Cordilleran Section - 115th Annual Meeting - 2019

Paper No. 33-1
Presentation Time: 8:05 AM


WEIS, Dominique, Department of Earth, Ocean and Atmospheric Sciences, University of British Columbia, 2020-2207 Main Mall, Vancouver, BC V6T-1Z4, Canada

The Cascadia subduction zone is one of the hottest in the world because of the very young subducting Juan de Fuca plate and an ideal location to examine compositional variations and the respective contributions of mantle heterogeneity, crustal contamination, and slab melting along the ~1300 km length of the arc.

Selective sampling of mainly primitive basalts from key volcanoes in the High Cascades and along the Garibaldi volcanic belt for high-precision geochemical and isotopic studies shows these two main segments of the arc to belong to different isotopic trends (Mullen & Weis, 2013, 2015; Mullen et al., 2017). The High Cascades lavas straddle a mixing trend between a depleted end-member akin to Juan de Fuca MORB and an enriched end-member so far best represented by Cascadia basin sediments (Carpentier et al., 2013, 2014), with no contribution by Astoria fan sediments. The Garibaldi volcanic belt has the least sediment input and also the influx of a distinct enriched mantle component at the northern slab edge, as also reflected by more homogeneous Tl isotopic compositions, around the mantle value (O’Connor et al., 2017).

At a finer scale, some volcanic centers present compositional and geochemical variations resulting from the longevity and evolution of the major volcanoes and reflecting the complex distribution of magma along the Cascades arc, especially in the Columbia segment (Wall et al., 2018): geochemically, Glacier Peak lavas belong to the High Cascades array, but with a lower proportion of sediments input, Mt. Adams basalts form a Pb-Pb linear trend that is oblique to the main High Cascades Array, and Goat Rocks volcanic complex lavas have geochemical characteristics intermediary between those of the High Cascades and Mt. Adams.

So far, the sedimentary component contributing to Cascades arc magmas remains poorly defined and, only Cascadia Basin sediments appear to be involved – new high-precision geochemical data on the Explorer (Cousens et al., 2017) and Juan de Fuca plates allow to identify the depleted end-member in the High Cascades lavas better. Altogether, the compositional heterogeneity of the primitive arc magmas is not so much related to the presence of multiple mantle components but instead to variability in the composition and quantity of slab contributions.