Trace Element and Pb Isotope Constraints on Dynamic Evolution of Earth Reservoirs
Abstract
Advances in interpretation of Pb isotope systematics provide constraints for modelling Earth evolution. Such improved understanding of Pb isotope systematics has coincided with advances in techniques for accurate Pb isotope ratio measurement by MC-ICPMS. Continental growth since 3.75 Ga has occurred at convergent margins via dehydration of subducted slabs and supra-subduction zone melting. Nb is preferentially retained in slabs relative to U and Th, which are lost to escaping fluids. Over time, the depleted upper mantle (DM) lost U and Th relative to Nb. Thus Nb/Th and Nb/U of UM mirror amount of continental crust present. Because Nb, Th and U are similarly incompatible during MORB melting, temporal Nb-Th-U systematics of mantle can be reconstructed from uncontaminated, depleted-mantle derived rocks1. Excellent agreement exists between crustal growth curve based on Nb/Th and those based on Pb isotope systematics2 and geophysics 3. Temporal variation of Nb/U reflects crustal extraction until 2 Ga. It then reflects preferential U recycling into DM, constraining timing of preservation of a pandemic oxygenated atmosphere. Increase in atmospheric O2 explains the second Pb paradox and refines understanding of DM evolution. Key to understanding mantle Pb isotope evolution is the realization that DM has highly dynamic U/Pb and Th/U ratios relative to undegassed lower mantle (LM). Thus, so-called OIB EM-1 reservoir could reflect LM4. Pb data for Phanerozoic and Proterozoic Gp 2 kimberlites from South Africa plot in thorogenic and uranogenic Pb space consistent with a LM source [4]. Mineralogically, chemically and isotopically different Gp 1 kimberlites, which are readily discernable in plots of PM normalized Ta/U and Nb/Th have very radiogenic 206Pb/204Pb and 208Pb/204Pb but relatively unradiogenic 207Pb/204Pb, compositions identical to HIMU OIB's. We have suggested in [4] that the HIMU isotopic composition can be derived from EM-1 during a transient <100 Ma stage of strong U/Pb fractionation. Recent discovery of LM baddeleyite provides the mineralogical rationale for this scenario. Subducted oceanic crust and continental sediment are unlikely candidates for OIB HIMU source, as trace element fractionation during subduction induced dehydration lowers U/Pb ratio of residual slabs. This has important consequences for genesis of lamproites and minettes. In an speculative model by [5], and supported by seismic tomography, TZ was interpreted as a graveyard for slabs containing high pressure mineralogies such as majorite, NAL phases and hollandite. Partial melts derived from such an environment yield alkaline rocks with Pb isotopic compositions plotting to the left of the Geochron. Significantly, this interpretation is now supported by Pb isotopic data for TZ macrocryst suite xenoliths. An unrelated, now extinct HIMU reservoir, is inferred from Pb isotopes in TTG gneisses in some Archean cratons. Evolution of this source is reflected in Pb isotopic data for galena from Isua that require source separation before 4.3 Ga. The only conceivable long-lived source would have been Hadean crust. Rare examples of pre-plate tectonics TTG gneisses with this isotopic memory occur in the NAC where feldspar Pb isotopes define rotated isochrons that intersect the transient HIMU evolution vector at the time of zircon crystallization of the gneiss protoliths. This transient early Archean HIMU reservoir was subsequently destroyed by subduction. 1Collerson&Kamber (1999) Science 283, 1519. 2Kramers&Tolstikhin (1997) Chem. Geol. 139, 75. 3Reymer&Schubert (1984) Tectonics 3, 63. 4Kamber&Collerson (1999) JGR 105, 25479. 5Ringwood (1994) Phys. Earth Planet. Int. 86, 5.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2001
- Bibcode:
- 2001AGUFM.V52B..12C
- Keywords:
-
- 1010 Chemical evolution;
- 1040 Isotopic composition/chemistry;
- 3670 Minor and trace element composition