Global Distribution of Tectonic Plates Revealed by the Cluster Analysis of Geodetic Data in the Angular Velocity Space

Abstract

The surface of the Earth is divided into a dozen or so rigid plates which are in relative motions. In this paper, we attempted to objectively estimate the distribution of plates using a cluster analysis of geodetically observed velocity data. Taking the spherical nature of the Earth into account, we have formulated a clustering algorithm of velocity data in the angular velocity space (AVS). Each data is represented as a straight line and intersection points of lines correspond to the Euler vectors of tectonic plates in the AVS. We developed a method to estimate intersection points of lines, and then applied a hierarchical clustering algorithm to the estimated points. The proposed method was applied to observed geodetic data obtained from all over the world, and successfully reproduced the distribution of major tectonic plates previously recognized based on geological and geophysical knowledge.