Non-stationary Mixed-phase Wavelet Estimation by Higher-order Statistics in One of Hydrocarbon Fields of Iran

Summary Seismic inversion is the process of removing the effects of propagated wavelet to recover the values of acoustic impedance. To estimate a reliable mixed-phase wavelet for a viscoelastic medium, we utilized higher-order statistics method. Non-minimum phase wavelet estimation based on cumulant matching is known as statistical wavelet estimation approach. To apply this approach, we windowed the seismic data in such a way that two consecutive windows were only one sample apart. No fixed wavelet was considered for any windows and we let the phase of each wavelet to rotate at each sample in the window. Comparison of the fourth-order cumulant of the whitened trace with fourth-order moment of the all-pass operator in each window yields a cost function which should be minimized. In this regard, we chose simulated annealing method. The proposed method was applied to synthetic and real data set from one of hydrocarbon fields of Iran. The results showed that the proposed method can estimate non-stationary mixed-phase wavelet in area with variable Q values and low signal-to-noise ratio. The highest correlation is about 89%.