GSA Annual Meeting in Indianapolis, Indiana, USA - 2018

Paper No. 116-2
Presentation Time: 9:00 AM-6:30 PM

JOURNEY TO THE SURFACE: INDEPENDENT RHYOLITIC CRYSTAL-POOR MAGMA EXTRACTION AND STORAGE PRECEDING CALDERA-FORMING SILICIC ERUPTIONS


BURCHIEL, Cole J., Department of Geology, Oberlin College, Oberlin, OH 44074, HARMON, Lydia J., Department of Earth and Environmental Sciences, Vanderbilt University, Nashville, TN 37235, GUALDA, Guilherme A.R., Earth and Environmental Sciences, Vanderbilt University, Nashville, TN 37235 and GRAVLEY, Darren M., Geological Sciences, University of Canterbury, Christchurch, 8041, New Zealand

Shallow bodies of rhyolitic magma are capable of generating large, explosive caldera-forming volcanic eruptions. Eruptible melt-rich, crystal-poor magma originates from deeper reservoirs where it is segregated from a sluggish crystal mush.

Extracted crystal-poor magma may be directly erupted or stored in a shallower magma body. If stored, storage may occur contiguously in a vertically batch atop the crystal mush, or may migrate further to a significantly shallower position within the crust before erupting. To precisely identify which process of extraction to eruption occurs within silicic pre-eruptive systems, we conducted quantitative EDS glass analysis and XRF whole rock analysis on 12 pumice samples from the Mamaku ignimbrite. This work ultimately illustrates the magmatic history preceding the Mamaku caldera-forming eruption in New Zealand’s Taupo Volcanic Zone.

Whole-rock and glass compositions were used as initial conditions in thermodynamic phase equilibrium models using rhyolite-MELTS to calculate extraction and storage pressures respectively. By comparing results from matching whole-rock and glass compositions from single pumice samples, we illustrate individual magmatic migrations from extraction to storage. Calculations were run between 1300 – 700 ºC at 1 ºC intervals and 400 – 25 MPa at 25 MPa intervals to identify the qtz+plag+opx cotectic conditions for each sample. We explored fO2 conditions (∆NNO = -3 – +1) and found equilibrium conditions at ∆NNO = -1 – 0.

Our sample suite as a whole displays shallow depths (124 – 211 MPa; ~4.8 – 8.2 km) of melt storage contrasting with significantly deeper depths (200 – 360 MPa; ~7.8 – 14.2 km) of extraction. Of significance, two distinct magma types were observed; one sample, GL2306-N, contains a less evolved mineralogy that matches a secondary, previously identified Mamaku magma type. Calculated extraction and storage depths are notably less than those of the dominant magma type. These depths show negligible differences in extraction and storage depths for GL2306-N, suggesting contiguous extraction and storage. Results from remaining samples support a hypothesis that rhyolitic magma migration from a crystal mush to pre-eruption storage may occur across a considerable vertical extent for some silicic volcanic systems.