Skip to the content

The AWI and FWI waveform inversion cost functions are applied to seismic data for velocity optimisation in a Tullow Oil operated exploration block in the Central Atlantic margin. The key objective is to extend the main benefits of wave equation inversion to below the diving wave zone whilst using the fewest number of iterations. Employing the two stated cost functions, a model is obtained which is well resolved over a kilometre below the diving wave zone as necessary to cover the target area for prospect maturation in the survey area.

High frequency model evolution

Iterations are run on 50m then 25m grids for long to short scale velocity recovery. 40 iterations are run upto 10Hz, then 30 iterations upto 20Hz. When we get to iteration 70 the target layer at 3km depth is being defined. We decide to push the inversion for 15 more iterations on a 12.5m grid with frequencies upto 40Hz. This requires 36 core EC2 instances - the least number of cores to handle the size of the computational domain.

The final model shows tilted fault block definition at target level with interbedded sand and shale layers. The fine scale detail also shows congruence with PSDM stack amplitude anomalies as the earth reflectivity is mapped into the velocity model successfully.

 

50.0m grid - iter 0

50.0m grid - iter 0

12.5m grid - iter 70

12.5m grid - iter 70

12.5m grid - iter 85

12.5m grid - iter 85

12.5m grid - iter 70

12.5m grid - iter 70

12.5m grid - iter 85

12.5m grid - iter 85

12.5m grid - iter 85

12.5m grid - iter 85

12.5m grid - iter 85 overlaid on migration

12.5m grid - iter 85 overlaid on migration

migration

migration

Model predicted traces

The updating of the traces predicted by the model in relation to the field recorded traces show the inversion is being driven by refractions and reflections.

 

iter 0

iter 0

iter 70

iter 70

field traces

field traces

S-Cube Cloud

Run XWI on the Cloud

Discover an unprecedented increase in the resolution of your velocity model.