2003 Seattle Annual Meeting (November 2–5, 2003)

Paper No. 11
Presentation Time: 10:40 AM

NATURE AND TIMING OF MAGMA INTERACTION PROCESSES IN ARC VOLCANIC SYSTEMS: DATA FROM ROCKS AND FROM EXPERIMENTS ON ZONING PRODUCTION IN FE-TI OXIDES


RUTHERFORD, Malcolm J., Brown Univ, PO Box 1846, Providence, RI 02912-1846 and DEVINE III, Joseph D., Geology Department, Brown Univ, Providence, macr@brown.edu

The ongoing eruption on Montserrat is producing a hybrid andesite with clear and abundant evidence for a mingled basalt component in pre-existing andesite. In addition to enclaves of basaltic composition, the magma has textural and compositional evidence for heating by the basalt, including Ti-rich rims on Ti-magnetite phenocrysts and Cpx reaction rims on hornblende and quartz. Pargasitic hornblende microphenocrysts in the andesite are identical to those in the basalt enclaves, and this pargasite can not be crystallized from the andesite bulk composition under any set of conditions in the hornblende stability field. The large, (1-1.5 cm) hornblende phenocrysts in the andesite are composed of up to seven growth cycles that are visible in sections parallel to c. Each growth zone begins with a sharp increase in Si and Mg (and a drop in Al, Fe and alkalis). The Al-rich composition is then gradually regained with increasing T. Experiments indicate the compositional cycles are produced by a temperature change from ~820 to 855 at the 130 MPa water pressure required by the phenocryst-melt assemblage. The breakdown of the groundmass pargasite is critical to growth of the thick cycles of new hornblende because the amount of hornblende that can be accommodated in the andesitic melt in going from 830 to 860 C is very limited. The Ti-enrichment profile in magnetite rims ranges from 0 to 30 microns in samples of the erupted hybrid andesite. Experimenting with a sample containing rim-free magnetite crystals, the rim zoning profile is produced in 22 days at 860 C, and as little as 2 days at 870. Hornblende suffers significant breakdown to form a Cpx rim in 2 days, however; and the general lack of this reaction in the erupting magma limits the time that any magma could have spent at 870 C or above. Based of the above data, each batch of erupting andesitic magma has experienced hybridization and an associated heating prior to eruption. The heating occurred over a period of less than 20 days prior to eruption assuming the hybrid magma was heated close to the upper temperature limit for hornblende stability in this magma.