Ca. 2.7 Ga ferropicritic magmatism: A record of Fe-rich heterogeneities during Neoarchean global mantle melting

Abstract

Although terrestrial picritic magmas with FeOTOT ⩾13wt.% are rare in the geological record, they were relatively common ca. 2.7Ga during the Neoarchean episode of enhanced global growth of continental crust. Recent evidence that ferropicritic underplating played an important role in the ca. 2.74–2.70Ga reworking of the Ungava craton provides the impetus for a comparison of ca. 2.7Ga ferropicrite occurrences in the global Neoarchean magmatic record. In addition to the Fe-rich plutons of the Ungava craton, volumetrically minor ferropicritic flows, pyroclastic deposits, and intrusive rocks form parts of the Neoarchean greenstone belt stratigraphy of the Abitibi, Wawa, Wabigoon and Vermillion domains of the southern and western Superior Province. Neoarchean ferropicritic rocks also occur on five other Archean cratons: West Churchill, Slave, Yilgarn, Kaapvaal, and Karelia; suggesting that ca. 2.7Ga Fe-rich magmatism was globally widespread.Neoarchean ferropicrites form two distinct groups in terms of their trace element geochemistry. Alkaline ferropicrites have fractionated REE profiles and show no systematic HFSE anomalies, broadly resembling the trace element character of modern-day ocean island basalt (OIB) magmas. Magmas parental to ca. 2.7Ga alkaline ferropicrites also had high Nb/YPM (>2), low Al2O3/TiO2 (<8) and Sc/Fe (⩽3×10−4) ratios, and were enriched in Ni relative to primary pyrolite mantle-derived melts. The high Ni contents of the alkaline ferropicrites coupled with the low Sc/Fe ratios are consistent with derivation from olivine-free garnet-pyroxenite sources. The second ferropicrite group is characterized by decisively non-alkaline primary trace element profiles that range from flat to LREE-depleted, resembling Archean tholeiitic basalts and komatiites. In contrast to the alkaline ferropicrites, the magmas parental to the subalkaline ferropicrites had flat HREE, lower Nb/YPM (<2), higher Al2O3/TiO2 (8–25) and Sc/Fe (⩾4×10−4) ratios, and were depleted in Ni relative to melts of pyrolitic peridotite; suggesting they were derived from garnet-free peridotite sources. Neodymium isotopic evidence indicates that the source of alkaline ferropicrites was metasomatically enriched shortly before magma generation (⩽3.0Ga), but the subalkaline ferropicrites do not show evidence of precursor metasomatism. The metasomatic enrichment of the alkaline ferropicrite sources may have been accompanied by conversion of Fe-rich peridotite to secondary garnet-pyroxenite.Melting experiments on “pyrolitic” compositions and consideration of the dependence of the density of silicate liquids on pressure and temperature, suggest that ferropicrites cannot originate by melting of normal terrestrial mantle (Mg-number=0.88–0.92) at high pressures and temperatures. The geochemical similarity between the subalkaline ferropicrites and the shergottite–nakhlite–chassigny (SNC) and howardite–eucrite–diogenite (HED) differentiated meteorites suggests, however, that the Fe-rich mantle may originate from the infall of Fe-rich chondritic meteorites. The occurrence of ca. 2.7Ga Fe-rich rocks on at least six cratons that are commonly coeval with the more ubiquitous komatiites and Mg-tholeiites is consistent with the existence of heterogeneous Fe-rich “plums” throughout the Neoarchean mantle. The paucity of ferropicrites in the post-2.7Ga geological record suggests that majority of these Fe-rich plums have been melted out during the global Neoarchean melting of the mantle.