Secular compositional changes in hydrated mantle: the record of arc-type basalts

Condie, Kent C.; Puetz, Stephen J.; Spencer, Christopher J.; Roberts, Nick M.W.. 2022 Secular compositional changes in hydrated mantle: the record of arc-type basalts. Chemical Geology, 607, 121010. 10.1016/j.chemgeo.2022.121010

Full text not available from this repository. (Request a copy)

Official URL: http://dx.doi.org/10.1016/j.chemgeo.2022.121010

Abstract/Summary

We use a filtered database that tracks the composition of hydrated mantle with time using arc-type basalts, including an unknown amount of plume-type mantle in the Archean. A factor of 2–10 geochemical variability, both between and within sample sites, characterizes arc-type basalts and especially so >3.5 Ga and < 2.0 Ga. Based on which geographic sites are available for sampling, this variability is probably due to an anomalously high proportion of enriched components in some arc-type mantle sources. The proportion of an enriched component in magma sources is related to the ratio of continental to oceanic arc sites in the database, and thus is at least in part controlled by tectonic setting sample biases. We suggest that online geochemical databases of mafic igneous rocks should not be used to track secular changes in mantle composition or temperature unless they have been carefully filtered for tectonic setting, alteration/metamorphism, reliable ages, rock type, and major element composition. If the relatively few geographic sites are representative, large variations in incompatible element distributions in arc-type basalts suggest poorly mixed mantle before 3.5 Ga. However, the uniformity of incompatible element distributions from 3.5 to 2.6 Ga supports the existence of well-mixed mantle sources during this time. Long-term changes in major and compatible element contents of arc-type basalt mantle sources from 3.5 to 1.0 Ga are consistent with decreasing mantle temperature and degree of melting during this time. Although enriched mantle sources for arc-type basalts appear in the Neoarchean, they do not become widespread until after 2.0 Ga. From 2.0 to 1.0 Ga, enriched mantle contributions are relatively high but variable, probably in response to some combination of increased enriched mantle in the magma sources, sampling biases from relatively few sites with an anomalously high proportion of enriched mantle, and recycling of oceanic and continental lithosphere into the mantle in response to global propagation of subduction. Superimposed on long-term geochemical trends are five geographically widespread geochemical anomalies (deviations of greater than ±2σ from a smoothed Gaussian curve) in incompatible element distributions (2.55, 2.15, 1.75, 0.65 and 0.25 Ga) and an additional anomaly at 3.6–3.3 Ga in some compatible elements. All five incompatible element anomalies occur near age peaks in the U/Pb zircon age record (at 2.5, 2.15, 1.85, 0.60, and 0.25 Ga in both detrital and igneous zircons) and the 2.55, 2.15 and 1.75 Ga anomalies are very close to peaks in Re/Os model ages for mantle ultramafic rocks. The simplest explanation for the incompatible element anomalies is episodic enrichment of mantle sources of arc-type basalts, perhaps related to widespread mantle events (overturn or/and mantle plume) that brought enriched components into subduction-related sources of arc-type basalts.

Item Type:	Publication - Article
Digital Object Identifier (DOI):	10.1016/j.chemgeo.2022.121010
ISSN:	00092541
Date made live:	09 Aug 2022 09:52 +0 (UTC)
URI:	https://nora.nerc.ac.uk/id/eprint/533033

Item Type:

Publication - Article

Digital Object Identifier (DOI):

10.1016/j.chemgeo.2022.121010

ISSN:

00092541

Date made live:

09 Aug 2022 09:52 +0 (UTC)

URI:

https://nora.nerc.ac.uk/id/eprint/533033