Elsevier

Chemical Geology

Volume 455, 20 April 2017, Pages 22-31
Chemical Geology

Multiple mantle sources of continental magmatism: Insights from “high-Ti” picrites of Karoo and other large igneous provinces

https://doi.org/10.1016/j.chemgeo.2016.08.034Get rights and content

Highlights

  • Reconstruction of primary melts of the Karoo igneous province using melt inclusions

  • “Hybrid” garnet pyroxenite in the lithospheric mantle source of high-Ti Karoo melts

  • Variable mantle sources and melting conditions for high-Ti continental suites worldwide

  • Contribution of subcontinental lithospheric mantle to some magmas in south Atlantic

Abstract

Magmas forming large igneous provinces (LIP) on continents are generated by extensive melting in the deep crust and underlying mantle and associated with break-up of ancient supercontinents, followed by formation of a new basaltic crust in the mid-oceanic rifts. A lack of the unifying model in understanding the sources of LIP magmatism is justified by lithological and geochemical complexity of erupted magmas on local (e.g. a cross-section) and regional (a single and different LIP) scales. Moreover, the majority of LIP rocks do not fit general criteria for recognizing primary/primitive melts (i.e. < 8 wt% MgO and absence of high-Fo olivine phenocrysts).

This study presents the mineralogical (olivine, Cr-spinel, orthopyroxene), geochemical (trace elements and Sr-Nd-Hf-Pb isotopes) and olivine-hosted melt inclusion compositional characteristics of a single primitive (16 wt% MgO), high-Ti (2.5 wt% TiO2) picrite with high-Mg olivine (up to 91 mol% Fo) from the Letaba Formation in the ~ 180 Ma Karoo LIP (south Africa). The olivine compositions (unusually high δ18O (6.17‰), high NiO (0.36–0.56 wt%) and low MnO and CaO (0.12–0.20 and 0.12–0.22 wt%, respectively)) are used to argue for a non-peridotitic mantle source. This is supported by the enrichment of the rock and melts in most incompatible trace elements and depletion in heavy rare earth elements (e.g. high Gd/Yb) that reflects residual garnet in the source of melting. The radiogenic isotopes resemble those of the model enriched mantle (EM-1) and further argue for a long-term enrichment of the source in incompatible trace elements.

The enriched high-Ti compositions, strongly fractionated incompatible trace elements, presence of primitive olivine and high-Cr spinel in the Letaba picrites are closely matched by olivine-phyric rocks from the ~ 260 Ma Emeishan (Yongsheng area, SW China) and ~ 250 Ma Siberian (Maimecha-Kotuy region, N Siberia) LIPs. However, many other compositional parameters (e.g. trace element and δ18O compositions of olivine phenocrysts, Fe2 +/Fe3 + in Cr-spinel, Sr-Nd-Hf isotope ratios) only partially overlap or even diverge. We thus imply that parental melts of enriched picritic rocks with forsteritic olivine from three major continental igneous provinces – Karoo, Emeishan and Siberia cannot be assigned to a common mantle source and similar melting conditions.

The Karoo picrites also exhibit some mineralogical and geochemical similarities with rocks and glasses in the south Atlantic Ridge and adjacent fracture zones. The geodynamic reconstructions of the continental plate motions since break-up of the Gondwanaland in the Jurassic support the current position of the source of the Karoo magmatism in the southernmost Atlantic. Co-occurrence of modern and recent anomalous rocks with normal mid-ocean ridge basalts in this region can be related to blocks/rafts of the ancient lithosphere, stranded in the ambient upper mantle and occasionally sampled by rifting-related decompressional melting.

Introduction

Large igneous provinces (LIPs) are a geological feature recorded throughout much of Earth history, with many conspicuous examples on all present-day continents and in several ocean basins. The main features of LIPs - exceptionally large volumes of magmas (in excess of 106 km3) erupted over relative short timescales (< 5 Ma) - require extraordinary rates of melt generation in the mantle sources of LIPs. Debate about LIP petrogenesis is intense (e.g. http://www.mantleplumes.org/; http://www.largeigneousprovinces.org/) and seeks to address issues such as commonalities in compositional characteristics of primary melts, their mantle sources and the relevant physical conditions of melt generation, e.g. temperature and pressure of melting and rheological properties of the mantle. The principal inferred magma sources of LIPs (plumes from the lower mantle; the subcontinental lithospheric mantle, SCLM) represent two endmembers in the spectrum of possible contributors to geochemical diversity recorded in LIPs. Other sources (e.g. convecting asthenosphere, continental crust) are often invoked to construct petrogenetic models that explain all observations in a given LIP. The lower mantle source, as a loosely defined “deep mantle plume”, is heterogeneous (e.g. “head & tail”) varying in composition and temperature, and comprising recycled oceanic and continental crust, as well as entrained asthenospheric mantle. Likewise, the SCLM is vertically and laterally heterogeneous, as inferred from the range of peridotite and eclogite lithologies brought to the surface as xenoliths by kimberlite and alkali basalt magmas. Asthenospheric mantle is thought to play an important role in both endmember scenarios (plume, SCLM), either as a contaminant of “deep mantle plumes” or as a source of heat and melts which modify, and ultimately lead to large-scale melting of, the overlying lithospheric mantle. LIPs emplaced into the continental crust are particularly difficult to interpret because contamination of hot mantle-derived magmas during ascent through the continental crust is inevitable.

The variety of inferred magma sources in LIPs reflects the lithological and geochemical complexity of LIP magmas on local (e.g. series of volcanic units) and regional (a single and different LIP) scales. The majority of LIP rocks contain < 8 wt% MgO and lack high-Fo olivine phenocrysts, and thus do not fit the criteria for recognizing primary/primitive melts. Modelling primary melt compositions in such cases is hampered by uncertainties in mantle source mineralogy, conditions of melting and unknown effects of chemical interaction as the magmas move away from their mantle source. This leaves such models poorly constrained and often redundant.

Here we present a case study of a single sample of picrite, from the Karoo LIP, for which an unusually broad range of petrologically relevant data are available (major and trace elements, Sr-Nd-Pb-Hf isotopes, mineral chemical and O-isotope compositions, melt inclusions). These results are compared to other Karoo eruptives, to high-Mg equivalents in the Emeishan and Siberian continental LIP provinces, and to basalts on the modern south Atlantic Ridge which share compositional characteristics with the Karoo picrite. The main conclusion of this study is that even the most primitive LIP magma and olivine compositions do not allow us to ‘see through’ the scatter of multiple sources and petrogenetic pathways. Apparently there is a wide range in LIP magma sources, implying that attempts to use one or several overlapping chemical characteristics in diverse LIPs to produce a unifying model of LIP magmatism at the planetary scale are likely to be futile. This conclusion is not a negative outcome but rather a call for compromise between proponents of conflicting models.

Section snippets

Olivine-phyric rocks in the Karoo Igneous Province

The Karoo LIP in southern Africa is a substantial part of the Early Jurassic magmatic-tectonic event associated with break-up of the Gondwana supercontinent. Extensively studied volcanic successions in south east Zimbabwe comprise at least three main lithological packages of variable but significant thickness (on average 2–3 km and up to 5 km), comprising stratigraphically early olivine-phyric basalts (picrites) overlain by olivine-free basalts and dolerites and, completing the sequence,

Results

The Letaba Formation olivine-phyric rocks used in this study have been thoroughly documented by Cox and Jamieson (1974) in Nuanetsi area, presently Mwenezi, (sample N356) and Bristow (1984) in northern Lebombo, (samples labelled KP, KS and KA; Supplementary Table S1). The Nuanetsi picrite N356 was selected for analysis of major and trace elements, radiogenic isotope ratios, compositions of olivine phenocrysts and olivine-hosted inclusions (melt, crystal, fluid), whereas the bulk rock

Constraints on primary melts and mantle sources

The dominance of primitive olivine phenocrysts (up to 91 mol% Fo) in picrite N356 supports their origin from primary or near-primary melts. Significant variations in the Fo content of the overall phenocryst population and within individual crystals (Figs. 1c, 2) suggest that their parental melts evolved in terms of Mg/Fe2 + due to olivine fractionation. The systematic behaviour of the minor elements in olivine (Fig. 2 b–d) advocates either a single primary melt or a batch of compositionally and

Conclusions

A comparison of enriched picritic rocks with highly magnesian olivine from three major continental basalt provinces (Karoo, Emeishan, Siberia) demonstrates that their primary melts cannot be attributed to a common mantle source and similar melting conditions. Inferences regarding mantle source composition, extent of melting and depth of melt segregation must be based on both whole-rock compositions and robust data on the earliest liquidus assemblage (i.e. olivine, Cr-spinel ± pyroxene).

Acknowledgements

We thank P. Robinson, S. Gilbert, K. McGoldrick, and K. Goemann for help with analytical works. The final manuscript paper benefitted from thorough and helpful reviews by Chris Harris, Marlina Elburg and Dejan Prelević and editorial handling by Sebastian Tappe. This study was initially (2003–2005) supported by the Alexander von Humboldt Foundation (Germany) in the form of the Wolfgang Paul Award to A. Sobolev and the Friedrich Wilhelm Bessel Award to V. Kamenetsky. Recent financial support was

References (63)

  • V.S. Kamenetsky et al.

    Olivine-enriched melt inclusions in chromites from a low-Ca boninite, Cape Vogel, Papua New Guinea: evidence for ultramafic primary magma, refractory mantle source and enriched components

    Chem. Geol.

    (2002)
  • L.N. Kogarko et al.

    Geochemical evidence for meimechite magma generation in the subcontinental lithosphere of Polar Siberia

    J. Asian Earth Sci.

    (2000)
  • P.J. le Roux et al.

    Mantle heterogeneity beneath the southern Mid-Atlantic Ridge: trace element evidence for contamination of ambient asthenospheric mantle

    Earth Planet. Sci. Lett.

    (2002)
  • M.T. McCulloch et al.

    Geochemical and geodynamical constraints on subduction zone magmatism

    Earth Planet. Sci. Lett.

    (1991)
  • D.R. Nelson et al.

    The origin of ultrapotassic rocks as inferred from Sr, Nd and Pb isotopes

    Geochim. Cosmochim. Acta

    (1986)
  • S.Y. O'Reilly et al.

    Ultradeep continental roots and their oceanic remnants: a solution to the geochemical “mantle reservoir” problem?

    Lithos

    (2009)
  • M.R. Perfit et al.

    Chemical characteristics of island-arc basalts: implications for mantle sources

    Chem. Geol.

    (1980)
  • R.L. Rudnick

    Composition of the continental crust

  • A. Schwindrofska et al.

    Origin of enriched components in the South Atlantic: evidence from 40 Ma geochemical zonation of the discovery seamounts

    Earth Planet. Sci. Lett.

    (2016)
  • A.V. Sobolev et al.

    Siberian meimechites: origin and relation to flood basalts and kimberlites

    Russ. Geol. Geophys.

    (2009)
  • N. Arndt et al.

    Two mantle sources, two plumbing systems: tholeiitic and alkaline magmatism of the Maymecha River basin, Siberian flood volcanic province

    Contrib. Mineral. Petrol.

    (1998)
  • I. Bindeman et al.

    The Mantle Sources of Continental Flood Basalts from Oxygen Isotope Composition of Primitive Olivine Phenocrysts

    (2009)
  • J.W. Bristow

    Picritic rocks of the north Lebombo and south-east Zimbabwe

  • J.W. Bristow et al.

    Volcanic rocks of the Lebombo-Nuanetsi-Sabi zone: classification and nomenclature

  • R.W. Carlson et al.

    A comparison of Siberian meimechites and kimberlites: implications for the source of high-Mg alkalic magmas and flood basalts

    Geochem. Geophys. Geosyst.

    (2006)
  • S.L. Chung et al.

    Plume-lithosphere interaction in generation of the Emeishan flood basalts at the Permian-Triassic boundary

    Geology

    (1995)
  • K.G. Cox

    Karoo igneous activity, and the early stages of the break-up of Gondwanaland

  • K.G. Cox et al.

    The petrology of the Karroo basalts of Basutoland

    Am. Mineral.

    (1966)
  • K.G. Cox et al.

    The olivine-rich lavas of Nuanetsi: a study of polybaric magmatic evolution

    J. Petrol.

    (1974)
  • K.G. Cox et al.

    The geology of the Nuanetsi Igneous Province

    Phil. Trans. R. Soc. London. Ser. A, Math. and Phys. Sci.

    (1965)
  • K.G. Cox et al.

    Geochemical and petrographic provinces in Karroo basalts of southern Africa

    Am. Mineral.

    (1967)
  • Cited by (43)

    • Periodicity of Karoo rift zone magmatism inferred from zircon ages of silicic rocks: Implications for the origin and environmental impact of the large igneous province

      2022, Gondwana Research
      Citation Excerpt :

      Consequently, the periodicity of magmatism inferred from the age data has important implications for the origin and environmental impact of the Karoo LIP. Overall, the chronological and geochemical data on the least-contaminated rock types point to tapping of depleted upper mantle during the early phase in the rift zone (e.g. Heinonen et al., 2010b; Luttinen et al., 2015) and different kinds of primitive and enriched lithospheric mantle or plume sources during the main phase magmatism in the Karoo basins and the late phase magmatism in the rift zone (Ellam et al., 1992; Sweeney et al., 1994; Ellam, 2006; Jourdan et al., 2007a; Kamenetsky et al., 2017; Howarth and Harris, 2017; Luttinen, 2018; Turunen et al., 2019; Ashwal et al., 2021). Here we focus on the implications of compositional variability in the different magmatic phases for the environmental impact of the Karoo LIP.

    View all citing articles on Scopus
    View full text