K-Semblance & K-Tomography

In-Depth Proprietary Technology

Semblance is a measure of the similarity or “likeness” of seismic traces. Semblance-based Tomography has been widely used for decades as a means of velocity modeling for depth imaging.

In-Depth innovates upon conventional tomography using our proprietary K-Semblance and K-Tomo, which uses the dip field to properly account for dipping geology when calculating trace similarity. This makes the resulting tomography dip-independent, converging to a more reliable result with fewer iterations. (Fig. 1)

With conventional tomography, these dipping errors can produce unwanted bullseyes and artificial undulations in the final velocity model (Fig. 2). Our proprietary K-Semblance and K-Tomography fully resolves this issue. (Fig. 3)

Fig. 1. Velocity model residuals calculated using conventional vertical semblance (left) and K-Semblance (right). After each velocity update, semblance is used to estimate the velocity correction necessary for the next iteration of tomography.

In the example shown above, the initial starting model underestimated the velocity by 10%. Vertical semblance accurately quantifies this error in regions with zero dip, but it consistently overcorrects in regions with steep dips. In contrast, the K-Semblance approach yields the correct velocity residual regardless of dip.

Fig. 2.

Artificial bullseyes and velocity undulations (arrow) produced by steep-dipping regions (circled) due to conventional tomography limitations.

These undulations seriously warp and damage target area seismic. Regions near or below steep dips will be most affected.

Fig. 3. Migrated Seismic overlaid with Legacy conventional tomography velocity model (left) and In-Depth’s proprietary K-Tomo (right). Conventional Tomo produces artificial velocity undulations and bullseyes in and below dipping regions, strongly warping the migrated seismic. In contrast, K-Tomo produces a stable velocity model without obvious artifacts. The resulting seismic is more focused, with clear geological structure and sharp faults.