Subgrain-Size Piezometer: A Recalibration and its Application in Natural Samples

 A subgrain-size piezometer is intended to be free from subsidiary effects of recrystallisation, such as phase mixing and pinning, unlike the classical grain-size piezometers, which are best limited to monomineralic samples to avoid these effects. Previously calibrated subgrain-size piezometers have a wide range of uncertainty in stress for a given intercept length. The log-log linear regression fits contribute to the large and impractical error ranges in linear space. The reason behind this behaviour could be the method applied to measure the representative intercept length of the experimental samples. We reanalysed the same calibration datasets used in the existing subgrain-size piezometer and observed that the distributions of intercept lengths are not log-normal. Instead of taking the arithmetic mean of such datasets, we propose that the median may be a better statistic to represent the central tendency of the datasets. Additionally, we have considered subgrains having misorientation angles from 2–10°. Removing 1–2° subgrain boundaries strikes a balance between data loss and noise reduction. Moreover, we propose a method whereby the measurement of subgrain intercepts is free from grain-boundary intercepts, which usually contribute to the largest values in the datasets. Care is taken to minimise the noise in the electron backscatter diffraction datasets whilst preserving the subgrain boundaries by conservatively choosing the halfQuadratic filter parameters. In this updated subgrain-size piezometer, the error ranges in the linear space are reduced from hundreds of megapascals to a few tens of megapascals. We compared the new calibration with the classical grain-size piezometers in two recrystallised monomineralic quartz-bearing natural rock samples. One sample is from a deformed quartzite in a shear zone and the other is from a sheared silicic vein inside a craton. Misorientation axes of subgrain boundaries indicate that basal and prism slip occurred in the respective samples, implying that the deformation temperatures are different. Recrystallisation regimes are confined to certain temperature ranges, and we tested the subgrain-size piezometer in two separate regimes. The range of the differential stress estimated from our recalibrated piezometer is narrowest amongst the available piezometers, for both samples, even when postdeformation grain growth is observed in one of them.