OPX lost at ~1GPa, 1250K-1800K?

[matrix]

Hi,
Recently, I'm working on partial melting of primitive mantle via pMELTS. The initial condition is set as: primitive Earth mantle composition (Sun and McDonough, 1989), isentropic, continuous melting with minimum melt fraction 0.2%, and starts at 4GPa, 1723K. As pMLETS yielded, OPX is stable within the following area P=0.9-1.2GPa and T>=1250K (of course, OPX is stable at other T and P condition). However, one paper ("Evolution of young oceanic lithosphere and the meaning of seafloor subsidence rate", Korenaga, 2016) shows that there is no OPX within 0.9-1.2 GPa and T>1250K, while CPX is much higher than pMELTS generally predicted in the same area (in their Fig.2b and 2c). What's more insteresting is, e.g., when P=1.0 GPa, T=1622K (which is normal P-T path within Earth upper mantle), I got identical bulk chemical composition to their results (their Table 2, row 1 GPa). (We have the same initial condition). So what is the problem with missing OPX while having the same bulk chemical composition? Possibly it's related to phase transitions. Somebody help me!


Thanks,
Mingming