“Fingerprinting” tectono-magmatic provenance using trace elements in igneous zircon

Abstract

Over 5300 recent SHRIMP-RG analyses of trace elements (TE) in igneous zircon have been compiled and classified based on their original tectono-magmatic setting to empirically evaluate “geochemical fingerprints” unique to those settings. Immobile element geochemical fingerprints used for lavas are applied with the same rational to zircon, including consideration of mineral competition on zircon TE ratios, and new criteria for distinguishing mid-ocean ridge (MOR), magmatic arc, and ocean island (and other plume-influenced) settings are proposed. The elemental ratios in zircon effective for fingerprinting tectono-magmatic provenance are systematically related to lava composition from equivalent settings. Existing discrimination diagrams using zircon U/Yb versus Hf or Y do not distinguish TE-enriched ocean island settings (i.e., Iceland, Hawaii) from magmatic arc settings. However, bivariate diagrams with combined cation ratios involving U–Nb–Sc–Yb–Gd–Ce provide a more complete distinction of zircon from these settings. On diagrams of U/Yb versus Nb/Yb, most MOR, ocean island, and kimberlite zircon define a broad “mantle-zircon array”; arc zircon defines a parallel array offset to higher U/Yb. Distinctly low U/Yb ratios of MOR zircon (typically 0.1, high Sc/Yb separates arc settings from low-Sc/Yb plume-influenced sources. The slope of scandium enrichment trends in zircon differ between MOR and continental arc settings, likely reflecting the involvement of amphibole during melt differentiation. Scandium is thus also critical for discriminating provenance, but its behavior in zircon probably reflects contrasting melt fractionation trends between tholeiitic and calc-alkaline systems more than compositional differences in primitive magmas sourced at each tectono-magmatic source.