ScienceDirect

Chemical Geology

j ourna l homepage: www.e lsev ie r .com/ locate /chemgeo

Helium isotope systematics in the vicinity of the Azores triple junction:Constraints on the Azores geodynamics

Pedro Madureira a,b,⁎, Manuel Moreira c, João Mata d, João Carlos Nunes e, Cécile Gautheron f, Nuno Lourenço g,Rosário Carvalho d, Manuel Pinto de Abreu h

a Estrutura de Missão para a Extensão da Plataforma Continental, Rua Costa Pinto, 165, 2770-047 Paço d'Arcos, Portugalb Universidade de Évora, Centro de Geofísica de Évora, Departamento de Geociências, Rua Romão Ramalho, 59, 7000-671 Évora, Portugalc Equipe de Géochimie et Cosmochimie, Institut de Physique du Globe de Paris, Sorbonne Paris Cité, CNRS (UMR 7154), 1 rue Jussieu, 75238 Paris Cedex, Franced Universidade de Lisboa, Faculdade de Ciências, Departamento de Geologia (GeoFCUL), Centro de Geologia da Universidade de Lisboa (CeGUL), Edifício C6, Campo Grande,1749-016 Lisboa, Portugale Universidade dos Açores, Departamento de Geociências, R. Mãe de Deus, Apartado 1422, 9501-801 Ponta Delgada, Açores, Portugalf UMR Interactions et Dynamique des Environnements de Surface-CNRS 8148, Université Paris Sud, 91405 Orsay, Franceg Instituto Português do Mar e da Atmosfera, Rua C do Aeroporto, 1749-077 Lisboa, Portugalh MAM, Terreiro do Paço, Lisboa, Portugal

⁎ Corresponding author: Estrutura deMissãopara a ExteRua Costa Pinto, 165, 2770-047 Paço d’Arcos, Portugal. Tel

E-mail addresses: pedrom@emepc-portugal.org (P. Ma(M. Moreira), jmata@fc.ul.pt (J. Mata), jcnunes@uac.pt (J.Ccecile.gautheron@u-psud.fr (C. Gautheron), nuno.lourencmdrcarvalho@fc.ul.pt (R. Carvalho), manuel.abreu@mam.

http://dx.doi.org/10.1016/j.chemgeo.2014.02.0150009-2541/© 2014 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 17 August 2013Received in revised form 2 February 2014Accepted 17 February 2014Available online 26 February 2014

Editor: David R. Hilton

Keywords:He isotopesTerceira RiftEurasia–Nubia boundaryAzores triple junction

We present new He isotopic data from subaerial and submarine samples collected along the Terceira Rift (AzoresPlateau) in order to constrain the sources involved in the Azoresmagmatism in the unique geodynamic context oftheAzores triple junction. Lower thanMORB 4He/3He isotopic ratioswere determinedonolivines from the TerceiraIsland and the adjacent Serreta submarine ridge (down to≈69,000;≈10.5 R/Ra), which in addition to the Ne iso-topes previously published inMadureira et al. (2005) [Primitive helium and neon isotopes in Terceira Island: Con-straints on the origin of the Azores archipelago. Earth and Planetary Science Letters 233, 429–440], support thegeochemical models that consider the involvement of the lower mantle in the Azores magmatism. The analysisof He data published for the Mid-Atlantic Ridge (MAR) points to the existence of chemically distinct mantledomains underneath Eurasia and Nubia lithospheric plates, which can result from an heterogeneous distributionof asthenosphere metasomatism during the initial stages of plume activity in the Azores region. At the TerceiraRift, lavas from Graciosa Island, as well as from the western end of São Miguel Island and D. João de Castro Bank,display 4He/3He ratios similar to those observed along the MAR segments located to the north of the Azores triplejunction area. Conversely, samples from the south Hirondelle Basin and also Faial display a He isotopic signaturesimilar to that of MORB erupted along MAR segments located to the south of the Azores Plateau. The TerceiraRift and the Azores triple junction area are thus characterized by the mingling of two different asthenosphericdomains referred as “Eurasia” and “Nubia” type and by the presence of mantle plume derived material havingprimitive helium signature.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Mid-oceanic ridges represent the largest volcanic feature on Earth,from which Mid-Ocean Ridge Basalts (MORB) erupt. Unlike OceanIsland Basalts (OIB), the composition of most MORB records a markeddepletion in incompatible elements inherited from their mantle source(e.g. Hofmann, 2003; Workman and Hart, 2005). However, MORB canalso display enriched signatures in areas characterized by anomalousdepths, higher heat flow and magma production, strong negative

nsão da PlataformaContinental,: +351 213 004 165.dureira), moreira@ipgp.fr. Nunes),o@ipma.pt (N. Lourenço),gov.pt (M.P. de Abreu).

mantle Bouguer anomaly and thicker oceanic crust (e.g. Klein andLangmuir, 1987; Niu and O'Hara, 2008). In the North Atlantic, anoma-lous depths of the Mid-Atlantic Ridge (MAR) are particularly evidentin Iceland and along the Azores bathymetric anomaly (~30°N and~50°N; Schilling et al., 1983; Dosso et al., 1999). At the Azores areathis has been interpreted as resulting from the higher magma produc-tion rates triggered by the injection of deeper mantle material suppliedby a plume (e.g. Adam et al., 2013 and references therein) also feedingthe OIB which erupted at the Azores islands (Schilling, 1975).

Helium has been successfully applied to discriminate betweenMORB and OIB reservoirs and is considered fundamental to constrainmantle's structure and convection models (Kurz et al., 1982; Moreiraet al., 1995; Moreira and Allègre, 1998; Sarda et al., 2000; Graham,2002; Kurz et al., 2005). Helium has two isotopes, 3He and 4He. The lat-ter is radiogenic, being produced mostly in the radioactive chains of

63P. Madureira et al. / Chemical Geology 372 (2014) 62–71

235U, 238U and 232Th. 3He is a primordial isotope, present in Earth sinceits accretion. The 4He/3He isotopic ratio (or also R/Ra where R is the3He/4He ratio, and Ra the atmospheric ratio of 1.384 10−6) is variablein oceanic basalts. It presents amean value of 90,000± 10,000 (R/Ra=8 ± 1) in depleted MORB (e.g. Allègre et al., 1995; cf. Graham, 2002)whereas it varies from 15,000 (R/Ra = 50) at Baffin Island (Stuartet al., 2003) up to 180,000 (R/Ra ~ 4) for some basalts which eruptedat São Miguel Island-eastern Azores archipelago (Moreira et al., 1999;Jean-Baptiste et al., 2009;Moreira et al., 2012). This dichotomybetweenMORB and OIB has been interpreted as reflecting the existence of tworeservoirs in the mantle; one, more homogeneous, being sampled byMORB and another, characterized by a higher heterogeneity, which issampled bymantle plumes andmirrored in OIB. Mantle plumes containprimordial material (low 4He/3He) mixed with variable amounts ofrecycled material (sediments and/or oceanic crust) having high4He/3He isotopic ratios. In order to generate the low 4He/3He ratiosobserved in OIB, their mantle source has to be richer in primordial3He than the MORB source and consequently it is assumed as lessdegassed. Therefore, any mantle plume rising under a mid-oceanridge should be easily detectedwith the helium isotopes, which is in-deed observed inmany on-ridge hotspots such as Iceland, Saint Paul-Amsterdam, Shona, Discovery, and Easter Island (e.g. Poreda, 1993;Moreira et al., 1995; Graham et al., 1999; Stuart et al., 2003).

However, in geodynamic settings dominated by plume–ridge inter-action and magma/source mixing processes, the erupted lavas can dis-play He isotopic ratios that are dominated by the MORB componentandnot by the plume end-member. This is particularly the case of the al-kaline basalts sampled at the Azores islandswheremost of themeasured4He/3He ratios are close to the typical MORB value. Despite the “High3He” signature that has been recognized under the central Azores archi-pelago, the primitive helium signal appears to be diluted by the ambientasthenospheric mantle (Moreira et al., 1999; Madureira et al., 2005).

In order to better understand the influence of theMORB source on thehelium signature expressed by the Azores lavas, we present new He iso-topic data from subaerial and submarine samples collected along the Ter-ceira Rift (TR). The TR represents the present-day plate boundarybetween Eurasia and Nubia plates, which is considered to behave as ahyper-slow spreading system and as a place where new oceanic crustcan be generated (Vogt and Jung, 2004). Here, we combine this datawith topography and helium data from the Mid-Atlantic Ridge (MAR)in order to address: 1) the origin of the helium signatures sampled byprimitive magmas erupted along this plate boundary; 2) its relationshipwith that observed in basalts erupted from the adjacent MAR and 3) theimplications to the Azoresmelting anomaly and triple junction dynamics.

2. Geological and He-isotope background

The Terceira Rift (TR) is located at the north-eastern border of theAzores plateau, the latter representing 400,000 km2 of elevated oceaniccrust roughly underlined by the 2000–2500 m isobaths (Fig. 1A;Lourenço et al., 1998). The plateau is part of the Azores platform,which corresponds to a morphological domain bounded by the Hayes(~33.5°N) and Maxwell (~47.5°N) Fractures Zones to the south andnorth, respectively. The TR is a 550 km long ESE trending structureinterpreted as the third arm of the RRR-type (Ridge–Ridge–Ridge)Azores triple junction where the Eurasia, Nubia and North America lith-ospheric plates meet (Machado, 1959; Krause andWatkins, 1970; Vogtand Jung, 2004). At the time of the formation of the Azores plateau(~20 Ma), the triple junction was defined by the MAR and the Picoand East Azores fracture zones (Gente et al., 2003; Luis and Miranda,2008). In more recent times the location of the Eurasia–Nubia plateboundary is thought to have migrated from SW to NE under the influ-ence of the absolutemotion of the Nubia and Eurasia lithospheric platesto SW (Gripp and Gordon, 2002; Vogt and Jung, 2004). Several islandsand seamounts are aligned along the TR, alternating with deep basinssometimes reaching depths of more than 3000 m (at Hirondelle basin,

Fig. 1B). The available geodetic data show that the relative displacementbetween Eurasia and Nubia is mostly accommodated by the TR and alsoalong the Faial–Pico lineament (Fernandes et al., 2006).

Until the present work, existing data on noble gases away fromthe adjacent MAR was limited to the Azores islands, either from phe-nocrysts collected in basaltic lavas or from thermal waters and gasemissions. Studies on olivine phenocrysts point to the presence of arelatively primitive mantle source feeding the volcanic activity atTerceira, Pico and S. Jorge islands (minimum 4He/3He ratios of≈63,500; 70,150 and 50,000 respectively; R/Ra ≈11.5; 10.3 and15, respectively) (Moreira et al., 1999; Moreira and Allègre, 2002),which in the former case it is also characterized by more primitivethan MORB Ne isotopic signatures (Madureira et al., 2005). Olivinephenocrysts from the central and eastern S. Miguel lavas are charac-terized by radiogenic signatures with 4He/3He ratios of up to 180,000(R/Ra ≈4). These radiogenic lavas are interpreted as resulting fromthe presence in their mantle source of ancient (~3 Ga) recycled oceaniccrust with seamounts or underplated basaltic magmas and (Moreiraet al., 2012), or subcontinental lithospheric mantle delaminated duringthe opening of the Atlantic ocean (Moreira et al., 1999). For the westernlavas of S. Miguel and the remaining islands of the Central Azores theobtained 4He/3He isotopic ratios are similar to the average N-MORBvalue within 2σ of uncertainty. Noble gas isotopic data from thermalwaters and gas emissions agree with those obtained from rock samplescollected in the same islands, but in the case of Graciosa, 4He/3He ratiosas low as≈64,500 (R/Ra≈ 11.2) measured in a bubblingmud pool arereferred by Jean-Baptiste et al. (2009).

3. Sampling and analytical procedures

In this workwe present new helium data obtained frommelt and/orfluid inclusions trapped in olivine phenocrysts, which were separatedfrom basaltic lavas (s.l.) collected at Faial, Graciosa, Terceira and S.Miguel islands, on the E-SE flank of the D. João de Castro Bank and atthe bottom of the South Hirondelle basin (Fig. 1A and B). We also ana-lyzed one sample collected in situ during the eruption that occurredon the Serreta submarine ridge, at the western flank of Terceira Island,from December 1998 to August 2001 producing floating lava balloonsfrom a vent located at depths of around 400 m to the sea surface(Gaspar et al., 2003). Despite the location of the Faial Island awayfrom the TR,we included this site in our study since several authors con-sidered intersection of the Pico–Faial lineament with theMAR as the lo-cation of the Azores triple junction (e.g. Luis et al., 1994; Fernandes et al.,2006). He data from Faial Island also include the analysis of an olivineseparate extracted from one dunite xenolith.

Submarine lavas were dredged or sampled by the Portuguese ROV“Luso” at depths ranging from 750 m (D. João de Castro Bank) downto ~2900 m (South Hirondelle basin) during the EMEPC/Açores/G3/2007, EMEPC/Açores/Luso/2008 and 2009 cruises. Subaerial lavaswere collected in situ during several campaigns, which occurredbetween 2008 and 2010. Sampling sites were selected in order to be arepresentative of the main active basaltic volcanic systems. Based onavailable geochronology, field observations, cartography and marinemagnetic data, both subaerial and submarine samples correspond tolavas younger than 1 Ma. This information is important in order to ex-clude the eventual He cosmogenic contribution to our isotopic analyses.

The choice of olivine, rather than clinopyroxene for He isotopic anal-ysis wasmade in order to: 1) avoid the effects of possible contaminationofmagmatic signatures by 4Heproduced in-situ by thedecay of U and/orTh hypothetically present in the pyroxene composition; 2) prevent theeffects of diffusive He loss, which according to Nuccio et al. (2008) ismore suitable to occur in clinopyroxene due to its crystalline structure(see also Trull and Kurz, 1993).

Lava samples were gently crushed and sieved and olivine crystalswere then handpicked with the aid of a table light and a binocularmicroscope. Homogeneous separates were prepared concerning the

Fig. 1. A) Geotectonic setting of the Terceira Rift showing the bathymetric chart of the Azores plateau (adapted from Lourenço et al., 1998). Upper left corner: Location of the Azores Ar-chipelago (black rectangle) in the Atlantic Ocean.White circles represent sampling sites located along the Terceira Rift and at Faial Island. G: Graciosa Island; T: Terceira Island; DJC: D. Joãode Castro Bank; SH: South Hirondelle Basin; SM: S. Miguel Island; F: Faial Island; NA: North America plate; ATJA: Azores triple junction area; MAR: Mid-Atlantic Ridge; NMAR and SMARrepresent the different He isotopic signatures found to the north of Kurchatov fracture zone and to the south of East Azores fracture zone, respectively (see text). B) Bathymetric and to-pographic profile along the Terceira Rift (with vertical exaggeration) showing the distribution of the sampling sites on this structure.

64 P. Madureira et al. / Chemical Geology 372 (2014) 62–71

olivine color, freshness and diameter (mostly ranging between 1.4 and2 mm). Mineral grains were then ultrasonically cleaned in distilledwater and ethanol and leachedwith HF for 15min to remove the exter-nal surface that could be in contact with the basaltic matrix. The finalstep of cleaning was achieved with ethanol. Volatiles were extractedby crushing in vacuum (150 to 200 strokes) because this methodmain-ly releasesmagmatic gases trapped inmelt inclusions, thus avoiding thecontribution of cosmogenic 3He or 4He produced by radioactive U andTh decay in the rock matrix. Helium data were obtained on ARESIBO Iand Noblesse (Nu instruments©) mass spectrometers, both housed atthe Institut de Physique du Globe de Paris. The analytical proceduresfor the ARESIBO I and for the new Noblesse line can be found inMadureira et al. (2005) and Moreira et al. (2011), respectively.

Major element and REE data from whole rock analyses were ob-tained at Activation Laboratories (Canada) by ICP-OES and ICP-MS

respectively (Annex 1). Reproducibility was generally better than5%. Additional information on analytical procedure can be found inMadureira et al. (2011) and http://www.actlabs.com.

4. Results and discussion

In order to obtain, asmuch as possible, information about the primaryHe signature of magmas wemostly restricted this study to basaltic rockswith MgO N6 wt.%, considered representative of magmas showing noevidence for significant crystal fractionation (Appendix 1). Three sam-ples from Terceira island are more evolved (MgO down to 3.5 wt.%)and affected by fractional crystallization of olivine, clinopyroxene andFe–Ti oxides (Madureira et al., 2011). Noble gas isotopic analyses fromthese samples are, however, consistentwith those obtained from prima-ry magmas and do not influence any of our conclusions.

Table 1He data from samples collected along the Terceira Rift and at Faial Island. R and Ra are respectively the measured and the atmospheric He isotope ratios; X is determined as X = (4He/20Ne)/(4He/20Ne)air; Rc/Ra is the air-corrected He isotope ratiodetermined as Rc/Ra = [(R/Ra x X) − 1]/(X− 1).

Sample Sample TASclassification

Latitude Longitude Site Depth (m) Analyzedphenocryst

Weight (g) R/Ra s 4He ccSTP/g 4He/3He s 20Ne X [4He]c ccSTP/g Rc/Ra

INZ-18 Basalt 39°00.78N 27°57.57W Graciosa Island Subaerial Ol 0.720 7.72 0.11 7.87E−09 93,593 1318 1.27E−11 1953 7.86.E−09 7.72INZ-19 Basalt 39°00.78N 27°57.57W Subaerial Ol 0.796 7.93 0.13 5.99E−09 91,073 1417 2.95E−11 638 5.98.E−09 7.94INZ-24 Basalt 39°2.15N 27°59.24W Subaerial Ol 0.803 7.64 0.20 2.25E−09 94,599 2420 1.60E−11 442 2.25.E−09 7.65PMT-Serr2 Basalt 38°48.30N 27°28.74W Serreta ridge Lava balloon (sea surface) Ol 0.766 10.47 0.39 2.47E−08 69,041 2458 3.78E−12 20,576 2.47.E−08 10.47PMT-71 Basalt 38°41.47N 27°21.30W Terceira Island Subaerial Ol 0.530 9.19 0.17 5.38E−09 78,622 1395 1.10E−10 154 5.35.E−09 9.24PMT-73 Mugearite 38°41.47N 27°21.30W Subaerial Ol 0.488 9.05 0.22 3.59E−09 79,825 1880 1.70E−11 662 3.58.E−09 9.06PMT-74A2 Mugearite 38°41.47N 27°21.30W Subaerial Ol 0.152 9.86 0.17 4.20E−08 73,256 1224 1.19E−11 11,144 4.20.E−08 9.86PMT-88 Basalt 38°39.78N 27°14.82W Subaerial Ol 0.467 9.18 0.08 4.12E−08 78,722 688 2.78E−11 4660 4.11.E−08 9.18PMT-89 Basalt 38°41.28N 27°14.28W Subaerial Ol 0.340 9.17 0.07 5.90E−08 78,810 593 2.78E−11 6676 5.90.E−08 9.17PMT-101 Basalt 38°48.18N 27°15.30W Subaerial Ol 0.450 9.49 0.09 2.80E−08 76,124 712 2.32E−11 3800 2.80.E−08 9.49PMT-414 Basalt 38°38.80N 27°05.67W Subaerial Ol 0.438 8.38 0.13 1.18E−08 86,269 1303 5.90E−11 632 1.18.E−08 8.39L08D7R2 Basalt 38°15.86N 26°33.59W D. João Castro Bank −1170 Ol 1.217 7.89 0.06 5.29E−08 91,608 691 3.27E−10 508 5.27.E−08 7.90L09D22R1 Basalt 38°14.06N 26°33.53W −786 Ol 0.795 7.44 0.07 4.38E−08 97,096 842 8.20E−11 1678 4.38.E−08 7.45L09D22R3 Havaite 38°14.01N 26°33.61W −747 Ol 1.033 7.35 0.06 3.61E−08 98,343 801 2.39E−11 4738 3.61.E−08 7.35D5-002 Basalt 37°56.23N 26°10.45W Hirondelle Basin −3200 Ol 0.701 8.37 0.07 2.72E−08 86,376 711 3.80E−11 2248 2.72.E−08 8.37D5-003 Basalt 37°56.23N 26°10.45W −3200 Ol 0.417 8.08 0.12 1.09E−08 89,375 1283 1.22E−11 2809 1.09.E−08 8.09D5-009 Basalt 37°56.23N 26°10.45W −3200 Light green Ol 0.363 8.24 0.08 2.75E−08 87,691 874 1.05E−11 8271 2.75.E−08 8.24

Basalt −3200 Brown Ol 0.694 8.41 0.10 1.28E−08 85,931 1019 1.10E−11 3664 1.28.E−08 8.41Basalt −3200 Ol 0.726 8.40 0.10 1.32E−08 86,044 1037 3.94E−11 1053 1.32.E−08 8.40

D5-011 Basalt 37°56.23N 26°10.45W −3200 Ol 0.683 8.07 0.13 9.37E−09 89,512 1367 1.27E−11 2327 9.37.E−09 8.08D5-012 Basalt 37°56.23N 26°10.45W −3200 Ol 0.504 8.36 0.06 4.26E−08 86,465 637 3.03E−11 4413 4.26.E−08 8.36D5-R100 Basalt 37°56.23N 26°10.45W −3200 Ol 0.317 8.00 0.16 1.05E−08 90,347 1804 1.39E−11 2384 1.05.E−08 8.00L09D21R3 Basalt 37°56.25N 26°11.09W −2919 Ol 0.456 8.98 0.07 7.07E−08 80,434 632 1.52E−11 14,650 7.07.E−08 8.98INZ-42 Basalt 37°51.56N 25°51.20W S. Miguel Island Subaerial Ol 0.483 7.49 0.16 6.04E−09 96,446 2054 1.18E−10 162 6.00.E−09 7.53INZ-44 Basalt 37°51.67N 25°51.03W Subaerial Ol 0.741 7.33 0.13 6.86E−09 98,543 1669 5.00E−11 432 6.85.E−09 7.35INZ-45 Basalt 37°51.67N 25°51.03W Subaerial Ol 0.583 8.09 0.07 3.30E−08 89,296 762 1.44E−10 722 3.30.E−08 8.10INZ-47 Basalt 37°49.36N 25°26.83W Subaerial Ol 0.810 5.01 0.09 7.01E−09 144,170 2644 6.55E−12 3366 7.01.E−09 5.01INZ-1E Basalt 38°33.98N 28°46.26W Faial Island Subaerial Ol 0.695 8.38 0.06 2.801E−08 86,250 650 3.18E−10 277 2.79.E−08 8.40INZ-2 Havaite 38°36.62N 28°45.53W Subaerial Ol 0.7776 7.95 0.08 1.349E−08 90,917 939 1.18E−10 359 1.35.E−08 7.97INZ-7 Havaite 38°35.89N 28°49.75W Subaerial Ol 0.6887 7.36 0.07 2.481E−08 98,204 908 1.77E−10 442 2.48.E−08 7.37INZ-10 Xenolith 38°36.72N 28°45.77W Subaerial Light green Ol 0.8103 8.57 0.03 1.549E−07 84,284 258 2.44E−10 1995 1.55.E−07 8.58INZ-12 Havaite 38°34.69N 28°46.83W Subaerial Ol 0.6375 8.15 0.09 1.877E−08 88,630 980 6.98E−11 846 1.87.E−08 8.16

65P.M

adureiraetal./Chem

icalGeology

372(2014)

62–71

66 P. Madureira et al. / Chemical Geology 372 (2014) 62–71

Helium concentrations and isotopic ratios measured from oliv-ine are presented in Table 1. 4He contents range between 2.25× 10−9 and 1.55 × 10−7 ccSTP/g (Table 1 and Fig. 2A). This intervalis within the range of He concentrations previously obtained forthe Azores (Moreira et al., 1999; Madureira et al., 2005). Thehighest He content was obtained on olivine xenocrysts being simi-lar to the results obtained byMoreira et al. (1999) for Faial and Picoxenoliths.

He isotopic ratios were corrected for air contamination and, accord-ingly, are also reported as Rc/Ra (Shaw et al., 2006) based on measured

A

B

1.00E-09

1.00E-08

1.00E-07

1.00E-06

27.0027.5028.0028.5029.00

4 He

[ ]

ccS

TP

/g

Longitude °W

xenoliths DC

Terceira

Graciosa

PicoFaial

Serretaridge

30000

50000

70000

90000

110000

130000

150000

170000

190000

210000

27.0027.5028.0028.5029.00

4 He/

3 He

Longitude °W

DC

Terceira

Graciosa

Pico

Faial

Serretaridge

Fig. 2.A) Plot of He concentrations in function of longitude. The highest He contentwas found foof 4He/3He isotopic ratios with longitude. Themost primitive isotopic ratios found in this work a4He/3He ratio presented by depleted MORB.

He/Ne ratios (Rc/Ra= [(R / Ra × X)− 1] / (X− 1) and X= (4He/20Ne) /(4He/20Ne)air). The similarity between air-corrected and non-correctedratios shows that air contamination processes were negligible in oursamples.

He concentrations and He isotopic ratios are highly variable alongthe Terceira Rift as can be observed in Fig. 2 where such parametersare plotted as a function of longitude. Our data show that the lowestHe concentrations are generally relatedwith samples erupted fromma-ture central volcanoes with summit calderas. However, to fully explorethe link between He concentrations and degassing processes occurring

25.0025.5026.0026.50

GraciosaSerreta ridgeTerceiraD. João de CastroHirondelleS. MiguelFaialMoreira et al. 1999; Madureira et al. 2005

S. Miguel

Hirondelle

. João astro

25.0025.5026.0026.50

S. Miguel

Hirondelle

. João astro

r the Faial xenoliths, similarly to the results obtained byMoreira et al. (1999). B) Variabilityre represented by the Terceira and Serreta samples. The horizontal line point to the typical

67P. Madureira et al. / Chemical Geology 372 (2014) 62–71

in distinct volcanic structures the use of other volatile species is re-quired, which will be done in the near future.

In accordance with previous studies, Terceira lavas (including thesubmarine Serreta ridge) show relatively primitive He ratios (down to≈69,000; ≈10.5 R/Ra; see also Madureira et al., 2005 and Moreiraet al., 1999 and Fig. 2B). The most radiogenic He ratio (4He/3He≈121,600 or R/Ra ≈6) was found in a lava sample from the central S.Miguel Island in agreement with data from Kurz et al. (1990) andMoreira et al. (1999, 2012). Graciosa Island, D. João de Castro Bankand the western part of S. Miguel Island share a common radiogenicHe isotopic signature with a mean 4He/3He ratio of ≈94,500 (R/Ra≈7.7). Samples collected at the bottom of Hirondelle Basin displayless radiogenic signatures than the western S. Miguel samples, buthigher 4He/3He ratios than those found in Terceira. He isotopic ratiosobtained from Faial samples cover most of the range of values foundat Hirondelle Basin, Graciosa, D. João de Castro and western S. Miguel.

La/Yb ratios have been used as a proxy for the extent of partialmelting contributing for the generation of alkaline basaltic magmas(e.g. George and Rogers, 2002). For the Azores archipelago a correlationhas been shown to exist between the distance to the MAR (i.e. the lith-osphere thickness) and the degree of partial melting (e.g. Beier et al.,2008, 2012). However, taking into account the observed variation ofthe He isotopic ratio with longitude (Fig. 2) and the absence of correla-tion between 4He/3He and La/Yb ratios (Fig. 3) we consider that theobtained He isotopic signatures are independent of the extent ofmantlemelting. The causes of such variability are addressed below.

4.1. He evidence for distinct Eurasia and Nubia mantle He isotopicsignatures

In order to better understand the variability of He isotopic signaturesalong the TRwewill first evaluate the available heliumdata collected onthe adjacent MAR basalts (Fig. 4; Kurz et al., 1982; Marty andOzima, 1986; Gautheron, 2002; Moreira and Allègre, 2002). Thecharacterization of the MORB component is crucial since the dilutionof the plume component by the MORB source can result in hotspotlavas characterized by 4He/3He ratios similar to the local upper mantleas illustrated by the hyperbolic binary mixing model of Madureiraet al. (2005). From Fig. 4, the presence of four zones (A to D)with differ-ent He isotopic signatures and/or MAR bathymetry is readilyinterpreted. Zone A, located at latitudes immediately to the south ofthe Azores triple junction area (ATJA, see also Fig. 1) is characterizedby less radiogenic He isotopic ratios, down to 75,000 (R/Ra ≈9.5) at38.5°N. These values are interpreted as representing the stronger influ-ence of the present-day Azores plume, which is also responsible for the

50000

100000

150000

0 5 10 15 20 25

4 He/

3 He

La/Yb

Faial

Graciosa

Serreta

Terceira

D. João Castro

Hirondelle

S. Miguel

R2 = 0.23

Fig. 3. A) Plot of whole rock La/Yb ratios against the olivine helium concentration. The ab-sence of correlation between the two deliverables is readily interpreted from the diagram.

higher magma productivity and elevated ridge topography in this zonewhere depths shallower than1500mare reached (Fig. 4). Zones B and Ccorrespond to MAR segments also included in the bathymetric, geo-chemical and gravimetric anomaly of the Azores platform and locatedto the south and to the north of zone A, respectively. Despite the similardepths (in excess of 2,000 m) of MAR axis at zones B and C, the latterzone is characterized by systematic higher 4He/3He ratios than thosein zone B (Fig. 4). A different area (zone D) located between 39°N and40.5°N and displaying identical 4He/3He ratios to those observed inzone B is also considered. Zone D also correlates with shallower MARsegments, arguing for higher magma productivity and a stronger in-fluence of the present-day plume sampled at 38.5° and at the Azoresarchipelago. This zone is therefore interpreted as the result of theprogressive interaction of the Azores plume source with the Eurasiaupper mantle source (zone C), which is responsible for the migrationof the Nubia–Eurasia plate boundary from its former position at theintersection with the East Azores fracture zone towards the TR(Vogt and Jung, 2004).

The presence of several outliers relative to the trends interpretedfrom these zones should be also noted . Most of them seem to be relatedwith local processes occurring along fracture zones (e.g. Langmuir andBender, 1984) such as Oceanographer and Kurchatov (34°N and40.5°N, respectively). The exception is the highest radiogenic valueobtained on samples collected at 39°N (Moreira and Allègre, 2002). Itcorresponds to a shallowoff-axis area characterized by very lowHe con-centrations, whichwas interpreted by these authors as the result of sev-eral cycles of degassing that could locally explain the occurrence of veryradiogenic He isotopic ratios.

One of themost interesting features emerging fromdata depicted byFig. 4, is the similar depth of the MAR axis within zones B and C,notwithstanding their distinct He isotope signatures. From geophysicalevidence, the Azores plume is thought to be deflected to the South(Yang et al., 2006), therefore offering an explanation for the moreunradiogenic signatures at zone B, where a stronger plume contributionwould be expected. However, in such a model a shallower topographyshould be observed as a consequence of a higher heat flux, which isnot the case. Thus we interpret the difference in the He isotopic ratiosas independent of the present-day plume influence.

Recent detailed studies on lavas (Gale et al., 2013) and melt in-clusions (Laubier et al., 2012) from the Mid-Atlantic Ridge to theSouth of the Azores (≈ zone B in Fig. 4) have highlighted that thealong ridge geochemical variability in the vicinity of the Azores canbe explained by a binary mixing model involving a depleted MORB-type source and an end-member enriched in incompatible traceelements. Interestingly, the enriched end-member is more enriched(e.g. higher Th/La andNb/La) than the hypothetical Azores plume signa-ture, yet characterized by similar Sr-, Nd- and Pb-isotope ratios. Thiswas interpreted by Gale et al. (2013) as reflecting previous shallowmantlemetasomatismby low-degreemelts stemming from the ascend-ing plume. This early plume-induced metasomatism would imprint theplume signature in the MORB source in areas not significantly affectedby the present-day thermal and buoyant effects of the ascendingplume, thus explaining the distinction in the He isotope signatures butsimilar bathymetry between the ridge sectors to the South (zone B)and North (zone C) of the Azores (see Fig. 4).

4.2. He isotopic signature of the mantle source sampled at the Terceira Rift

He isotopic data presented in this work allow an overlook on the en-tire TR including islands and submarine sites located along this struc-ture. The comparison of the obtained signatures with other existentgeochemical or geophysical information can provide new insights onthemantle processes coupled with the TR and triple junction dynamics.

In accordance with previous works on noble gas in the Azores, thelower 4He/3He isotopic ratios in this study were obtained for lavas col-lected at Terceira Island and on the Serreta ridge (Fig. 5). This was

-5000

-4500

-4000

-3500

-3000

-2500

-2000

-1500

-1000

-500

0

60000

70000

80000

90000

100000

110000

120000

30 32 34 36 38 40 42 44 46 48

Dep

th (

m)

4 He/

3 He

Latitude °N

Kurz et al., 1982; Marty & Ozima, 1986

Moreira & Allègre, 2002; Gautheron, 2002

SMAR

N MAR

B

C

D

A

Fig. 4.He isotopic ratio obtained frommodern volcanic glass sampled along theMid-Atlantic Ridge. Data fromKurz et al. (1982), Marty and Ozima (1986),Moreira andAllègre (2002) andGautheron (2002). Thedashed line links the sampling depth values ofMAR lavas. The area between 37.5°N and 40°N correlateswith the shallowestMAR segments and is interpreted as theridge interaction with the present-day Azores plume. Zones A to D reflect distinct MAR sections characterized by different He isotopic signatures and/or ridge axis depth.

68 P. Madureira et al. / Chemical Geology 372 (2014) 62–71

interpreted as a consequence of the Terceira Island proximity relative tothe center of the Azores plume (e.g. Moreira et al., 1999; Madureiraet al., 2005), a geochemically based percept lately supported by the to-mographic study of Yang et al. (2006). However, it should bementionedthat Jean-Baptiste et al. (2009) determined He isotopic ratios in gasescollected from a bubbling mud pool at Graciosa, which displays amore primitive signature than the lavas sampled at surface. Theseauthors measured a 4He/3He ratio of 64,340 (R/Ra≈11.3) in the formerwhereas the helium isotopic ratios in lavas range from 91,000 to 94,600(R/Ra between 7.6 and 7.9). Similar discrepancies between surfacegases and those trapped in melt and/or fluid inclusions have been de-scribed for other volcanic regions as the Mount Etna (e.g. Rizzo et al.,2006; Nuccio et al., 2008). Although kinetic fractionation of helium

50000

70000

90000

110000

130000

150000

0 50 100 150 200

4 He/

3 He

Distance

TS

GraciosaFaial

Zones B and D

Zone C

Fig. 5. He isotopic ratio variation along the Terceira Rift as obtained in this study. Zones B

during bubble formation (similar to that proposed by Ruzié andMoreira, 2010) can explain the data from the mud pool at Graciosa,we cannot discard the ability of thermal spring gases to portraymore ef-ficiently the plume fingerprint than the inclusions trapped in olivine.This hypothesis is also supported by a recent reinterpretation of tomo-graphic data, which point to the existence of two distinct mantle up-wellings beneath the central Azores and centered close to the Terceiraand Graciosa islands (Adam et al., 2013).

Notwithstanding the tendency to increase the 4He/3He ratios fromTerceira towards both ends of the TR as shown by our new isotopicdata, a relatively regular variation between relatively unradiogenicand more radiogenic He signatures spaced apart from each other by~50–80 km can be observed (Fig. 5). Sampled sites can therefore be

250 300 350 400 450

to the MAR (km)

S. Miguel

Hirondelle

D. João Castroerceira /

erreta Ridge

Present-dayAzores plume

, C and D correspond to those referred in Fig. 4, based on distinct He isotopic ratios.

69P. Madureira et al. / Chemical Geology 372 (2014) 62–71

grouped into two different classes characterized by 4He/3He ratioshigher (up to 98,500, excluding the sample from the centralS. Miguel Island) and lower (down to 80,000) than ≈90,000 (R/Ra≈8), matching with the different domains previously recognized inthe MAR segments located to the north and to the south of the ATJA.This variation can be explained considering the mixing between oneend-member considered as the Azores plume and the ambientMORB source following the model of Moreira et al. (2001), Moreiraand Allègre (2002) and Madureira et al. (2005).

For our samples we did not get analytical data for non-volatile isoto-pic systems. However, as shown by Moreira et al. (1999) for the Azoresislands and by Moreira and Allègre (2002) for MAR basalts, a co-variation exists between He and Pb isotopes reinforcing the plausibilityof such mixing model.

He isotope variability of samples collected along the MAR andthe TR, as well as the possible connection between the signaturesobserved along these two structures are better understood fromFig. 6. The outliers observed in Fig. 4 were removed for clarity. Asmentioned above, the distinct He signatures assigned to the Nubiaand Eurasia upper mantle and portrayed in MORB, are both well

OFZ

PFZ

Zone B

Zone A

Zon

Z

NorthAmerica plate

4 3He/ He > 900004 380000 < He/ He < 90000

4 3He/ He < 80000

Fig. 6. Location of sampleswithHe isotopic data along the TR (this study) and theMAR (Kurz etC andD correspond to those of Fig. 4. KFZ: Kurchatov Fracture Zone; PFZ: Pico Fracture Zone; EAFthe present-day Azores triple junction; G: Graciosa Island; T: Terceira Island; DJC: D. João de CAtlantic Ridge. Arrow stands for the absolute motion of Eurasia and Nubia plates relative to a h

preserved along the MAR, respectively to the south of the Pico/EastAzores fracture zone and to the north of Kurchatov fracture zone.Along the TR, lavas from the Graciosa Island, D. João de Castro andthe western S. Miguel Island are characterized by helium isotopicsignatures typically assigned to the Eurasia upper mantle (zone C;Figs. 4 and 6). Conversely, lavas from the Hirondelle Basin andmost of the new samples collected at Faial display 4He/3He ratiossimilar to those typical of Nubia (zones B and D; Figs. 4 and 6). TRis therefore interpreted as a plate boundary characterized by the“mingling” of Eurasia-type and Nubia-type (“metasomatized”)upper mantle domains, which may result from the progressive mi-gration of the Nubia–Eurasia plate boundary from the East Azoresfracture zone towards north. We must emphasized that for TR lavasthe possibility that such dichotomy is caused by a more complexmixing of variable proportions of the heterogeneous Azores plume(heterogeneous) and a relatively homogeneous depleted uppermantle source cannot be discarded.

Future geophysical and geochemical data will constrain the possibleexistent link between the plate kinematics along the TR and the differ-ent regional mantle domains inferred from helium data. However, we

KFZ

EAFZ

e D

one C

rcTe

eira Rift

Eurasiaplate

Nubiaplate

DJC

m

~20 m/yr

MA

R

F

SM

P

SH

G

T

Ser

SJ

ATJ ?

al., 1982;Marty andOzima, 1986; Gautheron, 2002;Moreira andAllègre, 2002). Zones A, B,Z: East Azores Fracture Zone;OFZ: Oceanographer Fracture Zone; ATJ: possible location forastro Bank; SH: South Hirondelle Basin; SM: S. Miguel Island; F: Faial Island; MAR: Mid-otspot reference frame (Gripp and Gordon, 2002).

70 P. Madureira et al. / Chemical Geology 372 (2014) 62–71

note that helium concentrations were already correlated with the de-gree of ductile deformation in oceanic peridotites (Kurz et al., 2009)and that fluid content is a major factor controlling mantle rheology(e.g. Hidas et al., 2013).

Along the TR, the present-day Azores plume component is bettersampled by Terceira lavas (including the Serreta ridge) and possiblyalso by thermal spring gases from Graciosa Island. Primitive He iso-topic ratios were also described for S. Jorge (4He/3He down to45,000 or R/Ra up to 16; Moreira and Allègre, 2002) and Pico(Moreira, unpublished data) islands. The distribution of these prim-itive signatures has been explained as a consequence of the plumedeflection from the area located NE of Terceira Island towardssouth (Yang et al., 2006; Adam et al., 2013) and of the southwest-ward motion of Eurasia and Nubia (236° from north) relative to ahotspot reference frame (Gripp and Gordon, 2002). This modelalso explains the northward migration of the plate boundary andthe creation of a Nubia-type He signature in Eurasia domains(zone D; Figs. 4 and 6).

Also important is the discussion on the location of the triple junc-tion, resulting from the intersection of Nubia–Eurasia boundarywith the MAR. Most models point either to the prolongation of theTR (e.g. Beier et al., 2008) or of the Pico–Faial ridge (e.g. Luis et al.,1994) to the MAR, placing the ATJA between 38.5 and 39.5°N.Based on He isotopic data, we argue that the location of the Azorestriple junction is possibly located further north, at 40.5°N in thenorthern end of zone D (Fig. 6).

5. Conclusions

The new helium data confirms the relatively primitive heliumsignature of the Azores hotspot, although diluted by the ambientasthenospheric mantle. According to our data, we interpret theTerceira Rift as a geochemical boundary between the Eurasia-typeand Nubia-type upper mantle domains, which are both sampled

Appendix 1

Site Sample SiO2 Al2O3 Fe2O3(T) MnO MgO CaO Na2O K2O TiO2

% % % % % % % % %

Graciosa INZ-18 49.37 16.1 10.3 0.148 6.92 9.13 3.69 1.21 2.269INZ-19 48.95 15.92 10.1 0.146 6.66 8.99 3.75 1.22 2.304INZ-24 45.42 16.85 10.92 0.150 7.73 10.33 2.94 1.05 3.219

Serreta PMT-Serr2 47.68 14.77 10.89 0.148 8.64 10.31 3.06 0.71 2.665Terceira PMT-71 48.45 16.35 11.80 0.168 4.64 9.84 3.72 1.22 3.699

PMT-73 51.12 15.86 11.09 0.183 3.55 7.68 4.68 1.76 2.918PMT-74 50.94 16.04 10.76 0.180 3.45 7.72 4.63 1.72 2.818PMT-88 47.49 13.89 13.48 0.211 6.21 9.72 3.35 1.25 3.813PMT-89 47.62 13.92 13.36 0.207 6.48 9.93 3.21 1.14 3.765PMT-101 47.44 14.12 11.83 0.176 8.51 10.52 3.07 0.86 3.062PMT-414 47.82 14.04 9.77 0.151 11.6 11.44 2.57 1.02 1.939

D. João deCastro

L08D7R2 47.36 11.04 9.41 0.149 15.2 12.65 2.11 0.84 1.458L09D22R1 47.82 14.64 10.98 0.174 8.12 10.52 3.38 1.52 2.196L09D22R3 47.75 15.07 11.31 0.174 7.95 10.29 3.59 1.58 2.263

Hirondelle D05-003 44.9 14.79 11.64 0.169 7.81 10.97 2.8 2.02 3.394D05-009 44.92 14.83 12.13 0.173 8.08 10.88 2.69 1.97 3.479D05-011 44.64 14.86 11.94 0.173 7.6 10.75 2.73 2.18 3.442D05-012 45.31 14.97 11.76 0.170 7.72 10.91 2.92 2.06 3.453L09D21R3 45.53 12.84 11.65 0.162 10.3 12.25 2.73 1.35 3.034

S. Miguel INZ-42 46.29 15.26 12.75 0.195 6.43 10.1 2.98 1.72 3.497INZ-44 47.65 13.2 11.67 0.176 8.69C 11.53 2.83 1.51 3.43INZ-45 44.95 10.77 11.7 0.170 12.24 13.69 1.83 0.95 3.192INZ-47 46.76 12.12 11.45 0.175 9.3 11.75 2.28 1.74 3.645

Faial INZ-1E 47.00 14.07 10.3 0.142 10.99 10.26 2.94 1.02 2.409INZ-2 48.22 15.62 10.4 0.141 7.61 9.88 3.63 1.59 2.613INZ-7 48.16 16.52 10.75 0.140 7.92 9.39 3.81 1.62 2.7INZ-12 48.38 16.29 10.33 0.141 6.45 9.33 3.87 1.72 2.729

Whole rock major element and REE data of sampled lavas.LOI: Loss on ignition. Major element concentration was determined by ICP-OES. REE concentration w

by lavas erupted at the Azores islands. He isotopic signatures ofthese domains are well preserved along the MAR, respectively tothe north of the Kurchatov fracture zone and to the south of theEast Azores fracture zone. This dichotomy probably resulted froma heterogeneous distribution of asthenosphere metasomatism dur-ing the initial stages of plume activity in the Azores region. Thepresent-day plume component is well sampled at the Azores archi-pelago by Terceira subaerial and submarine lavas. Along the MAR,this component is also sampled at 38.5°N, resulting from the south-ward bending of the mantle plume centered at NE of Terceira Is-land. This process, coupled with the motion of Eurasia and Nubiaplates (towards the azimuth 236°) relative to a hotspot referenceframe, was responsible for the migration of the TR from the EastAzores fracture zone to its present location during the last 20 Myrand also for the creation of a Nubia-type upper mantle Hesignature.

From a geochemical point of view, the presented He isotopicdata allowed us to suggest that the intersection between the TRand the MAR, i.e. the Azores triple junction, is possibly located at40.5°N, to the north of the area that has been considered by otherauthors.

Acknowledgments

This is a contribution to research projects INOGAZ (PTDC/CTE-GIN/68851/2006), OREAZ (PTDC/CTE-GIX/102061/2008) andTerRiftic (PTDC/MAR/111306/2009) projects funded by the FCTand FCT/COMPETE programs. PM acknowledges the EMEPC team,the navy crew and particularly the fantastic ROV team onboard theNRP Almirante Gago Coutinho. JM also acknowledges funding fromPest-OE/CTE/UI0263/2011. Constructive and valuable commentsand suggestions made by the Reviewers (A. Rizzo and an anony-mous reviewer) and also by the Editor (D. Hilton), greatly contribut-ed to the improvement of the manuscript.

P2O5 LOI Total La Ce Nd Sm Eu Tb Dy Yb Lu

% % % ppm ppm ppm ppm ppm ppm ppm ppm ppm

0.53 −0.44 99.24 30.80 61.60 31.50 6.91 2.35 1.00 5.41 2.29 0.320.49 −0.17 98.38 30.80 60.90 31.40 6.89 2.38 1.00 5.36 2.33 0.330.48 −0.45 98.64 27.40 58.90 29.80 7.07 2.36 0.94 5.11 2.02 0.290.34 −0.44 98.80 22.60 48.10 26.40 5.96 2.22 1.02 5.42 2.07 0.270.58 −0.30 100.17 47.10 92.90 45.60 9.99 3.33 1.40 7.01 2.54 0.330.79 −0.51 99.11 67.97 133.97 63.68 13.70 4.46 1.94 9.88 3.66 0.480.78 −0.47 98.57 62.53 121.75 58.10 12.44 3.97 1.72 8.83 3.29 0.431.14 −0.61 99.94 52.83 109.10 58.80 12.94 4.73 1.80 8.94 2.99 0.391.05 −0.89 99.78 41.69 88.88 51.41 10.75 4.23 1.69 8.15 2.82 0.390.83 −0.79 99.62 37.30 77.45 42.96 9.58 3.82 1.37 6.70 2.24 0.290.31 −0.22 100.4 22.10 44.00 20.60 5.04 1.68 0.77 4.28 1.73 0.250.23 0.15 100.6 14.00 29.64 15.33 3.48 1.19 0.55 2.92 1.24 0.190.44 0.39 100.2 29.74 57.54 27.07 5.49 1.77 0.78 4.37 1.90 0.290.39 0.26 100.6 29.63 57.92 25.64 5.38 1.81 0.78 4.45 1.87 0.260.7 0.68 99.87 43.40 91.70 42.00 8.34 2.71 1.09 5.37 1.93 0.270.76 0.7 100.6 44.40 94.00 44.70 8.99 2.81 1.15 5.81 2.05 0.280.79 1.03 100.1 44.80 96.30 45.00 8.80 2.90 1.14 5.75 2.07 0.290.69 0.68 100.7 44.50 95.70 44.80 8.84 2.82 1.13 5.60 2.07 0.290.47 0.45 100.8 32.07 68.03 35.55 7.29 2.36 0.90 5.09 1.67 0.240.86 −0.12 99.97 54.40 111.00 49.40 10.10 3.16 1.13 6.07 2.20 0.300.54 −0.32 100.9 48.70 92.30 44.20 8.79 2.69 1.03 5.35 1.92 0.240.38 −0.18 99.69 32.40 63.10 31.70 6.68 2.04 0.81 4.24 1.45 0.170.59 0.42 100.2 60.20 114.00 53.20 10.50 3.07 1.20 6.18 2.08 0.260.28 −0.31 99.1 25.40 54.20 28.00 6.07 1.96 0.85 4.61 1.76 0.240.51 −0.51 99.71 39.20 77.80 36.70 7.31 2.27 0.99 5.19 2.19 0.300.42 −0.53 100.9 33.60 69.70 32.70 6.76 2.07 0.93 4.81 1.98 0.280.56 −0.46 99.34 42.20 83.00 39.00 7.65 2.39 1.01 5.32 2.20 0.31

as determined by ICP-MS.

71P. Madureira et al. / Chemical Geology 372 (2014) 62–71

References

Adam, C., Madureira, P., Miranda, J.M., Lourenço, N., Yoshida, M., Fitzenz, D., 2013. Mantledynamics and characteristics of the Azores plateau. Earth Planet. Sci. Lett. 362,258–271.

Allègre, C.J., Moreira, M., Staudacher, T., 1995. 3He/4He dispersion and mantle convection.Geophys. Res. Lett. 22 (17), 2325–2328.

Beier, C., Haase, K.M., Abouchami, W., Krienitz, M.S., Hauff, F., 2008. Magma genesis byrifting of oceanic lithosphere above anomalous mantle: Terceira Rift, Azores.Geochem. Geophys. Geosyst. 9, Q12013. http://dx.doi.org/10.1029/2008GC002112.

Beier, C., Mata, J., Stöckhert, F., Mattielli, N., Brandl, P.A., Madureira, P., Genske, F.S.,Martins, S., Madeira, J., Haase, K.M., 2012. Geochemical evidence for melting of car-bonated peridotite on Santa Maria Island, Azores. Contrib. Mineral. Petrol. http://dx.doi.org/10.1007/s00410-012-0837-2.

Dosso, L., Bougault, H., Langmuir, C., Bollinger, C., Bonnier, O., Etoubleau, J., 1999. The ageand distribution of mantle heterogeneity along the Mid-Atlantic Ridge (31–41degrees N). Earth Planet. Sci. Lett. 170, 269–286.

Fernandes, R.M.S., Bastos, L., Miranda, J.M., Lourenço, N., Ambrosius, B.A.C., Noomen, R.,Simons, W., 2006. Defining the plate boundaries in the Azores region. J. Volcanol.Geotherm. Res. 156, 1–9.

Gale, A., Laubier, M., Escrig, S., Langmuir, C., 2013. Constraints on melting processes andplume–ridge interaction from comprehensive study of the FAMOUS and NorthFamous segments, Mid-Atlantic Ridge. Earth Planet. Sci. Lett. 365, 209–220.

Gaspar, J.L., Queiroz, G., Pacheco, J.M., Ferreira, T., Wallenstein, N., Almeida, M.H.,Coutinho, R., 2003. Basaltic lava balloons producing during the 1998–2001 SerretaSubmarine Ridge eruption (Azores). Geophys. Monogr. Ser. 140, 205–212.

Gautheron, C., 2002. Le manteau lithospherique sous continental contraint par lasystematique des gaz rares: apport a la geodynamique chimique des reservoirsterrestres. (PhD Thesis) University Paris VII, Paris, France (252 pp.).

Gente, P., Dyment, J., Maia, M., Goslin, L., 2003. Interaction between theMid-Atlantic Ridgeand the Azores hot spot during the last 85 Myr: emplacement and rifting of the hotspot-derived plateaus. Geochem. Geophys. Geosyst. 4, 8514. http://dx.doi.org/10.1029/2003GC000527.

George, R.M., Rogers, N.W., 2002. Plume dynamics beneath the African plate inferred fromthe geochemistry of the Tertiary basalts of southern Ethiopia. Contrib. Mineral. Petrol.144, 286–304.

Graham, D., 2002. Noble gas isotope geochemistry of mid-ocean ridge and ocean islandbasalts: characterization of mantle source reservoirs. In: Porcelli, D., Ballentine, C.J.,Wieler, R. (Eds.), Noble Gases in Geochemistry and Cosmochemistry, Reviews inMineralogy and Geochemistry. Mineralogical Society of America and GeochemicalSociety, 47, pp. 247–318. Mineralogical Society of America, Washington, D. C.,Washington D.C., U.S.A.

Graham, D.W., Johnson, K.T.M., Douglas Priebe, L., Lupton, J.E., 1999. Hotspot–ridge inter-action along the Southeast Indian Ridge near Amsterdam and St Paul islands: heliumisotope evidence. Earth Planet. Sci. Lett. 167, 297–310.

Gripp, A.E., Gordon, R.G., 2002. Young tracks of hotspots and current plate velocities.Geophys. J. Int. 150, 321–361.

Hidas, K., Garrido, C.J., Tommasi, A., Navarta, J.A.P., Thielmann, M., Konc, Z., Frets, E.,Marches, C., 2013. Strain localization in pyroxenite by reaction-enhanced softeningin the shallow subcontinental lithospheric mantle. J. Petrol. 54, 1997–2031.

Hofmann, A.W., 2003. Sampling mantle heterogeneity through oceanic basalts: isotopesand trace elements. In: Heinrich, D.H., Karl, K.T. (Eds.), Treatise on Geochemistry.Pergamon, Oxford, pp. 61–101.

Jean-Baptiste, P., Allard, P., Coutinho, R., Ferreira, T., Fourré, E., Queiroz, G., Gaspar, J.L.,2009. Helium isotopes in hydrothermal volcanic fluids of the Azores archipelago.Earth Planet. Sci. Lett. 281, 70–80.

Klein, E.M., Langmuir, C.H., 1987. Global correlation of ocean ridge basalt chemistry withaxial depth and crustal thickness. J. Geophys. Res. 92, 8089–8115.

Krause, D.C., Watkins, N.D., 1970. North Atlantic crustal genesis in the vicinity of theAzores. Geophys. J. R. Astron. Soc. 19, 261–283.

Kurz, M.D., Jenkins, W.J., Schilling, J.-G., Hart, S.R., 1982. Helium isotopic variation inthe mantle beneath the central North Atlantic Ocean. Earth Planet. Sci. Lett. 58,1–14.

Kurz, M.D., Kammer, D.P., Gulessarian, A., Moore, R.B., 1990. Isotopic variations withinoceanic islands: He, Sr and Pb isotopes in basalts from SãoMiguel, Azores. EOS 71,657.

Kurz, M.D., Moreira, M., Curtice, J.D., Lott III, E., Mahoney, J.J., Sinton, J.M., 2005. Correlatedhelium, neon, and melt production on the super-fast spreading East Pacific Rise near17°S. Earth Planet. Sci. Lett. 232, 125–142.

Kurz, M.D., Warren, J.M., Curtice, J., 2009. Mantle deformation and noble gases: heliumand neon in oceanic mylonites. Chem. Geol. 266, 10–18.

Langmuir, C., Bender, J.F., 1984. The geochemistry of oceanic basalts in the vicinity oftransform faults: observations and implications. Earth Planet. Sci. Lett. 69,107–127.

Laubier, M., Gale, A., Langmuir, C.H., 2012. Melting and crustal processes at the FAMOUSsegment (Mid-Atlantic Ridge): new insights from olivine-hosted melt inclusionsfrom multiple samples. J. Petrol. 53, 665–698.

Lourenço, N., Miranda, J.M., Luis, J.F., Ribeiro, A., Mendes Victor, L.A., Madeira, J.,Needham, H.D., 1998. Morpho-tectonic analysis of the Azores Volcanic Plateau

from a new bathymetric compilation of the area. Mar. Geophys. Res. 20,141–156.

Luis, J.F., Miranda, J.M., 2008. Reevaluation of magnetic chrons in the North Atlanticbetween 35°N and 47°N: implications for the formation of the Azores Triple Junctionand associated plateau. J. Geophys. Res. 113, B10105. http://dx.doi.org/10.1029/2007/JB005573.

Luis, J.F., Miranda, J.M., Galdeano, A., Patriat, P., Rossignol, J.C., Mendes Victor, L.A., 1994.The Azores triple junction evolution since 10 Ma from aeromagnetic survey of theMid-Atlantic Ridge. Earth Planet. Sci. Lett. 125, 439–459.

Machado, F., 1959. The fracture pattern of Azorean volcanoes. Bull. Volcanol. 17, 119–125.Madureira, P., Moreira, M., Mata, J., Allègre, C.J., 2005. Primitive helium and neon isotopes

in Terceira Island: constraints on the origin of the Azores archipelago. Earth Planet.Sci. Lett. 233, 429–440.

Madureira, P., Mata, J., Mattielli, N., Queiroz, G., Silva, P., 2011. Mantle source hetero-geneity, magma generation and magmatic evolution at Terceira Island (Azoresarchipelago): constraints from elemental and isotopic (Sr, Nd, Hf, and Pb) data.Lithos 126, 402–418.

Marty, B., Ozima, M., 1986. Noble gas distribution in oceanic basalt glasses. Geochim.Cosmochim. Acta 50, 1093–1098.

Moreira, M., Allègre, C.J., 1998. Helium–Neon systematics and the structure of the mantle.Chem. Geol. 147, 53–59.

Moreira, M., Allègre, C.J., 2002. Rare gas systematics on Mid Atlantic Ridge (37°–40°N).Earth Planet. Sci. Lett. 198, 401–416.

Moreira, M., Staudacher, T., Sarda, P., Schilling, J.-G., Allègre, C.J., 1995. A primitive plumeneon component in MORB: the Shona ridge-anomaly, South Atlantic. Earth Planet.Sci. Lett. 133, 367–377.

Moreira, M., Doucelance, R., Dupré, B., Kurz, M.D., Allègre, C.J., 1999. Helium and leadisotope geochemistry in the Azores archipelago. Earth Planet. Sci. Lett. 169,189–205.

Moreira, M., Breddam, K., Curtice, J., Kurz, M.D., 2001. Solar neon in the Icelandic mantle:evidence for an undegassed lower mantle. Earth Planet. Sci. Lett. 185, 15–23.

Moreira, M., Escartin, J., Gayer, E., Hamelin, C., Bezos, A., Guillon, F., Cannat, M., 2011. Raregas systematics on Lucky Strike basalts (37°N, North Atlantic): evidence for efficienthomogenization in a long-lived magma chamber system? Geophys. Res. Lett. 38,L08304. http://dx.doi.org/10.1029/2011GL046794.

Moreira, M., Kanzari, A., Madureira, P., 2012. Helium and neon isotopes in São Miguel is-land basalts, Azores Archipelago: new constraints on the “low 3He” hotspot origin.Chem. Geol. 322–323, 91–98.

Niu, Y., O'Hara, M.J., 2008. Global correlations of Ocean Ridge Basalt chemistry with axialdepth: a new perspective. J. Petrol. 49, 633–664.

Nuccio, M., Paonita, A., Rizzo, A., Rosciglione, A., 2008. Elemental and isotope covariationof noble gases in mineral phases from Etnean volcanics erupted during 2001–2005,and genetic relation with peripheral gas discharges. Earth Planet. Sci. Lett. 272,683–690.

Poreda, R., 1993. Helium isotope ratios in Easter microplate basalts. Earth Planet. Sci. Lett.119, 319–329.

Rizzo, A., Caracausi, A., Favara, R., Martelli, M., Paonita, A., Paternoster, M., Nuccio, P.M.,Rosciglione, A., 2006. New insights into magma dynamics during last two eruptionsof Mount Etna as inferred by geochemical monitoring from 2002 to 2005. Geochem.Geophys. Geosyst. 7. http://dx.doi.org/10.1029/2005GC001175.

Ruzié, L., Moreira, M., 2010. Magma degassing process during Plinian eruptions.J. Volcanol. Geotherm. Res. 192, 142–150.

Sarda, P., Moreira, M., Staudacher, T., Schilling, J.-G., Allègre, C.J., 2000. Rare gas systemat-ics on the southernmost Mid-Atlantic Ridge: constraints on the lower mantle and theDupal source. J. Geophys. Res. 105, 5973–5996.

Schilling, J.-G., 1975. Azores mantle blob: rare earth evidence. Earth Planet. Sci. Lett. 25,103–115.

Schilling, J.-G., Zajac, M., Evans, R., Johnston, T., White, W., Devine, J.D., Kingsley, R., 1983.Petrologic and geochemical variations along the Mid-Atlantic Ridge from 29 °N to 73°N. Am. J. Sci. 283, 510–586.

Shaw, A.M., Hilton, D.R., Fischer, T.P., Walker, J.A., de Leeuw, G.A.M., 2006. Helium isotopevariations in mineral separates from Costa Rica and Nicaragua: assessing crustal con-tributions, timescale variations and diffusion-related mechanisms. Chem. Geol. 230,124–139.

Stuart, F.M., Lass-Evans, S., Fitton, J.G., Ellam, R.M., 2003. High 3He/4He ratios in picriticbasalts from Baffin Island and the role of a mixed reservoir in mantle plumes. Nature424, 57–59.

Trull, T.W., Kurz, M.D., 1993. Experimental measurements of 3He and 4He mobility inolivine and clinopyroxene at magmatic temperatures. Geochim. Cosmochim. Acta57, 1313–1324.

Vogt, P., Jung, W., 2004. The Terceira Rift as hyper-slow, hotspot-dominated obliquespreading axis: a comparison with other slow-spreading plate boundaries. EarthPlanet. Sci. Lett. 218, 77–90.

Workman, R.K., Hart, S.R., 2005. Major and trace element composition of the depletedMORB mantle (DMM). Earth Planet. Sci. Lett. 231, 53–72.

Yang, T., Shen, Y., Van der Lee, S., Solomon, S.C., Hung, S.C., 2006. Upper mantle structurebeneath the Azores hotspot from finite-frequency seismic tomography. Earth Planet.Sci. Lett. 250, 11–26.