Crustal Deformation and Fault Mechanics

Deformation and seismicty, Miyakejima, Japan summer, 2000.

In the summer of 2000,  a sequence of eruptions and accompanying caldera subsidence occurred on Miyake Island, off the Izu Peninsula Japan.  An extremely energetic swarm of earthquakes extended northwest of Miyake Island indicating the injection of a 30 km  long dike.  During the first week of activity, beginning on June 26th, the seismicity propagated linearly away from the island.  The advance of the swarm stopped following a magnitude 6 earthquake of the coast of Kozushima Island.  In the following months seismicity and deformation (recorded by the GEONET continuous GPS network) continued steadily in the region around Miyakejima until August 23rd, 2000, including several magnitude 5 and 6 earthquakes.  The vast majority of the deformation observed by GPS occurred after the seismic swarm stopped extending.  The Miyakejima sequence raises a number of interesting questions: Why was there so little deformation accompanying the first week of the swarm? Why did the dike stop lengthening during a period when the GPS data indicate dramatic increase in extension across the dike?  Can we quantitatively relate the intrusion process to the dramatic earthquake sequence?

Figure 1. Arrows show unfiltered horizontal displacements measured by
continuous GPS stations from June 26th – August 31st, 2000.   Dots
represented relocated earthquake locations during the same time period:
purple = M. 6, green = M 5, and brown = <M 5.  Dashed lines mark
plate boundaries.

We are investigating these issues with Ph.D. student Gwyneth Hughes (GES).   Gwyn has relocated the earthquakes using double difference methods.  She is currently investigating geologically reasonable models to explain the GPS data and location of the earthquake swarm.