EVALUATION OF MINERAL THERMOMETERS AND BAROMETERS APPLICABLE TO GARNET LHERZOLITE ASSEMBLAGES

Abstract

We have tested the ability of some 12 different barometer and 20 different thermometer formulations to reproduce the experimental P-T equilibration conditions of natural multi-component garnet lherzolite assemblages. For natural rock compositions it is essential to take account of the influence of both Cr and Fe on the garnet-orthopyroxene Al exchange reaction customarily used as a barometer for such assemblages and accordingly our results demonstrate that the formulation of Nickel and Green (1985) is the most satisfactory. No single thermometer formulation was judged to be reliable throughout the P-T range of interest. In our view equilibration temperatures are best assessed by consideration of a combination of the most satisfactory thermometer formulations, based on the two-pyroxene solvus and the Fe2+-Mg2+ exchange reactions between mineral pairs. Our results further indicate that use of the barometer (MacGregor 1974) and thermometer formulations recommended by Finnerty and Boyd (1984 and 1986) will lead to inaccurate assessment of the temperatures and pressures of equilibration for most garnet lherzolite xenolith assemblages and hence to incorrect interpretation of their depth of derivation within the mantle.