Solving deeply buried statics, offshore West Africa

Reflection-based medium and long-wavelength statics corrections

In this example from offshore West Africa, variable lateral velocity zones (between 1550 and 1650m/s) buried below the sea floor create time domain "slumps" or variable transit times through the overburden. The resulting static effects, although relatively modest in magnitude, cause a significant time "sags", mis-stacking and degradation in event continuity at the target events deeper in the section. If these effects are ignored, the result is a poorly imaged volume that introduces greater uncertainty in structural interpretation and non-flat gathers unsuitable for subsequent AVO and inversion workflows.

The depth of burial of the slumps (between 200 and 300 ms two-way traveltime) was too great for refraction statics to resolve effectively; while the long spatial wavelength of the anomalies created static features longer than a single spread such that conventional reflection statics routines also failed to find a solution.

WesternGeco applied a long to medium wavelength statics workflow that utilized reflection data decomposition to solve for the long wavelength component of the deeply buried statics. This was followed by a secondary pass of conventional reflection based residual statics to compute the high spatial-frequency residual static effects. The 3D approach of these methods also improved resolution of geologically feasible static features in areas of more challenging data quality.

The results demonstrate a clear improvement in image quality across the survey area. Figure 1a shows a shallow time slice from the seismic volume within the static generating interval. Figure 1b maps the derived static anomalies at an equivalent depth, and indicates a good correlation with the geological features evident in Figure 1a. Deeper time slices show a corresponding improvement in reflector continuity, and a reduction in apparent 'noise' after application of these static corrections (Figure 2b) compared to without the corrections (Figure 2a), and this improvement is also shown in the comparison stack sections in Figure 3. The same deep buried static model was subsequently used to improve imaging for a number of very shallow targets.