No evidence for Archean eclogite-facies metamorphism.

Ning et al. (1) report Archean eclogite-facies metamorphic conditions (792 °C to 890 °C, 19.8 kbar to 24.5 kbar) whereby they argue that deep subduction and thus modern-style plate tectonics have operated since at least the Late Archean. However, the conclusion is compromised by flawed geobarometric estimates and petrography that is inconsistent with eclogite-facies metamorphism.

Firstly, their rock samples do not meet the requirements of the two conventional geobarometers (2, 3) the authors use to calculate the peak metamorphic pressures. The garnet clinopyroxene (Grt-Cpx) geobarometer (2), based on the model reaction of Grossular + Pyrope = Diopside + CaTs (Fig. 1), was calibrated at high temperatures (1,200 °C to 1,550 °C; Fig. 1), and was designed for mantle eclogites (2). The tetrahedral aluminum (Aliv) contents of Cpx in Ning et al. (1), given the analytical uncertainties, are indistinguishable from zero (<0.05) and are far outside of the calibration range of the geobarometer, since metamorphic temperatures are <1,200 °C (Fig. 1). To illustrate the consequences, we apply the geobarometer to published metamorphic studies. This geobarometer overestimates pressure by at least 7 kbar to 11 kbar for low Aliv (<0.05) metamorphic Cpx identical to that in Ning et al.’s peak pressure calculation (Fig. 1). The other rare earth element (REE)-in-Grt-Cpx geobarometer (3) requires high garnet Mg# (Mg/(Fe2+ + Mg) > 0.4 in molar ratio)—a threshold none of their samples meet. To illustrate the consequences, this geobarometer yields systematically overestimated pressures for published samples with garnet Mg# < 0.4 (Fig. 1).

Fig. 1.

(A) Model reaction of 2/3 Grs + 1/3 Prp = Di + CaTs for the Grt-Cpx geobarometer (2) (after ref. 5). (B) Comparison between experimental calibration range of the Grt-Cpx geobarometer (2) and Aliv contents of metamorphic Cpx in Ning et al. (1) (after ref. 2). (C) Validity test of the Grt-Cpx geobarometer (2) applied to published metamorphic rocks (6–8). (D) Validity test of the REE-in-Grt-Cpx geobarometer (3) applied to samples with low Mg# (<0.4) garnet (7, 9).

Furthermore, the petrographic features presented in Ning et al. (1) are inconsistent with eclogite-facies metamorphism. Neither garnet exsolution from clinopyroxene nor rutile exsolution from garnet emphasized by the authors is compelling evidence for eclogite-facies metamorphism. Garnet exsolution from clinopyroxene forms during pressure increase, or cooling from high temperatures, or a combination of both (Fig. 1). Rutile exsolution from garnet is a common phenomenon in granulite-facies rocks undergoing cooling (4). Thus, the two exsolution textures together indicate cooling from high temperatures. In addition, phase equilibria simulations for their garnet pyroxenite indicate >60 vol % garnet at >16 kbar (Fig. 2), in sharp contrast to the 30 vol % to 40 vol % garnet with limited decomposition textures in their rocks. Besides, plagioclases in their rocks are Na rich (>7 wt % Na2O), so there should be sodic clinopyroxene at eclogite-facies conditions. However, neither sodic clinopyroxene nor its decomposed product of clinopyroxene + plagioclase symplectite is observed.

Fig. 2.

(A) Phase equilibria diagram in the model system NCKFMASHTO calculated using bulk-rock composition of garnet pyroxenite (18SY19C-2) in Ning et al. (1). The thermodynamic dataset, solution models, Fe2O3/(Fe2O3+FeO) = 0.1, and water content constraint follow those of ref. 10. (B) Isopleths of garnet volume percentage and mineral assemblages in Ning et al. (1).

In summary, due to the misuse of geobarometers, the peak pressures are overestimated by at least 7 kbar to 11 kbar, which moves their peak metamorphic conditions squarely out of the eclogite-facies field. Without robust evidence for eclogite-facies metamorphism, the peak metamorphic pressures would not exceed 12 kbar at T > 890 °C, which reconciles with the observed mineral assemblages (Fig. 2) that are considered “retrograde” in Ning et al. (1). From our respective, the recalculated results are unremarkable relative to what is already known about predominantly high-temperature Archean metamorphic conditions. Thus, the samples of Ning et al. (1) do not support Archean eclogite-facies metamorphism, or the operation of modern-style plate tectonics in the late Archean.