Pre-eruption history of enriched MORB from the Menez Gwen (37°50′N) and Lucky Strike (37°17′N) hydrothermal systems, Mid-Atlantic Ridge

The Menez Gwen (37 degrees 50'N) and Lucky Strike (37 degrees 17'N) seafloor hydrothermal systems on the Mid-Atlantic Ridge are influenced by the Azores plume, which results in a geochemically enriched character of the host basalts. Two groups of highly plagioclase phyric lavas can be identified at Menez Gwen. Group I lavas are highly vesicular and highly enriched in incompatible elements with high (La/SM)(N), Nb/Zr, Ba/Nb, Zr/Y and Nb/Y ratios. The enriched group I lavas resemble those at the Lucky Strike site. Group II porphyritic lavas are less vesicular, less enriched in incompatible elements with low (La/SM)N, Nb/Zr. Ba/Nb, Zr/Y and Nb/Y ratios. These differences between group I and II lavas can be explained either by melting-induced mixing of a two-component mantle or mixing between two melts. Furthermore, the presence of identical assemblages of plagioclase phenocrysts in both groups is strong evidence for mantle heterogeneities. Silicate melt inclusions, hosted in plagioclase phenocrysts, exhibit small metallic precipitates of Fe + Ni + Cu +/- Zn (+/- Co +/- As) +/- S +/- Cl within the vapour bubbles. The heterogeneous trapping of distinct coexisting phases during plagioclase growth indicates that a high temperature ore-metal rich immiscible fluid must have been exsolved from the magma prior to eruption. On the other hand, lava vesicularities are remarkably different between the group I and group 11 lavas. They indicate ubiquitous degassing of magmatic volatiles, particularly in group I lavas that underwent significant pre-eruptive degassing. The presence of an ore-metal-rich fluid being exsolved from the magma chamber in conjunction with pre-eruptive degassing is a strong indicator of direct magmatic contribution of metals and volatiles to the seafloor hydrothermal systems. At the Menez Gwen site, timing of degassing and the enriched chemistry of lavas proved to be crucial factors contributing to the exsolution of a metal-rich fluid prior to magma eruption. (C) 2009 Elsevier B.V. All rights reserved.