Spectral Shaping Technology and its Application for Dense-Gas Dessert Prediction on Songliao Basin

In Anda-Songzhan area, the dense sandy gravel of Shahezi formation is generally gas-bearing, which is a key field of natural-gas reserve replacement in Songliao Basin. However, the resolution of deep seismic data is low, so it is difficult to meet the accuracy requirements of dessert prediction in unconventional reservoirs. On the basis of amplitude-preserving imaging processing, we count the global wavelet for poststack anti-Q filtered seismic records, get the global filter operator by shaping the global wavelet with the expected output wavelet, and obtain the seismic records of frequency extension by making convolution the operator with the poststack data. After processing, the main frequency of data is increased by 14 Hz, and the average coincidence rate of sandstone “dessert area” selected by combining multiple attributes with relative wave impedance inversion is increased from 67 to 80%. It supports the deployment of 7 exploration wells in this area and provides strong technical support for increasing storage and production in unconventional fields.