HR: 17:00h
AN: V52F-05    [PDF]
TI: U-Th-Pa Constraints on Melting Beneath Theistareykir, Northern Iceland
AU: * Stracke, A
EM: stracke@mpch-mainz.mpg.de
AF: Max-Planck-Institut fAŸ’'A+ƒ_TAŸƒ_ÿA›ƒ,ªƒ,›AŸ’'A›ƒ,ª�­AŸƒ_sA,A¬r Chemie, Abteilung Geochemie, J. J. Becher Weg 27, Mainz, 55128 Germany
AU: Bourdon, B
EM: bourdon@ipgp.jussieu.fr
AF: Laboratoire de GAŸ’'A+ƒ_TAŸƒ_ÿA›ƒ,ªƒ,›AŸ’'A›ƒ,ª�­AŸƒ_sA,Acochimie et Cosmochimie IPGP-CNRS UMR7579, 4 Place Jussieu, Paris, 75252 France
AU: McKenzie, D
EM: mckenzie@madingley.org
AF: Bullard Laboratories, Department of Earth Sciences, Cambridge University, Madingley Road, Cambridge, CB30EZ United Kingdom
AB: New U-Pa disequilibrium data on a suite of 12 high-MgO basalts (MgO $>$ 8 wt.%) from Theistareykir, northern Iceland, provide further constraints on the timescales and dynamics of melting than has heretofore been possible based on U-Th disequilibrium data alone. 10 out of 12 basalts from Theistareykir have ($^{231}$Pa/$^{235}$U) between 1.83 and 2.19 at a range of ($^{230}$Th/$^{238}$U) from 1.19 to 1.38. Two basalts originating from the same lava flow (Langaviti) have ($^{231}$Pa/$^{235}$U) of 2.41 and 2.71 with ($^{230}$Th/$^{238}$U) of 1.29 and 1.31, respectively. While changes in ($^{230}$Th/$^{238}$U) alone could be explained by changes in melt extraction velocity only, combined ($^{230}$Th/$^{238}$U) - ($^{231}$Pa/$^{235}$U) systematics require changes in both upwelling and melt velocity. The varying ($^{230}$Th/$^{238}$U) at relatively constant ($^{231}$Pa/$^{235}$U) for 10 out of 12 samples require a coupling between melt and upwelling velocity (lower upwelling velocities are associated with lower melt velocities) in order to keep the relatively large range in ($^{230}$Th/$^{238}$U) and create the small range in ($^{231}$Pa/$^{235}$U). There is a general correlation between ($^{231}$Pa/$^{235}$U) and major element concentrations and heavy rare earth concentrations but not light rare earth concentrations and other elemental or isotopic parameters. The samples with the highest ($^{231}$Pa/$^{235}$U) tend to have, for example, the highest MgO, lowest Na$_{2}$O, and Lu concentrations. Thus, the least differentiated samples have the highest ($^{231}$Pa/$^{235}$U). This effect is most obvious considering the two samples with the highest ($^{231}$Pa/$^{235}$U) from the Langaviti flow. This suggests that decay during magma evolution potentially has a second order effect on the ($^{231}$Pa/$^{235}$U) ratios; at least for some of the Theistareykir samples. For the two samples with the highest ($^{231}$Pa/$^{235}$U) from Langaviti (2.41 and 2.71), $^{231}$Pa excess similar to the rest of the samples (($^{231}$Pa/$^{235}$U) of about 2) will result from decay during magma evolution within 20-30,000 years, which is within recent estimates for magma differentiation times. Even for these samples, however, ($^{230}$Th/$^{238}$U) ratios are relatively little influenced because of the approximately two times longer half life of 230Th compared to $^{231}$Pa (within 30,000years, ($^{230}$Th/$^{238}$U) decrease from about 1.32 to 1.24, for example). The potential influence of decay during magma evolution imposes significant uncertainty on absolute estimates of melting parameters such as upwelling velocity, residual porosity and melt extraction velocity. Furthermore, the selected mineral-melt partition coefficients have a strong influence on the inferred melting parameters, independent of the choice of the melting model. Some first-order estimates using a near-fractional (dynamic) melting model, however, suggest that upwelling velocities $<$ 2cm/yr (in good agreement with the total spreading rate of 18mm/yr), melt extraction velocities $>$ 1m/yr and residual porosity $<$ 0.1% are most likely.
DE: 1025 Composition of the mantle
DE: 1040 Isotopic composition/chemistry
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting