Seismic data processing using matching filter based cost function optimization

1. A method for processing seismic data from a subsurface, the method comprising:

obtaining observed seismic data of the subsurface;

generating a reflectivity model of the subsurface, the reflectivity model yielding modeled seismic data of the subsurface when forward modeled;

using least squares migration over a cost function comprising a function of a matching filter containing a mapping between the observed seismic data and the modeled seismic data by iteratively updating the reflectivity model to yield an updated reflectivity model that when forward modeled yields updated modeled seismic data that match the observed seismic data; and

using geophysical parameters from the updated reflectivity model for structural analysis of the subsurface to locate oil and gas reservoirs.

2. The method of claim 1 , wherein generating the reflectivity model further comprises generating the reflectivity model comprising a stacked reverse time migration image.

3. The method of claim 1 , wherein using least squares migration further comprises:

defining the matching filter to map at least one of the observed seismic data to the modeled seismic data and the modeled seismic data to the observed seismic data; and

defining the cost function over the reflectivity model as a function of the matching filter.

4. The method of claim 3 , wherein the cost function comprises a ratio of a least squares norm of the matching filter multiplied by a temporal lag and the least squares norm of the matching filter.

5. The method of claim 3 , wherein using least squares migration further comprises updating the reflectivity model to drive the matching filter to a unit spike at zero lag.

6. The method of claim 3 , wherein:

defining the matching filter comprises defining a plurality of matching filters; and

using least squares migration further comprises updating the reflectivity model to the plurality of matching filters to yield a unitary matching filter at zero lag.

7. The method of claim 3 , wherein using least squares migration further comprises maximizing the cost function or minimizing the cost function.

8. The method of claim 1 , wherein the matching filter comprises a Wiener filter.

9. The method of claim 1 , further comprising pre-processing at least one of the observed seismic data, the modeled seismic data and the reflectivity model, wherein pre-processing comprises at least one of scaling, filtering, noise attenuation, event selection, smoothing and re-datuming.

10. The method of claim 1 , further comprising:

encoding the observed seismic data and the modeled seismic data by convolution with a set of encoding functions; and

altering the encoding functions for each iterative update of the reflectivity model.

11. A method for processing seismic data from a subsurface, the method comprising:

computing a common image gather from the observed seismic data;

estimating a velocity model of the subsurface;

using wave equation migration velocity analysis based on a cost function comprising a function of a matching filter containing a mapping between a reference trace at a given location and a considered trace in the common image gather by iteratively updating the velocity model to yield an updated velocity model corresponding to a desired flatness in the common image gather as indicated by the cost function; and

using geophysical parameters from the updated velocity model for structural analysis of the subsurface to locate oil and gas reservoirs.

12. The method of claim 11 , wherein the reference trace comprises a stack of common image gather over a given parameter or a given trace selected from the common image gather.

13. The method of claim 11 , wherein using wave equation migration velocity analysis further comprises:

defining the matching filter to map at least one of the reference trace to the considered trace and the considered trace to the reference trace; and

defining the cost function over the velocity model as a function of the matching filter.

14. The method of claim 13 , wherein the cost function comprises a weight function over all receiver locations and all parameters used to define the common image gather and the matching filter normalized to a given exponent and summed across all depths.

15. The method of claim 13 , wherein using wave equation migration velocity analysis further comprises updating the velocity model to drive the matching filter to a unit spike at zero lag.

16. The method of claim 13 , wherein:

defining the matching filter comprises defining a plurality of matching filters; and

using wave equation migration velocity analysis further comprises updating the velocity model to the plurality of matching filters to a unitary matching filter at zero lag.

17. The method of claim 13 , wherein using wave equation migration velocity analysis further comprises maximizing the cost function or minimizing the cost function.

18. The method of claim 11 , wherein the matching filter comprises a Wiener filter.

19. The method of claim 11 , further comprising varying the matching filter according to recording time in the observed seismic data.

20. A method for processing seismic data from a subsurface, the method comprising:

identifying two traces from observed seismic data of the subsurface;

obtaining a cross-correlation for the two traces; and

using the cross-correlation to determine a time shift between the two traces by estimating a center of cross-correlation by identifying the time shift between the two traces that minimizes a weighted integral with respect to frequency of the derivative of the phase squared of the time-shifted cross-correlation, wherein the weighted integral permits calculations to be conducted entirely in a time domain,

wherein the time shift between the two traces is employed in estimating static corrections, comparing a base and a repeat survey, performing a demultiplex application with data and multiple model, multi component warping, performing cable interpolation to provide a measure of cross-line slowness of events between streamers and/or full waveform inversion minimizing difference between data and models,

wherein updated models enable a structural analysis of the subsurface to locate oil and gas reservoirs.