Seismic wavelet phase estimation by semi-automatic seismic-well tying

Seismic wavelet estimation is important for the seismic well tie and seismic inversion. Unfortunately, it is a tedious and time-consuming work. Huge efforts have been spent on seismic wavelet estimation and most of them focus on the amplitude spectrum estimation and ignore the phase determination. In this paper, we develop an algorithm to automatically determine the phase of estimated wavelets. Our workflow begins with statistical wavelet estimation and employing the statistical wavelet to perform automatic seismic-well tying. We then compute a new seismic wavelet by using the well and seismic data together. To obtain the best phase for the wavelet computed using well and seismic data, we rotate the phase of the wavelet by using a user defined increment and perform the automatic seismic-well tying for each phase rotated wavelets. The phase which has the maximum correlation coefficient between synthetic and seismic data is regarded as the best phase for wavelets in each iteration. We next update the time-depth relation according the best seismic-well tying (the maximum correlation coefficient). The wavelet estimation using well and seismic data, phase rotation, and automatic seismic well tying procedures are repeated until the different of wavelets and time-depth relationships in current and previous iteration are smaller than a user-defined threshold.