- journal_title：Geophysics
- Contributor：Hongmei Cao ; Jing Xie ; Young Kim ; Hua-wei Zhou
- Publisher：Society of Exploration Geophysicists
- Date：2008-
- Format：text/html
- Language：en
- Identifier：10.1190/1.2953088
- journal_abbrev：Geophysics
- issn：0016-8033
- volume：73
- issue：5
- firstpage：VE217
- section：VELOCITY MODEL BUILDING
Multiscale tomography (MST) can constrain depth-imaging artifacts that appear in the velocity field inverted using conventional single-scale tomography (SST). Depth-imaging artifacts are poorly constrained velocity results that cause defocusing and/or incorrect structures in the migrated stack. The quality of an inverted velocity distribution from a tomographic approach depends heavily on ray coverage. In areas of poor ray coverage with a small ray count or parallel rays, the velocity distribution can be nonunique because of the lack of constraints. In the case of large residual moveouts from prestack migration, which most likely could happen in the presence of a complex geologic setting, serious depth-imaging artifacts can appear in the velocity field inverted using SST. To solve this problem, we tested the ability of MST to constrain artifacts in a case study. MST parameterizes the study area into overlapping submodels of different grid sizes. The inverted velocity-perturbation value for a specific location is the superposition of velocity-perturbation values from those submodels. The submodel with a large grid size provides a stable regional value that helps constrain local artifacts from submodels of smaller grid sizes. Artifacts using SST are suppressed significantly by MST. Migrated stack and common-image gathers indicated improvements in the velocity field from MST compared with the initial velocity field and that from SST.