Analysis of groundwater flow line of the Goshiki-numa pond community using elevation-considered principal component analysis.

Abstract

To reveal the dynamics of underground water flows, measuring many factors and chemical components are usually required. However, it is difficult for the human sense to find the right answers using the many chemo-data in different factors. Although principal component analysis is one of the useful means in a multivariate analysis (chemometrics) that can reduce the multivariable to only 2-D or 3-D and can successfully categorize quantitative multidata into several groups based on the similarity of water quality, underground water flows are difficult to elucidate owing to the lack of continuous information. This paper presents the dynamics of groundwater around the Goshiki-numa pond community (Goshiki-numa), National park of Japan, using multi-chemical component analysis and elevation-considered principal component analysis. Despite the difficulty of understanding the dynamics of groundwater flows of the pond community using limited factors thus far, an elevation-considered principal component analysis (e-PCA) presented herein has revealed the underground water flows around Goshiki-numa ponds by inputting 19 factors × 102 water (total 1,938 data) collected from 2011 to 2014 and 2016. This e-PCA proved to be an effective chemometrics technique for revealing underground water flows. It is believed to apply not only to analytical sciences but also to environmental sciences, civil engineering, and other fields dealing with multidata of water quality.