著者

Kenta YOSHIDA Tatsu KUWATANI Atsushi YASUMOTO Satoru HARAGUCHI Kenta UEKI Hikaru IWAMORI

出版者

Japan Association of Mineralogical Sciences

雑誌

Journal of Mineralogical and Petrological Sciences (ISSN:13456296)

巻号頁・発行日

pp.171127, (Released:2018-06-07)

被引用文献数

3

---

Conventional clustering algorithms such as k–means and fuzzy c–means (FCM) cluster analysis do not fully utilize the spatial distribution information of geologic samples. In this paper, we propose GEOFCM, a new clustering method for geochemical datasets with location coordinates. A spatial FCM algorithm originally constructed for image segmentation was modified for application to a sparse and unequally–spaced dataset. The proposed algorithm evaluates the membership function of each sample using neighboring samples as a weighting function. To test the proposed algorithm, a synthetic dataset was analyzed by several hyper–parameter settings. Applying this algorithm to a geochemical dataset of granitoids in the Ina–Mikawa district of the Ryoke belt shows that samples collected from the same geological unit are likely to be classified as the same cluster. Moreover, overlapping geochemical trends are classified consistently with spatial distribution, and the result is more robust against noise addition than standard FCM analysis. The proposed method is a powerful tool to use with geological datasets with location coordinates, which are becoming increasingly available, and can help find overviews of complicated multidimensional data structure.

著者

Satoru Haraguchi Kenta Ueki Kenta Yoshida Tatsu Kuwatani Mika Mohamed Shunsuke Horiuchi Hikaru Iwamori

出版者

The Geological Society of Japan

雑誌

地質学雑誌 (ISSN:00167630)

巻号頁・発行日

vol.124, no.12, pp.1049-1054, 2018-12-15 (Released:2019-03-15)

参考文献数

11

被引用文献数

5

---

A geochemical database for the basement rocks of the Japanese islands has been constructed. These data were previously published in Japanese domestic journals and bulletins not readily accessible to the international community. The database includes 5818 samples from 224 articles, and provides major and trace element concentrations, isotopic ratios, geographical coordinates (latitude, longitude, and altitude) of sampling points, and geological and lithological information. These data are provided in a unified and consistent format as an Excel spreadsheet, which can be efficiently utilized for statistical analyses and data-driven approaches in geochemistry, geology, and petrology.

著者

Tatsu KUWATANI Kenji NAGATA Kenta YOSHIDA Masato OKADA Mitsuhiro TORIUMI

出版者

Japan Association of Mineralogical Sciences

雑誌

Journal of Mineralogical and Petrological Sciences (ISSN:13456296)

巻号頁・発行日

pp.170923, (Released:2018-04-19)

被引用文献数

5

---

Geothermometry and geobarometry are used to study the equilibration of mineral inclusions and their zoned host minerals, which provide information on the P–T conditions of inclusions at the time of their entrapment. However, reconstructing detailed P–T paths remains difficult, owing to the sparsity of inclusions suitable for geothermometry and geobarometry. We developed a stochastic inversion method for reconstructing precise P–T paths from chemically zoned structures and inclusions using the Markov random field (MRF) model, a type of Bayesian stochastic method often used in image restoration. As baseline information for P–T path inversion, we introduce the concepts of pressure and temperature continuity during mineral growth into the MRF model. To evaluate the proposed model, it was applied to a P–T inversion problem using the garnet–biotite geothermometer and the garnet–Al2SiO5–plagioclase–quartz geobarometer for mineral compositions from published datasets of host garnets and mineral inclusions in pelitic schist. Our method successfully reconstructed the P–T path, even after removing a large part of the inclusion dataset. In addition, we found that by using a probability distribution of the most probable P–T path, rather than a single solution, an objective discussion of the validity of the thermodynamic analysis is possible.