Skip to content

One small step in the crust, one giant leap for magma: Insights into magma differentiation from basalt to rhyolite at Cordón Caulle derived from rhyolite-MELTS simulations

Liam J. Kelly 1, Anna Ruefer2, Edgar L. Carrillo1,3, Sarah Hickernell2, Sarah Ward1, Guilherme A.R. Gualda1, Heather Winslow^4^, Philipp Ruprecht4

  • Affiliations: 1Department of Earth and Environmental Sciences, Vanderbilt University, Nashville, TN, USA; 2Department of Earth and Planetary Sciences, Stanford University, Stanford, CA, USA; 3Department of Earth Science, University of Oregon, Eugene, OR, USA; 4Department of Geological Sciences and Engineering, University of Nevada, Reno, Reno, NV, USA; \now at USGS Hawaii Volcano Observatory

  • Presentation type: Poster

  • Presentation time: Friday 16:30 - 18:00, Room Poster Hall

  • Poster Board Number: 9

  • Programme No: 1.5.16

  • Theme 1 > Session 5

Abstract

Magma mush systems are commonly invoked as the source from which crystal-melt segregation produces rhyolites, but these systems are rarely observed. The 2011-2012 VEI 4 eruption of Cordón Caulle produced rhyolite lavas which scavenged basaltic enclaves containing interstitial glass similar to their host rhyolitic lava. These enclaves offer a window into an active, shallow basaltic mush system. This mush was previously proposed as the source from which crystallization generates high-silica rhyolite in a single step. Here, we use rhyolite-MELTS to determine whether this is thermodynamically realistic. First, we use melt geobarometry to establish that enclave-derived pressures (~25-200 MPa) are consistent with those previously determined for the lavas. We then simulate isobaric crystallization using a range of initial starting water concentrations. We find that, using an initial melt matching the whole-rock composition of the enclave, it is possible to produce via fractional crystallization a rhyolite that matches the composition of the natural enclave glass. We also explore the physical consequences of crystallization and fluid exsolution during magma evolution. Lower initial water simulations (0.5-1.0 wt.% H2O), at pressures of 100-200 MPa produce changes in volume most consistent with pre-eruptive ground deformation signals. We determine the timescales of heat loss from the crystallizing basaltic magma to be ~8-25 ka, but it is plausible that heat loss occurred in <5 ka, which is broadly consistent with repose times of the system. The application of rhyolite-MELTS to an actively monitored system offers multifaceted insights into the geochemical, thermal, and physical evolution of magmas.