HR: 1340h
AN: T33D-0586    [Abstracts]
TI: Geochemistry of Dikes and Lavas Recovered from `Tectonic Windows' into the Upper Ocean Crust
AU: * Klein, E M
EM: ek4@duke.edu
AF: Nicholas School of the Environment and Earth Sciences, Duke University, Durham, NC 27708-0227 United States
AU: Pollock, M A
EM: map16@duke.edu
AF: Nicholas School of the Environment and Earth Sciences, Duke University, Durham, NC 27708-0227 United States
AU: Karson, J A
EM: jkarson@duke.edu
AF: Nicholas School of the Environment and Earth Sciences, Duke University, Durham, NC 27708-0227 United States
AU: 2005 Scientific Party, P
EM: ek4@duke.edu
AF: Nicholas School of the Environment and Earth Sciences, Duke University, Durham, NC 27708-0227 United States
AB: Geochemical variations among lavas sampled on- and off-axis at mid-ocean ridges are commonly used to infer diverse magmatic processes occurring at depth during crustal accretion. Examples include the assumption that along-axis chemical discontinuities represent boundaries between crustal magma bodies, or that chemical variations observed along flow-lines represent the temporal evolution of a steady-state crustal reservoir. Accumulating evidence derived from `tectonic windows', rare vertical exposures of in situ ocean crust, suggests however that surface sampling may offer a biased view of the true diversity of magma compositions erupted in close spatial and temporal proximity. Our recent Alvin and Jason II mapping and sampling survey of the lavas, dikes and gabbros exposed at the Pito Deep (PD), adds to a growing data base on crustal exposures that includes samples previously recovered from the Hess Deep (HD; Stewart et al., 2002, 2003) and Blanco Transform (BT; Pollock et al., in press). Based on the HD and BT results, and our initial analyses of PD samples, the crust generated at these intermediate- (BT), fast- (HD) and super-fast (PD) spreading ridges show a wide diversity of lava and dike compositions (e.g., Mg no.) in close proximity and with depth, and without clear spatial or temporal trends. In addition, the relationships between lava and dike compositions differ between the HD and PD. At PD, initial analyses suggest a similar range in compositions between dikes and lavas. In contrast, at HD, the lavas show clear evidence of crystal accumulation and associated changes in chemical composition, likely leading to preferential eruption; HD dikes, however, showed little evidence of this effect, suggesting the presence within the dike unit of a range of magma (melt+ crystals) compositions that never erupt at the surface. These chemical differences are particularly intriguing in view of the differences in thicknesses of dike and lava units at the two sites: lavas are much thinner ( approx. 400 m) and dikes are much thicker (700-1000 m) at PD compared to HD (approx. 500-800 m for lavas and 500-800 m for dikes). An examination of the origin of the chemical variations observed at these tectonic windows, combined with information from other exposures of ocean crust with depth (drill cores, ophiolites) may lead to a better understanding of the factors that govern the partitioning of magma within the upper crust and eruption.
UR: http://www.nicholas.duke.edu/pitodeep/index.html
DE: 1021 Composition of the oceanic crust
DE: 1032 Mid-oceanic ridge processes (3614, 8416)
DE: 1036 Magma chamber processes (3618)
DE: 1065 Major and trace element geochemistry
DE: 8178 Tectonics and magmatism
SC: Tectonophysics [T]
MN: Fall Meeting 2005