Focal mechanisms of small earthquakes beneath the whole Japan Islands
based on first-motion polarities picked using deep learning
Takahiko Uchide

National Institute of Advanced Industrial Science and Technology, National Institute of Advanced Industrial Science and Technology, National Institute of Advanced Industrial Science and Technology, National Institute of Advanced Industrial Science and Technology, National Institute of Advanced Industrial Science and Technology
Corresponding Author:t.uchide@aist.go.jp
Author ProfileAbstract
Knowledge of crustal stress field is essential for understanding
tectonics and earthquake generation. One way to estimate the crustal
stress field is based on focal mechanisms of earthquakes. This study
investigated focal mechanisms of ~ 110 thousand
microearthquakes in Japan Islands shallower than 20 km based on the
first-motion polarities picked by a simple neural network model, which
was trained using two data sets: moderate to large earthquakes all over
Japan and microearthquakes in two regions in Japan. The threshold of the
confidence score from the neural network model was so chosen as to
maximize the overall quality of focal mechanism solutions. The P- and
T-axes of the numerous focal mechanism solutions provide more detailed
distributions of crustal stress field. For example, in Chugoku region,
there exist slight differences in the trend of P-axes azimuths between
the northern and southern areas are observed, corresponding to the
geodetic observations in space.