Full waveform inversion approach to building an S-wave velocity model using PS data

A seismic data processing apparatus, comprising:

• an interface configured to obtain seismic data acquired using multicomponents sensors able to detect P-waves and S-waves emerging from a subsurface formation after P-waves have been injected therein, the detected P-waves being recorded as PP observed data and the detected S-waves being recorded as PS observed data, respectively;
• and
• a data processing unit connected to the interface and configured
• to separate the PS observed data from the seismic data;
• to generate PS synthetic data using acoustic approximation equations in anisotropic media with a P-wave model, a current S-wave velocity model and a reflectivity model as inputs;
• and
• to update the current S-wave velocity model using a full waveform inversion, FWI, to minimize an amplitude-discrepancy-mitigating, ADM, cost function quantifying difference between the PS observed data and the PS synthetic data while alleviating an amplitude mismatch between the PS observed data and the PS synthetic data due to the use of the acoustic approximation equations,
• wherein the updated S-wave velocity model is usable to assess presence of natural resources in the subsurface formation and/or to assist extraction of the natural resources from the subsurface formation.

A non-transitory computer readable medium storing executable codes which, when executed on a computer make the computer perform a method for seismic exploration comprising:

• obtaining seismic data acquired using multicomponent sensors able to detect P-waves and S-waves emerging from a subsurface formation after P-waves have been injected therein, the detected P-waves being recorded as PP observed data and the detected S-waves being recorded as PS observed data, respectively;
• separating the PS observed data from the seismic data;
• generating PS synthetic data using acoustic approximation equations in anisotropic media with a P-wave model, a current S-wave velocity model and a reflectivity model as inputs;
• updating the current S-wave velocity model using a full waveform inversion, FWI, to minimize an amplitude-discrepancy-mitigating, ADM, cost function quantifying difference between the PS observed data and the PS synthetic data while alleviating an amplitude mismatch between the PS observed data and the PS synthetic data due to the use of the acoustic approximation equations;
• and
• using the updated S-wave velocity model to assess presence of natural resources in the subsurface formation and/or to assist extraction of the natural resources from the subsurface formation.