Mantle Source Components Research Articles

Dawn of Phanerozoic orogeny in the North Atlantic tract; Evidence from the Seve-Kalak Superterrane, Scandinavian Caledonides

Despite their Early Phanerozoic age, the Scandinavian Caledonides provide a remarkably rich record of the continental break-up and development of the continent-ocean transition as well as the ensuing subduction and imbrication of the pristine plate margin, all emplaced on land and accessible. We first organize the evidence for Baltoscandian rift basin formation and magmatism, now scattered in two major nappe complexes (and by semantics) in terms of a superterrane, the Seve-Kalak Superterrane. Extensive (1000 km) mafic dyke swarms and partly sheeted dyke complexes are interpreted as fragments of a Large Igneous Province. While attempted break-up and some tholeiitic magmatism took place already at c. 800 Ma, successful rifting occurred first in connection with intense, partly alkaline dyking and emplacement of ultramafic complexes between c. 620 and 550 Ma. This late magmatism is markedly enriched as compared to MORB, interpreted to reflect mixing between an enriched mantle source component and depleted mantle. The evidence for Early Ordovician imbrication and subduction of the rifted and intruded margin is then reviewed. In order to explain the preservation of parts of the continent-ocean transition and rift basins, we suggest early detachment and thrusting towards the foreland, by analogy with the emplacement of the Semail Ophiolite. Intercalated garnet peridotites require incorporation of subcontinental mantle fragments during imbrication. Structures and fabrics of eclogites and their host rocks suggest that extensional tectonics was important during their Early Ordovician exhumation. If so, how much of Scandian extensional tectonics is instead inherited Finnmarkian?

Isotopic and trace element constraints on the genesis of a boninitic sequence in the Thetford Mines ophiolitic complex, Quebec, Canada

The Mont Ham Massif (part of the Thetford Mines ophiolite, south Quebec) represents a magmatic sequence made up of tholeiitic and boninitic derived products. A geochemical study confirms the multicomponent mixing models that have been classically advanced for the source of boninites, with slab-derived components added to the main refractory harzburgitic peridotite. An isochron diagram of the boninitic rocks is interpreted as a mixing trend between two components: (i) a light rare earth element (LREE) enriched component (A), interpreted as slab-derived fluid–melts equilibrated with sedimentary materials (εNd = −3, 147Sm/144Nd = 0.140), and (ii) a LREE-depleted component (B) (0.21 < 147Sm/144Nd < 0.23), interpreted as slab-derived fluid–melt equilibrated with recycled Iapetus oceanic crust and equated to the Nd-isotope characteristics of the Iapetus mantle (εNd = 9). A multicomponent source is also necessary to explain the Nd-isotope and trace element composition of the tholeiites, which are explained by the melting of a more fertile, lherzolitic mantle and (or) mid-ocean ridge basalt source (component C), characterized by a large-ion lithophile element depleted pattern and an Iapetus mantle Nd isotopic composition (εNd = 9), mixed in adequate proportions with the two previously infered slab-derived components (A and B). The genesis of the boninites of Mont Ham is not significantly different from those of boninites located in the Pacific. An intraoceanic subduction zone appears to be an appropriate geodynamic environment for the Mont Ham ophiolitic sequence.

Petrogenesis of Devonian lamprophyre and carbonatite minor intrusions, Kandalaksha Gulf (Kola Peninsula, Russia)

Minor magmatic intrusions (dykes and explosion pipes) of lamprophyric and carbonatitic compositions occur on several islands in the Gulf of Kandalaksha (White Sea, Kola Peninsula, Russia). The lamprophyre dykes yielded K-Ar ages of 368 ± 15 Ma and 360 ± 16 Ma, similar to the majority of alkaline rocks from the Kola Alkaline Province. Mineralogical data (presence of perovskite and sodalite, absence of amphibole phenocrysts) and geochemical data (low SiO 2 and A1 2O 3, high MgO) indicate an ultramafic lamprophyre affinity for the investigated silicate rocks. The lamprophyres contain a wide variety of xenoliths including hornblende- and biotite-rich cumulate ultramafic rocks. The carbonatite intrusions have ferrocarbonatite affinities and one of them contains xenoliths of coarse-grained Si-rich calciocarbonatites, together with abundant hornblendites and glimmerites which resemble those in the lamprophyres. The calciocarbonatite xenoliths themselves contain fragments of mica- and hornblende-rich rocks. 40Ar- 39Ar ages on phlogopite and amphibole from calciocarbonatite and hornblende-rich cumulate xenoliths are between 386 ± 1.0 Ma and 395.6 ± 4.4 Ma, indicating an early Devonian age and suggesting a close relationship between the calciocarbonatite xenoliths and the ultramafic cumulate xenoliths. Thus, the ferrocarbonatite host magma may have disrupted an older calciocarbonatite-hornblendite-glimmerite intrusion at depth and incorporated xenoliths from it. The presence of hornblendite and glimmerite xenoliths with similar parageneses and identical ages in both the ultramafic lamprophyres and ferrocarbonatites suggests a close relationship between the ferrocarbonatite and lamprophyric magmas. REE patterns of the lamprophyric dykes and calciocarbonatite xenoliths show strong similarities, indicating a petrogenetic relationship. The ultramafic lamprophyres have REE patterns which are indistingishable from the contemporaneous kimberlites and melilitites from the nearby Terskii Bereg area, south Kola Peninsula. Age-corrected Sr and Nd isotope compositions demonstrate that the calciocarbonatite xenoliths have close affinities with other Devonian carbonatites from Kola and Karelia, whereas the lamprophyres are similar to the Kola kimberlites and melilitites. Differences between the Sr and Nd isotopic ratios of the silicate and carbonatite magmas throughout the Kola Alkaline Province are probably due to different mantle-source components. The Kola carbonatites mainly show a depleted mantle signature whereas the lamprophyres, melilitites and kimberlites were derived from a more enriched mantle. However, some degree of assimilation of lower continental crust and late-stage hydrothermal alteration of the silicate magmas may also have occurred.

Geochemical evolution of rift magmas by progressive tapping of a stratified mantle source beneath the Ross Sea Rift, Northern Victoria Land, Antarctica

Source compositions of Neogene-Quaternary volcanic rocks from the McMurdo Volcanic Group of the Ross Sea Rift in Northern Victoria Land, Antarctica are constrained by NdSrPb isotopes and trace element ratios in near-primary basalts. The rocks erupted along the western rift margin (Victoria Land Basin) and the western rift shoulder (Transantarctic Mountains). Near-primary basalts show no evidence of crustal contamination, suggesting that their initial NdSrPb isotopes reflect the composition of their mantle sources. The initial isotope ratios of near-primary basalts range from about 87 Sr 86 Sr = 0.70281 to 0.70504 and 143 Nd 144 Nd = 0.51269 to 0.51291 ( ϵ Nd(t) = 1.3–5.5). The 206 Pb 204 Pb ratios vary between 19.3 and 20.1. Our results, in conjunction with data from the literature [1–3], suggest the involvement of three mantle source components (or their mixtures) during the formation of Ross Sea magmas: depleted MORB-type mantle, an enriched mantle component (EM), and a component akin to HIMU. The involvement of these mantle components during magma genesis correlates with tectonic setting: whereas MORB-type compositions are restricted to Recent within-rift basalts, EM and HIMU isotope signatures dominate off-rift magmas from the western rift shoulder and Marie Byrd Land respectively. Basalts from the western rift shoulder show temporal isotopic and trace element variations from EM towards HIMU-type signatures between 15 and 5 m.y. On the other hand, within-rift and Marie Byrd Land basalts changed from HIMU-type towards MORB-type compositions through time. These temporal geochemical variations together with the respective tectonic settings of magmatism suggest that the mantle beneath the Ross Sea Rift is stratified in the order MORB- to HIMU- to EM-type sources. It appears that during rift evolution the EM- and HIMU-type sources are consumed and depleted asthenospheric MORB-type mantle rises progressively into the melting region. This suggests that EM resides in the mantle lithosphere. The HIMU-type component may be related to the head of an active or, alternatively, ‘fossilized’ mantle plume attached to the base of the lithosphere beneath the Ross Sea area.

Petrology and geochemistry of late Tertiary/Quaternary mafic alkaline volcanism in Romania

Alkaline volcanic activity occurred in the Persani Mountains (eastern Transylvanian Basin) and Banat (eastern Pannonian Basin) regions of Romania between 2.5 Ma and 0.7 Ma. This volcanism followed an extended period of subduction-related mostly andesitic and dacitic magmatism in the Eastern Carpathian arc. The Persani Mts. alkaline activity coincided with the last phase of subduction-related activity. Several lava flows and pyroclastic deposits in the Persani Mts. carry peridotitic mantle xenoliths and amphibole megacrysts. Major- and trace-element geochemistry indicates that the alkaline magmas are primitive, silica-undersaturated alkali basalts and trachybasalts (7.8–12.3 wt.% MgO; 119–207 ppm Ni; 210–488 ppm Cr) which are LREE-enriched. Mantle-normalised trace-element diagrams reveal an overall similarity to continental intraplate alkali basalts, but when compared with a global average of ocean island basalts (OIB), the Banat lavas are similar to average OIB, whereas the Persani Mts. basalts have higher Rb, Ba, K and Pb and lower Nb, Zr and Ti. These features slightly resemble those of subduction-related magmas, particularly those of a basaltic andesite related to the nearby older arc magmas. With 87 Sr 86 Sr varying from 0.7035-0.7045 and 143 Nd 144 Nd from 0.51273-0.51289, the Romania basalts are indistinguishable from those of the western Pannonian basin (Hungary and Austria) and Neogene alkali basalts throughout Europe. Amphibole megacrysts have similar isotopic signatures, and their REE patterns indicate derivation by crystallisation from a mafic alkaline magma. The age-corrected Sr and Nd isotopic compositions of a calc-alkaline basaltic andesite related to the preceeding period of subduction also lies within the field of the younger alkaline magmas. Pb isotopic ratios of the Romanian alkali basalts do not lie on the NHRL, but overlap the field of Tertiary alkali basalts from the western Pannonian basin, and have unusually high 207 Pb 204 Pb at a given 206 Pb 204 Pb . Thus it is probable that, although the Romanian alkali basalts have a strong asthenospheric (i.e. OIB-type mantle source) component, their Pb isotopic characteristics were derived from mantle which was affected by the earlier subduction.

Mafic alkaline magmatism associated with the European Cenozoic rift system

Primitive mafic alkaline volcanic rocks from the Cenozoic rift system of western and central Europe have major and trace element and Sr-Nd-Pb isotopic characteristics which suggest the involvement of both lithospheric and asthenospheric mantle source components in their petrogenesis. The geochemical characteristics of the lithospheric component are in part constrained by those of spinel lherzolite and mafic granulite xenoliths entrained within the magmas. This component appears to be the product of partial melting of phlogopite/amphibole bearing mantle which was metasomatised as a consequence of magmatic activity during and preceeding the Hercynian orogeny and during subsequent phases of Permo-Carboniferous and Cenozoic extension. Partial melting of phlogopite is required to account for the generation of potassic magmas (leucitites and leucite nephelinites) with K 2O Na 2O ratios > 1. This component appears to differ between the individual Hercynian terrane blocks of Europe, reflecting, in part, their previous magmatic histories. The asthenospheric component has affinities with the source of HIMU OIB. This could represent a zone of enriched mantle in the thermal boundary layer at the base of the subcontinental lithosphere which is preferentially partially melted during extension. Alternatively it could, in part, be 500-400 Ma oceanic lithosphere subducted during the Hercynian orogeny. Although the data do not preclude the existence of deep mantle plumes, there is no need to invoke them in order to explain the HIMU characteristics.

Fluorine and boron geochemistry of an ensialic marginal basin volcano: Deception Island, Bransfield Strait, Antarctica

We present the initial results of a quantitative investigation of the volatile geochemistry of Deception Island, an active volcano situated near the spreading axis of a Quaternary ensialic marginal basin (Bransfield Strait, northern Antarctic Peninsula). Fluorine contents in Deception Island magmas (112–461 ppm) are comparable with lavas from a range of tectonic environments but F-K 2O relationships most closely compare with continental flood basalts and lavas from island arcs and some marginal basins. Boron contents are high (4.3–16.3 ppm) and the values overlap with those of arc lavas; they provide strong support for the presence of a mantle source component derived from the slab subducted at the coeval trench (by melting at the slab/wedge interface and/or during slab dehydration). Both F and B acted incompatibly in Deception Island magmas but there is significant variation in incompatible-element ratios such as K/F, K/B, P/F, P/B, which strongly suggests that the magmatic system was open to some or all of these elements during differentation. The variations in these ratios also provide evidence for the presence of at least two stages in the magmatic evolution of the volcano. During pre-caldera times, mafic magma was emplaced into the upper crust where it evolved and may have reacted with the crustal envelope, thus changing the contents of some or all of the elements F, B, K and P and their inter-element ratios. A later, large influx of hot, mafic magma into the chamber may have been responsible for a major eruption that ultimately led to the formation of the caldera. K/F, K/B, etc, ratios in the magma chamber were “reset” and subsequently continued to change, possibly by further crustal interaction during melt evolution in post-caldera times.

Mantle heterogeneity in northeastern Africa: evidence from Nd isotopic compositions and hygromagmaphile element geochemistry of basaltic rocks from the Gulf of Tadjoura and southern Red Sea regions

Basaltic rocks from the Gulf of Tadjoura and southern Red Sea regions have been analysed for their Nd isotopic compositions and major and trace element concentrations. The wide variation in isotopic and geochemical compositions of the basaltic rocks is best explained by the mixing phenomenon involving a variety of mantle source components. To test the mixing hypothesis, a combined use of Nd isotopes and hygromagmaphile elemental ratios is proven very powerful. Three reservoirs have been identified as minimum components in their petrogenesis: (1) DMM (depleted MORB mantle), a mantle source depleted in light rare earth elements (LREE), which is the principal component of the N-MORB type basalts of this region; (2) REC (Ramad enriched component), equivalent to the hot-spot type of source detected in the south of Red Sea; (3) TEC (Tadjoura enriched component), a rather unique component located in the region of Tadjoura Gulf; it is characterised by a relative depletion in Rb, K, Th and U in a primitive mantle- or chondrite-normalised incompatible element pattern; this component could have been produced by mantle metasomatism of an originally depleted mantle. Mixing in various proportions of the above components is considered to be the principal mechanism for the formation of basalts with such diverse isotopic and trace element compositions.

30. Evaporation and sputtering

We report the Sr, Nd and Pb isotopic ratios of clinopyroxene separates from spinel peridotite xenoliths hosted by late Cenozoic alkali basaltic rocks from Baengnyeong Island, Jogokni, and Jeju Island, all near the southeastern margin of the Sino-Korean Craton in South Korea. Measured Sr, Nd and Pb isotopic compositions in the clinopyroxene separates show, in general, simple trends between depleted MORB-like mantle (DMM) and relatively enriched components, which can be explained by old event(s) of infiltration by metasomatic melts/fluids that enriched a precursor time-integrated depleted mantle in incompatible trace elements and overprinted its isotopic composition. Comparison of the xenoliths with the host basalts suggests that infiltration of carbonatitic melts can explain the characteristics of the Baengnyeong xenoliths while reaction with residual slab-derived melt/fluid subjected to amphibole and phlogopite fractionation explains the Jogokni xenoliths best, ruling out direct influence of the magmatism that produced the host basalts. On the other hand, the Jeju clinopyroxene separates display a mixing hyperbola between DMM and the host basalts in Sr–Pb and Nd–Pb isotope correlation diagrams, suggesting some influence by liquids similar to the host basalts. The Korean xenoliths tend to be similar in the overall range of their mineralogic modes, elemental compositions, and Sr–Nd–Pb isotopic ratios to those of the fertile peridotite xenoliths derived from the post-Archean lithospheric mantle of eastern China, and yet without any compositional evidence for a refractory Archean mantle root. Overall comparison of the measured isotopic compositions for these Korean xenoliths with available data on lavas from eastern China, indicates that the dominant enriched mantle source components for the Cenozoic intraplate volcanism in the East Asia (EM1 for northeast China, and EM2 for Southeast Asia) are absent in the lithospheric mantle beneath the region. The Cenozoic intraplate volcanism, therefore, must have originated in the asthenosphere.