Cenozoic magmatism in the western Ross Embayment: Role of mantle plume versus plate dynamics in the development of the West Antarctic Rift System

Geochemical and chronological data for Cenozoic plutons and dikes from northern Victoria Land (Antarctica), were used to propose a tectonic‐magmatic model for this portion of the West Antarctic Rift System (WARS). The seven major plutons are compositionally bimodal, with gabbroic and syenitic portions. Among the 180 studied dikes, most are 1 m thick and have alkali basalt‐basanite‐tephrite compositions, along with minor intermediate rocks. Trachytic‐rhyolitic dikes (up to 50 m thick) are by far less common. The 40Ar‐39Ar data for dikes indicate middle Eocene to early Oligocene ages, the oldest found to date for igneous activity throughout the WARS. The geochronological‐structural framework provides evidence for coeval emplacement of dikes on two main NW‐SE and N‐S striking trends, whereas plutonic‐subvolcanic activity occurred in adjacent crustal sectors at different times. Mafic dikes display major and trace element distribution typical of basalts from oceanic islands, such as high ratios of Nb and Ta to large ion lithophile and heavy rare earth elements, coupled with prominent negative K and Pb anomalies in the primitive mantle‐normalized multielement diagrams. Initial isotopic compositions are within the ranges of 0.70299–0.70372 for 87Sr/86Sr and 4.2–6.3 for εNd. These features are shared by younger Neogene lavas. Geochemical modeling for both mafic dikes and lavas indicates an enriched magma source characterized by a residual potassic hydrous phase, thus pointing out a mantle source nearly uniform throughout 50 m.y. Comparable geochemical signature are reported for the magma source of the other Cenozoic volcanic provinces of the WARS and the whole Antarctic Plate. The new data for the early igneous rocks of the rift put in evidence a chronologic‐structural link between magmatic evolution, regional tectonics and plate dynamics which suggests a model for WARS magmatism that is alternative to current plume hypotheses. We propose that magma genesis and emplacement are related to reactivation of preexisting translithospheric faults, which promoted local decompression melting of an enriched mantle that was previously veined during the decompression episode associated with the amagmatic late Cretaceous extensional rift phase.