We develop a method for imaging subsurface faults using teleseismic waveform data. The target of our method is not an entire velocity structure, but location and size of a subsurface fault. The method consists of the following steps:
(1) pre-processing of observed records,
(2) selection of trial model (initial model of structure),
(3) estimation of incident wavefield from the bottom of the studied region,
(4) forward calculation,
(5) calculation of the steepest descent direction of the model parameters,
(6) image analysis and identification of faults.
The idea of the method has been originally proposed by TAKENAKA et al. (1996). In the present paper we examine the feasibility of the approach using the synthetic data and several trial models in the two dimensional SH cases. We also here propose a practical technique to estimate the incident wavefield from the observed data. In the numerical examination, we could recover the image of the subsurface fault for all trial models we used, which indicates our method has potentiality for sensing real subsurface faults and encourages further development of the method.