Application of Higher Order Statistics to Seismic Inversion

Seismic inversion is an effective method for study of reservoir distribution, but it can hardly satisfy the demand when available seismic data are in low resolution. Whether the resolution of inversion data is high or low depends on the seismic wavelet to a large extent. This paper proposes the higher order statistics method by which the seismic wavelet taken is more similar to the real seismic wavelet taken by the conventional method. By comparison between the two methods, the seismic data inversed with higher order statistics method are obviously improved in seismic data continuity and resolution, showing that it has a good application potential for improving seismic data resolution.