The Ni-in-garnet geothermometer: calibration at natural abundances

Experiments to study the partitioning of Ni between olivine and garnet at natural abundances were performed at 1200 degrees C - 3 GPa, 1300 degrees C - 3, 5 and 7 GPa and 1500 degrees C - 5 GPa using piston-cylinder and multianvil apparatus. The experiments were intended to provide a further test of the Ni-in-garnet geothermometer calibrated previously using enhanced Ni levels. The run products were analysed by electron microprobe using special operating conditions to enhance counting statistics. Ni-Mg and Fe-Mg exchange between olivine and garnet was tightly reversed. Results of the experiments suggest that Ni substitution in garnet behaves according to Henry's Law up to levels of similar to 3000 ppm. The Ni-in-garnet geothermometer derived from reversed experiments in this study is consistent (within error) with that derived from previous experiments at enhanced levels of Ni; neither of these experimental calibrations is consistent with any empirically derived Ni-in-garnet geothermometer. Sources of the discrepancy between the empirical and experimental calibrations of the Ni-in-garnet geothermometer are examined in detail and it is concluded that the main reason for the differences involves errors in the calculation of temperatures in the empirical version. These errors are clearly demonstrated using a recently published data set for garnet peridotite xenoliths in which both Ni-in-garnet and more conventional multi-equilibrium thermobarometry was employed.