Challenges in constraining theP–Tconditions of mafic granulites: An example from the northern Trans‐North China Orogen

Abstract

Some mafic granulites in the Sanggan area of the northern Trans‐North China Orogen (TNCO) have a relatively simple mineralogy with low energy grain shapes that are compatible with an assumption of equilibrium, but the rock‐forming minerals show variations in composition that create challenges for thermobarometry. The mafic granulites, which occur as apparently disrupted dyke‐like bodies in tonalite–trondhjemite–granodiorite gneisses, are divided into two types based on petrography and chemical composition. Type 1 mafic granulites are fine‐ to medium‐grained with an equilibrated texture and an assemblage of plagioclase+clinopyroxene+garnet+magnetite+ilmenite and sometimes minor hornblende±orthopyroxene. Type 2 mafic granulites are coarse‐grained and hornblende bearing with a peak assemblage of garnet+clinopyroxene+plagioclase+hornblende and variably developed coronae and symplectites of plagioclase+hornblende+orthopyroxene partially replacing porphyroblastic garnet±clinopyroxene. SIMS U–Pb dating of metamorphic zircon from two type 1 mafic granulites yields metamorphic ages of c. 1.84 and 1.83 Ga, consistent with published ages of the type 2 mafic granulites. Based on phase equilibrium modelling, we use the common overlap of P–T fields defined by the mineral assemblage limits, and the mole proportion and composition isopleths of different minerals in each sample to quantify the metamorphic conditions. For type 1 granulites, overlap of the mineral proportion and composition fields for each of three samples yields similar P–T conditions of 710–880°C at 0.57–0.79 GPa, 820–850°C at 0.59–0.63 GPa and 800–860°C at 0.59–0.68 GPa. For the type 2 granulites, overlaying the peak assemblage fields for three samples yields common P–T conditions of 870–890°C at 1.1–1.2 GPa. For the retrograde assemblage, overlap of the mineral proportion and composition fields for each sample yields similar P–T conditions of 820–840°C at 0.85–0.88 GPa, 860–880°C at 0.83–0.86 GPa and 880–930°C at 0.89–0.95 GPa. The P–T conditions appear distinct between the two types of mafic granulite, with the mineralogically simple type 1 mafic granulites recording the lowest pressures. However, there are significant uncertainties associated with these results. For the granulites, there are uncertainties related to the determination of modes and composition of the equilibration volume, particularly estimation of O and H2O contents, and in the phase equilibrium modelling there are uncertainties that propagate through the calculation of mole proportions and mineral compositions. The compound uncertainties on pressure and temperature for high‐T granulites are large and the results of our study show that it may be unwise to rely on P–T conditions determined from the simple intersection of calculated mineral composition isopleths alone. Since the samples in this study are from a limited area—a few hundred square metres—we infer that they record a single P–T path involving both decompression and cooling. However, there is no evidence of the high‐P granuli...

A comparison of observed and thermodynamically predicted phase equilibria and mineral compositions in mafic granulites

“…Figure depicts plotting positions of hornblende, garnet, orthopyroxene, and clinopyroxene in a CFM diagram; it shows that the predicted Fe–Mg partitioning between amphibole and clinopyroxene ( X Mg in hb > aug) is the reverse of widely established patterns in natural rocks in which clinopyroxene is more magnesian than coexisting amphibole (cf. Hartel & Pattison, ; Huang et al., ; Kumar & Chacko, ; Loock, Stosch, & Seck, ; Pattison, ; Zhao, Wilde, Cawood, & Liu, ). This discrepancy pertains whether Fe 3+ is included or not.…”

Integrating X‐ray mapping and microtomography of garnet with thermobarometry to define the P–T evolution of the (near) UHP Międzygórze eclogite, Sudetes, SW Poland

“…The P – T calculations were performed using phase equilibrium modelling, conventional thermobarometry, and average P – T . As recently summarized by Huang, Brown, Guo, Piccoli, and Zhang (), there is a large number of uncertainties associated with any P – T estimate. Among the most important are the sum of errors resulting from the analytical techniques used, uncertainties on the mineral mode calculations (2D scans vs. 3D geometry of natural rocks) and the imperfection of the thermodynamic data sets and activity models.…”

“…The best-preserved sample M2 has been chosen for the P-T estimates. The P-T calculations were performed using phase equilibrium modelling, conventional thermobarometry, and average P-T. As recently summarized by Huang, Brown, Guo, Piccoli, and Zhang (2018), there is a large number of uncertainties associated with any P-T estimate.…”

Wet or dry? The difficulty of identifying the presence of water during crustal melting

“…A limitation of the models presented here is that they are based on mineral equilibria using the simplified KFMASHTO system. A critical aspect is the correct selection of a starting composition to address the problem at hand, since an equilibrium assemblage can be difficult to identify, as was recently discussed in detail in Huang, Brown, Guo, Piccoli, and Zhang () or in Guevara and Caddick (). Different layers represent different domains in a rock and averaging compositions across layers can lead to inaccurate results.…”