This research was conducted in order to address important questions regarding why volcanic eruptions take place.  Volcanic eruptions, especially violent and large eruptions that have the potential to globally disrupt climate and cause severe economic and human hardship, are driven by the exsolution (or phase separation) of a gas phase out of the magma, which by expansion drives the magma out of the volcanic edifice to cause an eruption.  The generation of a magmatic gas phase is associated with decompression and/or phase change;  phase change refers to the crystallization of the magma to form rock-forming minerals, and this process, like decompression, lowers the solubility of the gas in the magma, forcing it to bubble out of solution and form a distinct phase. Crystallization is driven by cooling. Therefore, as a magma cools it generally releases gas.  The premise of our research is that without external triggering such as seismic activity, or other tectonic affects that would open a depressurizing conduit between a magma chamber and a volcanic edifice, a magma can build up a sufficient quantity of a gas phase as it cools in order to achieve a critical over-pressure, creating a dynamically unstable condition that leads to spontaneous eruption.  Our research is aimed at quantifying this process.  We developed  thermodynamic models to describe the solubility of volcanic gases in magmatic liquids and we coupled these models to others for solid and liquid phases in order to create a computational thermodynamic framework that permits us to calculate exactly how much solid and gas exolves from a magma as it cools.  We focused our research on silicic magmas of the type responsible for large scale supereruptions that have occurred in southeastern California, Yellowstone and New Zealand.   In the process, we developed ancillary modeling techniques that allow us to estimate the pressure and other physical conditions present in a magma chamber prior to eruption by examining the products of eruption in the rock record.  A number of important outcomes emerge from this research: (1) internal triggering of volcanic eruption associated with volatile gas separation accompanying phase change is possible, (2) sufficient over-pressurization is more likely during intervals in the cooling history of a magma that correspond to rapid crystallization over a narrow temperature interval (a so-called invariant point on the solid-liquid phase diagram), (3) the degree of over-pressurization associated with gas exsolution depends on where the magma body resides in the crust, a consequence of the affect of ambient pressure on the compressibility of the gas: deeper magma bodies are less likely to over-pressurize, shallower bodies are much more prone to immediately release gas as it accumulates, (4) there is a window of eruptibility for silicic magma that corresponds to magma chamber depths between about 1 and 15 kms, and (5) application of the modeling techniques developed under this grant to real systems (Long Valley caldera in southeastern California, the volcanoes of the North Island of New Zealand) provide estimates of the eruptive potential and longevity of large shallow magma reservoirs that are precursors of supereruptions.  All of the modeling tools and results developed under this grant are freely available at the web portal melts.ofm-research.org.

Last Modified: 09/27/2016
Modified by: Mark S Ghiorso