A critical evaluation of two-pyroxene thermometry in Adirondack granulites

Temperatures have been calculated from the composition of sixty-five coexisting ortho- and clinopyroxene pairs from the Adirondacks, New York, using calibrations proposed by Wood & Banno (1973), Ross & Huebner (1975), Saxena (1976) and Wells (1977), in order to test the accuracy and precision of pyroxene thermometers in metamorphic granulites. Calculated temperatures are highly scattered ($\text{<600–900°}\text{C}$) and show no systematic variation within the Adirondack terrane. Several pyroxene pairs from very localized areas near Colton (N.W. Adirondacks) and Tupper Lake (C. Adirondacks) yield similar scattered temperatures (600–900°C), which disagree with feldspar and oxide thermometry and are inconsistent with the stability of various silicate and/or sulfide assemblages. All pyroxene thermometer models are sensitive to ferrous-ferric ratios with variations of up to 50°C when Fe³⁺ is estimated from pyroxene stoichiometry of pyroxene compositions obtained by microprobe analysis. These data suggest that the present pyroxene thermometers are sufficiently inconsistent, inaccurate and imprecise that they should not be used as quantitative thermometers in metamorphic rocks.