Decadal time gaps between mafic intrusion and silicic eruption obtained from chemical zoning patterns in olivine

TitleDecadal time gaps between mafic intrusion and silicic eruption obtained from chemical zoning patterns in olivine
Publication TypeJournal Article
Year of Publication2004
AuthorsCosta F, Chakraborty S
JournalEarth and Planetary Science Letters
Volume227
Pagination517-530
Date PublishedNov
ISBN Number0012-821X
Accession NumberWOS:000225124900026
Abstract

Intrusion of mafic magma in a silicic crustal reservoir has been reported in numerous studies of subduction-related volcanoes and is considered a main process for triggering explosive eruptions and for transferring geochemical signatures from the mantle to the crust. Here we determine the timescales between mafic-silicic encounters and eruption by exploiting the fact that during such interactions, magmas exchange crystals, and their mineral compositions will strive to equilibrate to the new environmental conditions. Specifically, we model the Fe-Mg zoning patterns of olivine found in the dacites, andesite, and quenched mafic inclusions of the Holocene zoned eruption of Volcan San Pedro (36degreesS, Chilean Andes). The San Pedro eruption is an example where mixing between silicic and matic magmas produced intermediate compositions (e.g., andesite) at depth, in the magma reservoir. For modeling the compositional profiles, we took into account the anisotropy of Fe-Mg diffusion in olivine and also the effects of diffusion in multiple dimensions, aspects that have been largely neglected in previous works. The retrieved timescales for olivine-melt equilibration by diffusion display an array ranging from similar to1 year for the most silica-rich dacite to similar to50 years for the andesite. This array of timescales, rather than a single value, provides new insight into the details of mafic-silicic mixing at depth. They reveal processes that occur on a longer timescale than the days to weeks time gap for cases where the mafic intrusion triggered the eruption and they might be associated with the instances where seismic activity or significant inflation of the volcanic edifice was not immediately followed by eruption. On the other hand, the year to decade timescales that we report for producing magmas of intermediate composition (e.g. andesites) are much shorter than the 10(3)-10(5) years inferred from U-series disequilibria data for igneous differentiation by closed-system fractionation to produce similar magma compositions at subduction zones. (C) 2004 Elsevier B.V. All rights reserved.