USING THERMODYNAMIC MODELING TO PREDICT FRACTIONAL CRYSTALLIZATION PATHS OF ANDESITE TO TRACHYDACITE

The Isom Formation is a package of distinctive trachydacite ignimbrites in the Oligocene Indian Peak caldera complex along the Utah-Nevada border. Similar high-alkali rocks are also found in other Cenozoic caldera complexes and imply their generation is a fundamental part of the evolution of large volcanic systems. Best and Christiansen (1989) hypothesized that the magmas are the result of fractional crystallization of andesite; in contrast, other large volume ignimbrites appear to involve significant mixing and crustal contamination. This hypothesis has been tested by using compositions of andesites in the vicinity of the Isom tuffs and modeling their fractional crystallization paths using MELTS (Ghiorso 1995). Different environmental parameters were examined to see if a path exists that leads from andesite to the distinctive Isom-type compositions. Initial compositions of parent andesites were taken from two different andesite flows--samples CHUCK-1A and BAN-10. For CHUCK-1A to be a parent of an Isom-type magma, pressure would have to be about 8 kb with 2% water and oxygen fugacity of QFM+1. These conditions produce a trachydacite with a mineral assemblage (cpx+plg+mt+ap) that is similar to the Isom at 950°C, its eruption temperature. In addition, it reproduces compositional trends for Si, Ti, Ca, Al, and Mg found in the Isom tuffs, but poorly matches the trends for Fe and alkalis. Some of the deviations of the modeled trends from the real magmas may be the result of errors in the MELTS thermodynamic database. For example, MELTS calculations produce multiple coexisting pyroxenes with unusual compositions. Tests using BAN-10 as the parent andesite produced Isom-like fractional crystallization trends and mineral assemblages at 4 kb, 2% water, and oxygen fugacity of QFM+1. Compositional variations match best for Si, Ti, Ca, Al, and Mg but are still low for alkalis and Fe. Calculations at lower pressures, using either starting composition, do not fit the trends as well, with mismatches for Ca and Al. Thus, taken at face value the implication that high pressure crystallization is necessary to make Isom-type magmas is consistent with the apparent absence of calderas caused by a shallow (~2 kb) magma chamber.