DI44A-03
Melt Zone Fringes and the Geochemistry of Earth’s Mantle

Thursday, 17 December 2015: 16:30
303 (Moscone South)
Donald J DePaolo, University of California Berkeley, Berkeley, CA, United States
Abstract:
Small amounts of water in Earth’s mantle, coupled with mantle convection and subduction zone processes, may be responsible for explaining much of the geochemical diversity of Hot Spot basalts and mid-ocean ridge (MOR) basalts. One of the inconsistencies in modeling mantle evolution is that the average trace element composition of the MOR mantle suggests it has been affected by past melt extraction events involving only small melt fractions – less than about 1%. However, petrological studies indicate that most basalt magmas (or their more olivine-rich precursors) are formed by larger degrees of melting, in the range of 5-20%. A possible explanation for this inconsistency is that the mantle we associate with MOR’s has been produced not in the main melting regions where basalt is produced, but in the fringes of those melt zones, where the presence of water causes a large volume of mantle to be only incipiently melted. Mid-ocean ridges may be where the largest volume of incipiently depleted lherzolite (IDL) is produced; the presence of 500-700 ppm H2O results in a large volume of mantle passing through the incipient melt regime. IDL is also produced at intraplate hotpots, where higher temperatures induce deeper melting and the melting zone contains a large proportion of IDL because of radial geometry and higher water contents. Mid-ocean ridge ‘cold spots’ may also produce a large proportion of IDL. In general, IDL produced in melt zone fringes at MOR’s does not become part of the cold upper oceanic lithosphere. Consequently, the IDL may not be deeply subducted and could be preferentially retained in the upper mantle, producing a form of steady-state geochemical layering in the mantle over billions of years. The expected layering would be more MOR-like in the upper mantle and more OIB-like in the deeper mantle that is more strongly affected by the accumulated addition of the heterogeneous and moderately “enriched” oceanic crust and upper lithosphere. Regardless of the geometry of mantle heterogeneity, any model for the origin, location, size, and history of the MOR source must include the creation and fate of low-melt fraction regions that are inexorably linked to magmatism on a planet with a moderate water-content mantle.