[Objective]Conventional seismic attribute analysis in the time domain is based on the conversion from prestack depth migration data to time domain data,which will cause the loss of effective high-frequency informa-tion.To make full use of the advantage of the high imaging accuracy of depth domain data,it is necessary to carry out the attribute analysis of depth domain data.Because the wavenumber in the depthdomain is related to the fre-quency and wave velocity,obtaining a high-resolution depth wavenumber spectrum is the key to seismic attribute a-nalysis in the depth domain.[Methods]In this paper,based on the spectral decomposition method of sparse in-version,an overcomplete wavelet dictionary in the depth domain is established,and the orthogonal matching pursuit algorithm is used to improve the computational resolution of the depth wavenumber spectrum.By calculating the at-tributes of the depth wavenumber spectrum of the theoretical model and comparing them with the attributes of the time-frequency spectrum,the variation characteristics of the depth wavenumber spectrum of the hydrocarbon reser-voir are analyzed.Through the application of depth wavenumber spectral attribute analysis of field data,the practi-cability of using the depth wavenumber spectrum to predict oil and gas reservoirs is verified.[Results]The results show that the depth wavenumber spectral decomposition method based on the orthogonal matching pursuit algorithm has high resolution and can be used as a high-precision method for hydrocarbon reservoir prediction in the depth do-main.[Conclusion]The application of field data shows that the low-wavenumber shadow appears below the oil and gas reservoir in the deep wavenumber spectrum,which can be used as a sign to indicate the existence of oil and gas reservoirs in the depth domain.The depth wavenumber spectral decomposition based on orthogonal matching pursuit can effectively identify the low-wavenumber shadow anomaly,which enables to predict the oil and gas reser-voirs by use of the depth domain seismic data.
维普
万方数据