Accurate velocity modeling is essential for depth imaging, pore pressure prediction, and prospect evaluation. We can achieve accurate velocity modeling for lateral heterogeneity and vertical gradients by using geostatistical methodologies, including well data calibration and integration, and reflection tomography. We offer a complete cycle of velocity model building for all purposes.
Velocity modeling cycle: original (upper-right), kriging, (upper-left),
calibrated (lower-left), and salt flooded model (lower-right).
Integrated velocity modeling, or “iDEPTHing”, is a SeisLink technique which removes blunders in velocity picks and determines lateral heterogeneity (structural anisotropy). iDEPTHing integrates all existing data such as seismic velocities, check-shot data, sonic data, well control data, well tops, geologic maps, interpreted horizons, 2D hazard data, and salt models including salt entry and exit points to produce highly geologically accurate velocity models. Accurate velocity modeling results in E&P risk reduction, which translates into higher returns.
Sample Input and Output:
Seismic & Well Control Data (Input) Earth Model (Output)
Velocity interpolation based on variogram models provides adequate smoothing and simulate lateral heterogeneity.
Regional calibration of well data (above figure) provides not only the calibration at the well locations but also the calibration between wells. Calibrated velocities are emerging with prominent regional geologic features.
Cell-based vs. Layer-based Tomography
Pre-stack depth migration will generate common image point gathers (CIG). Residual velocity errors are scanned for tomographic inversion. For given common image point gathers (CIG), ray tracing computes the path lengths of cells along the ray, and residual time errors are used as data for tomographic inversion. Skeletons were auto-picked instead of waiting for interpreted horizons. RMO were auto-picked at every skeleton location.
Skeleton picks Inline dips
In-line and cross-line dips were estimated. Ray tracing and inversion solves for velocity update. RMO Quality control routines are key for the success of tomography.
The demigration and parsimonious migration loop makes iteration efficient by updating not only velocities but also skeletons and RMOs for the follow-up inversion. It reduces the number of iteration which involves time-consuming prestack migration.