Journal of Disaster Research
Online ISSN : 1883-8030
Print ISSN : 1881-2473
ISSN-L : 1881-2473
Special Issue on Integrated Research for Volcano Hazards Mitigation
Principal Component Analysis of the Color of Bulk Ash Samples at Sakurajima Volcano, Southwest Japan
Taketo Shimano Atsushi YasudaMasato Iguchi
Author information
JOURNAL OPEN ACCESS
J-STAGE Data

2025 Volume 20 Issue 3 Pages 308-316

Details
Abstract

We conducted principal component analysis on the colorimetric data of dried bulk ash samples collected daily between 2009 and 2015 at Sakurajima Volcano, southwest Japan. The CIE L*a*b* color system describes the colorimetric values. The analysis was performed for different sets of values: Set 1, average L*a*b* values alone; Set 2, average and ratios of color values; and Set 3, average, ratios, and their standard deviations (Std). The results show that 89% of data in Set 3 consists of five principal components. The first principal component (PC1) reflects Stds rather than averages of color value assigned to the second principal component (PC2). The third to fifth principal components (PC3 to 5) reflect ratios of color values (b/a, L/a). These results show that the major variation in color of the ashfall samples at Sakurajima Volcano is the scattering, or heterogeneity, rather than the average or representing color values. The temporal variation of the PCs synchronizes with some aspects of eruptive phenomena. PC1 (heterogeneity) increases with the frequency of explosions, whereas PC2 (chroma and brightness) decreases with the discharge rate. PC3 and PC4 reflect changes in mixing end components of ash particles, which may indicate the renewal of ascending magma or the shift in vent location or ascent path. As colorimetric measurement is a quick and quantitative method, it can be a candidate technique for monitoring as a proxy of petrological data, although the color dependence on the size distribution requires consideration.

Content from these authors

This article cannot obtain the latest cited-by information.

© 2025 Fuji Technology Press Ltd.

This article is licensed under a Creative Commons [Attribution-NoDerivatives 4.0 International] license (https://creativecommons.org/licenses/by-nd/4.0/).
The journal is fully Open Access under Creative Commons licenses and all articles are free to access at JDR official website.
https://www.fujipress.jp/jdr/dr-about/#https://creativecommons.org/licenses/by-nd
Previous article Next article
feedback
Top