HR: 17:30h
AN: V52F-07    [PDF]
TI: $^{226}$Ra-$^{230}$Th-$^{238}$U and $^{231}$Pa-$^{235}$U Isotopic Disequilibria at the Galapagos Spreading Centre: Inferences About Mechanisms of Plume-Ridge Mass Transfer
AU: Kokfelt, T F
EM: tfk@dlc.ku.dk
AF: GEOMAR, Wischhofstr. 1-3, Kiel, 24148 Germany
AU: Kokfelt, T F
EM: tfk@dlc.ku.dk
AF: Danish Lithosphere Centre, Oster Voldgade 10, L, Copenhagen, 1350 K Denmark
AU: * Hoernle, K A
EM: khoernle@geomar.de
AF: GEOMAR, Wischhofstr. 1-3, Kiel, 24148 Germany
AU: Lundstrom, C C
EM: lundstro@staff.uiuc.edu
AF: Dept. of Geology, University of Illinois at Urbana Champaign, 1301 W Green St., Urbana, IL 61801 United States
AU: Hauff, F
EM: fhauff@geomar.de
AF: GEOMAR, Wischhofstr. 1-3, Kiel, 24148 Germany
AU: Werner, R
EM: rwerner@geomar.de
AF: Thetys Geoconsulting, Wischhofstr. 1-3, Kiel, 24148 Germany
AB: Mass transfer between a plume and an adjacent spreading ridge may occur either as movement of solid mantle or movement of melt. Recent modeling quantifies the limiting conditions at which long-distance melt transport can occur. Braun and Sohn [2003; EPSL 113: 417-430] estimate that Darcian transport to a ridge through sub-lithospheric channels could occur for several hundreds of kilometers, {\it provided} that channel porosity is appreciably high ($\geq$ 3%). Using the Galapagos Spreading Center (GSC) as an example, these authors argued that transport by melt migration could best account for the distribution of the enriched plume-like signature on the GSC directly north of the hotspot, as a solid flow model should produce a skewed (eastward) enrichment pattern, due to the general flow direction of the upper mantle. \\ Our new U-series data from 86$\deg$W to 92.3$\deg$W on the GSC provide an independent test of this melt migration model. Assuming U-series disequilibria are unsupported during lateral transport, the disequilibria will decrease with transport time (and distance). If plume-melts mix with more depleted (zero-aged) ridge-centered MORB melts to generate the GSC basalts, `aged' plume-derived melts delivered to the GSC would have significantly lower U-series disequilibria than the same melts erupted directly above the hotspot. In order to minimize radioactive decay during transport, we assume fast lateral melt transport within 70 kyr from plume to the ridge, corresponding to 100 km distance at 10% channel porosity (Braun and Sohn, 2003). To explain the observed GSC mixing relations at these conditions, the initial ($^{230}$Th)/($^{232}$Th) in the enriched melt component must be $\sim$1.15 and ($^{230}$Th)/($^{238}$U) and ($^{231}$Pa)/($^{235}$U) must be $\sim$50% and $\sim$300%, respectively. The short half-life of $^{226}$Ra requires that all ($^{226}$Ra)/($^{230}$Th) disequilibria observed at the GSC must exclusively originate from the MORB melt component. The inferred initial $^{231}$Pa and $^{230}$Th excesses in the plume-melt component appear to be unrealistically high compared to global OIB database and need to be tested with data from the Galapagos islands directly above the inferred plume stem. If the inferred values prove to be too high, this would imply that the model may only work for shorter transport times, and to get shorter transport times appears to require high channel porosities in excess of 10% that are probably unrealistic.
DE: 4860 Radioactivity and radioisotopes
DE: 8120 Dynamics of lithosphere and mantle--general
DE: 8400 VOLCANOLOGY
DE: 8434 Magma migration
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting