HR: 10:35h
AN: MR22A-02    [Abstracts]
TI: Post-Perovskite Double-Crossing, Partial Melting, and the Thermal Structure of Earth's D" Layer
AU: * Hernlund, J W
EM: hernlund@ess.ucla.edu
AF: Department of Earth and Space Sciences, UCLA, 595 Charles Young Drive East, Box 951567, Los Angeles, CA 90095-1567 United States
AU: Thomas, C
EM: tine@liv.ac.uk
AF: Department of Earth and Ocean Sciences, University of Liverpool, Jane Herdman Laboratories 4 Brownlow Street, Liverpool, CA L69 3GP United Kingdom
AU: Tackley, P J
EM: ptackley@ess.ucla.edu
AF: Department of Earth and Space Sciences, UCLA, 595 Charles Young Drive East, Box 951567, Los Angeles, CA 90095-1567 United States
AU: Tackley, P J
EM: ptackley@ess.ucla.edu
AF: Institute of Geophysics and Planetary Physics, UCLA, 595 Charles Young Drive East, Box 951567, Los Angeles, CA 90095-1567 United States
AB: The recently discovered post-perovskite (post-Pv) phase transition has been proposed to explain the occurrence of a seismic discontinuity at the top of D". Recent seismic migration techniques have revealed the presence of an even deeper discontinuity that accompanies the discontinuity at the top of D" beneath Eurasia and the Caribbean region. We show that both discontinuities can be explained by the post-Pv phase transition as the result of a double-crossing between the geotherm in the mantle's lower thermal boundary layer and the post-Pv phase boundary, consistent with current estimates for the phase diagram of MgSiO3 and higher estimates of core-mantle boundary (CMB) temperatures. This double-crossing model predicts that perovskite (Pv), rather than post-Pv, is stable in the lowermost D" layer, while post-Pv can only exist as a layer that does not extend to the CMB. Furthermore, the thickness of such a post-Pv layer will be greatest in cool regions of D", and can become entirely absent for high enough mantle temperatures. Thus the model can explain the lack of detection of a D" discontinuity in some regions as the result of a geotherm that is not cool enough to dip into the post-Pv stability field. Additionally, any partial melting of the lowermost D" layer takes place from the Pv stability field, and implies an intersection between the mantle solidus with the post-Pv phase boundary at outer core pressures. The proximity to this triple point provides a viable explanation for seismic and experimental inferences of dense melting that does not solely depend on partitioning of heavy elements into the melt phase. Using numerical models, we show that dense partial melting in a thermally convecting mantle can explain the observed features of ultra-low velocity zones, and combined with porous flow/segregation implies the existence of a silicate compaction boundary layer at the CMB interface that might explain a core-mantle transition zone, or "fuzzy CMB." Thus a simple model of a phase boundary double-crossing and dense partial melting can explain a wide variety of features in D". Finally, applying the model to the thermal boundary layer structure, current quantitative constraints for the post-Pv Clapeyron slope imply a minimum heat flux of 50 to 80 mW/$m^{2}$ where a D" discontinuity exists, and a global CMB heat flow on the order of 10 TW or greater.
UR: http://geodyn.ess.ucla.edu/~hernlund/double-cross.html
DE: 7207 Core and mantle
DE: 3630 Experimental mineralogy and petrology
DE: 3924 High-pressure behavior
DE: 1212 Earth's interior--composition and state (8105)
DE: 1213 Earth's interior--dynamics (8115, 8120)
SC: Mineral and Rock Physics [MR]
MN: 2004 AGU Fall Meeting