Oxygen isotopes in MORB revisited insights into the nature of the

							   Recycled crust in the upper
 mantle: insights from combining
isotopic data and geodynamics


        Kari M. Cooper, UC Davis

     (collaboration with John M. Eiler, Caltech,
   Kenneth W.W. Sims, WHOI, Charles H. Langmuir,
           Harvard, Louise Kellogg, UCD)
        CIDER 2006: Mixing in the
                mantle

• CIDER group:
  Stan Hart,
  Louise Kellogg,
  Liz Vanacore,
  Amber Harris,
  Julie Prytulak
• AGU poster
  2006



    Oxygen isotopes and radiogenic isotopes in MORB
Sampling mantle heterogeneities




                       Hunt & Kellogg, 2001


Kellogg et al., 2002




                        Brandenburg & van Keken, 2007
        Oxygen isotope composition
                 of MORB
                                               • Correlations between
                                                 laser data for 18O and
                                                 other geochemical
                                                 indices
                                               • Major element:
                                                 ‘buffered’ composition
                                               • Balanced by strong
                                                 isotopic contrasts
                                                 between crust and
                                                 mantle
Harmon&Hoefs,
   1995; Eiler, 2001
                       Cooper et al. in prep
                          (see poster)
                                               • Mixing relations
                                                 relatively simple to
                                                 define
18O in MORB by ocean basin
                   Oxygen isotopes have the
         = 5.52
         = 0.10     same distribution in each
                     ocean basin.
                   • How faithfully does the
         = 5.52
                     normal distribution of data
         = 0.11     record mantle variations?
                   • What are the implications
                     for the physical distribution
                     and sampling of upper
         = 5.49
         = 0.10     mantle heterogeneities?
        Monte Carlo simulations




How much of the variance can be attributed to
analytical error and to small sample size?
Summary and future directions
• Distribution of 18O-enriched
  domains in upper mantle is the
  similar beneath all ocean basins
• Next step: comparison with
  geodynamic simulations -
  implications of observed distribution
• Forum like CIDER is important in
  developing broad thinking about a
  specific problem