2015 GSA Annual Meeting in Baltimore, Maryland, USA (1-4 November 2015)

Paper No. 145-4
Presentation Time: 9:00 AM-6:30 PM


DREYER, Brian M., University of California, Santa Cruz, Institute of Marine Sciences, Santa Cruz, CA 95064, GILL, James B., Department of Earth and Planetary Sciences, University of California, Santa Cruz, Santa Cruz, CA 95064 and CLAGUE, David A., Monterey Bay Aquarium Research Institute, 7700 Sandholdt Road, Moss Landing, CA 95039, bdreyer@ucsc.edu

Recent work has shown that the aggregate chemistry of mid-ocean ridge (MOR) basalts cannot be reproduced using closed system fractional crystallization and therefore open system magmatic replenishment is required (O’Neill and Jenner, 2012; Coogan and O’Hara, 2015; Shorttle, 2015). However, these inferences are based on global data sets and so do not consider geological/volcanological context, geochemical variability, and temporal parameters on the scale of individual lava flows. Improved understanding of MOR magmatic processes and compositional scales requires greater sampling density and geological control (Rubin et al., 2001; 2009).

The Endeavour segment of the Juan de Fuca MOR is underlain by an AMC, and aggregate ranges in minor and trace element concentrations and ratios relative to MgO are also too great to be explained by closed system fractional crystallization. We evaluate the scales of magmatic openness through an examination of Endeavour lavas that seem to be comagmatic or coeruptive based on robust geological (Clague et al., 2014), geochemical (Gill et al., 2015), and geochronological (Jamieson et al., 2013; Clague et al., 2014) evidence. This approach is similar to that used for historical MOR eruptions of Rubin et al. (2001). We identified 14 “chemomagmatic units” that are spatially proximate and chemically relatable and separable that collectively represent eruptions since ~11ka. Some units may be single lava flows. Other units appear to have erupted intermittently over hundreds to thousands of years during which chemically dissimilar lava may also erupt. Consistent differences in trace element ratios between units argue against intermixing.

Melt evolution was modeled using MELTS for units that have reasonably broad major element variation; starting melt compositions were those of the most mafic sample for each unit. We find that simple models of fractional crystallization can adequately reproduce most of the major and incompatible trace element behavior. Thus, within the context of refined temporal and spatial scales, magmatic batches typically lie within analytical resolution of closed systems, despite growing evidence that magmatic systems are open at the segment scale.