HR: 11:20h
AN: V51K-05    [PDF]
TI: Why is the Ocean Heavy?
AU: * Aubaud, C
EM: aubau001@umn.edu
AF: Dept. Geology and Geophysics, 310 Pillsbury Drive, Minneapolis, MN 55455 United States
AU: Hirschmann, M M
EM: Marc.M.Hirschmann-1@umn.edu
AF: Dept. Geology and Geophysics, 310 Pillsbury Drive, Minneapolis, MN 55455 United States
AB: The hydrogen isotopic ratio of the oceans ($\delta$D=0$\permil$) is heavier than that of the bulk Earth ($\delta$D=-40$\pm$20$\permil$) and the upper mantle ($\delta$D=-75$\pm$5$\permil$). To investigate the process responsible for making this isotopic difference between the oceans and the Earth s interior, we have calculated forward models of water exchange between the mantle and the surface through Earth history. We use two functions for oceanic crust production through time: (1) proportional to radioactive heat production (low mantle overturn rates) and (2) proportional to the square of the heat production (high overturn rates). The volume of the mantle exchanging with the surface is varied from the upper mantle (above 660 km) to the whole mantle. The calculations start with all the water initially present in the mantle. Early rapid degassing is required to satisfy constraints given by xenon isotopes. Fluxes of mantle degassing at mid-ocean ridges and regassing at subduction zones must approach steady state for at least the Phanerozoic, as constrained by near-constancy of continental freeboard during this time. In our calculations, degassing of one ocean mass is achieved rapidly (in $\sim$1 Ga), except in the case of whole mantle convection at low overturn rates, in which case the xenon and freeboard constraints are not met. Calculated $\delta$D of the ocean also reaches steady state after the first Ga. The constancy of the D/H ratio of the ocean through time is consistent with $\delta$D profiles of modern to Jurassic oceanic crust and ophiolites of variable age. Results are insensitive to assumptions about initial distributions of water; i.e., they also hold if all water is initially at the surface or if there is large initial isotopic disequilibrium between the surface and the mantle. Model results show that the ocean/upper mantle difference in $\delta$D is set by fractionation at subduction zones, reflecting vigorous exchange between the two reservoirs. They also require that dehydration of subducted altered oceanic crust ($\delta$D=-40$\permil$) is accompanied by a decrease of 35$\pm$5$\permil$ in $\delta$D. This places a strong constraint on the dehydration process and the residual hydrous phases carrying water into the mantle. Our calculations strongly suggest that both the upper mantle and oceans have been profoundly affected by recycling of water and that water in the upper mantle is dominated by a recycled component. They also may imply that oceanic basalts with high $\delta$D relative to typical MORB have a larger component of primordial H$_{2}$O.
DE: 1030 Geochemical cycles (0330)
DE: 1040 Isotopic composition/chemistry
DE: 3035 Midocean ridge processes
DE: 3337 Numerical modeling and data assimilation
DE: 4825 Geochemistry
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting