Prestack acoustic full-waveform inversion (FWI) can provide long-wavelength components of the P-wave velocity by using low frequencies and long-offset direct/diving/refracted waves, which could be simulated via a large space grid, and it is weakly sensitive to density. Poststack impedance inversion can usually quickly yield high-resolution impedance, and it is sensitive to density. Therefore, we have combined these two methods to develop an FWI-driven impedance inversion. Our method first uses FWI to obtain the long-wavelength velocity with a guaranteed overlap between the high frequencies of the velocity and the low frequencies of the poststack data. Then, the fitting rock-physics relationship between the density and the velocity is adopted to translate the FWI velocity into the low-frequency impedance. Finally, the resulting low-frequency impedance is used to construct an a priori constraint for poststack impedance inversion. The method has the ability to solve the overlap between the FWI-based converted prior impedance model and poststack data, and it can thereby yield a broadband absolute impedance result. We adopt a Marmousi II model example and a real data case to test the performances of the FWI-driven impedance inversion and indicate its advantages compared with the conventional well-driven impedance inversion that uses well logs and interpreted horizons to build the prior impedance model. The synthetic data example demonstrates that well-driven impedance inversion produces a result with a relatively large deviation to the true impedance model at complex structure zones. However, FWI-driven impedance inversion favorably recovers all interesting sediment layers at complex structure zones. The real data example illustrates that well-driven impedance inversion yields a result with a distinct footprint of the prior model created from well logs and horizons. On the other hand, we find that FWI-driven impedance inversion yields a geologically reasonable solution, which not only conforms to the time-space variation trend of the well logs, but it also reveals a basin structural-depositional evolution.

Low-frequency information is important in reducing the nonuniqueness of absolute impedance inversion and for quantitative seismic interpretation. In traditional model-driven impedance inversion methods, the low-frequency impedance background is from an initial model and is almost unchanged during the inversion process. Moreover, the inversion results are limited by the quality of the modeled seismic data and the extracted wavelet. To alleviate these issues, we have investigated a double-scale supervised impedance inversion method based on the gated recurrent encoder-decoder network (GREDN). We first train the decoder network of GREDN called the forward operator, which can map impedance to seismic data. We then implement the well-trained decoder as a constraint to train the encoder network of GREDN called the inverse operator. Besides matching the output of the encoder with broadband pseudowell impedance labels, data generated by inputting the encoder output into the known decoder match the observed narrowband seismic data. The broadband impedance information and the already-trained decoder largely limit the solution space of the encoder. Finally, after training, only the derived optimal encoder is applied to unseen seismic traces to yield broadband impedance volumes. Our approach is fully data driven and does not involve the initial model, seismic wavelet, and model-driven operator. Tests on the Marmousi model illustrate that our double-scale supervised impedance inversion method can effectively recover low-frequency components of the impedance model, and we determine that low frequencies of the predicted impedance originate from well logs. Furthermore, we apply the strategy of combining the double-scale supervised impedance inversion method with a model-driven impedance inversion method to process field seismic data. Tests on a field data set indicate that the predicted impedance results not only reveal a classic tectonic sedimentation history but also match the corresponding results measured at the locations of two wells.

Nonstationary seismograms can be mathematically expressed by nonstationary convolution model (NSCM) which integrates equivalent Q-value (Q_e), source wavelet and acoustic impedance (AI) into a single formula. Performing nonstationary AI inversion based on NSCM requires subsurface Q_e field distribution and source wavelet, which are all challenging works. Constant Q_e field and extracted wavelet obtained from the early part of seismograms are commonly used as alternatives, consequently, inaccurate AI inversion results will be easily yielded. To deal with this problem, we decompose nonstationary seismic inversion into three subproblems. Firstly, Q_e curves are estimated at well locations using a novel scheme. Since Gabor transform can induce approximate factorization of NSCM, the time-varying wavelet amplitude spectrum (TVWAS) can be obtained by dividing time-frequency amplitude spectrum (TFAS) of the seismic signal by the TFAS of reflection coefficient. Stable Q_e curves at well locations then can be estimated by the compensation-based Q analysis method whose input data are TVWAS. Secondly, having estimation of Q_e and AI logging information at well locations, source wavelet inversion objective function with wavelet transform domain sparse constraint is proposed based on time domain NSCM. This inverse problem is solved efficiently using projected fast iterative soft-thresholding algorithm (pFISTA). Finally, AI inversion objective function with low frequency constraint and isotropic total variation (ITV) transform domain sparse constraint is presented based on frequency domain NSCM and split Bregman method for minimizing this objective function is proposed. Benchmark model synthetic tests and field data application indicate the effectiveness of our nonstationary AI inversion algorithm by demonstrating the improved accuracy and resolution of the AI inversion results.

Amplitude variation with incidence angle (AVA) analysis is an essential tool for discriminating lithology in hydrocarbon reservoirs. Compared to traditional AVA inversion using only compressional wave (P-wave) information, joint AVA inversion using PP and PS seismic data provides better estimation of rock properties (e.g., density, P- and shear wave [S-wave] velocities). Currently, the most used AVA inversions depend on the approximations of the Zoeppritz equations (e.g., the Shuey and Aki-Richards approximations), which are not suitable for formations with strong contrast interfaces and seismic data with large incidence angles. Based on the previous derivation of the accurate Jacobian matrix, we have found that the sign of each partial derivative of reflection coefficient with respect to the P-, S-wave velocities and density changes across the interface represents a good indicator for the reflection interfaces. Accordingly, we adopt an adaptive stratified joint PP and PS AVA inversion using the accurate Jacobian matrix that can automatically obtain the layer information and can be further used as a constraint in the inversion of in-layer rock properties (density and P- and S-wave velocities). Due to the use of the exact Zoeppritz equations and accurate Jacobian matrix, this proposed inversion method is more accurate than traditional AVA inversion methods, has higher computational efficiency, and can be applied to seismic wide-angle reflection data or seismic data acquired for formations with strong contrast interfaces. The model study shows that this proposed inversion method works better than the classic Shuey and Aki-Richards approximations at estimating reflection interfaces and in-layer rock properties. It also works well in handling a part of the complex Marmousi 2 model and real seismic data.