The Origin and Evolution of the Kaapvaal Cratonic Lithospheric Mantle

A detailed petrological and geochemical study of low-temperatureperidotite xenoliths from Kimberley and northern Lesotho ispresented to constrain the processes that led to the magmaphileelement depletion of the Kaapvaal cratonic lithospheric mantleand its subsequent re-enrichment in Si and incompatible traceelements. Whole-rocks and minerals have been characterized forRe–Os isotope compositions, and major and trace elementconcentrations, and garnet and clinopyroxene for Lu–Hfand Sm–Nd isotope compositions. Most samples are characterizedby Archaean Os model ages, low Al, Fe and Ca contents, highMg/Fe, low Re/Os, very low (< 0·1 x chondrite) heavyrare earth element (HREE) concentrations and a decoupling betweenNd and Hf isotope ratios. These features are most consistentwith initial melting at 3·2 Ga followed by metasomatismby hydrous fluids, which may have also caused additional meltingto produce a harzburgitic residue. The low HREE abundances ofthe peridotites require that extensive melting occurred in thespinel stability field, possibly preceded by some melting inthe presence of garnet. Fractional melting models suggest that30% melting in the spinel field or 20% melting in the garnetfield followed by 20% spinel-facies melting are required toexplain the most melt-depleted samples. Garnet Nd–Hf isotopecharacteristics indicate metasomatic trace element enrichmentduring the Archaean. We therefore suggest a model includingshallow ridge melting, followed by metasomatism of the Kaapvaalupper mantle in subduction zones surrounding cratonic nuclei,probably during amalgamation of smaller pre-existing terranesin the Late Archaean (2·9 Ga). The fluid-metasomatizedresidua have subsequently undergone localized silicate meltinfiltration that led to clinopyroxene ± garnet enrichment.Calculated equilibrium liquids for clinopyroxene and their Hf–Ndisotope compositions suggest that most diopside in the xenolithscrystallized from an infiltrating kimberlite-like melt, eitherduring Group II kimberlite magmatism at 200–110 Ma (Kimberley),or shortly prior to eruption of the host kimberlite around 90Ma (northern Lesotho). KEY WORDS: Kaapvaal craton; lithospheric mantle; metasomatism; Nd–Hf isotopes; Re–Os isotopes     

Magmatic Evolution and Ascent History of the Aries Micaceous Kimberlite, Central Kimberley Basin, Western Australia: Evidence from Zoned Phlogopite Phenocrysts, and UV Laser 40Ar/39Ar Analysis of Phlogopite-Biotite

The Neoproterozoic Aries kimberlite was emplaced in the centralKimberley Basin, Western Australia, as a N–NNE-trendingseries of three diatremes infilled by lithic-rich kimberlitebreccias. The breccias are intruded by hypabyssal macrocrysticphlogopite kimberlite dykes that exhibit differentiation toa minor, high-Na–Si, olivine–phlogopite–richteritekimberlite, and late-stage macrocrystic serpentine–diopsideultramafic dykes. Mineralogical and geochemical evidence suggeststhat the high-Na–Si, olivine–phlogopite–richteritekimberlite was derived from the macrocrystic phlogopite kimberliteas a residual liquid following extended phlogopite crystallizationand the assimilation of country rock sandstone, and that themacrocrystic serpentine–diopside ultramafic dykes formedas mafic cumulates from a macrocrystic phlogopite kimberlite.Chemical zonation of phlogopite–biotite phenocrysts indicatesa complex magmatic history for the Aries kimberlite, with theearly inheritance of a range of high-Ti phlogopite–biotitexenocrysts from metasomatized mantle lithologies, followed bythe crystallization of a population of high-Cr phlogopite phenocrystswithin the spinel facies lithospheric mantle. A further oneto two phlogopite–biotite overgrowth rims of distinctcomposition formed on the phlogopite phenocrysts at higher levelsduring ascent to the surface. Ultra-violet laser ⁴⁰Ar/³⁹Ar datingof mica grain rims yielded a kimberlite eruption age of 815·4± 4·3 Ma (95% confidence). ⁴⁰Ar/³⁹Ar laser profilingof one high-Ti phlogopite-biotite macrocryst revealed a radiogenic⁴⁰Ar diffusive loss profile, from which a kimberlite magma ascentduration from the spinel facies lithospheric mantle was estimated(assuming an average kimberlite magma temperature of 1000°C),yielding a value of 0·23–2·32 days for thenorth extension lobe of the Aries kimberlite. KEY WORDS: ⁴⁰Ar/³⁹Ar; diamond; kimberlite; mantle metasomatism; phlogopite–biotite     

Olivine in the Udachnaya-East Kimberlite (Yakutia, Russia): Types, Compositions and Origins

Olivine is the principal mineral of kimberlite magmas, and isthe main contributor to the ultramafic composition of kimberliterocks. Olivine is partly or completely altered in common kimberlites,and thus unavailable for studies of the origin and evolutionof kimberlite magmas. The masking effects of alteration, commonin kimberlites worldwide, are overcome in this study of theexceptionally fresh diamondiferous kimberlites of the Udachnaya-Eastpipe from the Daldyn–Alakit province, Yakutia, northernSiberia. These serpentine-free kimberlites contain large amountsof olivine (50 vol.%) in a chloride–carbonate groundmass.Olivine is represented by two populations (olivine-I and groundmassolivine-II) differing in morphology, colour and grain size,and trapped mineral and melt inclusions. The large fragmentalolivine-I is compositionally variable in terms of major (Fo_85–94)and trace element concentrations, including H₂O content (10–136ppm). Multiple sources of olivine-I, such as convecting andlithospheric mantle, are suggested. The groundmass olivine-IIis recognized by smaller grain sizes and perfect crystallographicshapes that indicate crystallization during magma ascent andemplacement. However, a simple crystallization history for olivine-IIis complicated by complex zoning in terms of Fo values and traceelement contents. The cores of olivine-II are compositionallysimilar to olivine-I, which suggests a genetic link betweenthese two types of olivine. Olivine-I and olivine-II have oxygenisotope values (+ 5·6 ± 0·1 VSMOW, 1 SD)that are indistinguishable from one another, but higher thanvalues (+ 5·18 ± 0·28) in ‘typical’mantle olivine. These elevated values probably reflect equilibriumwith the Udachnaya carbonate melt at low temperatures and ¹⁸O-enrichedmantle source. The volumetrically significant rims of olivine-IIhave constant Fo values (89·0 ± 0·2 mol%),but variable trace element compositions. The uniform Fo compositionsof the rims imply an absence of fractionation of the melt'sFe²⁺/Mg, which is possible in the carbonatite melt–olivinesystem. The kimberlite melt is argued to have originated inthe mantle as a chloride–carbonate liquid, devoid of ‘ultramafic’or ‘basaltic’ aluminosilicate components, but becameolivine-laden and olivine-saturated by scavenging olivine crystalsfrom the pathway rocks and dissolving them en route to the surface.During emplacement the kimberlite magma changed progressivelytowards an original alkali-rich chloride–carbonate meltby extensively crystallizing groundmass olivine and gravitationalseparation of solids in the pipe. KEY WORDS: kimberlite; olivine; partial melting; carbonatitic melt; oxygen isotopes; H₂O     

Overlap of Karoo and Ferrar Magma Types in KwaZulu-Natal, South Africa

A suite of mafic dykes from the Underberg region of southernKwaZulu-Natal (South Africa) were intruded at 178 Ma, coincidentin age with the major Okavango Dyke Swarm of Botswana, and alsocoincident with minor Karoo-related intrusions of the northernand central Lebombo. The dykes are all low-Ti–Zr tholeiites,they trend NW–SE and are presumed to continue into theKaroo central area of the Lesotho Highlands. In many respects,the Underberg dykes are similar to the majority of the low-Ti–Zrvolcanic and subvolcanic intrusions of the Karoo; however, their⁸⁷Sr/⁸⁶Sr and Nd isotope ratios are either ‘Ferrar-like’(⁸⁷Sr/⁸⁶Sr 0·710; Nd < –3) or transitional betweenKaroo low-Ti–Zr and Ferrar low-Ti magmas. A potentialFerrar source for at least some of the Underberg dykes is supportedby anisotropy of magnetic susceptibility analyses of the dykesuite, which demonstrate absolute flow direction from the SEto the NW, consistent with Gondwana reconstructions. The roleof crustal contamination and combined fractional crystallizationis also demonstrated to have played a key role in the petrogenesisof the Underberg dykes, involving a local upper crust contaminant.However, the composition of the ‘Ferrar-like’ dykescannot be easily explained by AFC processes, but they do demonstratethat melting of a lithospheric mantle source enriched to a smalldegree by subduction-derived fluid was also important. KEY WORDS: dyke; basalt; crustal contamination; large igneous province     

Origin of Paleoproterozoic Komatiites at Jeesiorova, Kittila Greenstone Complex, Finnish Lapland

Komatiites from the 2 Ga Jeesiörova area in Finnish Laplandhave subchondritic Al₂O₃/TiO₂ ratios like those in Al-depletedkomatiites from Barberton, South Africa. They are distinct inthat their Al abundances are higher than those of the Al-depletedrocks and similar to levels in Al-undepleted komatiites. Moderatelyincompatible elements such as Ti, Zr, Eu, and Gd are enriched.Neither majorite fractionation nor hydrous melting in a supra-subductionzone setting could have produced these komatiites. Their highconcentrations of moderately incompatible elements may haveresulted from contamination of their parental melt through interactionwith metasomatic assemblages in the lithospheric mantle or enrichmentof their mantle source in basaltic melt components. Re–Osisotope data for chromite from the Jeesiörova rocks yieldan average initial ¹⁸⁷Os/¹⁸⁸Os of 0·1131 ± 0·0006(2), _Os(I) = 0·1 ± 0·5. These data, coupledwith an initial _Nd of +4, indicate that melt parental to thekomatiites interacted minimally with ancient lithospheric mantle.If their mantle source was enriched in a basaltic component,the combined Os–Nd isotopic data limit the enrichmentprocess to within 200 Myr prior to the formation of the komatiites.Their Os–Nd isotopic composition is consistent with derivationfrom the contemporaneous convecting upper mantle. KEY WORDS: Finnish Lapland; Jeesiörova; komatiites; mantle geochemistry; petrogenesis; redox state; Re/Os isotopes; Ti enrichment     

Early-Middle Jurassic Dolerite Dykes from Western Dronning Maud Land (Antarctica): Identifying Mantle Sources in the Karoo Large Igneous Province

A suite of dolerite dykes from the Ahlmannryggen region of westernDronning Maud Land (Antarctica) forms part of the much moreextensive Karoo igneous province of southern Africa. The dykecompositions include both low- and high-Ti magma types, includingpicrites and ferropicrites. New ⁴⁰Ar/³⁹Ar age determinationsfor the Ahlmannryggen intrusions indicate two ages of emplacementat 178 and 190 Ma. Four geochemical groups of dykes have beenidentified in the Ahlmannryggen region based on analyses of60 dykes. The groups are defined on the basis of whole-rockTiO₂ and Zr contents, and reinforced by rare earth element (REE),⁸⁷Sr/⁸⁶Sr and ¹⁴³Nd/¹⁴⁴Nd isotope data. Group 1 were intrudedat 190 Ma and have low TiO₂ and Zr contents and a significantArchaean crustal component, but also evidence of hydrothermalalteration. Group 2 dykes were intruded at 178 Ma; they havelow to moderate TiO₂ and Zr contents and are interpreted tobe the result of mixing of melts derived from an isotopicallydepleted source with small melt fractions of an enriched lithosphericmantle source. Group 3 dyke were intruded at 190 Ma and formthe most distinct magma group; these are largely picritic withsuperficially mid-ocean ridge basalt (MORB)-like chemistry (flatREE patterns, ⁸⁷Sr/⁸⁶Sr_i 0·7035, Nd_i 9). However, theyhave very high TiO₂ (4 wt %) and Zr (500 ppm) contents, whichis not consistent with melting of MORB-source mantle. The Group3 magmas are inferred to be derived by partial melting of astrongly depleted mantle source in the garnet stability field.This group includes several high Mg–Fe dykes (ferropicrites),which are interpreted as high-temperature melts. Some Group3 dykes also show evidence of contamination by continental crust.Group 4 dykes are low-K picrites intruded at 178 Ma; they havevery high TiO₂–Zr contents and are the most enriched magmagroup of the Karoo–Antarctic province, with ocean-islandbasalt (OIB)-like chemistry. Dykes of Group 1 and Group 3 aresub-parallel (ENE–WSW) and both groups were emplaced at190 Ma in response to the same regional stress field, whichhad changed by 178 Ma, when Group 2 and Group 4 dykes were intrudedalong a dominantly NNE–SSW strike. KEY WORDS: flood basalt; depleted mantle; enriched mantle; Ahlmannryggen; Karoo dyke     

Mineralogy and Petrology of the Aries Diamondiferous Kimberlite Pipe, Central Kimberley Block, Western Australia

Aries is a deeply weathered micaceous kimberlite pipe (820 Ma)consisting of four lobes: South, Central, North, and North Extension.It is the largest ( 18 ha) and most diamondiferous of the fewkimberlites currently known on the Australian continent, andis rich in country-rock (dolerite and quartzite) xenoliths.Three textural varieties of Aries kimberlites can be recognized,together with autoclastic breccias: (1) macrocrystalmedium-grained;(2) aphanitic (5 vol. % olivine macrocrysts); and (3) macrocrystalsegregated. The kimberlites contain two generations of olivinepseudomorphs (30–40 vol. %), and two of phlogopite (upto 60 vol. %), in a groundmass of apatite, calcite, diopside,sphene, spinels, serpentine, talc, and accessory groundmassminerals including aeschynite [(Ce, Ca) (Ti, Nb)₂O₆], barite,ilmenite, monazite, rutile, siderite, and unidentified Nb-Fe-titanates.Phlogopite zoning is complex and differs from lobe to lobe,but general compositions and trends resemble phlogopites fromkimberlites (TiO₂ 0–5–4 wt. %, A1₂O₃ 9–16%);tetraferriphlogopite substitution is indicated by low Al insome grains. Diopside is low in Cr, Al, Na, and Ti, with highmg-number [molecular Mg/(Mg + Fe²⁺) 93]. Apatite contains upto 17–5% SrO, calcite up to 1–7% SrO but littleMgO or FeO, sphene up to 1.5% Nb₂O₅, and ilmenite 2.6% Nb2O5and 16% MnO but no detectable MgO. Extremely complex moqftiological, textural, and compositionalvariations are present in spinels. They can be divided intofive textural-genetic types: cognate Groundmass chromian spinels(Type G); Inclusions of chromian spinels in olivine macrocrysts(Type I), probably representing either early phenocrysts ormantle xenocrysts: Macrocryst chromian spinels (Type M), probablyrepresenting xenocrysts; late-stage groundmass Fenian spinels(Type F), derived from serpentinization of olivine; Alterationferrian spinels (Type A), found as inclusions associated withsiliceous melt inclusions, in Types I and M, and probably representinginteraction of these earlier types with late-stage melts. Someof these, particularly Types M and F, show further texturalsub-types with no obvious genetic significance. The pipe formed from several magma-pulses. All four lobes maycontain at least one pulse in common, but Central and SouthLobes include additional pulse(s) which yielded distinctivephlogopite zoning, whereas North Lobe and North Extension includepulsc(s) which may have originated at higher mantle levels andyielded more evolved phlogopites. Aries most resembles South African Group II kimberlites mineralogically,certain West African micaceous kimberlites geochemically, andGroup I kimberlites isotopically. A distinctive mantle source-regionis implied by high Nb/U, Ce/Sr, Ce/P, Rb/Ba, and especiallyNb/Zr ratios. Similar anomalous geochemical signatures are sharedwith two other contemporaneous (800 Ma) lampro-phyric intrusionsin the east Kimberley (at Maude Creek and Bow Hill), suggestingthat a scattered alkaline province exists in the Kimberley Block,generated from a regionally anomalous mantle source.     

Geochemistry of South African On- and Off-craton, Group I and Group II Kimberlites: Petrogenesis and Source Region Evolution

Bulk-rock geochemical compositions of hypabyssal kimberlites,emplaced through the Archaean Kaapvaal craton and ProterozoicNamaqua–Natal belt, are used to estimate close-to-primarymagma compositions of Group I kimberlites (Mg-number = 0·82–0·87;22–28 wt % MgO; 21–30 wt % SiO₂; 10–17 wt% CaO; 0·2–1·7 wt % K₂O) and Group II kimberlites(Mg-number = 0·86–0·89; 23–29 wt %MgO; 28–36 wt % SiO₂; 8–13 wt % CaO; 1·6–4·6wt % K₂O). Group I kimberlites are distinguished from GroupII by their lower Ba/Nb (<12), Th/Nb (<1·1) andLa/Nb (<1·1) but higher Ce/Pb (>22) ratios. Thedistinct rare earth element patterns of the two types of kimberlitesindicate a more highly metasomatized source for Group II kimberlites,with more residual clinopyroxene and less residual garnet. Thesimilarity of Sr and Nd isotope ratios and diagnostic traceelement ratios (Ce/Pb, Nb/U, La/Nb, Ba/Nb, Th/Nb) of Group Ikimberlites to ocean island basalts (OIB), but more refractoryMg-numbers and Ni contents, are consistent with derivation ofGroup I kimberlites from subcontinental lithospheric mantle(SCLM) that has been enriched by OIB-like melts or fluids. Sourceenrichment ages and plate reconstructions support a direct associationof these melts or fluids with Mesozoic upwelling beneath southernAfrica of a mantle plume(s), at present located beneath thesouthern South Atlantic Ocean. In contrast, the geochemicalcharacteristics of both on- and off-craton Group II kimberlitesshow strong similarity to calc-alkaline magmas, particularlyin their Nb and Ta depletion and Pb enrichment. It is suggestedthat Group II kimberlites are derived from both Archaean andProterozoic lithospheric mantle source regions metasomatizedby melts or fluids associated with ancient subduction events,unrelated to mantle plume upwelling. The upwelling of mantleplumes beneath southern Africa during the Mesozoic, at the timeof Gondwana break-up, may have acted as a heat source for partialmelting of the SCLM and the generation of both Group I and GroupII kimberlite magmas. KEY WORDS: kimberlite; geochemistry; petrogenesis; mantle plumes; South Africa     

Geochemical and Isotopic Heterogeneities along an Island Arc-Spreading Ridge Intersection: Evidence from the Lewis Hills, Bay of Islands Ophiolite, Newfoundland

This study focuses on the origin of magma heterogeneity andthe genesis of refractory, boninite-type magmas along an arc–ridgeintersection, exposed in the Lewis Hills (Bay of Islands Ophiolite).The Lewis Hills contain the fossil fracture zone contact betweena split island arc and its related marginal oceanic basin. Threetypes of intrusions, which are closely related to this narrowtectonic boundary, have been investigated. Parental melts inequilibrium with the ultramafic cumulates of the PyroxeniteSuite are inferred to have high MgO contents and low Al₂O₃,Na₂O and TiO₂ contents. The trace element signatures of thesePyroxenite Suite parental melts indicate a re-enriched, highlydepleted source with 0·1 x mid-ocean ridge basalt (MORB)abundances of the heavy rare earth elements (HREE). Initial_Nd values of the Pyroxenite Suite range from -1·5 to+0·6, which overlap those observed for the island arc.Furthermore, the Pyroxenite Suite parental melts bear strongsimilarities to boninite-type equilibrium melts from islandarc-related pyroxenitic dykes and harzburgites. Basaltic dykessplit into two groups. Group I dykes have 0·6 x MORBabundances of the HREE, and initial _Nd values ranging from +5·4to +7·5. Thus, they have a strong geochemical affinitywith basalts derived from the marginal basin spreading ridge.Group II dykes have comparatively lower trace element abundances(0·3 x MORB abundances of HREE), and slightly lower initial_Nd values (+5·4 to +5·9). The geochemical characteristicsof the Group II dykes are transitional between those of GroupI dykes and the Pyroxenite Suite parental melts. Cumulates fromthe Late Intrusion Suite are similarly transitional, with _Ndvalues ranging from +2·9 to +4·6. We suggest thatthe magma heterogeneity observed in the Lewis Hills is due tothe involvement of two compositionally distinct mantle sources,which are the sub-island lithospheric mantle and the asthenosphericmarginal basin mantle. It is likely that the refractory, boninite-typeparental melts of the Pyroxenite Suite result from remeltingof the sub-arc lithospheric mantle at an arc–ridge intersection.Furthermore, it is suggested that the thermal-dynamic conditionsof the transtensional transform fault have provided the prerequisitefor generating magma heterogeneity, as a result of mixing relationshipsbetween arc-related and marginal basin-related magmas. KEY WORDS: Bay of Islands ophiolite; transform (arc)–ridge intersection; boninites; rare earth elements, Nd isotopes     

Primitive Magma From the Jericho Pipe, N.W.T., Canada: Constraints on Primary Kimberlite Melt Chemistry

We report the first estimates of primary kimberlite melt compositionfrom the Slave craton, based on samples of aphanitic kimberlitefrom the Jericho kimberlite pipe, N.W.T., Canada. Three samplesderive from the margins of dykes where kimberlite chilled againstwall rock (JD51, JD69 and JD82) and are shown to be texturallyconsistent with crystallization from a melt. Samples JD69 andJD82 have geochemical characteristics of primitive melts: theyhave high MgO (20–25 wt %), high mg-numbers (86–88),and high Cr (1300–1900 ppm) and Ni (800–1400 ppm)contents. They also have high contents of CO₂ (10–17 wt%). Relative to bulk macrocrystal kimberlite, they have lowermg-numbers and lower MgO but are enriched in incompatible elements(e.g. Zr, Nb and Y), because the bulk kimberlite compositionsare strongly controlled by accumulation of mantle olivine andother macrocrysts. The compositions of aphanitic kimberlitefrom Jericho are similar to melts produced experimentally bypartial melting of a carbonate-bearing garnet lherzolite. Onthe basis of these experimental data, we show that the primarymagmas from the Jericho kimberlite could represent 0·7–0·9%melting of a carbonated lherzolitic mantle source at pressuresand temperatures found in the uppermost asthenosphere to theSlave craton. The measured CO₂ contents for samples JD69 andJD82 are only slightly lower than the CO₂ contents of the correspondingexperimental melts; this suggests that the earliest hypabyssalphase of the Jericho kimberlite retained most of its originalvolatile content. As such these samples provide a minimum CO₂content for the primary kimberlite magmas from the Slave craton. KEY WORDS: kimberlite; melt; primitive; primary magma; Slave craton     

Origin of CFB Magmatism: Multi-tiered Intracrustal Picrite-Rhyolite Magmatic Plumbing at Spitzkoppe, Western Namibia, during Early Cretaceous Etendeka Magmatism

Early Cretaceous tholeiitic picrite-to-rhyolite dykes aroundSpitzkoppe, western Namibia, are part of the extensive HentiesBay–Outjo swarm, penecontemporaneous with 132 Ma Etendekalavas 100 km to the NW. Although only intermediate to rhyoliticdykes contain clinopyroxene phenocrysts, the behaviour of Ca,Al and Sc in the dyke suite shows that liquidus clinopyroxene—togetherwith olivine—was a fractionating phase when MgO fell to9 wt %. Both a plot of CIPW normative di–hy–ol–ne–Qand modelling using (p)MELTS show that a mid-crustal pressureof 0·6 GPa is consistent with this early clinopyroxenesaturation. Sr, Nd, Hf and Pb isotope variations all show trendsconsistent with AFC contamination (assimilation linked to fractionalcrystallization), involving Pan-African Damara belt continentalcrust. The geochemical variation, including isenthalpic AFCmodelling using (p)MELTS, suggests that the picrites (olivine-richcumulate suspensions) were interacting with granulite-faciesmetamorphic lower crust, the intermediate compositions withamphibolite-facies middle crust, and the rhyolitic dykes (anda few of the basalts) with the Pan-African granites of the uppercrust. The calculated densities of the magmas fall systematicallyfrom picrite to rhyolite and suggest a magmatic system resemblinga stack of sills throughout the crust beneath Spitzkoppe, withthe storage and fractionation depth of each magma fraction controlledby its density. Elemental and isotopic features of the 20 wt% MgO picrites (including Os isotopes) suggest that their parentalmelts probably originated by fusion of mid-ocean ridge basalt(MORB) source convecting mantle, followed by limited reactionwith sub-continental lithospheric mantle metasomatized justprior to the formation of the parental magmas. Many of the distinctivefeatures of large-volume picritic–basaltic magmas maynot be derived from their ultimate mantle sources, but may insteadbe the results of complex polybaric fractional crystallizationand multi-component crustal contamination. KEY WORDS: flood basalts; Spitzkoppe; picrite; trace elements; hafnium isotopes; Etendeka     

The Geochemistry of the Arabian Lithospheric Mantle--a Source for Intraplate Volcanism?

We present trace element and Sr–Nd–Hf–Pb isotopecompositions for clinopyroxenes from anhydrous spinel peridotiteand garnet ± spinel pyroxenite xenoliths of Pan-Africanlithospheric mantle from Jordan, including the first high-precisiondouble-spike Pb isotope measurements of mantle clinopyroxene.Clinopyroxenes from the peridotites are variably Th–U–LILE–LREEenriched and display prominent negative Nb, Zr and Ti anomalies.MREE–HREE abundances can generally be modelled as partialmelting residues of spinel lherzolite with primitive-mantle-likecomposition after extraction of 5–10% melt, whereas theenrichments in Th–U–LILE–LREE require a Pan-Africanor later metasomatic event. The large range of Nd, Sr, Pb andHf isotope ratios in both peridotites and pyroxenites (e.g.Nd 1·4–17·5; ²⁰⁶Pb/²⁰⁴Pb 17·2–20·4;Hf 0·6–164·6) encompasses compositionsmore radiogenic than mid-ocean ridge basalt (MORB), and Pb isotopescover almost the entire range of oceanic basalt values. Hf valuesare some of the highest ever recorded in mantle samples andare decoupled from Nd in the same samples. Marked correlationsbetween Sr–Nd–Pb isotopes, LILE–LREE enrichmentsand HFSE depletion suggest that the metasomatizing agent wasa carbonatitic-rich melt and isotopic data suggest that metasomatismmay have been related to Pan-African subduction. The metasomaticmelt permeated depleted upper mantle (<16 kbar) during Pan-Africansubduction at 600–900 Ma, and the variably metasomatizedmaterial was then incorporated into the Arabian lithosphericmantle. There is no evidence for recent metasomatism (<30Ma) related to the Afar plume like that postulated to have affectedsouthern Arabian lithospheric mantle. Hf isotopes in the mantleclinopyroxenes are unaffected by metasomatism, and even somestrongly overprinted lithologies record ancient (>1·2Ga) pre-metasomatic Lu–Hf signatures of the depleted uppermantle that was the protolith of the Arabian lithospheric mantle.The ‘resistance’ of the Lu–Hf isotopic systemto later metasomatic events resulted in the development of extremelyheterogeneous Hf isotopic signatures over time that are decoupledfrom other isotopic systems. No mantle sample in this studyexactly matches the chemical and isotopic signature of the sourceof Jordanian intraplate basalts. However, the xenolith compositionsare broadly similar to those of the source of Arabian intraplatebasalts, suggesting that the numerous Cenozoic intraplate volcanicfields throughout Arabia may be the product of melting uppermantle wedge material fertilized during Pan-African subductionand incorporated into the Arabian lithospheric mantle. We proposea model whereby the proto-Arabian lithospheric mantle underwenta major melting event in early Proterozoic–late Archeantimes (at the earliest at 1·2 Ga). Island-arc volcanismand major crust formation occurred during the Pan-African orogeny,which liberated fluids and possibly small-degree melts thatmigrated through the mantle creating zones of enrichment forcertain elements depending upon their compatibility. Immobileelements, such as Nb, were concentrated near the base of themantle wedge providing the source of the Nb-rich Jordanian volcanicrocks. More mobile elements, such as LILE and LREE, were transportedup through the mantle and fertilized the shallow mantle sourceof the Jordanian xenoliths. Following subduction, the mantlewedge became fossilized and preserved distinct enriched anddepleted zones. Lithospheric rifting in the Miocene triggeredpartial melting of spinel-facies mantle in the lower lithosphere,which mixed with deeper asthenospheric garnet-facies melts asrifting evolved. These melts entrained segments of variablycarbonatite-metasomatized shallow lithospheric mantle en routeto the surface. KEY WORDS: Arabian lithospheric mantle; Jordan; mantle xenoliths; Sr–Nd–Hf–Pb isotopes     

Comment on 'Mexican Peridotite Xenoliths and Tectonic Terranes: Correlations among Vent Location, Texture, Temperature, Pressure, and Oxygen Fugacity' by J. F. Luhr & J. J. Aranda-Gomez (1997)

This comment addresses the interpretation of oxygen fugacitydata for spinel peridotite xenoliths from five Mexican volcanicfields presented by Luhr & Aranda-Gomez (Journal of Petrology,38, 1075–1112, 1997). The postulated east–west increaseof the FMQ (‘relative oxygen fugacity’, where FMQis fayalite–magnetite–quartz) values is inherentto the method and therefore of questionable geological significance.Increases in FMQ do not necessarily mirror oxidation processesin the mantle controlled by subduction-related fluids. KEY WORDS: mantle metasomatism; Mexico; peridotite xenoliths; relative oxygen fugacity     

High-Pressure Melting of Eclogite and the P-T-X History of Tonalitic to Trondhjemitic Zoisite-Pegmatites, Munchberg Massif, Germany

Zoisite-bearing high-pressure pegmatites from the MünchbergMassif, Germany, provide an excellent example of the characteristicsof the onset of metabasite melting at eclogite-facies conditions.The pegmatites were derived by partial melting of a mid-oceanridge basalt (MORB)-like eclogite at T 680°C/2·3GPa to 750°C/3·1 GPa, which produced small amountsof tonalitic to trondhjemitic melt. The melt concentrated locallyin isolated, small melt pockets and crystallized primary zoisiteas liquidus phase at P 2·3 GPa/680°C to 2·1GPa/750°C. Compositional zoning of pegmatite zoisite recordsan ensuing multi-stage uplift history with successive, discretecrystallization events at 1·4 ± 0·2 GPa/650–700°Cand 1·0 ± 0·1 GPa/620–650°C.Resorption textures indicate reheating and thermal perturbationof the whole system prior to each successive crystallizationevent. Final solidification of zoisite-pegmatites occurred at0·9 ± 0·1 GPa/620–650°C. Thedata suggest that isolated melt + zoisite crystal mush pocketsformed an integral part of the eclogite throughout uplift frommelt formation at T 680°C/2·3 GPa to 750°C/3·1GPa to final solidification at 0·9 GPa/620–630°C;that is, over a depth range of 45–60 km. The entire pegmatite-formingprocess was probably fluid conserving: fluid present duringmelt formation was trapped by fully or nearly water-saturatedsiliceous melts, whereas fluid liberated during pegmatite crystallizationinteracted with dehydrated eclogite-facies assemblages to formamphibolite-facies hydrous minerals. A set of empirical D^{melt/eclogite}values based on mean zoisite-pegmatite and eclogite compositionwere used to model the onset of partial high-pressure meltingof metabasites. KEY WORDS: adakite; high-pressure melting; pegmatite; trondhjemite; zoisite     

Fenitizing Processes Induced by Ferrocarbonatite Magmatism at Swartbooisdrif, NW Namibia

The southeastern margin of the anorthositic Kunene IntrusiveComplex, NW Namibia, has been subsequently invaded by Mesoproterozoicsyenite, nepheline syenite and ferrocarbonatite dykes alongNE- and SE-trending faults. The first generation of carbonatiteintrusions frequently contains fenitized anorthositic wall-rockfragments set in a ferrocarbonatite matrix; later, subordinateveins of massive ferrocarbonatite are almost xenolith-free andcut through the main carbonatite dykes. A mantle source forboth carbonatite generations is constrained by their respectiveoxygen and carbon isotope compositions of ankerite (¹⁸O_SMOW8·91–9·73; ¹³C_PDB –6·98 to–6·76). Na-rich fluids, released from the meltparental to the ferrocarbonatites, caused the fenitization ofboth the incorporated anorthosite xenoliths and the borderinganorthosite, syenite and nepheline syenite. This process ismainly characterized by the progressive transformation of Ca-richplagioclase, K-feldspar and nepheline into albite and/or sodalite.The changing mineral modes indicate that the fenitizing fluidswere sodium-rich and strongly Si-deficient solutions, whichalso contained significant amounts of Sr, Ba, Nb and the lightrare earth elements. On the basis of mineral equilibria studies,it is possible to reconstruct the temperature conditions forcarbonatite emplacement (c. 830 ± 200°C) and recrystallization(c. 480 ± 130°C), and for the metasomatic formationof sodalite (c. 700 ± 70°C). KEY WORDS: anorthosite; fenitization; ferrocarbonatite; sodalite; stable isotopes     

Experimental Constraints on the Role of Garnet Pyroxenite in the Genesis of High-Fe Mantle Plume Derived Melts

The anhydrous phase relations of an uncontaminated (primitive),ferropicrite lava from the base of the Early Cretaceous Paraná–Etendekacontinental flood basalt province have been determined between1 atm and 7 GPa. The sample has high contents of MgO (14·9wt %), FeO^* (14·9 wt %) and Ni (660 ppm). Olivine phenocrystshave maximum Fo contents of 85 and are in equilibrium with thebulk rock, assuming a of 0·32. A comparison of our results with previous experimental studiesof high-Mg rocks shows that the high FeO content of the ferropicritecauses an expansion of the liquidus crystallization field ofgarnet and clinopyroxene relative to olivine; orthopyroxenewas not observed in any of our experiments. The high FeO contentalso decreases solidus temperatures. Phase relations indicatethat the ferropicrite melt last equilibrated either at 2·2GPa with an olivine–clinopyroxene residue, or at 5 GPawith a garnet–clinopyroxene residue. The low bulk-rockAl₂O₃ content (9 wt %) and high [Gd/Yb]_n ratio (3·1)are consistent with the presence of residual garnet in the ferropicritemelt source and favour high-pressure melting of a garnet pyroxenitesource. The garnet pyroxenite may represent subducted oceaniclithosphere entrained by the upwelling Tristan starting mantleplume head. During adiabatic decompression, intersection ofthe garnet pyroxenite solidus at 5 GPa would occur at a mantlepotential temperature of 1550°C and yield a ferropicriteprimary magma. Subsequent melting of the surrounding peridotiteat 4·5 GPa may be restricted by the thickness of theoverlying sub-continental lithosphere, such that dilution ofthe garnet pyroxenite melt component would be significantlyless than in intra-oceanic plate settings (where the lithosphereis thinner). This model may explain the limited occurrence offerropicrites at the base of continental flood basalt sequencesand their apparent absence in ocean-island basalt successions. KEY WORDS: continental flood basalt; ferropicrite; mantle heterogeneity; mantle melting; phase relations; pyroxenite     

Hafnium Isotope and Trace Element Constraints on the Nature of Mantle Heterogeneity beneath the Central Southwest Indian Ridge (13{degrees}E to 47{degrees}E)

Hafnium isotope and incompatible trace element data are presentedfor a suite of mid-ocean ridge basalts (MORB) from 13 to 47°Eon the Southwest Indian Ridge (SWIR), one of the slowest spreadingand most isotopically heterogeneous mid-ocean ridges. Variationsin Nd–Hf isotope compositions and Lu/Hf ratios clearlydistinguish an Atlantic–Pacific-type MORB source, presentwest of 26°E, characterized by relatively low _Hf valuesfor a given _Nd relative to the regression line through all Nd–Hfisotope data for oceanic basalts (termed the ‘Nd–Hfmantle array line’; the deviation from this line is termed_Hf) and low Lu/Hf ratios, from an Indian Ocean-type MORB signature,present east of 32°E, characterized by relatively high _Hfvalues and Lu/Hf ratios. Additionally, two localized, isotopicallyanomalous areas, at 13–15°E and 39–41°E,are characterized by distinctly low negative and high positive_Hf values, respectively. The low _Hf MORB from 13 to 15°Eappear to reflect contamination by HIMU-type mantle from thenearby Bouvet mantle plume, whereas the trace element and isotopiccompositions of MORB from 39 to 41°E are most consistentwith contamination by metasomatized Archean continental lithosphericmantle. Relatively small source-melt fractionation of Lu/Hfrelative to Sm/Nd, compared with MORB from faster-spreadingridges, argues against a significant role for garnet pyroxenitein the generation of most central SWIR MORB. Correlations between_Hf and Sr and Pb isotopic and trace element ratios clearly delineatea high-_Hf ‘Indian Ocean mantle component’ that canexplain the isotope composition of most Indian Ocean MORB asmixtures between this component and a heterogeneous Atlantic–Pacific-typeMORB source. The Hf, Nd and Sr isotope compositions of IndianOcean MORB appear to be most consistent with the hypothesisthat this component represents fragments of subduction-modifiedlithospheric mantle beneath Proterozoic orogenic belts thatfoundered into the nascent Indian Ocean upper mantle duringthe Mesozoic breakup of Gondwana. KEY WORDS: mid-ocean ridge basalt; isotopes; incompatible elements; Indian Ocean     

Cr-Saturation Arrays in Concentrate Garnet Compositions from Kimberlite and their Use in Mantle Barometry

The spinel–garnet transition in Cr/Al-enriched peridotiticbulk compositions is known from experimental investigationsto occur at 20–70 kbar, within the pressure range sampledby kimberlites. We show that the Cr₂O₃–CaO compositionsof concentrate garnets from kimberlite have maximum Cr/Ca arrayscharacterized by Cr₂O₃/CaO 0·96–0·81, andinterpret the arrays as primary evidence of chromite–garnetcoexistence in Cr-rich harzburgitic or lherzolitic bulk compositionsderived from depth within the lithosphere. Under Cr-saturatedconditions on a known geotherm, each Cr/Ca array implicitlydelineates an isobar inside a garnet Cr₂O₃–CaO diagram.This simplification invites a graphical approach to calibratean empirical Cr/Ca-in-pyrope barometer. Carbonaceous chromite–garnetharzburgite xenoliths from the Roberts Victor kimberlite tightlybracket a graphite–diamond constraint (GDC) located atCr₂O₃ = 0·94CaO + 5·0 (wt %), representing a pivotalcalibration corresponding to 43 kbar on a 38 mW/m² conductivegeotherm. Additional calibration points are established at 14,17·4 and 59·1 kbar by judiciously projecting garnetcompositions from simple-system experiments onto the same geotherm.The garnet Cr/Ca barometer is then simply formulated as follows(in wt %):

if Cr₂O₃ 0·94CaO + 5, then P₃₈ (kbar) = 26·9+ 3·22Cr₂O₃ – 3·03CaO, or

if Cr₂O₃ <0·94CaO + 5, then P₃₈ (kbar) = 9·2+ 36[(Cr₂O₃+ 1·6)/(CaO + 7·02)].

A small correction to P₃₈ values, applicable for 35–48mW/m² conductive geotherms, is derived empirically by requiringconventional thermobarometry results and garnet concentratecompositions to be consistent with the presence of diamondsin the Kyle Lake kimberlite and their absence in the Zero kimberlite.We discuss application of the P₃₈ barometer to estimate (1)real pressures in the special case where chromite–garnetcoexistence is known, (2) minimum pressures in the general casewhere Cr saturation is unknown, and (3) the maximum depth ofdepleted lithospheres, particularly those underlying Archaeancratons. A comparison with the P_Cr barometer of Ryan et al.(1996, Journal of Geophysical Research 101, 5611–5625)shows agreement with P₃₈ at 55 ± 2 kbar, and 6–12%higher P_Cr values at lower P₃₈. Because the P_Cr formulationsystematically overestimates the 43 kbar value of the GDC by2–6 kbar, we conclude that the empirical Cr/Ca-in-garnetbarometer is preferred for all situations where conductive geothermsintersect the graphite–diamond equilibrium. KEY WORDS: Cr-pyrope; chromite; P₃₈ barometer; mantle petrology; lithosphere thickness     

Genesis of Ultramafic Lamprophyres and Carbonatites at Aillik Bay, Labrador: a Consequence of Incipient Lithospheric Thinning beneath the North Atlantic Craton

Numerous dykes of ultramafic lamprophyre (aillikite, mela-aillikite,damtjernite) and subordinate dolomite-bearing carbonatite withU–Pb perovskite emplacement ages of 590–555 Ma occurin the vicinity of Aillik Bay, coastal Labrador. The ultramaficlamprophyres principally consist of olivine and phlogopite phenocrystsin a carbonate- or clinopyroxene-dominated groundmass. Ti-richprimary garnet (kimzeyite and Ti-andradite) typically occursat the aillikite type locality and is considered diagnosticfor ultramafic lamprophyre–carbonatite suites. Titanianaluminous phlogopite and clinopyroxene, as well as comparativelyAl-enriched but Cr–Mg-poor spinel (Cr-number < 0.85),are compositionally distinct from analogous minerals in kimberlites,orangeites and olivine lamproites, indicating different magmageneses. The Aillik Bay ultramafic lamprophyres and carbonatiteshave variable but overlapping ⁸⁷Sr/⁸⁶Sr_i ratios (0·70369–0·70662)and show a narrow range in initial _Nd (+0·1 to +1·9)implying that they are related to a common type of parentalmagma with variable isotopic characteristics. Aillikite is closestto this primary magma composition in terms of MgO (15–20wt %) and Ni (200–574 ppm) content; the abundant groundmasscarbonate has ¹³C_PDB between –5·7 and –5,similar to primary mantle-derived carbonates, and ¹⁸O_SMOW from9·4 to 11·6. Extensive melting of a garnet peridotitesource region containing carbonate- and phlogopite-rich veinsat 4–7 GPa triggered by enhanced lithospheric extensioncan account for the volatile-bearing, potassic, incompatibleelement enriched and MgO-rich nature of the proto-aillikitemagma. It is argued that low-degree potassic silicate to carbonatiticmelts from upwelling asthenosphere infiltrated the cold baseof the stretched lithosphere and solidified as veins, therebycrystallizing calcite and phlogopite that were not in equilibriumwith peridotite. Continued Late Neoproterozoic lithosphericthinning, with progressive upwelling of the asthenosphere beneatha developing rift branch in this part of the North Atlanticcraton, caused further veining and successive remelting of veinsplus volatile-fluxed melting of the host fertile garnet peridotite,giving rise to long-lasting hybrid ultramafic lamprophyre magmaproduction in conjunction with the break-up of the Rodinia supercontinent.Proto-aillikite magma reached the surface only after coatingthe uppermost mantle conduits with glimmeritic material, whichcaused minor alkali loss. At intrusion level, carbonate separationfrom this aillikite magma resulted in fractionated dolomite-bearingcarbonatites (¹³C_PDB –3·7 to –2·7)and carbonate-poor mela-aillikite residues. Damtjernites maybe explained by liquid exsolution from alkali-rich proto-aillikitemagma batches that moved through previously reaction-lined conduitsat uppermost mantle depths. KEY WORDS: liquid immiscibility; mantle-derived magmas; metasomatism, Sr–Nd isotopes; U–Pb geochronology     

Lower Crustal Magma Genesis and Preservation: a Stochastic Framework for the Evaluation of Basalt-Crust Interaction

We present a quantitative assessment of the thermal and dynamicresponse of an amphibolitic lower crust to the intrusion ofbasaltic dike swarms in an arc setting. We consider the effectof variable intrusion geometry, depth of intrusion, and basaltflux on the production, persistence, and interaction of basalticand crustal melt in a stochastic computational framework. Distinctmelting and mixing environments are predicted as a result ofthe crustal thickness and age of the arc system. Shallow crustal(30 km) environments and arc settings with low fluxes of mantle-derivedbasalt are likely repositories of isolated pods of mantle andcrustal melts in the lower crust, both converging on daciticto rhyodacitic composition. These may be preferentially rejuvenatedin subsequent intrusive episodes. Mature arc systems with thickercrust (50 km) produce higher crustal and residual basaltic meltfractions, reaching 0·4 for geologically reasonable basaltfluxes. The basaltic to basaltic andesite composition of bothcrustal and mantle melts will facilitate mixing as the networkof dikes collapses, and Reynolds numbers reach 10^–4–1·0in the interiors of dikes that have been breached by ascendingcrustal melts. This may provide one mechanism for melting, assimilation,storage and homogenization (MASH)-like processes. Residual mineralassemblages of crust thickened by repeated intrusion are predictedto be garnet pyroxenitic, which are denser than mantle peridotiteand also generate convective instabilities where some of thecrustal material is lost to the mantle. This reconciles thethinner than predicted crust in regions that have undergonea large flux of mantle basalt for a prolonged period of time,and helps explain the enrichment of incompatible elements suchas K₂O, typical of mature arc settings, without the associatedmass balance problem. KEY WORDS: crustal anatexis; delamination; lower crust; magma mixing; thermal model