How Depleted is the MORB mantle?

Knowledge of the degree of mantle depletion of highly incompatible elements is critically important for assessing Earth's internal heat production and Urey number. Current views of the degree of MORB source depletion are dominated by Salters and Stracke (2004), and Workman and Hart (2005). The first is based on an assessment of average MORB compositions, whereas the second considers trace element data of oceanic peridotites. Both require an independent determination of one absolute concentration, Lu (Salters & Stracke), or Nd (Workman & Hart). Both use parent-daughter ratios Lu/Hf, Sm/Nd, and Rb/Sr calculated from MORB isotopes combined with continental-crust extraction models, as well as "canonical" trace element ratios, to boot-strap the full range of trace element abundances. We show that the single most important factor in determining the ultimate degree of incompatible element depletion in the MORB source lies in the assumptions about the timing of continent extraction, exemplified by continuous extraction versus simple two-stage models. Continued crust extraction generates additional, recent mantle depletion, without affecting the isotopic composition of the residual mantle significantly. Previous emphasis on chemical compositions of MORB and/or peridotites has tended to obscure this. We will explore the effect of different continent extraction models on the degree of U, Th, and K depletion in the MORB source. Given the uncertainties of the two most popular models, the uncertainties of U and Th in DMM are at least ±50%, and this impacts the constraints on the terrestrial Urey ratio. Salters, F.J.M. and Stracke, A., 2004, Geochem. Geophys. Geosyst. 5, Q05004. Workman, R.K. and Hart, S.R., 2005, EPSL 231, 53-72.