CEmin: A MATLAB‐Based Software for Computational Phenocryst Extraction and Statistical Petrology. (13th April 2018)

• Record Type: Journal Article
• Authors: Zeng, Ling; Cheng, Lilu; Costa, Fidel; Herrin, Jason
• Is Part Of: Geochemistry, geophysics, geosystems. Volume 19:Number 4(2018)
• Page Start: 1378
• Page End: 1392
• Publication Date: 2018-04-13
• Subjects: plagioclase -- software -- crystal size distribution -- volcano; Geochemistry -- Periodicals; Geophysics -- Periodicals; Earth sciences -- Periodicals; 550.5
• DOI: 10.1002/2017GC007346
• Languages: English
• ISSNs: 1525-2027

Abstract

Abstract: It is nowadays common to collect large‐area backscattered electron images and X‐ray maps of entire standard petrographic thin sections. These images can be calibrated for compositions of some minerals (e.g., plagioclase) with a small number of electron microprobe analyses, and thus provide a wealth of quantitative data for hundreds of crystals. However, to effectively make use of the textures and compositions of large numbers of crystals we need to be able to efficiently outline and segregate the crystals of interest from the rest of the sample. Here we present CEmin, a set of MATLAB routines that are user‐friendly and allow users to semiautomatically separate plagioclase crystals in grayscale images of volcanic rocks for further processing. These data can then be used for textural and chemical zoning studies. Efficiently extracting large amounts of crystal data allows for identification of plagioclase populations that are indicative of magmatic processes (e.g., closed versus open system) and statistical comparison to thermodynamic models. Plain Language Summary: It is possible nowadays to acquire large data sets of crystal and chemical information of magmatic rocks in a quick and efficient manner. Here we present a new user‐friendly software that allow the extraction of the size and compositions of hundreds of crystals from a single petrographic thin section. Such information opens the era of big data in petrological studies and allows to investigate volcanic and plutonic processes using statistical approaches. Key Points: It is nowadays possible to collect a large amount of petrological information from thin sections in a short time We present a user‐friendly software that allows to extract size and composition of large numbers of phenocrysts in a semi‐automatic manner Such large amount of crystal data allows to investigate magmatic processes using statistical approaches