|
| | Abstract | | Seismic waveform inversion is a novel technique of imaging subsurface structures. It reconstructs a model of physical parameters that explains waveforms of observed seismic data by using a nonlinear least-squares inversion. In this study, we investigate an application of the method to reflection seismic data. The problems are (1) singular nature of sensitivity near sources and receivers at the surface, (2) attenuation of sensitivity in deeper part of the section, and (3) contamination of surface wave. We used phase misfit as the object function, and introduced a weight of the gradient increasing with depth, and a near-offset trace mute to reduce the effects of the problems listed above. Using synthetic waveform data numerically generated from a given structure model, we successfully retrieved a clear structure image after iterative reconstructions. We applied the method to wide-angle reflection survey data acquired in the Fujikawa-kako fault zone - ISTL seismic reflection survey in 2012 (2012FIST) to reveal the detailed structure of the Omiya fault. Although the reconstructed velocity structure is consistent with the recent interpretation that the Omiya fault is a reverse fault, it was far from convergence mainly due to the insufficient number of seismic sources used in the survey. |
|
|
|
| |
|