Geochemistry of lavas from Aegina and Poros (Aegean Arc, Greece): Distinguishing upper crustal contamination and source contamination in the Saronic Gulf area

Abstract

Pliocene (Phase 1) deposits from the volcanic centres of Aegina and Poros (western Aegean Arc) consist of andesitic-dacitic lava flows and domes with mafic enclaves, whereas basaltic-andesitic and rhyolitic lava flows were emplaced during the early Pleistocene Phase 2 and its precursor activity. Fifty-nine new whole rock analyses of lavas and enclaves show geochemical characteristics in keeping with their continental arc setting, with a significant spread in Sr-Nd-Hf-Pb isotope compositions and a strong crustal imprint (e.g. 87Sr/86Sr 0.7041–0.7076). Unlike other Aegean volcanic centres, the Phase 1 deposits show high Ba concentrations and high Ba/Th ratios. These are unlikely to be a signature of fluid-mediated metasomatism of the mantle wedge; subduction of barite-enriched sediments appears a more plausible option. The amount of sediment added to the mantle wedge is ca. 4%. High 208Pb/204Pb relative to 206Pb/204Pb, especially for the Poros lavas, points towards contamination with continental crust similar to the basement of Ios. First-order modelling shows that up to 35% crustal contamination took place. Additionally, crystal fractionation, magma mixing and crystal accumulation shaped the whole rock geochemistry. Similar to the neighbouring island of Methana, the widest isotopic variation is seen in the early Pleistocene deposits, reflecting remobilisation of magma from the previous volcanic episode by isotopically more primitive mafic magmas after a period of volcanic quiescence. The isotopic contrast between the volcanic centres of the Saronic Gulf and those of Santorini and Kos-Nisyros further west is likely related to the geometry and physical characteristics of the subducting slab and mantle wedge, rather than the actual amount of sediments subducted or their composition.