Digital Appendix to "Role of chamber replenishment in the formation of the Merensky Reef and its footwall anorthosite"

Supplementary material to "Role of chamber replenishment in the formation of the Merensky Reef and its footwall anorthosite" submitted to Journal of Petrology in 2024. The archive contains: (1) element maps from electron probe microanalysis; (2) raw data files and processed data from electron backscatter diffraction analysis; (3) raw CSDCorrections files for crystal size distribution analysis; (4) input and output thermodynamic modelling files from Magma Chamber Simulator.\nLineage: The EPMA map of chromite crystals in the lower chromitite was produced using a Cameca SX-Five EPMA at the Centre de Microcaractérisation Raimond Castaing (University Paul Sabatier, Toulouse). The accelerating voltage was 15 kV, the beam current 40 nA, the step size 5 microns, and the electron beam diameter was 1 µm. We used a counting time of 0.5 s and only Al2O3, Cr2O3, FeO, MgO and TiO2 were measured. Analyses with a Cr2O3 content of at least 35 wt.% and a total oxide content above 95 wt.% were considered to isolate chromite crystals and to remove the silicate fraction and alteration products (e.g., magnetite). Pixels missing data were interpolated using a moving median filter and the resulting data were smoothed using a mean filter over a rectangle of 7 x 7 pixels.\n\nThe electron backscatter diffraction (EBSD) maps were produced at Cardiff University using a Zeiss Sigma HD Analytical Field Emission Gun Scanning Electron Microscope equipped with two Oxford Instruments 150 mm2 energy dispersive spectrometers and a Nordlys EBSD detector inserted to 191 mm. For EBSD, colloidal silica-polished samples were mounted at 70° to the incident electron beam at a working distance of 20 mm. Multifield large area maps were collected with a step size of 20 µm, using 2 x2 binning with an exposure time of ~ 60 ms with a Hough resolution of 60 and a minimum number of bands of 8. Removing of wild spikes and interpolation of EBSD data using 8, 7, 6 and 5 identical neighbours was performed using the Channel5 software. Data with a mean angular deviation above 1.3 were discarded together with crystals with less than 10 pixels to limit indexing errors. The crystallographic preferred orientation (CPO) of pyroxene is visualized using an orientation density function (ODF) calculated with the mean orientation of each pyroxene crystal and a 10° halfwidth. Due to the prevalence of twinning in plagioclase, their CPO is visualized using an ODF calculated from all available measurements. The fabric strength for each crystallographic axis of interest is quantified by the J-indices.\n\nThe physical properties of chromite crystals were acquired from the EBSD data using MTEX software. Crystal size distribution (CSD) profiles of chromite were determined using the program CSDCorrections v1.6 (Higgins 2000). Profiles were computed using a box length measurement, a 1:1:1 isometric shape, a roundness of 0, and a massive fabric. The profiles were corrected using the volumetric phase abundance determined from the EBSD data. \n\nThermodynamically constrained forward models simulating the reaction between cumulate rocks and replenishing magma(s) were produced using the Magma Chamber Simulator (Bohrson et al. 2014). These models were assimilation-fractional crystallization models that were completed at 2 kilobars using 2-5°C temperature decrements. Prior to modelling, FeO/Fe2O3 was calculated for each footwall composition at 800°C and ΔFMQ, and hydrous equivalents (denoted as h-) with ~ 2 wt.% H2O were also produced. Furthermore, each floor rock was set slightly above solidus temperature so that 8-10 wt.% of interstitial melt was present.\n\nBohrson WA, Spera FJ, Ghiorso MS, Brown GA, Creamer JB, Mayfield A (2014) Thermodynamic model for energy-constrained open-system evolution of crustal magma bodies undergoing simultaneous recharge, assimilation and crystallization: The magma chamber simulator. J Petrol 55:1685–1717\nHiggins MD (2000) Measurement of crystal size distributions. Am Mineral 85:1105–1116

2024年投稿至"Journal of Petrology"（《岩石学杂志》）的论文《岩浆房补给对梅林斯基礁及其下盘斜长岩形成的作用》的补充材料。本存档包含以下内容：(1) 电子探针显微分析（electron probe microanalysis, EPMA）的元素面分布图；(2) 电子背散射衍射（electron backscatter diffraction, EBSD）分析的原始数据文件与处理后数据；(3) 用于晶体尺寸分布分析的CSDCorrections原始文件；(4) 基于岩浆房模拟器（Magma Chamber Simulator）的热力学模拟输入与输出文件。

数据溯源：下部铬铁矿层中铬铁矿晶体的EPMA面分布图，于图卢兹保罗·萨巴捷大学雷蒙·卡斯坦微表征中心（Centre de Microcaractérisation Raimond Castaing）使用Cameca SX-Five型电子探针显微分析仪完成测试。测试参数为：加速电压15 kV，束流40 nA，步长5微米，电子束直径1 µm；计数时长0.5 s，仅测量Al₂O₃、Cr₂O₃、FeO、MgO及TiO₂。将Cr₂O₃含量不低于35 wt.%且总氧化物含量高于95 wt.%的分析结果用于分离铬铁矿晶体，剔除硅酸盐组分与蚀变产物（如磁铁矿）。缺失数据的像素采用移动中值滤波进行插值，随后使用7×7像素矩形窗口的均值滤波器对结果数据进行平滑处理。

电子背散射衍射（EBSD）面分布图由卡迪夫大学完成测试，所用设备为配备两台牛津仪器150 mm²能谱仪及Nordlys EBSD探测器的Zeiss Sigma HD场发射扫描电子显微镜，探测器安装深度为191 mm。EBSD测试前，样品采用胶体二氧化硅抛光，并以与入射电子束成70°的角度安装，工作距离为20 mm。采集多视场大面积面分布图时，步长设为20 µm，采用2×2像素合并模式，曝光时长约60 ms，霍夫分辨率为60，最小带数为8。使用Channel5软件去除野值尖峰，并采用8、7、6、5个相邻晶粒对EBSD数据进行插值处理。剔除平均角偏差高于1.3的数据以及像素数少于10的晶体，以限制索引误差。辉石的晶体优选取向（crystallographic preferred orientation, CPO）通过取向密度函数（orientation density function, ODF）可视化，该函数基于每个辉石晶体的平均取向计算得到，半高宽为10°。由于斜长石中普遍存在孪晶，其CPO通过所有可用测量数据计算得到的ODF进行可视化。目标各晶轴的组构强度通过J指数（J-indices）量化。

铬铁矿晶体的物理性质基于EBSD数据，使用MTEX软件获取。铬铁矿的晶体尺寸分布（crystal size distribution, CSD）曲线采用CSDCorrections v1.6程序（Higgins 2000）测定。曲线计算采用盒长测量法、1:1:1等轴形状、圆度为0以及块状组构，随后通过EBSD数据得到的体积相丰度对曲线进行校正。

采用岩浆房模拟器（Magma Chamber Simulator，Bohrson et al. 2014）构建受热力学约束的正演模型，模拟堆晶岩与补给岩浆之间的反应。此类模型为同化-分离结晶模型，在2千巴压力下以2~5℃的温度降完成模拟。模拟前，针对800℃、ΔFMQ条件下的各下盘岩石组分计算FeO/Fe₂O₃比值，并制备约含2 wt.% H₂O的含水等价物（记为h-）。此外，将每个底板岩石的温度设置略高于固相线温度，以保留8~10 wt.%的间隙熔体。

参考文献：
Bohrson WA, Spera FJ, Ghiorso MS, Brown GA, Creamer JB, Mayfield A (2014) 模拟同时经历补给、同化与结晶的地壳岩浆体能量约束开放系统演化的热力学模型：岩浆房模拟器. 《岩石学杂志》55:1685–1717
Higgins MD (2000) 晶体尺寸分布的测量. 《美国矿物学家》85:1105–1116