Converted-Wave Processing and Imaging

Getting the most out of shear waves

Many of the key steps in the seismic data processing workflow require non-standard approaches for mode-converted data. These include common conversion point (CCP) binning, non-hyperbolic converted-wave moveout, layer-stripping corrections for shear-wave splitting Azimuthal Anisotropic Analysis, converted-wave migration, and model building.

Multicomponent data must be carefully analyzed for vector fidelity and effective separation of P-wave and S-waves. In addition, all converted-wave data must be processed with appropriate handling of the different P and S velocities and anisotropy - all of which are critical for optimal imaging.

Anisotropic PS Kirchhoff PSTM & PSDM is the most appropriate imaging workflow for multicomponent data incorporates a prestack migration algorithm that accounts for the different intrinsic anisotropy of P and S waves and can also compensate for the different source and receiver datum in marine OBC or land surveys.

PP/PS Anisotropic Prestack Depth Imaging is a workflow designed to handle the challenges associated with depth imaging PP and PS datasets together while ensuring accurate event ties.

Multicomponent Adaptive Noise Elimination (MANE) is a prestack noise attenuation method specifically designed to enhance the signal-to-noise ratio of multicomponent data.

Multicomponent cross-equalization (MCCE)

For OBC and land multicomponent data, it is important that the horizontal detector components have optimal vector fidelity for subsequent vector-wavefield processing (e.g., wavefield decomposition, rotation, and layer-stripping analysis).
In marine environments, MCCE cross-equalizes the shear-wave signals detected by the two horizontal detector components to make their amplitude and phase spectra consistent for further vector-wavefield processing. One of the components is assumed to be well coupled to the land surface or water bottom and is used as a reference for shear waves.

In addition, 3C orientation and rotation analysis provides a robust compensation for vector infidelity in situations when fixed (non-gimbaled) cables have been deployed.

Finally, the vertical component may contain inappropriate shear-wave energy from cross leakage or sensor rotation, and this can be included in the determination of the cross-equalization vector operators and optionally removed from the vertical component by an adaptive subtraction algorithm.

Mixed-mode processing

More complex mixed modes such as PSS can be processed with ray-based tools such as SeisCal (part of the Well-Driven Seismic toolbox) - a comprehensive display environment allowing the analysis and display of any pure or mixed-mode data in both time and depth. The combination of these different modes often gives the most accurate solutions to velocity model building and calibration of surface seismic to borehole data.

Wavefield Separation

It is often assumed that the pressure waves are detected primarily by the hydrophones and the vertical detector, and that any shear waves are detected by the horizontal components. However, this may not always be valid, and for true amplitude analysis, the data should be decomposed into separate P and S wavefields.

A simple acoustic up-down separation similar to PZ Processing and Imaging link to PZ Processing and Imaging performed above or below the seabed requires only knowledge of the properties of the water layer. However, an elastic decomposition of up and down wavefields for both P and S data requires additional properties of the seabed. WesternGeco uses the Schalkwijk (2004) adaptive wavefield decomposition method to combine pressure, horizontal, and vertical components to obtain the up and down P- and S-wavefields.

Scholte wave inversion

Scholte waves are surface waves generated at a fluid-solid interface, and can be particularly strong in shallow water. They can be used to compute the near-surface S-velocity models and long-period S-wave static corrections for multicomponent-OBC surveys. Our toolkit includes parametric transforms for aliased data and automatic picking and inversion of the principal modes.

Converted-wave binning and moveout portfolio

Processing multicomponent data requires non-conventional processing options for common converted-point (CCP) binning and non-hyperbolic converted-wave moveout. Converted-wave binning is available using asymptotic, or time-variant CCP binning, using Tessmer and Behle (1988), and Thomsen (1999). Converted-wave moveout is available using Taner and Koehler, Thomsen (1999), Cheret, Bale, and Leaney (2000), Yuan, Li, and Ziolkowski (2001). Both binning and moveout for converted waves (including mixed modes) is available using curved-ray VTI ray-trace-based methods.

Request More Information

Using PS Data

Identification of bright spots using PS data.
PrevNextZoom1 of 1