Data for the paper "Whyte et al, Quantifying water diffusivity and metamorphic reaction rates within mountain belts, and their implications for the rheology of cratons", published in Geochemistry, Geophysics, Geosystems.

The petrographic methods we used depended on whether the sample thin section had a cover slip, and whether previous whole-rock and/or mineral compositions were available. Mineral proportions were acquired by two methods. For thin sections without a cover slip, the samples were analysed by Quantitative Evaluation of Materials by Scanning Electron Microscopy (QEMSCAN) using the Quanta650F Scanning Electron Microscope (SEM) at the University of Cambridge, using 10--15 micron pixel sizes. Thin sections with a cover slip were analysed by point-counting scanned images of the thin sections using the software JMicroVision. Mineral compositional data were also acquired from two sources. Where available, the electron microprobe data from St-Onge & Lucas (1995) and St-Onge & Ijewliw (1996) were used. This dataset was supplemented using a Cameca SX-100 electron microprobe at the University of Cambridge, with analyses carried out using a 20 keV acceleration voltage, 20 nA probe current and a 1 micron spot size. Representative mineral compositional data for the samples analysed in the subsequent pseudosection modelling are given in Table~S2. Mineral abbreviations follow Whitney & Evans (2010), and mineral cation proportions were calculated using AX (Holland, 2009; accessed January 2020), which provides an estimate of X_{Fe}^{3+}} where stoichiometric criteria can be applied. Anhydrous mineral compositions were calculated to standard numbers of oxygen per formula unit (pfu), whereas biotite analyses were calculated to a total of 11, amphibole to 23 and epidote to 25 oxygen pfu. Finally, whole-rock data were acquired from two sources, either using the reported X-ray fluorescence data from St-Onge & Lucas (1995), or from combining the mineral proportions with the mineral compositions to generate a thin-section-specific bulk composition. Where both sources were available for a given sample, no significant (all within 1 wt.%) or systematic differences were observed, and the thin-section-specific bulk composition is listed in Table~S3.