“…[4] Although geochemical evidence indicates that the transition zone should be extremely dry (<0.00075 wt%) [Dixon et al, 2002;Rupke et al, 2006], more and more experimental evidence [e.g., Kohlstedt et al, 1996;Locke et al, 2002;Hirschmann et al, 2005;Smyth et al, 2006] and seismological observations [e.g., Revenaugh and Sipkin, 1994;van der Meijde et al, 2003;Song et al, 2004;Blum and Shen, 2004] suggest the existence of substantial deep water (>0.1 wt%) in the transition zone. The water solubility of olivine increases a hundred times relative to the ambient conditions at pressure above 10 GPa (∼300 km), and wadsleyite and ringwoodite can incorporate up to ∼2.9 wt% water in the transition zone [Kohlstedt et al, 1996;Mosenfelder et al, 2006].…”