Light oxygen isotopes in mantle-derived magmas reflect assimilation of sub-continental lithospheric mantle material

Abstract

Oxygen isotope ratios in mantle-derived magmas that differ from typical mantle values are generally attributed to crustal contamination, deeply subducted crustal material in the mantle source or primordial heterogeneities. Here we provide an alternative view for the origin of light oxygen-isotope signatures in mantle-derived magmas using kimberlites, carbonate-rich magmas that assimilate mantle debris during ascent. Olivine grains in kimberlites are commonly zoned between a mantle-derived core and a magmatic rim, thus constraining the compositions of both mantle wall-rocks and melt phase. Secondary ion mass spectrometry (SIMS) analyses of olivine in worldwide kimberlites show a remarkable correlation between mean oxygen-isotope compositions of cores and rims from mantle-like 18O/16O to lower ‘crustal’ values. This observation indicates that kimberlites entraining low-18O/16O olivine xenocrysts are modified by assimilation of low-18O/16O sub-continental lithospheric mantle material. Interaction with geochemically-enriched domains of the sub-continental lithospheric mantle can therefore be an important source of apparently ‘crustal’ signatures in mantle-derived magmas.

Highly oxidized intraplate basalts and deep carbon storage

“…Consistent with this inference, the metasomatized mantle xenolith PIC (Phlogopite-Ilmenite-Clinopyroxene) has a light O isotopic composition, located at the low-δ 18 O endmember in the trends shown in Fig. 5 (46). A recent olivine-δ 18 O investigation of global kimberlites from different cratons also shows that the metasomatized subcontinental lithospheric mantle is an important reservoir of isotopically light oxygen (46).…”

“…5). The hypothetical HOME is inferred to have mantle-like to higher olivineδ 18 O values, consistent with its carbonated origin (46). The low-δ 18 O endmember may be related to the subcontinental lithospheric mantle, which has experienced extensive metasomatism by fluid/melts derived from hydrothermally altered subducted oceanic crust (47).…”

Geodynamic and Isotopic Constraints on the Genesis of Kimberlites, Lamproites and Related Magmas From the Finnish Segment of the Karelian Craton

“…If the latter represent the “archetypal” case of kimberlite derivation from partial melting of a deep and potentially ancient mantle source (Giuliani et al., 2021; Nakanishi et al., 2021; Woodhead et al., 2019), then the Kaavi‐Kuopio kimberlites may reflect genesis from a more heterogeneous source under potentially more complex geodynamic circumstances. An additional complexity to consider is the role of a heterogeneous mantle lithosphere, particularly variations in lithospheric thickness, because kimberlites are well known to extensively interact with and assimilate lithospheric mantle material (e.g., Giuliani et al., 2020b; Russell et al., 2012; Soltys et al., 2016; Tovey et al., 2021; Xu et al., 2021). The latter has been shown to have a pivotal role in the major element make‐up of primitive melts parental to the Kaavi‐Kuopio kimberlites (Dalton, Giuliani, O'Brien, et al., 2020).…”

“…Though it is recognised that mantle assimilation alters major element compositions of kimberlitic melts ( e.g., Dalton, Giuliani, O'Brien, et al, 2020;Giuliani et al, 2020a;Xu et al, 2021 ), it remains to be seen if this same process can account for variations in incompatible trace elements and hence the Hf isotope variations in the Kaavi-Kuopio kimberlites. In support of this hypothesis are the statistically significant correlations observed between bulk-rock or perovskite εHf ( i ) and the Mg# of olivine xenocryst cores ( proxy for assimilated mantle material; R 2 = 0.94, p = 0.03 ) as well as various proxies for primitive melt composition, such as olivine rim Mg# ( R 2 = 0.92, p = 0.04 ) and chromite Mg# ( R 2 = 0.67, p = 0.03 ) and chromite TiO 2 contents ( R 2 = 0.71, p = 0.02; Figure 5 ).…”

Lithospheric thinning beneath the Tengchong volcanic field, Southern China: Insight from Cenozoic calc-alkaline basalts

“…The second model is decompression melting caused by lithospheric mantle removal and asthenospheric upwelling (Hoernle et al, 2006;Xu et al, 2008;Timm et al, 2009;Zhu et al, 2012). The third model is buoyant upwellings driven by deep subducted slabs (Davies and Bunge, 2006;Chen et al, 2017;Kuritani et al, 2019;Xu et al, 2021). The fourth model is edgedriven convection caused by lithospheric edge associated with plate movement (King and Ritsema, 2000;Dai et al, 2021).…”