Enhancing mantle conductivity

Karato bases his idea on diffusivity measurements of hydrogen7, assuming that the hydrogen contributes as charged ions, presumably protons, to electrical conduction in olivine. Because hydrogen is mobile in the [100] crystallographic direction of olivine, only a modest amount of hydrogen - H/Si content between about 200-2,000 parts per million - would be sufficient to account for the high-conductivity layer. [...]we need estimates of the hydrogen content of the upper mantle and the relative partitioning of hydrogen among co-existing mantle minerals. [...]we need to evaluate what influence the considerable anisotropy of hydrogen diffusivity in olivine, spanning two orders of magnitude7, may have on the amount of hydrogen required for enhanced conduction.