GSA Annual Meeting in Phoenix, Arizona, USA - 2019

Paper No. 241-8
Presentation Time: 10:40 AM

A NEW MODEL FOR THE FORMATION OF THE MEGACRYSTIC ANORTHITE “XENOCRYSTS” FOUND IN SUBDUCTION ZONE MAGMAS


CATALANO, Tristan, Dept of Geosciences, Stony Brook University, Stony Brook, NY 11794-2100 and NEKVASIL, Hanna, Department of Geosciences, Stony Brook University, Stony Brook, NY 11794

Highly calcic plagioclase crystals with cores up to An99 have been ejected from many subduction zone volcanoes. These centimeter-size megacrysts are generally homogenous until the rim (which is more albitic). The crystals contain inclusions of magnesian olivine, aluminous spinel, below 1 GPa and occasionally glass. At subduction zones, however, magmas formed by flux melting of the mantle wedge are too sodic to produce anorthite. The discovery of azeotropic behavior of anorthitic plagioclase in which isobaric melting begins at lower temperatures than more sodic compositions through an aluminous spinel+L producing reaction in the pressure regime 1-0.5 GPa, may provide a new model for the formation of anorthite. Our experimental phase equilibrium data in the system Ca-rich plagioclase-olivine indicates that the bulk composition of the products of plagioclase+olivine melting, that is, L+spinel, is more anorthitic that the coexisting plagioclase. Separation of L+spinel and their ascent together would result in decompression crystallization through a peritectic reaction and the formation of olivine+anorthite. We propose that the highly anorthitic crystals are formed independently from subduction zones magmas and that instead, they form from the depleted mantle in a process where upwelling of depleted mantle at the wedge margin causes depression melting as the source ascends. Ascent would induce L+spinel to recombine through a peritectic reaction to crystallize anorthite+olivine that is eventually carried to the conduits of subduction zone magmas by horizontal component of mantle flow. Importantly, in this model, hydration of the mantle is not the cause of the highly anorthitic crystals; rather, it is the depleted nature of the residual mantle that allows this. We are conducting further experiments to constrain or reject this process.