Online Journal for E&P Geoscientists

van Borselen, Roald ^*1; Barnes, Simon ¹; Salazar, Humberto ²; Vázquez, Alfredo ²; Rubén Ronzón, Israel ²
(1) PGS, Leiden, Netherlands. (2) Subdirección Técnica de Exploración, Pemex Exploración y Producción, Villahermosa, Mexico.

Summary

The processing strategy for the 3D prediction and subsequent elimination of long period surface-related multiple energy (SRME) on a 3D non-orthogonal land seismic data is presented. It is demonstrated that the new development of 4D Fourier regularization is critical in the pre-conditioning prior to the application of 3D SRME.

Pre-conditioning before SRME

In the surface multiple removal process, the data itself is used as the multiple prediction operator. For land data, arrivals that do not contribute to the multiples of the body waves have to be removed first as well as applying appropriate surface consistent scaling and deconvolution. Residual statics are also applied but relative statics are removed after regularization at the multiple prediction stage.

Results

The long period multiples for this case study are difficult to suppress in the common image gather domain unless they are removed in the pre-processing. If the lateral homogeneity assumption of the overburden that includes the Macuspana carbonates is relaxed, then a 1D CMP-orientated SRME approach can be used. However, if the 3D surface-related multiples are predicted, it can be seen that the 1D assumption is not valid. A critical pre-conditioning step prior to the 3D SRME the 4D Fourier regularization of the survey that is robust even when the survey is so sparse.

Conclusions

The successful application of 3D surface-related multiple prediction and subtraction has been demonstrated to a 3D onshore data set. Even though the geological complexity of the subsurface was relatively modest, the 3D multiple prediction flow clearly outperformed a 1D application, because the azimuth of the original input data is honored and subtle 3D geological features were accounted for properly. Multi-dimensional Fourier regularization, as well as the removal of surface waves, and accounting for statics have all been important prerequisites to the successful application of 3D SRME to this onshore 3D data set.