Generalized Matching Pursuit

Wavefield separation technique for deghosting over-under or dual-measurement marine data

WesternGeco texture hero

Efficient algorithm for wavefield reconstruction and 3D receiver-side deghosting

Generalized matching pursuit (GMP) performs high-fidelity wavefield reconstruction and full 3D receiver-side deghosting of multimeasurement marine streamer data simultaneously in a single, efficient algorithm. It enables us to extend the usable bandwidth of your data while interpolating between acquisition cables.

Deghosts and interpolates your data in one step

GMP relates the pressure (P), vertical (Z), and crossline (Y) measurements from the streamer to the unknown deghosted pressure. The input data is iteratively decomposed into its constituent components or basis functions until the original signal is described. Once the basis functions are calculated, the pressure wavefield can be reconstructed at any point within the streamer spread, resulting in a continuous, reliable representation of the deghosted wavefield in all directions.

Get a high-resolution dataset for a range of scenarios: shallow hazard analysis, time-lapse seismic processing, and reservoir characterization.

Simulates finer streamer spacing

A key benefit to GMP is it can handle all components of the seismic wavefield with a significant degree of propagation across the streamer spread. Using GMP, any spatially aliased data arriving from the crossline direction can be sampled appropriately in both time and space. The result is the simulation of finer streamer spacing than typically possible with traditional marine acquisition methods—providing you with a more detailed image of the subsurface.

Enhances image detail and clarity

Reconstruction and deghosting are performed jointly and efficiently within one algorithm. When used early in the processing workflow, all subsequent steps are enhanced due to the quality of the broader-spectrum deghosted wavelet and the spacing of interpolated data between cables. The high-fidelity wavelet enhances the performance of our multiple attenuation and velocity analysis algorithms. Additionally, the recovered low frequency content in the data improves full-waveform inversion (FWI) capabilities.

The result is a more accurate picture of subsurface geologic features—enabling you to

  • perform highly detailed shallow hazard analysis
  • monitor your reservoir more effectively over time
  • improve planning to extend the life of your reservoir.