CONSTRAINING THE GEOCHEMICAL RESERVOIRS CONTRIBUTING TO VOLCANISM IN THE KEWEENAW LIP

The ca. 1.1 Ga Keweenaw Large Igneous province (LIP) in North America represents one of the most complete pre-Phanerozoic flood basalt sequences on the planet. However, the thickness of the volcanic pile (ca. 5 km), and long duration of volcanism (ca. 30 Ma) is anomalous compared to other LIPs. Many of the peculiarities of the Keweenaw LIP can be traced to the contemporaneously active Mid-Continent Rift. The interaction of an upwelling plume with this pre-existing extensional structure gives rise to the unusual geometry and thickness of the volcanic pile. However a reason for the long duration of volcanism remains unclear. Previous studies proposed the influence of a mantle plume throughout the 30 Ma duration of volcanic activity, but rapid plate motion during this period complicates such a model. Establishing the geochemical reservoirs contributing to the Keweenaw LIP provides a potential constraint as to the origin of this long-lived magmatism. The goal of this contribution is to establish the relative role of the depleted mantle, continental lithospheric mantle, Archean crust, and mantle plume in the Keweenaw lavas. We present a new Hf isotope and trace element dataset for a well-characterized suite of samples from the Mamainse Point section of the Keweenaw LIP. We supplement these data with a Sr-Nd-Pb-Hf isotopic dataset from the contemporaneous Coldwell Alkaline Complex and lamprophyres in order to constrain the composition of the regionally metasomatized lithospheric mantle. We find that the lithospheric-mantle derived samples deviate significantly from the mantle array in εNd-εHf space. However, the sample suite from Mamainse Point plots along the mantle array, consistent with mass contribution from a mantle plume, the depleted mantle, and Archean crust. We observe differences in the isotopic characteristics of primitive samples erupted before and after a hiatus in basaltic volcanism (ca. 1100.4 Ma). We explore the potential origin of these observations in the context of the plume model and discuss their implications.