多変量解析による温泉の分類の試み [in Japanese] A new application of multivariate analysis to classification of hot springs [in Japanese]
Access this Article
Search this Article
To classify 252 hot springs in Akita Prefecture, We used a combination of factor analysis and cluster analysis. The factor analysis showed that six factors governed evaporated residue, temperature, and concentration of thirteen chemical components of the hot springs. Geochemical meanings of the six factors are mostly explainable. As the result of cluster analysis based on correlation of scores of the six factors, in which hot springs diluted from one source can be always group into same cluster, the 252 hot springs were classified into nine groups. Relationship of the six factors in each group suggested the genesis of hot spring. Thus, the combination of factor analysis and cluster analysis is useful for the classification of hot springs.
Chikyukagaku 22(1), 29-37, 1988
The Geochemical Society of Japan