|
| | Abstract | | Reliability analysis by the initial model randomization is one of the useful methods to evaluate the reliability/uncertainty of the velocity model estimated by tomographic inversion. This method provides the velocity structure model and the spatial distribution of velocity error range by calculating the average and the standard deviation of all results of the tomography from hundreds initial models. In this paper, we applied this reliability analysis to the refraction traveltime tomography using seismic data acquired on a 42km-long survey line with 1,152 receivers and 712 shots across the western marginal faults of Kitakami lowland, northeast Japan. The reconstructed velocity structure from 500 random initial models shows reasonable spatial variation with high reliability above 3-4km depth, which is comparable with the seismic section. In addition, distribution pattern of the deviation implicitly indicates the structural character such as faults and velocity boundaries. Convergences of the deviation and the velocity show that at least 200-300 initial models are required for the stable reliability analysis. |
|
|
|
| |
|