Trends in crystal chemistry recorded across the Alaska-Aleutian volcanic arc
Matthew Loewen1 , Sloane Kennedy2, Jordan Lubbers1, Frank Tepley2, Adam Kent2, Pavel Izbekov3
Affiliations: 1U.S. Geological Survey, Alaska Volcano Observatory, Anchorage, USA; 2Oregon State University, Corvallis, USA; 3University of Alaska Fairbanks, Alaska Volcano Observatory, Fairbanks, USA
Presentation type: Poster
Presentation time: Thursday 16:30 - 18:30, Room Poster Hall
Poster Board Number: 257
Programme No: 1.1.42
Abstract
The chemical composition of mineral phases reflects both the crystallization conditions and composition of melt they crystallize from. Crystal growth can thus record a diversity of melt compositions and conditions reflecting complex magma plumbing and crustal storage histories, while glass reflects the final magma amalgamation preceding eruption. Trace elements, especially incompatible trace element ratios, have particularly strong potential to capture melt compositional history at a volcano even from the crystal record of a single sample. Here, we take a broad look at mineral compositions across the entire Alaska-Aleutian volcanic arc, looking thus far at amphibole from 17 samples at 12 volcanoes. We use amphibole trace-element compositions to estimate the compositions of host melts from which they crystallized. The results extend to less and sometimes more evolved compositions than captured in their host sample bulk or glass composition. These trends are found in samples with a range of bulk compositions including basalts, andesites, and rhyolites. Sr diffusion profiles can provide further context to these preserved compositions within the zoning profiles of individual crystals. Other phases have potential to record complementary information. Plagioclase and pyroxene, while more limited in utility with trace element partitioning, are ubiquitous phases found in nearly all arc volcanic samples. Fe-Ti oxide pairs provide a relatively robust tool for calculating eruption temperature and fO2. Our purpose here is to show how integrated arc-wide studies of disparate mineral composition can provide a powerful complement to similar compilation efforts for whole rock and glass compositional data.