2017 Volume 21 Issue 6 Pages 297-308
The slow slip event (SSE) is an aseismic transient slip on a fault, and is an important seismological event in the subduction zone. In the field of seismology, a method is needed to accurately estimate the spatio-temporal slip distribution and duration of the SSE using crustal deformation data around the ground. However, conventional methods for estimating the spatio-temporal slip distribution and duration have been developed separately, and have employed models that cannot accurately capture the spatio-temporal behavior of the SSE. We propose a method for estimating the spatio-temporal slip distribution and duration of the SSE simultaneously by introducing a switching model that takes three forms corresponding to three periods. In the first and third periods, the fault is fixed, and in the second period, the fault is slipping slowly. The time points at which the model changes as well as the parameters of the switching model are estimated by the maximum likelihood method using the Expectation-Maximization algorithm. The proposed method thus improves the estimation accuracy of the slip distribution and duration of the SSE by applying appropriate models for specific situations. We compare the method to conventional approaches by applying each to synthetic data and real strain data.