2005 Salt Lake City Annual Meeting (October 16–19, 2005)

Paper No. 9
Presentation Time: 3:30 PM


SOMMER, Holger, O'NEILL, Craig, LEE, Cin-Ty and LENARDIC, Adrian, Keith Wiess School of Earth Science, Rice University, P.O. Box 1892, Houston, TX 77251-1892, somhol@rice.edu

Dehydration of subducting oceanic lithosphere can deliver free water into the mantle wedge and base of the continental lithosphere, profoundly affecting the rheology and mechanical coupling of these regions. It has been suggested that the thermal structure of shallow-dipping slabs would enable efficient water delivery to areas far inland (~1000km) from the arc. One place where this has been suggested is the North American Cordillera, wherein low angle subduction of the Farallon plate during the Laramide is believed to have hydrated much of the North American lithosphere, and hence, may have influenced post-Laramide continental deformation and magmatism. Such a hypothesis, however, has not been fully tested. The purpose of this study is to quantify the amount and spatial distribution of water released from subducting slabs for a variety of slab dips, velocities, and ages, with particular emphasis on low-angle subduction. After initial release of pore waters, much of the water bound in slabs is likely to be associated with hydrothermal serpentinization of the lithosphere prior to its arrival at a trench. We consider the dehydration of a completely serpentinized lherzolite (13% wt H2O). We model the thermal structure of the slab using a Lagrangian integration point finite-element code (Ellipsis), and construct PT paths for points at various depths in the slab. We use these PT relationships to derive pseudosections to calculate the different stability fields of the dominant water-bearing minerals in subducting slabs, using the Perplex toolbox (Connolly, 1990). The prograde metamorphic reactions (including the dewatering reactions) are themselves dependent on the water content of the slab, and we consider a fractional dehydration model (e.g., dynamic loss of water) to derive dewatering curves for different depths in the slab for flat and steeply dipping subduction zones. We find that a flat slab subduction can deliver water up to 800-1000 km inboard of the trench, compared to 200–400 km for steeply dipping subduction zones.